frametop.conf.example said HANDS_SWAP_SIDES=0, and desktops.sh copies it on a
fresh install, so every new install forced ft-camd's side camera names and
turned the hand tracker's own side check off. Some SteamVR restarts swap those
names, and then hands land beside their cutouts and recordings are mislabelled.
The example now says auto. scripts/conf-migrate.sh, run by install.sh and
hands/rec/install.sh, replaces the old line only where it's still exactly as
the example wrote it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With HANDS_SWAP_SIDES set to 0 or 1, ft-hands kept the forced naming as the
published truth even after the hands showed it was backwards. The hand recorder
took that as the session's decision, so the export labelled slam_left and
slam_right the wrong way round (dataset PR #4: every take swapped).
ft-hands now publishes the hands' answer once they disagree (state "forced,
disagrees"); tracking keeps the forced names. sides.read_live corrects the same
case from an ft-hands built before this, and the recorder's log says to use auto.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The known-issue notice becomes a fixed notice: the fix (98bd6ce, 662919b,
67a3c02) reaches main with this release. Anyone who hit "2 of 4 mono
cameras" is pointed at Update.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
PR #29 was squash-merged, so main and experimental last shared history at
1ebf369 and 11 files conflicted. Main's copy of each was the PR #29 head
(ec870d2), which experimental already has, so experimental's side is kept
for all of them. The result is experimental plus main's get.sh and
hand-recorder.md, and main can take experimental with a merge commit again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Turning gaze mode on wakes our tracker, so its eyes come back just as the
first calibration opens. DON_DELAY later, "the headset went on" re-armed the
automatic calibration while it was still open, and when it ended, ft-eyes'
status was still a moment old (not calibrated), so a second one opened at
once. Seen live on 2026-10-05: 26 of 27 dots, then 27 more.
The headset going on re-arms it only when no check is open.
first-calibration-test.py now has the headset go on mid-calibration and
checks that only one opens.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Through the ISP (no colour module), the near-black exposures come out
all zeros, identical every time, so their dequeues change no buffer and
ft-camd never finished learning the side cameras' buffers ("can't tell
which buffers are this camera's yet"): the side pair published nothing.
Such an index is now left unmapped and its frames counted as dark.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 89ec1608d9)
Without the Arcturus module, XRService runs the side cameras through the
ISP ("ISP enabled for tracking cameras (main VFE available)"): NV12 on
vfe0 and vfe1, pitch 1152. ft-camd published only grey cameras, so it
dropped them, and the recorder stopped at "ft-camd publishes only 2 of 4
mono cameras" whatever was restarted.
- ft-camd reads a tracking camera's NV12 luma as its grey image.
- ft-hands and check_sides name the cameras by XRService's numbering in
its log (TrackingCameraInit index 0-3), not by capture pipe, which
moves with the module; a camera whose size isn't its calibration's is
left out. sides.json says which device each camera was.
- camcheck reports the camera map, the ISP routing and which cameras
ft-camd is missing; the recorder says so instead of "restart it",
restarts its own ft-camd once when that's the only problem, and keeps
the map in session.json.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 4db5266cf7)
Headsets without the Arcturus color passthrough module stop at the
camera check (ft-camd publishes only 2 of 4 mono cameras). Say so on
the install page, with the fix due by October 6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
uninstall.sh replaces the README's thirteen uninstall commands. Run from a
Frametop desktop, those commands stopped the input relay second, which took
the keyboard and mouse away before the rest could be typed. And deleting
~/frametop first left Launch a program -> Desktop pointing at a missing
script.
The script works in two steps. Step 1 stops Frametop from starting: the
launcher entry, the user services (disabled, not stopped), the SteamVR
driver, the menu entries, and the eye grabber and Bluetooth fixes under
/etc (one sudo). Everything running keeps running until the headset
restarts, and the script offers to restart it. Step 2 runs once none of
Frametop's programs run. It deletes the code, and asks before deleting the
settings, recordings, and the dev container.
It doesn't use the rest of the repo, so the Pages one-liner works even when
~/frametop is gone. --dry-run shows each command without running it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
uninstall.sh replaces the README's thirteen uninstall commands. Run from a
Frametop desktop, those commands stopped the input relay second, which took
the keyboard and mouse away before the rest could be typed. And deleting
~/frametop first left Launch a program -> Desktop pointing at a missing
script.
The script works in two steps. Step 1 stops Frametop from starting: the
launcher entry, the user services (disabled, not stopped), the SteamVR
driver, the menu entries, and the eye grabber and Bluetooth fixes under
/etc (one sudo). Everything running keeps running until the headset
restarts, and the script offers to restart it. Step 2 runs once none of
Frametop's programs run. It deletes the code, and asks before deleting the
settings, recordings, and the dev container.
It doesn't use the rest of the repo, so the Pages one-liner works even when
~/frametop is gone. --dry-run shows each command without running it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A gaze problem couldn't be told from a report: it had no gaze section. A user's report of
2026-10-05 showed only that the gaze service ran and our tracker was picked.
The report now has a Gaze section: whether the gaze service and our tracker's frame grabber
are enabled and running, when SteamVR's correction and our tracker's calibration were made
(or "none"), and the service's status (ft-gazectl status, on one line: the tracker in use,
whether it's awake and why not, checks.problem). Then the last 10 checks and calibration dots
(checks.jsonl: each dot's reply and whether it was taken), the gaze service's last 60 lines
without podman's exec lines, and the frame grabber's last 20. Nothing in them is an eye image.
Tested on this Frame: each section fills in, and the status is one line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The desktop has its own XDG_CONFIG_HOME (~/.config/frametop), and SteamVR's tools and OpenVR
read their path registry from there. So from a terminal in the desktop:
- pointer/driver/install.sh (install.sh step 4, every reinstall or update from Konsole) wrote
a new registry, ~/.config/frametop/openvr/openvrpaths.vrpath, with only our driver and
"runtime": null, and never registered the driver with SteamVR. That stray file then hid
SteamVR from every OpenVR program started in the desktop.
- scripts/update-check.py (and so doctor.sh and report.sh) reported OpenVR "can't connect to
SteamVR as a background app" (VRInitError_Init_PathRegistryNotFound, or
InstallationNotFound with the stray file) and "vrcmd --overlays lists no overlays". A user's
report of 2026-10-05 showed both, with SteamVR and Frametop's services fine.
- ft-layout's vrcmd calls (the gamescope backend) found no overlays.
They now run with XDG_CONFIG_HOME=~/.config: the driver installer (install, uninstall, probe,
aimhere), update-check.py (for all its checks: nothing there reads the desktop's own config),
report.sh's vrpathreg show, and ft-layout's vrcmd. report.sh doesn't set it for the whole
report, since session/fix-panels.py --check reads the desktop's Plasma config through it.
The driver installer also removes the stray registry (only vrpathreg's, with no runtime), and
update-check.py warns about one. And vrpathreg runs `xdg-open vrmonitor://driverinstalled` to
tell SteamVR's desktop monitor, which the Frame doesn't have: in the desktop that showed "could
not read file vrmonitor://driverinstalled" on every install. A stand-in xdg-open on its PATH
takes it, so install.sh no longer filters the step's output.
Tested in a fake HOME with a copy of SteamVR's registry, a stray one, and the desktop's
XDG_CONFIG_HOME: the new installer removes the stray file, registers the driver in the copy,
and never reaches xdg-open; the old one wrote the stray file, left the copy alone, and ran
xdg-open. update-check.py from that environment: OpenVR and vrcmd ok (the old one fails both);
its stray-registry warning fires on a seeded file. ft-layout's vrcmd: 119 overlays (was 0).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A fresh install that chose our tracker (the installer's default since 12ad93d) could never
calibrate it, so gaze mode never worked. ft-eyes maps pupils to a gaze only once it has a
calibration, so before its first one ft-gaze's "own" source is empty. ft-gazed then never
counted a sample, Checks.can_run() stayed false, and the calibration refused to open: "Not
calibrated, and the calibration can't open: the eye tracker isn't sending", or "the headset is
off or the tracker isn't sending" from Calibrate. This Frame never hit it, because its
calibration came from the gaze probe. Reported 2026-10-05 on SteamOS stable.
- With our tracker in use, running, and uncalibrated ("blind"), SteamVR's tracker seeing an
eye is enough to open the full calibration (ft-eyes answering is what makes calibrated()
False rather than None).
- Each dot stands in for the gaze, so a click takes the CHECK_WINDOW up to it. ft-eyes already
checks, in calib-point, that each pupil was seen in enough frames and held still then, and
its reason for a refused dot shows in the panel's note as before.
- A quick or five-dot check is refused until then ("use Calibrate"): ours can't take a click
without a calibration.
- With no eyes seen, the reason given is that, not "the eye tracker isn't sending".
gaze/test/first-calibration-test.py runs the service against a fake ft-gaze, ft-eyes and
pointer helper, in a temp HOME: the calibration opens by itself when gaze mode comes on, each
click sends ft-eyes the 0.6 s before it, a refused dot's reason reaches the panel, and
calib-fit leaves the tracker calibrated with gaze mode still on. It passes here and fails on
the previous code. gaze/test/idle-test.py still passes. Not yet tested in the headset.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
JakeGreen2145's PR plus our fix 6df6f97: the watcher checks its buses
every 5 seconds instead of every second (about 1% of a core), and
design.md says the watcher is the only cleanup when the VR launcher
runs the desktop in steam.service.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each check starts two gdbus processes (about 5.5 ms of CPU each on the
Frame), so checking once a second cost about 1% of a core for as long as
the desktop ran. On SteamOS 0.3.0 native activation fails, so the watcher
always runs. A registry left behind now goes within 5 seconds instead of 1.
design.md also says where the watcher matters: started from the VR
launcher, the desktop runs in steam.service, which doesn't stop with it,
so keep-apps.sh and the unit's stop don't clean up there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
hand-recorder.md, served at deejanuz.github.io/frametop/hand-recorder.html:
the Konsole commands to install, update and uninstall the Hand Recorder,
who can take part, and what it needs. Frametop comes first for now, since
the recorder runs in its desktop.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Input relay: typing with the pointer helper down no longer ends the relay
Typing on a pass-through keyboard tells the helper "typing". With the
helper not running (SteamVR off), that send raised ConnectionRefusedError
and the relay exited, dropping every grab until systemd restarted it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* ft-steam: open Steam's menu through Steam's own UI
steam/ft-steam menu opens the SteamVR dashboard on Steam's menu, or closes
the dashboard if it's up, without pointer mode: it asks Steam's UI over its
debugging port to show its dashboard overlay (ShowVROverlay, what Steam
calls itself) and focus the Steam frame's left menu (MenuStore.OpenMainMenu).
ft-steam check says whether those calls still exist, and update-check.py
runs it, since a Steam client update can rename them.
The CDP client moves from display-settings/steam_settings.py to
steam/steamui.py, so both use it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Shortcuts: Steam menu, commands, and modifier taps
Key combinations (and mouse and controller buttons) get two new actions:
- steam_menu: Open Steam menu / close dashboard (steam/ft-steam menu).
- command:CMD: run CMD with sh -c, as the relay's service, with layout/,
float/ and steam/ on its PATH. Input Settings offers it for key
combinations as Run a command...
Both work without pointer mode.
A modifier on its own is now a key combination too: a tap, pressed and
released with no other key, mouse button, or scroll in between. A bound
tap sends the desktop F24 before the release, so Plasma's launcher stays
shut. The defaults gain a Meta tap for the Steam menu; this replaces
META_DASHBOARD, which only worked in pointer mode.
input/test/keys-test.py runs the relay against fake devices with every
outgoing socket renamed, so it's safe next to the live relay.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Relay: share a key combination's Meta release with frame-voice
A Meta+key combination (Meta+J for gaze_left, say) hides Meta's release
from the desktop, and the relay skipped share_key for it too. frame-voice
saw Meta go down on @frametop_keys and never come up, so it held all
dictated text back, waiting for that release. Keys of grabbed keyboards
are now shared as pressed, before key_binding() decides what the desktop
gets.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: ft-cutouts, the hand cutouts without pinches and grips
hands/ft-cutouts on|off|status starts ft-camd and ft-hands as transient
user units with ft-hands' new --no-gestures: hands are published for
ft-screens' cutouts, but no pinch or grip is detected, so nothing clicks
or drags and a closing hand doesn't raise the tracking rate. It needs a
build and ft-camd's capabilities, not hands/run.sh install. Its units
conflict with ft-handsctl's, so each stops the other, and they stop with
SteamVR.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* ft-cutouts status: only the current run's tracker lines
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Gaze: say why gaze mode can't work yet, and open the calibration whenever it's missing
Turning gaze mode on without a calibration opened Calibrate only on the
off-to-on change, and only if it could open right then. With the headset
off, the eye tracker silent, or the panel not built, or with gaze mode
already on when the gaze service started, nothing opened and nothing said
why: the pointer just stayed a mouse.
- The gaze service now checks every second: gaze mode on, no calibration
for the tracker in use, eyes seen -> the full calibration opens. One
that closes unfinished opens again only after the headset comes off and
on, gaze mode off and on, or Calibrate, so it doesn't loop. A start that
fails retries every 10 s.
- Its status says why gaze mode can't work yet (checks.problem): not
calibrated and opening, open, closed unfinished, or can't open and why.
- Input Settings shows that under the Gaze pointer switch, along with the
gaze service not installed or not running and our tracker missing its
frame grabber.
- ft-gazectl on notes a missing calibration.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Gaze: install our eye tracker, prefer it, and say why a calibration dot wasn't taken
A user on a fresh install got "Calibration failed: only 0 of 21 dots" with
no reason. The installer never installed our tracker, so gaze used
SteamVR's, and the only way SteamVR's tracker rejects a dot is losing an
eye for most of the look. The panel just showed a red ring.
- install.sh: step 9/10 installs our tracker (gaze/tracker/install.sh)
after gaze mode, yes by default; it needs sudo, so --yes runs it only
when sudo won't prompt. If it fails, gaze keeps SteamVR's tracker.
Configs that still say GAZE_TRACKER=steam (the old template) are asked
whether to switch.
- GAZE_TRACKER=auto, the new default: ours when it's installed (the
frame grabber, its unit, and ft-eyes' Python), else SteamVR's.
ft-gazed rechecks every second, so installing it switches over. Input
Settings lists Own tracker first as recommended, and says how to
install it when it's missing (checking the host's /etc through
/run/host from the dev container).
- The calibration panel has a note line, orange over the instructions:
why a dot wasn't taken (an eye lost, a blink, the eyes disagreeing for
SteamVR's tracker, from steady_samples' new drop counts; ft-eyes' reply
for ours), what a click is still waiting for after 1.5 s, and a failed
calibration's most common reason, which the Gaze page shows too.
steady_samples keeps the same samples as before (checked on 2037
windows of recordings); a lost eye is named before a blink, since its
openness reads 0.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* README: link the Frametop Discord
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Wait for a new container to finish setting up before entering it
container-up.sh starts the dev container in a scope of its own, so
distrobox enter finds it running and skips its wait for distrobox-init.
On a fresh install, init was still setting up passwordless sudo when
dev-container.sh ran sudo dnf install, and sudo asked for a password
with no terminal to read it from. container-up.sh now waits for
container_setup_done itself, and the container's sudo calls use -n,
so a password prompt fails at once with a clear message.
Fixes#9
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Prevent small desktop overlay pointer movements from starting a drag
* Clear reported drag state on controller release
* Click stability: only a hand controller's press starts it
The 3D mouse drives SteamVR's laser through the ft_pointer virtual
controller, so its events reach the screens the same way a controller's
do. The filter held every press, which turned the mouse's short drags
(selecting a character or two, nudging a slider) into clicks. Mark
button events from hand controllers and start the filter only on those.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Input relay: retry a new device until udev gives it to the input group
A new /dev/input node is root:root 0600 until udev applies GROUP=input. The
scan probed each new node once and marked it seen even when the open failed, so
a node caught in that gap was never opened. Behind a KVM, a switch brings back a
hub of devices at once: on the Frame, four nodes failed with EACCES in one switch,
the keyboard was never grabbed, and its keys went to gamescope instead of the
desktop screens. A node that isn't readable yet now waits for the next scan.
* Screens: take a screen's overlays from one copy in the catcher
While a button pressed on a screen is held, UpdateCatcher checks every tick
whether the laser is still on one of the screen's overlays. It built that list
from s.All().begin() and s.All().end(), but All() returns a std::array by
value: iterators into two different temporaries, which is undefined behaviour.
A clang build of ft-screens got a garbage length, threw std::length_error, and
aborted on the first click, taking KWin and the desktop with it.
* Gaze: leave SteamVR's gaze action alone during VR games
From curiousjtuber's PR #13: with the gaze service running, SteamVR
restarted its eye tracker every 10 to 13 s of Beat Saber, as if the
headset came off, and each restart took input focus from the game.
The PR stopped every read in a game. Only the action path reaches
SteamVR (UpdateActionState on the gaze set at overlay-global priority,
then GetEyeTrackingDataRelativeToNow); the mmap and our tracker are
read-only files. So only the action is skipped while a scene app runs,
and gaze keeps moving the pointer over the dashboard in a game. The
action source is only used with --source action.
Co-Authored-By: CuriousJ <curious.j.tuber@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: --record-hz, and the hand recorder's design (hands/rec/DESIGN.md)
ft-hands --record-hz N records at most N frame sets a second, for the hand recorder (10).
DESIGN.md lays out the recorder: the headset panel, the session runner and its script,
the files, review and export, consent.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: ft-handpanel, the hand recorder's headset panel
A head-locked SteamVR overlay for the hand recorder (hands/rec/DESIGN.md):
1.2 m ahead, 12 degrees up, 36 degrees wide, drawn with stb_truetype
into three shared DMA-BUFs as ft-gazepanel does. It shows the title,
step, wrapped instruction, note, countdown, hand chips, near/far bar
and a "Paused" cover, driven over @ft_handpanel.
It also places the touch target, a 2 cm dot in its own overlay fixed
in the room where the head was at the first command for that point,
and logs head and controller poses to poses.jsonl at 250 Hz from a
thread of its own. Both threads take one lock around OpenVR calls.
--no-vr prints each picture's state to stdout (and --dump writes the
pictures), for testing without a headset. hands/rec/build.sh builds it
in the dev container.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: the recorder's worn check goes by the panel's backlight, as frame-job does
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: the hand recorder's session runner and guided script
hands/rec/session.py runs a recording session from script.json: it starts
ft-camd and a tracking ft-hands as transient units only if they aren't
running, records each section as one take (ft-hands --record-only at
10 sets/s, a new sets-N.bin after each pause), drives ft-handpanel, and
writes session.json, calibration.json (identifying fields removed),
prompts.jsonl and take.json. Feedback comes from the live hands file and,
in the controller sections, from the panel's device poll. It also runs
from the command line (--dry-run, --speed, --ring, --no-start).
hands/rec/script.json: 11 sections, about 9 minutes without the object
and controller sections.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: the hand recorder's window, review and export
hands/rec/ft_handrec.py + main.qml (Kirigami, dev container; host launcher
hands/rec/ft-handrec): consent (CONSENT.md, asked again when its version
changes; profile.json with a random contributor id), the before-you-start
checklist with the lighting and free-space checks, the session controls
(Space pauses, Esc stops), review with a frame-set viewer that deletes
ranges, takes and sessions, export with progress and cancel (warns while
the headset is worn), and the upload page (UPLOAD.md, the
huggingface-cli command; HF_DATASET is a placeholder). --dry-run runs
sessions without processes, for testing.
hands/rec/takes.py (standard library): indexes sets.bin and sets-N.bin
without reading pixels, reads one set's cameras, keeps deleted ranges in
take.json, and exports: deleted sets left out, zstd -10 -T2 at nice 19,
manifest.json and SHA256SUMS, nothing left behind on cancel.
CONSENT.md and UPLOAD.md are drafts pending a legal review; the window
says contributions aren't open yet. The dev container gains zstd.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: upload from the hand recorder's window, export checks, a rehearsal
hands/rec/validate.py (standard library; Linux and Windows, Python 3.12+)
checks an export before upload and when it's received: SHA256SUMS, an
allow-list of files, the manifest's schema and keys, the consent version,
a uuid4 contributor, no identifying fields in calibration.json or
device.json, every sets.bin.zst decompressed to its end as a stream with
each FHSET01 header checked against the manifest, jsonl lines, the total
size. It decompresses with compression.zstd, zstandard or the zstd
program. validate.py DIR [--json].
hands/rec/hub.py uploads an export with huggingface_hub, as a pull
request to contributions/<contributor>/<session>: validate first, refuse
a repeat of the same export, check the login (whoami) and access
(auth_check), upload_folder(create_pr=True) with the manifest summary as
the description, then record the PR under "uploads" in session.json.
Errors are explained (terms not accepted, not found, 401/403, network).
--dry-run makes no network calls. FT_HANDREC_DATASET overrides
HF_DATASET (DeeJanuz/frametop-hands); while the texts are drafts a real
upload needs FT_HANDREC_ALLOW_UPLOAD=1.
The Upload page shows the login with "Check again" and how to run
hf auth login in a terminal (the token never enters the window), then
Upload with a phase, progress and Cancel (hub.py as a child process), the
PR link, and a warning for an export uploaded before. The manual
command stays as the fallback. ft-handrec --hub-dry-run.
session.py also saves device.json: cv.cad_from_cal and head from
/persist/device_config.json, the labeller's shape, nothing identifying;
export copies it. Session ids with a -N suffix are accepted everywhere.
hands/rec/rehearse.sh runs it all without the headset: ft-ringplay plays
30 s of a capture into a ring, session.py records a short test script
with ft-handpanel --no-vr and a tracker, then export, validate and a
dry-run upload (--repo ID uploads for real). It runs in one frame-job
scope, deletes its data and stops its processes, also on Ctrl+C.
hands/rec/tests/test_validate.py covers good and broken exports and hub.py
without the network. The dev container gains python3-huggingface-hub.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* README: Frametop doesn't work on the SteamOS beta yet
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 072a294941)
* Hands: step mode and pose pictures for the hand recorder
The first real session moved on every 5 s with text only, too fast to follow.
Each step now waits for Next (Space or the window's button), counts down 3-2-1
while recording, then holds. P pauses, R redoes a step, S skips a section;
"Advance by itself" (--auto) keeps the old timed flow. Nothing records while
a step waits: each step is its own recording part.
The panel and the window show a picture of each pose (hands/rec/poses,
generated by make_poses.py from a parametric hand, MIT) and a diagram of where
to hold the hands and how far out. prompts.jsonl gains ready, wait and redo
events; session.json gains mode.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: the headset button as Next, clearer push steps
The headset's right-side click button (KEY_SELECT on gpio-keys, read without
a grab) now works the session: Next while a step waits, pause during a hold,
resume while paused. With no mouse connected the hints lead with it.
The push sections say plainly to push straight out from the headset and pull
back, with a side-view picture of the head, the headset and the arrow, and
the bar's ends read "At your chest" and "Arm out". The bar labels are sent
as one field, so labels with spaces no longer split.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* ft-floatd: a launch for a missing app no longer kills the control socket
Gio.DesktopAppInfo.new() returns NULL for a desktop file that doesn't
exist, and PyGObject raises TypeError ("constructor returned NULL")
rather than returning None, so launch()'s `if info is None` never ran.
The exception escaped the control socket's GLib callback, GLib dropped
the watch, and ft-floatd stopped answering everything: the float key,
dock, Launch as Standalone, and profiles, until the desktop restarted.
Found on the Frame (2026-10-02): a profile saved with RustDesk's
Flatpak open records its window's app id, com.carriez.flutter_hbb,
which has no desktop file (the Flatpak's is com.rustdesk.RustDesk).
`ft-layout use` on that profile asked ft-floatd to launch it, and
ft-floatd went silent. With this, that launch replies "error no app
com.carriez.flutter_hbb" and the profile's other apps open.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* ft-floatd: profiles keep and relaunch Flatpak apps whose window names another app id
An X11 window in a Flatpak can give KWin an app id with no desktop file:
RustDesk's says com.carriez.flutter_hbb (its GTK application id), and the
Flatpak's desktop file is com.rustdesk.RustDesk. A profile recorded that
id, so it couldn't relaunch the app (PR #16 keeps that from killing
ft-floatd's socket). And the window's pid is the sandbox's own, so a
launched window matched neither by process nor by app id, and didn't
float.
desktop_name() finds the desktop file whose StartupWMClass names the
window's class (or app id) when the app id has none. Capture records
that name, a profile claims open windows by it, and a launch's window
matches by it.
Profiles saved before this keep the old id; save them again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: check the tracking cameras before recording, and watch for losing them
After the headset wakes, XRService sometimes fails to load the colour module's
VCINT FPGA image; then only the two side cameras run, without the IR light,
and the tracker finds no hands. hands/camcheck.py reads XRService's log, the
video nodes it holds and ft-camd's ring, and says ok, degraded or unknown.
The recorder won't start while degraded (--ignore-cameras overrides it), offers
a confirmed SteamVR restart, and stops the first hand-size step when the
tracker sees no hand at all. ft-camwatch (unit file only, not enabled) follows
the log, notifies, and with CAMWATCH_AUTO_RESTART=1 restarts SteamVR when the
headset isn't worn and nothing else uses VR.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: shorter recording sessions with pose sweeps
The second real session took 16 minutes, half of it 36 still poses. Labels
come from the auto-labeller, so what matters is variety, not clean holds.
A sweep step shows a strip of pose pictures and lights one every 4 s while
the hands move slowly near and far; each cue is a prompt event with
"cue": true. The core session is now 15 steps, about 5 minutes recorded.
The pose groups, the one-hand sweeps' groups and the cue order are shuffled
per session, seeded from its id and saved in session.json. A quick round
(--quick, or the checklist's choice) is about 2 minutes for extra lighting.
Touch the dot has 6 dots, the push sections two heights.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Input relay test: let the fake devices past the udev permission check
Since the relay leaves a node it can't read yet for the next scan (12f2e84, PR #12), it
checks os.access first, and the test's fake /dev/input paths don't exist, so the relay never
opened them and every key check failed. The fake os now says they're readable.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pause Frametop for VR games
Frametop kept using the headset during games: with gaze mode off, our eye tracker still took
about 60% of a core, remote desktop about 2 cores while on, and KWin kept drawing hidden
screens because ft-screens sent their frame callbacks at 90 Hz. Pausing gives that back, and
resuming brings back only what pausing stopped. It's also a way to keep the gaze service and
our eye tracker off during games, which PR #13 asked for.
Paused (input/game_pause.py, run by the input relay):
- frametop-gaze stops (ft-eyegrab then idles by itself), and hand tracking and remote desktop
stop if they run
- the desktop hides and slows down: ft-screens "pause on" hides every panel and sends KWin a
frame callback once a second; or, with pause_desktop "close", the desktop closes and starts
again on resume
- the relay lets go of the 3D mouse, typing goes to Steam, and mapped buttons and key
combinations do only pause_toggle, steam_menu and commands
Toggled by both thumbsticks clicked together twice (configurable), read passively from
vrserver's web socket (input/vrws.py) so it works in games and takes nothing from them; by the
new pause_toggle action; by input/ft-pause; and, with pause_auto (default on), by a VR game
starting and ending, which the pointer helper now reports ("vrgame 1|0"). Frametop Input
Settings has a Games page for it. update-check.py checks the web socket, and doesn't count a
paused gaze service as failed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: ft-hands works out which side camera is which
ft-camd tells the side cameras apart by XRService's buffer allocation order,
which some XRService starts reverse; both of 2026-10-02's starts did, so the
cutouts missed the hands. HANDS_SWAP_SIDES=auto (the default) has ft-hands
vote from hands seen in both side cameras: the landmark rays meet in front
of both cameras only under the right naming. While undecided it probes the
exchanged naming with the landmark model. It decides in about 2 s of hands
(right on all 7 recordings replayed), swaps the views in place, and publishes
sides.json. 0 and 1 still force it, with a warning when the hands disagree.
Recordings carry each part's naming and the session's decision; review,
export, validate and ft-handreplay put the names right, and takes.py sides
records a decision by hand.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Gaze: idle while the gaze isn't used
The gaze service ran ft-gaze and our own eye tracker all the time: with gaze mode off, ft-eyes
still took about 60% of a core, and ft-eyegrab, ft-gaze and ft-gazed 3 to 4% each. Now ft-gaze
and our tracker run only while gaze mode is on and someone wears the headset, while a check or
the calibration is open or asked for, or under a "wake" lease, which the Gaze page of Frametop
Input Settings renews while it's open. 30 s after the last use they stop, and the frame grabber
idles with our tracker.
- The pointer helper answers "gaze ? headset" with worn|away (SteamVR's activity level for the
headset); an older helper answers it as before, and the service then goes by gaze mode alone.
- A quick check, calibration, or fit check asked for while idle wakes the tracker and opens once
it sends; the automatic calibration waits quietly while it starts.
- Status has "awake" and "idle" (why), and the Gaze page shows it.
- gaze/test/idle-test.py runs the service with a fake helper and ft-gaze, offline.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* update-check: a still controller isn't a broken web socket
vrserver sends a controller's state only when something on it changes, and one lying still
or asleep may not even send its first one. The check subscribed to one controller and failed
after 3 s of silence. It now subscribes to all, and silence after a good handshake is a skip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: the recorder measures the light itself
The checklist page measures the light when it opens, starting ft-camd if
nothing runs it (and stopping it on quit), instead of saying the cameras
aren't running. The round's lighting defaults to what the cameras measure:
daylight or indoor, from the mono cameras' ambient infrared. Lamps give off
little infrared, so dim and normal rooms read alike; picking dim, room or
daylight still overrides it. session.json gets source, measured and
ambient_ir.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: the recorder's host commands run from the home folder
host-spawn starts a host command in the caller's folder. Started from /tmp,
the app's folder in the container is /run/host/tmp, which the host doesn't
have, so starting ft-camd (and every other host command) exited 127.
host_command now runs from home (env -C), and the launcher cds there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Input Settings: the Games page is Game optimization
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: push steps say to follow the hollow circle, not the blue dot
The dot is the current tracker's distance guess, often wrong; the ring is
where the hands should be.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Eye tracker: one thread for OpenCV and numpy
Nothing called cv2.setNumThreads, so OpenCV kept a pool of one worker per core for
pupil windows of 140 to 240 px. Live, its idle workers spun and yielded about 14,000
times a second each, about a quarter of a core, next to SteamVR's compositor. numpy's
OpenBLAS also started 8 threads that never had work.
eyes_pupil.py now sets OpenCV to one thread, and ft-eyes sets OPENBLAS_NUM_THREADS and
OMP_NUM_THREADS to 1 before numpy loads (a value already in the environment wins).
Replaying fit1 into a scratch share (ft-eyes-replay, 14 s measured, capped at one core
with the replay): threads 13 -> 1, involuntary context switches 3,812/s -> 430/s,
system time 6.9% -> 1.6% of a core. Under that cap the frames it kept up with went from
21-36 to 57-69 a second per eye.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pointer: read the overlay list every 20 s, not every second
The helper ran `vrcmd --overlays` once a second while the pointer was awake, and in gaze
mode the pointer never sleeps. Each run is a shell plus vrcmd, a new SteamVR client, about
26 to 30 ms of CPU, so about 3% of a core all the time.
The list is now read every 20 seconds, and at once (at most once a second) when it may have
changed: the pointer waking, the dashboard opening or closing or creating an overlay, the
scene app changing, an "overlays" request, and a left click that hit nothing, which may be
on a panel that came up since. The thread waits on a condition variable instead of waking
every 100 ms, so it sleeps while paused. The main loop looks the keys up again as soon as a
new list is in, rather than at its next 1 s tick. Overlays already on the list still show
and hide within 50 ms, from the IsOverlayVisible poll.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Gaze: ft-eyes and ft-gaze below SteamVR's priority
ft-eyes and ft-gaze run in the dev container through distrobox, so they live in podman's
libpod scope: frametop-gaze.service's limits never reach them, and they ran at nice 0
next to vrcompositor and vrserver, also at nice 0.
- ft-eyes sets itself to nice 10 and SCHED_BATCH at start, before its threads. Batch
turns off wakeup preemption, so a frame ft-eyes wakes up for can wait out a running
compositor's turn; a few ms late costs the gaze little.
- ft-gaze sets nice 5 before its threads start, but stays SCHED_OTHER: each sample goes
on to the pointer, and batch would add the same wait to every one of them.
- Both only ever lower their priority (a higher nice already set wins), and a failure
is logged and ignored. Checked in the dev container: nice 0 -> 10, policy 0 -> 3
(SCHED_BATCH) without any capability.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pointer: sleep on the command socket while the pointer is off
The main loop slept a fixed 8 ms, about 116 wakeups a second, whether the pointer was awake
or not, and every second it looked up every overlay's handle and read a string property from
all 64 device slots to find its own device.
With the pointer off and hand gestures off, the loop now waits in poll() on its command
socket for up to 250 ms, or 20 ms while mapped Frame controller buttons are being read
(SteamVR input has no event to wait for). A mouse command ends the wait at once. The
headset's activity level, the game check, and the "vrgame" and "gazeawake" repeats keep
going at that pace. The 50 ms visibility poll and the 1 s handle lookups run only while the
pointer is awake, and waking forces both. The device index is looked for only while it's
unknown, and again after SteamVR activates or deactivates a device. The HMD pose history is
kept only with hand gestures on, its one user.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Remote desktop: connect FreeRDP only while a VNC viewer is connected
vnc-bridge.sh kept FreeRDP connected to krdpserver from the moment remote desktop
started, so krdp captured and H.264-encoded every KWin redraw in software (openh264)
with nobody watching: krdpserver 55-78% of a core, xfreerdp 16-27%, Xvnc 6-11%, with 0
clients on :5900. krdp 6.7 creates its screencast session per RDP connection and drops
it when the connection closes, so krdpserver itself idles without one and stays up.
The bridge now counts established connections to Xvnc's port with ss, starts FreeRDP
when a viewer appears (the desktop shows about 3 s later; the VNC screen is black until
then) and stops it 45 s after the last one leaves (VNC_IDLE_SEC). Xvnc has no client
hook, so its log output, which it writes for every connection, wakes the bridge early;
otherwise it looks every 5 s while idle (0.1% of a core measured, against 0.9% for ss
once a second) and every second while FreeRDP runs. The layout check runs only while
FreeRDP runs.
While a viewer is connected the bridge sends "watch 15" to ft-screens (@ft_screens) at
once and every 5 s, so screens at a reduced frame rate (out of view, headset on a
stand) stream at full rate; it lapses by itself if the bridge dies, and an ft-screens
without the command just answers an error. The window search after starting FreeRDP
now ends when FreeRDP exits instead of polling for 30 s.
krdp on 127.0.0.1 with a fresh password, VNC on the tailnet address with VncAuth, and
remote-ctl.sh start/stop (pause and resume) are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Remote desktop: read the layout only after it changes
While FreeRDP ran, vnc-bridge.sh called ft-layout remote-view every 5 s, which scans
all of /proc for plasmashell and runs kscreen-doctor -j: about 4.4% of a core for a
layout that rarely changes.
It now stats the two files the answer depends on, the nested KWin's
~/.config/frametop/kwinoutputconfig.json (positions, scales, primary) and
~/.config/frametop-layout.json (screen sizes), once a second while FreeRDP runs. After
either changes it reads the layout every 2 s for 10 s, since KWin's outputs follow the
file a few seconds later; otherwise once a minute, in case a change touched neither.
With no VNC viewer connected nothing runs (previous commit).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Eye tracker: ft-eyes sleeps until the next frame is due
ft-eyes looked for new frames about 1,000 times a second: each pass of its loop asked the
control socket with a non-blocking recvfrom (a BlockingIOError nearly every time), read
both cameras' counters, and slept 1 ms. The frames come every 11.1 ms per camera, and only
as counters in ft-eyegrab's shared memory, so there's no fd to wait on.
Now each pass ends in select() on the control socket, with a timeout until 2 ms before the
next frame of either camera is due (from when its last one was seen), then every 1 ms until
it comes. A command wakes it at once. A camera with no frame for 0.1 s (headset off,
grabber idle) isn't waited for, and with both stopped it looks every 20 ms. Waiting for the
frame grabber's file uses the same select, 0.2 s at a time, instead of sleeping through
commands.
A new frame is still seen within about 1 ms of when it lands. On a synthetic share at 90 Hz
per camera, on a heavily loaded headset (load average 23, so ft-eyes rarely sat idle), its
waits went from 178 to 81 a second; unloaded, the old loop's 1 ms sleeps add up to about
1,000.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Screens: frame rates by attention, ticks in step with the display
ft-screens gave KWin a frame callback for every committed screen on each tick, and the tick
was an 11 ms timer set again after each run, so it slid through the display's frame and came
about 85 times a second at 90 Hz: the desktop repeated a frame several times a second (judder
in scrolling and video), and KWin drew every screen in one burst at a random point of
vrcompositor's frame. Hidden screens got the same 90 Hz unless Frametop was paused for a game.
- Ticks run on a timerfd at absolute times, once per display frame, 1 ms after the vsync
(IVRSystem::GetTimeSinceLastVsync and the HMD's display frequency, read once a second), so
KWin gets its callbacks early in the frame. Measured with --no-vr: 91 wakeups a second
instead of about 85. On the Frame the vsync times SteamVR reports lie on a 90 Hz grid.
- Each screen's callbacks come at a rate for how much of it you see (vr.cpp,
UpdateAttention): every frame while focused (within 12 degrees of where your head points,
a laser or the mouse on it in the last 1.5 s, carried, or typed on), 15 a second for the
rest of what you see (within 60 degrees), and 1 a second when hidden, behind you, or
paused. Levels rise at once and fall after 1.5 s (focused) or 0.5 s (in view). KWin draws
a screen only after its callback and its apps wait for theirs, so this throttles the apps
too. A screen where nothing changes costs nothing at any rate, as before.
- A video in view keeps every frame: 8 commits in a row that each redraw 6% or more of the
screen, at 10 a second or more, count as one (from the surface's buffer damage).
- "rates F V H" / --rates set the three rates (default 0 15 1, 0 = every frame), "rates?"
shows them and each screen's level, "watch S" gives everything full rate for S seconds for
a remote viewer (vnc-bridge.sh renews it), and "phase MS" moves the ticks for tuning.
- ft-screens' main thread runs at nice -5 after the session starts: SteamOS allows down to
-8 once the soft RLIMIT_NICE is raised, and KWin waits on these ticks. It had spent nearly
3 times as long waiting to run as running.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pointer: skip unchanged work while the pointer is awake
Every frame (about 116 a second) the helper tested the cursor ray against every visible
overlay twice with ComputeOverlayIntersection, set the dot's alpha, width, transform and
visibility (five calls into SteamVR), and sent the driver a pose datagram, even with the
mouse and the head still.
Now a frame reuses the last collision result when the mouse, the anchor (1 mm) and the
eye (5 mm) haven't moved and no overlay showed, hid, or changed handle. The passes still run
at least every 100 ms, since overlays move on their own (a floating window's controls follow
it), and always while dragging. The dots' setters go to SteamVR only when their value changes:
the placement when the dot moved 0.2 mm or the eye 5 mm, which turns or resizes it by well
under 1%, and the width on a 0.5% change. The plain pose goes to the driver only when the
laser's origin moved 0.2 mm or its direction 0.04 deg (0.1 mm where it lands, 15 cm on), and
at least every 100 ms; the driver keeps the last pose and reports it every frame. A tilt's
pose, a placement, or waking sends the next one regardless.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Session: blur, background contrast and animations off by default
The nested kwinrc had no [Plugins] group, so KWin ran its default blur and background
contrast effects, and kdeglobals had no AnimationDurationFactor, so animations ran at
full length. KWin renders through zink on Turnip, on the GPU vrcompositor needs, and
blur re-renders what's behind every translucent panel and menu; each animation frame is
another frame for KWin and ft-screens.
Before KWin starts, the session script now writes [Plugins] blurEnabled=false and
contrastEnabled=false to $XDG_CONFIG_HOME/kwinrc and [KDE] AnimationDurationFactor=0 to
its kdeglobals, each only if the desktop's own file has no value for it. It does this
once and records that in $XDG_CONFIG_HOME/frametoprc ([Defaults] effects=1), because
System Settings deletes a key put back to its default: without the marker, turning blur
back on wouldn't survive a restart. The ids blur and contrast are the built-in effects
of KWin 6.2.5 on SteamOS (both enabled by default in its plugin metadata). Tested
against a temporary XDG_CONFIG_HOME: fresh config, an existing blurEnabled=true kept,
and a deleted key not rewritten.
README and docs/reference.md say how to turn them back on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Session: don't autostart Discover's notifier or IBus in the desktop
The nested Plasma session runs the system's XDG autostart entries. Discover's update
notifier (/etc/xdg/autostart/org.kde.discover.notifier.desktop) started
plasma-discover --mode update inside it, 520-620 MB resident and about 9% of a core,
with flatpak-system-helper and AppStream downloads behind it. IBus started a nested
ibus-daemon with kimpanel and ibus-extension-gtk3, which no app in the desktop can use:
KWin's input method is ft-textinput (zwp_input_method_v1, focus reports only; the VR
keyboard types through ft-screens' seat), and the session already drops QT_IM_MODULE,
GTK_IM_MODULE and XMODIFIERS. Nothing in Frametop talks to IBus.
Before Plasma starts, the session script copies both entries into
$XDG_CONFIG_HOME/autostart with Hidden=true, which plasma-session honours for that
desktop only. It does this once ([Defaults] autostart=1 in frametoprc) and skips a name
the user already has a file for, so deleting the copy brings the program back. The
geoclue demo agent stays (it answers apps' location requests outside GNOME and idles at
0%), and orca's entry is OnlyShowIn GNOME-family desktops, so it never ran. Tested
against a temporary XDG_CONFIG_HOME, including an existing user ibus.desktop left alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Input relay: never block on the pointer helper's socket
The relay sent to @ft_pointer_helper on a blocking socket. When the helper stalled, a
layout placement or grabprobe holds it for seconds while ft-gazed keeps filling its socket at
90 Hz, the relay's one loop blocked with it: keyboards, the volume keys (which must never
reach gamescope), and pausing all stopped until the helper read again.
The socket is non-blocking now. A command the helper doesn't take (EAGAIN) waits in a queue,
and everything after it queues behind it so the order holds; tick() sends what it can on each
loop, and the select timeout drops to 20 ms while anything waits. Mouse moves add up into one
queued move. A scroll notch is dropped rather than queued, since scrolling seconds late is no
use; its release still goes. Presses, releases, show, hide, and the rest are kept, so no
button stays down. The queue holds at most 512 commands. While paused, the configured
pointer's queue still drains, so the releases and "hide" from standing down arrive. A
"vrbind" that hits a full socket is sent again on the next loop instead of being lost.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pointer driver: parse outside the lock, report only changes
Handle() held the state lock through a chain of up to a dozen sscanf calls per command, and
RunFrame, which vrserver calls every frame, takes the same lock, so a burst of commands
(about 116 poses a second, plus moves and buttons) could hold up vrserver's frame. Commands
are now parsed into locals first, and the lock is held only to store the result.
RunFrame also called UpdateBooleanComponent six times and UpdateScalarComponent twice every
frame, and TrackedDevicePoseUpdated every frame even while disconnected. Components now go to
SteamVR only when they change (all of them on the first frame). The pose still goes out every
frame while the device is connected, as a tracked device's should; the disconnected pose goes
out once. The helper now sends a pose only when it changes, so the comment says the driver
keeps the last one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* ft-powerd: ask SteamVR every 100 ms, not on every input event
The loop polled the input devices with a 100 ms timeout and then, on every wake, did
SteamVR's part: PollNextEvent, the headset's activity level and every device's pose, all
IPC calls to vrserver. Input wakes it at once so the displays come on with the first
key or motion, but a moving mouse sends hundreds of events a second, so moving the mouse
meant hundreds of rounds of IPC a second instead of 10.
Every wake still drains the input devices and the control socket and counts input as
use straight away; SteamVR's part, and the backlight read that goes with it, now run
only when 100 ms have passed since the last time, and poll sleeps until then. Built in
the dev container (power/build.sh, no warnings); not run, since the live ft-powerd holds
@ft_powerd.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Eye tracker: ft-eyegrab checks only the slot each camera writes next
While copying, ft-eyegrab woke every 300 us (about 1,500 to 3,000 times a second) and
fingerprinted all eight slots each time: 8 x 256 strided reads from DMA-BUF memory.
- Each look now checks only the slot each camera writes next. The order is known (camera 0
3,0,1,2; camera 1 7,5,4,6,5,7,6,4), and the next slot follows from the last two; the table
starts from those orders and learns from every frame, so a SteamVR update that changes
them costs a few seconds of full scans, not frames.
- A camera with nothing in its expected slot 1.5 frames after its last one, or with no
order yet, gets all four slots checked, as before. A frame that turns up in an unexpected
slot means full scans for that camera for 2 s.
- A slot's fingerprint is taken again when it stops being one of the two in use, so a later
check sees only a new frame. A frame is still passed on when its camera starts the frame
after next.
- Between frames it sleeps until 2.5 ms before the next is due, then looks every 1 ms, with
0.5 ms of timer slack (PR_SET_TIMERSLACK, --share only). With no frames from either
camera for 0.5 s (headset off) it looks every 4 ms.
- --rec keeps its 0.3 ms polls (and the expected-slot checks), for its timestamps.
Tested offline by building poll_frames against simulated cameras that write each frame in
four bursts, the last after the next frame starts (6 s, both cameras): 1,076 frames passed
on, none torn or skipped, with the known orders and with camera 1 in a different order.
Wakeups 1,486/s -> 207/s, the poller's CPU 3.6% -> 0.8% of a core (in plain memory; the real
DMA-BUF reads cost more), and a frame's start is seen 1.35 ms after it begins on average
instead of 0.76. With a camera stalling 15 ms every 2 s, the old poller passed on 22 torn
frames and the new one 8 or fewer.
Built (glibc 2.38 symbols at most, the host has 2.39), not installed: it runs as root from
/etc/frametop, so it takes effect only after gaze/tracker/install.sh (sudo).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: upload from the headset, then plug in and leave it
Export and upload are done in the headset now: the export page only notes
that VR may stutter a little. Upload opens the pull request first (a
draft) and shows its link, telling the person to plug in the headset and
leave it until it says Uploaded; the files then go to refs/pr/N, and the
pull request is marked open at the end. A retry of the same export goes on
in the same pull request. While an export or upload runs, a host unit holds
a logind sleep inhibitor so the Frame stays awake with the headset off.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Remote Access: check the status every 5 s instead of every 2 s
While its window was open, Frametop Remote Access ran remote-ctl.sh status every 2 s,
and each run spawns bash, curl (the tailnet name from tailscaled) and python3 to parse
it. It now checks every 5 s, plus when the window comes to the front and once more 2 s
after turning remote access on or off or changing the password, so a change still
shows within a couple of seconds. A check doesn't start while one is still running.
Doing the check in-process would duplicate remote-ctl.sh's idea of "running", which
the session and the pause code share.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* update-check: KWin's blur and contrast effect ids, retest hints
The session now turns KWin's blur and contrast effects off by id (blurEnabled and
contrastEnabled in the desktop's kwinrc), and a KWin that renamed them would quietly
leave them on. The check looks for their built-in factories (KWin::blur_factory,
KWin::contrast_factory) in kwin_wayland, which it already reads for --output-count,
and warns if one is gone. The kwin and plasma-workspace retest hints gain the blur and
the hidden autostart entries.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pause gesture: look the controllers up every 30 s, not every 3 s
The gesture reader fetched vrserver's /input/getstate.json over HTTP every 3 seconds, the
whole time the relay runs, to notice a controller's root path changing when the 3D mouse
takes or gives back its hand role.
It now looks them up when it connects, when a message comes from a device path it doesn't
know (at most every 3 s; the device is read from the message with two string searches, not
a JSON parse of all 160 a second), 1.5 s after the relay's 3D mouse connects or lets go (the
relay tells it through GamePause.controllers_changed), and otherwise every 30 s. The keys
test's pause stub gets the new method.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Input relay: send mouse motion at most every 4 ms
The relay sent the helper one "move" per SYN_REPORT, so a 1000 Hz mouse sent 1000 datagrams
a second to a helper whose loop runs every 8 ms, and each one went through a dozen sscanf
and strncmp tests in the helper before reaching the move handler. In a 200 ms test at
1000 Hz, 149 reports now make 45 moves with the same total.
flush() on a report now sends only once 4 ms have passed since the last move; tick() sends
the rest when due, and the select timeout shrinks to match. Buttons and the gaze
keys still flush first, unconditionally, so a click lands where the pointer was. In the
helper, "move" is now tested first in the command dispatch, and its handling is one lambda.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Gaze: ft-gaze prints and reads only the sources in use
ft-gaze computed and printed all six sources for every sample: about 1.3 KB of JSON a line
with our tracker (practice2), 120 KB a second through podman's stdio relay for ft-gazed to
json.loads 90 times a second. It also read SteamVR's gaze action for every sample outside
games (UpdateActionState and GetEyeTrackingDataRelativeToNow, two calls into vrserver, 180 a
second), though with our tracker ft-gazed only uses own and mmap1.
- ft-gaze takes --sources LIST (action, mmap1, mmap2, left, right, own, and eye for the EYE
object; all by default, so the probe and ft-eyes-session are unchanged), and with
--watch-stdin a line "sources LIST" on stdin switches them. A source left out isn't read
and prints as {"ok":0} ("eye" as null), so every line keeps the same keys. An older
ft-gaze ignores both, and prints everything as before.
- ft-gazed asks for what it reads: own,mmap1 with our tracker; left,right,mmap1 with
SteamVR's eyes; the source plus mmap1 and mmap2 on the older one-source path. While a
check or the calibration runs or waits to open, all of them, since checks record every
source (the calibration fits the action's correction too) and the fit check reads "eye".
It switches as soon as that changes, well inside the check's 0.45 s settle.
- So the action is read only during checks, or with --source action.
On recorded samples, a line with own and mmap1 is about 700 bytes instead of 1,300
(practice2), and one with left, right and mmap1 about 550 instead of 940 (test1).
gaze/test/idle-test.py now checks that ft-gaze starts with every source for a check and is
then switched to those in use, without the action or own; all its checks pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pointer: don't put a vanished panel back in the visibility map
The panel-edge test read visible[edgeKey], and when the last panel the cursor touched was
gone from the overlay list, that added it back as hidden. The map then had more entries than
there are handles, which made the 50 ms visibility poll run every frame, and since the last
commit it also counted as a visibility change each time, so unchanged frames were never
reused. The edge test now looks the key up without adding it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Gaze: ft-gaze's loop runs every 4 ms instead of 2
ft-gaze's loop slept 2 ms, so 500 times a second it read the head pose
(GetDeviceToAbsoluteTrackingPose), checked the eye tracker's counter, and drained SteamVR's
events, for samples that come 90 times a second.
It now sleeps 4 ms. A new sample is printed within 4 ms of appearing, 2 on average (was 1),
and the pose history still has a pose within 2 ms of any sample's time, which keeps the
head-pose error under 0.2 degrees for a head turning 100 degrees a second. Sleeping until
the next sample is due would have thinned the pose history to 11 ms.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Gaze: the hidden panel waits for a command instead of waking every 50 ms
The calibration panel runs for as long as the gaze service does, hidden nearly all the time,
and it woke 20 to 30 times a second to look at its socket and SteamVR's events: about 0.9% of
a core, the main cost left with gaze idle.
It now waits in poll() on its command socket: up to a second while hidden, and up to 10 ms
while shown, as before (it still drains SteamVR's events each pass, so a quit is acknowledged
within a second while hidden). A command wakes it at once, so "show" draws sooner than
before. With --watch-stdin, its stdin closing wakes it as well, so stopping it doesn't wait.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Pointer: ft-screens announces new panels to the helper
The helper now reads SteamVR's list of panels every 20 s instead of every second, so a panel
made in between (a floating window's menu, frametop.float.N.sub.K, or the Frametop keyboard
the first time it opens) couldn't be clicked with the mouse until the next read. ft-screens
now sends "overlay <key>" to @ft_pointer_helper right after it makes one, and the helper adds
it to its list at once (only frametop.* keys). An older helper ignores it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Hands: fix Export doing nothing, and show it's busy at once
af2ea7c put _stay_awake between exportSession and its @Slot, so the
window's Export button called a method QML couldn't see. A new test checks
every backend call in main.qml against Backend's slots and properties.
Export and Upload now say Exporting…/Uploading… with a spinner the moment
they're pressed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Lazy susan: Meta+Alt+Tab spins the panels around you
- ft-screens "spin next|prev|<degrees>": every unpinned screen and
floating window turns together about a vertical axis through your
head (0.3 s, eased), so the next panel on the right or left comes to
straight ahead; the arrangement stays as it is. Taps during a spin
add to it, from where the panels are headed; grabbing a panel or
placing it (ft-layout, ft-floatd) takes it out of the spin
- when a spin settles, the panel in front gets the pointer (recenter),
typing (as after a click), and KWin's active window: its floating
window, or the top window on a screen (ft-floatd "front N", the KWin
script's activate-output). KWin's outputs follow the screens'
new places (ft-layout scale), as after a move
- the input relay: spin_next and spin_prev actions, Meta+Alt+Tab and
Meta+Alt+Shift+Tab by default; Frametop Input Settings lists them.
Not Meta+Tab: that's Cmd+Tab on a Mac reached through a remote
desktop like RustDesk, and the relay would take the Mac's app
switcher. Meta+Alt+Tab (Cmd+Option+Tab) is unused on macOS,
Windows, and KDE
Used on the Frame (SteamOS 0.3.0 build 20260922) with one screen and
three or four floating windows, through RustDesk to a Mac.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* hand recorder: login command works from Frametop's Konsole
Frametop's Konsole sets XDG_RUNTIME_DIR=/run/user/UID/frametop, where podman finds
no container state, so 'distrobox enter dev -- hf auth login' failed with a crun
error. The command the Upload page shows now sets the real runtime folder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* hand recorder: log in from the Upload page, three-step page, no terminal
The Upload page is now three numbered steps: choose the export, log in to
Hugging Face, upload. Log in runs hub.py login, huggingface_hub's browser
login (OAuth device code, as hf auth login does): the link opens in the
browser and the page shows the code to enter, with Copy code and Cancel.
hub.py saves the token; the window never sees one, and nobody pastes one.
The terminal upload and the login command are gone from the page, and
UPLOAD.md is now a short 'About uploading' under the steps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* hand recorder: take.json keeps camera clock samples
sets.bin's capture_ns is CLOCK_MONOTONIC_RAW; poses.jsonl and prompts.jsonl are
CLOCK_MONOTONIC. On 2026-10-03 the two were 0.80 s apart during a session and
1.11 s apart five hours later, so images can't be paired with poses by
capture_ns. take.json now samples RAW minus MONOTONIC as each recording part
starts and stops (as ft-hands' raw_minus_mono_ns), so readers can put each
exposure on the poses' clock.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* hand recorder export: no controller poses without controllers, nothing in deleted ranges
When the checklist says no controllers, exported poses.jsonl has left and right
null and feedback lines carry no controller state: controllers left switched on
still get tracked (one wandered 2 m in a real session) and would read as the
hands' ground truth. Poses and live-tracker feedback inside deleted ranges are
left out too, as the images there are.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* hand recorder: final consent text (2026-10-03), residency check, installer
Consent 2026-10-03, after a non-lawyer review: who runs this and how to reach
them, the dataset is public (Hugging Face, possibly abroad), the Hugging Face
username shows next to the contributor id, purposes (no identification), safety,
the maintainer grant passes to whoever maintains Frametop next, withdrawal
before and after merge, rights such as the GDPR's, and what a new version means.
Residents of Illinois, Texas and Washington can't take part for now (biometric
privacy laws): a third checkbox, profile consent.region_ok, checked by
validate.py from this consent version on. The DRAFT banners are gone, so uploads
no longer need FT_HANDREC_ALLOW_UPLOAD.
hands/rec/install.sh installs the recorder on a Frame with Frametop: container
packages, hand tracking and panel builds, ft-camd's capabilities, menu entry.
test_qml_backend also checks each call's argument count against the slots.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Screens: a reset button next to the grab bar, clickable in VR games
Each desktop screen gets a reset button left of its bar (a reticle). It
puts every screen back in its layout around where you are now, like
Meta+Shift+R (ft-layout apply).
In a VR game the screens leave the controllers to the game (the
outside_games and dashboard modes), so a controller couldn't click any
of their controls. Aiming a hand controller at the reset button now sets
MakeOverlaysInteractiveIfVisible on that button's overlay alone, so the
trigger clicks it; the flag clears half a second after the aim leaves a
zone twice as wide, and the game gets the controllers back. The aim
comes from the laser poses ft-screens already reads to show the controls.
The ft-layout spawn is now RunLayout(cmd), shared with the arrange.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Click stability: 32 logical pixels by default, not 8
8 is about 0.2 degrees on a 3.4 m wide 3440-pixel screen 2 m away, so a
trigger press turned into a drag unless the hand was very still. 32
(about 0.9 degrees) felt much better in the headset (2026-10-03).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Screens: in games, pointing a controller at a panel turns its laser on
SteamVR's own floating windows take the laser while a controller points
at them in a game and give it back when it points away. Frametop's
panels didn't: with the controllers left to the game (outside_games, the
default, or dashboard), they couldn't be clicked without the dashboard.
ft-screens now sets MakeOverlaysInteractiveIfVisible on a screen or
floating window while a hand controller's laser pose meets it, its
controls, or its popups (UpdateAim; curved screens hit on their
cylinder), and clears it 0.3 s after the aim leaves a wider margin. A
drag or a held button keeps it on. The keyboard, one overlay, uses
ComputeOverlayIntersection and now follows the mode when a game starts
or ends while it's open. This replaces the reset button's own aim zone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Screens: find the controllers' laser tip during VR games too
GetComponentStateForDevicePath with no input source handle fails for
every render model component while a VR game runs (checked 2026-10-03
with a game up: all 21 components of frame_controller_right). TipOffset
then fell back to the controller's pose, which aims 40 degrees above the
Frame controller's laser. In games, pointing at a screen's middle missed
it and pointing below it hit, so the new aim-to-laser only worked from
the bottom; the controls' reveal and pin/roll aim were off the same way.
GetComponentState still answers then, with the same tip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* input-relay: Add mute key as volume key
Add KEY_MUTE as volume key. It will be mapped to KEY_MACRO28 and
use wpctl to toggle the mute of the default audio sink. The toggle of
the mute state will be done once when the key is pressed instead of
continuously toggling it when it is held down.
This allows the mute button on keyboards to work properly.
Signed-off-by: SuperTuxii <123881249+SuperTuxii@users.noreply.github.com>
* Session: bring back a taskbar saved on a screen the desktop doesn't have
Plasma 6.2.5 keeps a panel on a screen number and never moves one whose
number is past the screen count, so a taskbar saved on a spare output
(#18, lastScreen=8 with three screens) or on a screen a smaller layout
dropped stayed hidden. Before Plasma starts, session/fix-panels.py moves
such a panel and its tray's containment to screen 0 (the primary),
keeping its widgets, unless screen 0 already has a panel on that edge.
doctor.sh checks the panels' screens, and report.sh lists them with the
live outputs and panels.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* keys-test: the spin bindings (Meta+Alt+Tab, Meta+Alt+Shift+Tab), not while paused
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Menu entries: own programs for Reset Screen Layout and Hide/Show Screens
Reset Screen Layout and Hide/Show Screens both ran ft-layout. Steam lists
entries by program, so Hide/Show launched Reset. Each gets a wrapper.
From PR #17 (only this part of 9618be8; its host_command change is for the
Nix packages).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Screens: release a held button that can't come up on a screen
Pausing, or hiding the screen a button went down on, took the laser off
it mid-click; the pause gesture's second thumbstick click does that.
SteamVR's laser mouse then forgot the button ("Mouse down count is 1 but
states are all false"), no release came, and the catcher kept showing
whenever the pressing laser was off the panels, even while paused. Being
interactive, it kept the VR game's controllers from it until the
desktop restarted (2026-10-04, Beat Saber).
ft-screens now releases a held button when it's paused, when the screen
it went down on is hidden, or, during VR games, when the pressing hand
controller has held nothing for a second. GetControllerState answers
overlay apps only while a game runs; outside games a hold is never cut.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Signed-off-by: SuperTuxii <123881249+SuperTuxii@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Codex <codex@localhost>
Co-authored-by: CuriousJ <curious.j.tuber@gmail.com>
Co-authored-by: Patrick McDavid <fusionjunky@gmail.com>
Co-authored-by: SuperTuxii <123881249+SuperTuxii@users.noreply.github.com>
Co-authored-by: John Murray <5672686+JRMurr@users.noreply.github.com>
Pausing, or hiding the screen a button went down on, took the laser off
it mid-click; the pause gesture's second thumbstick click does that.
SteamVR's laser mouse then forgot the button ("Mouse down count is 1 but
states are all false"), no release came, and the catcher kept showing
whenever the pressing laser was off the panels, even while paused. Being
interactive, it kept the VR game's controllers from it until the
desktop restarted (2026-10-04, Beat Saber).
ft-screens now releases a held button when it's paused, when the screen
it went down on is hidden, or, during VR games, when the pressing hand
controller has held nothing for a second. GetControllerState answers
overlay apps only while a game runs; outside games a hold is never cut.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conflicts with experimental's pause_toggle, steam_menu and command: actions and its
AnnounceOverlay: both kept. The spin bindings join the Meta tap in the defaults, spinning
doesn't need pointer mode, and like other actions it does nothing while Frametop is paused.
AnnounceOverlay now sends through the PR's SendPointer.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Plasma 6.2.5 keeps a panel on a screen number and never moves one whose
number is past the screen count, so a taskbar saved on a spare output
(#18, lastScreen=8 with three screens) or on a screen a smaller layout
dropped stayed hidden. Before Plasma starts, session/fix-panels.py moves
such a panel and its tray's containment to screen 0 (the primary),
keeping its widgets, unless screen 0 already has a panel on that edge.
doctor.sh checks the panels' screens, and report.sh lists them with the
live outputs and panels.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
GetComponentStateForDevicePath with no input source handle fails for
every render model component while a VR game runs (checked 2026-10-03
with a game up: all 21 components of frame_controller_right). TipOffset
then fell back to the controller's pose, which aims 40 degrees above the
Frame controller's laser. In games, pointing at a screen's middle missed
it and pointing below it hit, so the new aim-to-laser only worked from
the bottom; the controls' reveal and pin/roll aim were off the same way.
GetComponentState still answers then, with the same tip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
SteamVR's own floating windows take the laser while a controller points
at them in a game and give it back when it points away. Frametop's
panels didn't: with the controllers left to the game (outside_games, the
default, or dashboard), they couldn't be clicked without the dashboard.
ft-screens now sets MakeOverlaysInteractiveIfVisible on a screen or
floating window while a hand controller's laser pose meets it, its
controls, or its popups (UpdateAim; curved screens hit on their
cylinder), and clears it 0.3 s after the aim leaves a wider margin. A
drag or a held button keeps it on. The keyboard, one overlay, uses
ComputeOverlayIntersection and now follows the mode when a game starts
or ends while it's open. This replaces the reset button's own aim zone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
8 is about 0.2 degrees on a 3.4 m wide 3440-pixel screen 2 m away, so a
trigger press turned into a drag unless the hand was very still. 32
(about 0.9 degrees) felt much better in the headset (2026-10-03).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Each desktop screen gets a reset button left of its bar (a reticle). It
puts every screen back in its layout around where you are now, like
Meta+Shift+R (ft-layout apply).
In a VR game the screens leave the controllers to the game (the
outside_games and dashboard modes), so a controller couldn't click any
of their controls. Aiming a hand controller at the reset button now sets
MakeOverlaysInteractiveIfVisible on that button's overlay alone, so the
trigger clicks it; the flag clears half a second after the aim leaves a
zone twice as wide, and the game gets the controllers back. The aim
comes from the laser poses ft-screens already reads to show the controls.
The ft-layout spawn is now RunLayout(cmd), shared with the arrange.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Consent 2026-10-03, after a non-lawyer review: who runs this and how to reach
them, the dataset is public (Hugging Face, possibly abroad), the Hugging Face
username shows next to the contributor id, purposes (no identification), safety,
the maintainer grant passes to whoever maintains Frametop next, withdrawal
before and after merge, rights such as the GDPR's, and what a new version means.
Residents of Illinois, Texas and Washington can't take part for now (biometric
privacy laws): a third checkbox, profile consent.region_ok, checked by
validate.py from this consent version on. The DRAFT banners are gone, so uploads
no longer need FT_HANDREC_ALLOW_UPLOAD.
hands/rec/install.sh installs the recorder on a Frame with Frametop: container
packages, hand tracking and panel builds, ft-camd's capabilities, menu entry.
test_qml_backend also checks each call's argument count against the slots.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When the checklist says no controllers, exported poses.jsonl has left and right
null and feedback lines carry no controller state: controllers left switched on
still get tracked (one wandered 2 m in a real session) and would read as the
hands' ground truth. Poses and live-tracker feedback inside deleted ranges are
left out too, as the images there are.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
sets.bin's capture_ns is CLOCK_MONOTONIC_RAW; poses.jsonl and prompts.jsonl are
CLOCK_MONOTONIC. On 2026-10-03 the two were 0.80 s apart during a session and
1.11 s apart five hours later, so images can't be paired with poses by
capture_ns. take.json now samples RAW minus MONOTONIC as each recording part
starts and stops (as ft-hands' raw_minus_mono_ns), so readers can put each
exposure on the poses' clock.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Upload page is now three numbered steps: choose the export, log in to
Hugging Face, upload. Log in runs hub.py login, huggingface_hub's browser
login (OAuth device code, as hf auth login does): the link opens in the
browser and the page shows the code to enter, with Copy code and Cancel.
hub.py saves the token; the window never sees one, and nobody pastes one.
The terminal upload and the login command are gone from the page, and
UPLOAD.md is now a short 'About uploading' under the steps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Frametop's Konsole sets XDG_RUNTIME_DIR=/run/user/UID/frametop, where podman finds
no container state, so 'distrobox enter dev -- hf auth login' failed with a crun
error. The command the Upload page shows now sets the real runtime folder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
af2ea7c put _stay_awake between exportSession and its @Slot, so the
window's Export button called a method QML couldn't see. A new test checks
every backend call in main.qml against Backend's slots and properties.
Export and Upload now say Exporting…/Uploading… with a spinner the moment
they're pressed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The helper now reads SteamVR's list of panels every 20 s instead of every second, so a panel
made in between (a floating window's menu, frametop.float.N.sub.K, or the Frametop keyboard
the first time it opens) couldn't be clicked with the mouse until the next read. ft-screens
now sends "overlay <key>" to @ft_pointer_helper right after it makes one, and the helper adds
it to its list at once (only frametop.* keys). An older helper ignores it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The calibration panel runs for as long as the gaze service does, hidden nearly all the time,
and it woke 20 to 30 times a second to look at its socket and SteamVR's events: about 0.9% of
a core, the main cost left with gaze idle.
It now waits in poll() on its command socket: up to a second while hidden, and up to 10 ms
while shown, as before (it still drains SteamVR's events each pass, so a quit is acknowledged
within a second while hidden). A command wakes it at once, so "show" draws sooner than
before. With --watch-stdin, its stdin closing wakes it as well, so stopping it doesn't wait.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ft-gaze's loop slept 2 ms, so 500 times a second it read the head pose
(GetDeviceToAbsoluteTrackingPose), checked the eye tracker's counter, and drained SteamVR's
events, for samples that come 90 times a second.
It now sleeps 4 ms. A new sample is printed within 4 ms of appearing, 2 on average (was 1),
and the pose history still has a pose within 2 ms of any sample's time, which keeps the
head-pose error under 0.2 degrees for a head turning 100 degrees a second. Sleeping until
the next sample is due would have thinned the pose history to 11 ms.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The panel-edge test read visible[edgeKey], and when the last panel the cursor touched was
gone from the overlay list, that added it back as hidden. The map then had more entries than
there are handles, which made the 50 ms visibility poll run every frame, and since the last
commit it also counted as a visibility change each time, so unchanged frames were never
reused. The edge test now looks the key up without adding it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ft-gaze computed and printed all six sources for every sample: about 1.3 KB of JSON a line
with our tracker (practice2), 120 KB a second through podman's stdio relay for ft-gazed to
json.loads 90 times a second. It also read SteamVR's gaze action for every sample outside
games (UpdateActionState and GetEyeTrackingDataRelativeToNow, two calls into vrserver, 180 a
second), though with our tracker ft-gazed only uses own and mmap1.
- ft-gaze takes --sources LIST (action, mmap1, mmap2, left, right, own, and eye for the EYE
object; all by default, so the probe and ft-eyes-session are unchanged), and with
--watch-stdin a line "sources LIST" on stdin switches them. A source left out isn't read
and prints as {"ok":0} ("eye" as null), so every line keeps the same keys. An older
ft-gaze ignores both, and prints everything as before.
- ft-gazed asks for what it reads: own,mmap1 with our tracker; left,right,mmap1 with
SteamVR's eyes; the source plus mmap1 and mmap2 on the older one-source path. While a
check or the calibration runs or waits to open, all of them, since checks record every
source (the calibration fits the action's correction too) and the fit check reads "eye".
It switches as soon as that changes, well inside the check's 0.45 s settle.
- So the action is read only during checks, or with --source action.
On recorded samples, a line with own and mmap1 is about 700 bytes instead of 1,300
(practice2), and one with left, right and mmap1 about 550 instead of 940 (test1).
gaze/test/idle-test.py now checks that ft-gaze starts with every source for a check and is
then switched to those in use, without the action or own; all its checks pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The relay sent the helper one "move" per SYN_REPORT, so a 1000 Hz mouse sent 1000 datagrams
a second to a helper whose loop runs every 8 ms, and each one went through a dozen sscanf
and strncmp tests in the helper before reaching the move handler. In a 200 ms test at
1000 Hz, 149 reports now make 45 moves with the same total.
flush() on a report now sends only once 4 ms have passed since the last move; tick() sends
the rest when due, and the select timeout shrinks to match. Buttons and the gaze
keys still flush first, unconditionally, so a click lands where the pointer was. In the
helper, "move" is now tested first in the command dispatch, and its handling is one lambda.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The gesture reader fetched vrserver's /input/getstate.json over HTTP every 3 seconds, the
whole time the relay runs, to notice a controller's root path changing when the 3D mouse
takes or gives back its hand role.
It now looks them up when it connects, when a message comes from a device path it doesn't
know (at most every 3 s; the device is read from the message with two string searches, not
a JSON parse of all 160 a second), 1.5 s after the relay's 3D mouse connects or lets go (the
relay tells it through GamePause.controllers_changed), and otherwise every 30 s. The keys
test's pause stub gets the new method.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The session now turns KWin's blur and contrast effects off by id (blurEnabled and
contrastEnabled in the desktop's kwinrc), and a KWin that renamed them would quietly
leave them on. The check looks for their built-in factories (KWin::blur_factory,
KWin::contrast_factory) in kwin_wayland, which it already reads for --output-count,
and warns if one is gone. The kwin and plasma-workspace retest hints gain the blur and
the hidden autostart entries.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
While its window was open, Frametop Remote Access ran remote-ctl.sh status every 2 s,
and each run spawns bash, curl (the tailnet name from tailscaled) and python3 to parse
it. It now checks every 5 s, plus when the window comes to the front and once more 2 s
after turning remote access on or off or changing the password, so a change still
shows within a couple of seconds. A check doesn't start while one is still running.
Doing the check in-process would duplicate remote-ctl.sh's idea of "running", which
the session and the pause code share.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Export and upload are done in the headset now: the export page only notes
that VR may stutter a little. Upload opens the pull request first (a
draft) and shows its link, telling the person to plug in the headset and
leave it until it says Uploaded; the files then go to refs/pr/N, and the
pull request is marked open at the end. A retry of the same export goes on
in the same pull request. While an export or upload runs, a host unit holds
a logind sleep inhibitor so the Frame stays awake with the headset off.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
While copying, ft-eyegrab woke every 300 us (about 1,500 to 3,000 times a second) and
fingerprinted all eight slots each time: 8 x 256 strided reads from DMA-BUF memory.
- Each look now checks only the slot each camera writes next. The order is known (camera 0
3,0,1,2; camera 1 7,5,4,6,5,7,6,4), and the next slot follows from the last two; the table
starts from those orders and learns from every frame, so a SteamVR update that changes
them costs a few seconds of full scans, not frames.
- A camera with nothing in its expected slot 1.5 frames after its last one, or with no
order yet, gets all four slots checked, as before. A frame that turns up in an unexpected
slot means full scans for that camera for 2 s.
- A slot's fingerprint is taken again when it stops being one of the two in use, so a later
check sees only a new frame. A frame is still passed on when its camera starts the frame
after next.
- Between frames it sleeps until 2.5 ms before the next is due, then looks every 1 ms, with
0.5 ms of timer slack (PR_SET_TIMERSLACK, --share only). With no frames from either
camera for 0.5 s (headset off) it looks every 4 ms.
- --rec keeps its 0.3 ms polls (and the expected-slot checks), for its timestamps.
Tested offline by building poll_frames against simulated cameras that write each frame in
four bursts, the last after the next frame starts (6 s, both cameras): 1,076 frames passed
on, none torn or skipped, with the known orders and with camera 1 in a different order.
Wakeups 1,486/s -> 207/s, the poller's CPU 3.6% -> 0.8% of a core (in plain memory; the real
DMA-BUF reads cost more), and a frame's start is seen 1.35 ms after it begins on average
instead of 0.76. With a camera stalling 15 ms every 2 s, the old poller passed on 22 torn
frames and the new one 8 or fewer.
Built (glibc 2.38 symbols at most, the host has 2.39), not installed: it runs as root from
/etc/frametop, so it takes effect only after gaze/tracker/install.sh (sudo).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The loop polled the input devices with a 100 ms timeout and then, on every wake, did
SteamVR's part: PollNextEvent, the headset's activity level and every device's pose, all
IPC calls to vrserver. Input wakes it at once so the displays come on with the first
key or motion, but a moving mouse sends hundreds of events a second, so moving the mouse
meant hundreds of rounds of IPC a second instead of 10.
Every wake still drains the input devices and the control socket and counts input as
use straight away; SteamVR's part, and the backlight read that goes with it, now run
only when 100 ms have passed since the last time, and poll sleeps until then. Built in
the dev container (power/build.sh, no warnings); not run, since the live ft-powerd holds
@ft_powerd.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Handle() held the state lock through a chain of up to a dozen sscanf calls per command, and
RunFrame, which vrserver calls every frame, takes the same lock, so a burst of commands
(about 116 poses a second, plus moves and buttons) could hold up vrserver's frame. Commands
are now parsed into locals first, and the lock is held only to store the result.
RunFrame also called UpdateBooleanComponent six times and UpdateScalarComponent twice every
frame, and TrackedDevicePoseUpdated every frame even while disconnected. Components now go to
SteamVR only when they change (all of them on the first frame). The pose still goes out every
frame while the device is connected, as a tracked device's should; the disconnected pose goes
out once. The helper now sends a pose only when it changes, so the comment says the driver
keeps the last one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The relay sent to @ft_pointer_helper on a blocking socket. When the helper stalled, a
layout placement or grabprobe holds it for seconds while ft-gazed keeps filling its socket at
90 Hz, the relay's one loop blocked with it: keyboards, the volume keys (which must never
reach gamescope), and pausing all stopped until the helper read again.
The socket is non-blocking now. A command the helper doesn't take (EAGAIN) waits in a queue,
and everything after it queues behind it so the order holds; tick() sends what it can on each
loop, and the select timeout drops to 20 ms while anything waits. Mouse moves add up into one
queued move. A scroll notch is dropped rather than queued, since scrolling seconds late is no
use; its release still goes. Presses, releases, show, hide, and the rest are kept, so no
button stays down. The queue holds at most 512 commands. While paused, the configured
pointer's queue still drains, so the releases and "hide" from standing down arrive. A
"vrbind" that hits a full socket is sent again on the next loop instead of being lost.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The nested Plasma session runs the system's XDG autostart entries. Discover's update
notifier (/etc/xdg/autostart/org.kde.discover.notifier.desktop) started
plasma-discover --mode update inside it, 520-620 MB resident and about 9% of a core,
with flatpak-system-helper and AppStream downloads behind it. IBus started a nested
ibus-daemon with kimpanel and ibus-extension-gtk3, which no app in the desktop can use:
KWin's input method is ft-textinput (zwp_input_method_v1, focus reports only; the VR
keyboard types through ft-screens' seat), and the session already drops QT_IM_MODULE,
GTK_IM_MODULE and XMODIFIERS. Nothing in Frametop talks to IBus.
Before Plasma starts, the session script copies both entries into
$XDG_CONFIG_HOME/autostart with Hidden=true, which plasma-session honours for that
desktop only. It does this once ([Defaults] autostart=1 in frametoprc) and skips a name
the user already has a file for, so deleting the copy brings the program back. The
geoclue demo agent stays (it answers apps' location requests outside GNOME and idles at
0%), and orca's entry is OnlyShowIn GNOME-family desktops, so it never ran. Tested
against a temporary XDG_CONFIG_HOME, including an existing user ibus.desktop left alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The nested kwinrc had no [Plugins] group, so KWin ran its default blur and background
contrast effects, and kdeglobals had no AnimationDurationFactor, so animations ran at
full length. KWin renders through zink on Turnip, on the GPU vrcompositor needs, and
blur re-renders what's behind every translucent panel and menu; each animation frame is
another frame for KWin and ft-screens.
Before KWin starts, the session script now writes [Plugins] blurEnabled=false and
contrastEnabled=false to $XDG_CONFIG_HOME/kwinrc and [KDE] AnimationDurationFactor=0 to
its kdeglobals, each only if the desktop's own file has no value for it. It does this
once and records that in $XDG_CONFIG_HOME/frametoprc ([Defaults] effects=1), because
System Settings deletes a key put back to its default: without the marker, turning blur
back on wouldn't survive a restart. The ids blur and contrast are the built-in effects
of KWin 6.2.5 on SteamOS (both enabled by default in its plugin metadata). Tested
against a temporary XDG_CONFIG_HOME: fresh config, an existing blurEnabled=true kept,
and a deleted key not rewritten.
README and docs/reference.md say how to turn them back on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every frame (about 116 a second) the helper tested the cursor ray against every visible
overlay twice with ComputeOverlayIntersection, set the dot's alpha, width, transform and
visibility (five calls into SteamVR), and sent the driver a pose datagram, even with the
mouse and the head still.
Now a frame reuses the last collision result when the mouse, the anchor (1 mm) and the
eye (5 mm) haven't moved and no overlay showed, hid, or changed handle. The passes still run
at least every 100 ms, since overlays move on their own (a floating window's controls follow
it), and always while dragging. The dots' setters go to SteamVR only when their value changes:
the placement when the dot moved 0.2 mm or the eye 5 mm, which turns or resizes it by well
under 1%, and the width on a 0.5% change. The plain pose goes to the driver only when the
laser's origin moved 0.2 mm or its direction 0.04 deg (0.1 mm where it lands, 15 cm on), and
at least every 100 ms; the driver keeps the last pose and reports it every frame. A tilt's
pose, a placement, or waking sends the next one regardless.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ft-screens gave KWin a frame callback for every committed screen on each tick, and the tick
was an 11 ms timer set again after each run, so it slid through the display's frame and came
about 85 times a second at 90 Hz: the desktop repeated a frame several times a second (judder
in scrolling and video), and KWin drew every screen in one burst at a random point of
vrcompositor's frame. Hidden screens got the same 90 Hz unless Frametop was paused for a game.
- Ticks run on a timerfd at absolute times, once per display frame, 1 ms after the vsync
(IVRSystem::GetTimeSinceLastVsync and the HMD's display frequency, read once a second), so
KWin gets its callbacks early in the frame. Measured with --no-vr: 91 wakeups a second
instead of about 85. On the Frame the vsync times SteamVR reports lie on a 90 Hz grid.
- Each screen's callbacks come at a rate for how much of it you see (vr.cpp,
UpdateAttention): every frame while focused (within 12 degrees of where your head points,
a laser or the mouse on it in the last 1.5 s, carried, or typed on), 15 a second for the
rest of what you see (within 60 degrees), and 1 a second when hidden, behind you, or
paused. Levels rise at once and fall after 1.5 s (focused) or 0.5 s (in view). KWin draws
a screen only after its callback and its apps wait for theirs, so this throttles the apps
too. A screen where nothing changes costs nothing at any rate, as before.
- A video in view keeps every frame: 8 commits in a row that each redraw 6% or more of the
screen, at 10 a second or more, count as one (from the surface's buffer damage).
- "rates F V H" / --rates set the three rates (default 0 15 1, 0 = every frame), "rates?"
shows them and each screen's level, "watch S" gives everything full rate for S seconds for
a remote viewer (vnc-bridge.sh renews it), and "phase MS" moves the ticks for tuning.
- ft-screens' main thread runs at nice -5 after the session starts: SteamOS allows down to
-8 once the soft RLIMIT_NICE is raised, and KWin waits on these ticks. It had spent nearly
3 times as long waiting to run as running.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ft-eyes looked for new frames about 1,000 times a second: each pass of its loop asked the
control socket with a non-blocking recvfrom (a BlockingIOError nearly every time), read
both cameras' counters, and slept 1 ms. The frames come every 11.1 ms per camera, and only
as counters in ft-eyegrab's shared memory, so there's no fd to wait on.
Now each pass ends in select() on the control socket, with a timeout until 2 ms before the
next frame of either camera is due (from when its last one was seen), then every 1 ms until
it comes. A command wakes it at once. A camera with no frame for 0.1 s (headset off,
grabber idle) isn't waited for, and with both stopped it looks every 20 ms. Waiting for the
frame grabber's file uses the same select, 0.2 s at a time, instead of sleeping through
commands.
A new frame is still seen within about 1 ms of when it lands. On a synthetic share at 90 Hz
per camera, on a heavily loaded headset (load average 23, so ft-eyes rarely sat idle), its
waits went from 178 to 81 a second; unloaded, the old loop's 1 ms sleeps add up to about
1,000.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
While FreeRDP ran, vnc-bridge.sh called ft-layout remote-view every 5 s, which scans
all of /proc for plasmashell and runs kscreen-doctor -j: about 4.4% of a core for a
layout that rarely changes.
It now stats the two files the answer depends on, the nested KWin's
~/.config/frametop/kwinoutputconfig.json (positions, scales, primary) and
~/.config/frametop-layout.json (screen sizes), once a second while FreeRDP runs. After
either changes it reads the layout every 2 s for 10 s, since KWin's outputs follow the
file a few seconds later; otherwise once a minute, in case a change touched neither.
With no VNC viewer connected nothing runs (previous commit).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
vnc-bridge.sh kept FreeRDP connected to krdpserver from the moment remote desktop
started, so krdp captured and H.264-encoded every KWin redraw in software (openh264)
with nobody watching: krdpserver 55-78% of a core, xfreerdp 16-27%, Xvnc 6-11%, with 0
clients on :5900. krdp 6.7 creates its screencast session per RDP connection and drops
it when the connection closes, so krdpserver itself idles without one and stays up.
The bridge now counts established connections to Xvnc's port with ss, starts FreeRDP
when a viewer appears (the desktop shows about 3 s later; the VNC screen is black until
then) and stops it 45 s after the last one leaves (VNC_IDLE_SEC). Xvnc has no client
hook, so its log output, which it writes for every connection, wakes the bridge early;
otherwise it looks every 5 s while idle (0.1% of a core measured, against 0.9% for ss
once a second) and every second while FreeRDP runs. The layout check runs only while
FreeRDP runs.
While a viewer is connected the bridge sends "watch 15" to ft-screens (@ft_screens) at
once and every 5 s, so screens at a reduced frame rate (out of view, headset on a
stand) stream at full rate; it lapses by itself if the bridge dies, and an ft-screens
without the command just answers an error. The window search after starting FreeRDP
now ends when FreeRDP exits instead of polling for 30 s.
krdp on 127.0.0.1 with a fresh password, VNC on the tailnet address with VncAuth, and
remote-ctl.sh start/stop (pause and resume) are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The main loop slept a fixed 8 ms, about 116 wakeups a second, whether the pointer was awake
or not, and every second it looked up every overlay's handle and read a string property from
all 64 device slots to find its own device.
With the pointer off and hand gestures off, the loop now waits in poll() on its command
socket for up to 250 ms, or 20 ms while mapped Frame controller buttons are being read
(SteamVR input has no event to wait for). A mouse command ends the wait at once. The
headset's activity level, the game check, and the "vrgame" and "gazeawake" repeats keep
going at that pace. The 50 ms visibility poll and the 1 s handle lookups run only while the
pointer is awake, and waking forces both. The device index is looked for only while it's
unknown, and again after SteamVR activates or deactivates a device. The HMD pose history is
kept only with hand gestures on, its one user.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ft-eyes and ft-gaze run in the dev container through distrobox, so they live in podman's
libpod scope: frametop-gaze.service's limits never reach them, and they ran at nice 0
next to vrcompositor and vrserver, also at nice 0.
- ft-eyes sets itself to nice 10 and SCHED_BATCH at start, before its threads. Batch
turns off wakeup preemption, so a frame ft-eyes wakes up for can wait out a running
compositor's turn; a few ms late costs the gaze little.
- ft-gaze sets nice 5 before its threads start, but stays SCHED_OTHER: each sample goes
on to the pointer, and batch would add the same wait to every one of them.
- Both only ever lower their priority (a higher nice already set wins), and a failure
is logged and ignored. Checked in the dev container: nice 0 -> 10, policy 0 -> 3
(SCHED_BATCH) without any capability.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The helper ran `vrcmd --overlays` once a second while the pointer was awake, and in gaze
mode the pointer never sleeps. Each run is a shell plus vrcmd, a new SteamVR client, about
26 to 30 ms of CPU, so about 3% of a core all the time.
The list is now read every 20 seconds, and at once (at most once a second) when it may have
changed: the pointer waking, the dashboard opening or closing or creating an overlay, the
scene app changing, an "overlays" request, and a left click that hit nothing, which may be
on a panel that came up since. The thread waits on a condition variable instead of waking
every 100 ms, so it sleeps while paused. The main loop looks the keys up again as soon as a
new list is in, rather than at its next 1 s tick. Overlays already on the list still show
and hide within 50 ms, from the IsOverlayVisible poll.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Nothing called cv2.setNumThreads, so OpenCV kept a pool of one worker per core for
pupil windows of 140 to 240 px. Live, its idle workers spun and yielded about 14,000
times a second each, about a quarter of a core, next to SteamVR's compositor. numpy's
OpenBLAS also started 8 threads that never had work.
eyes_pupil.py now sets OpenCV to one thread, and ft-eyes sets OPENBLAS_NUM_THREADS and
OMP_NUM_THREADS to 1 before numpy loads (a value already in the environment wins).
Replaying fit1 into a scratch share (ft-eyes-replay, 14 s measured, capped at one core
with the replay): threads 13 -> 1, involuntary context switches 3,812/s -> 430/s,
system time 6.9% -> 1.6% of a core. Under that cap the frames it kept up with went from
21-36 to 57-69 a second per eye.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The dot is the current tracker's distance guess, often wrong; the ring is
where the hands should be.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
host-spawn starts a host command in the caller's folder. Started from /tmp,
the app's folder in the container is /run/host/tmp, which the host doesn't
have, so starting ft-camd (and every other host command) exited 127.
host_command now runs from home (env -C), and the launcher cds there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The checklist page measures the light when it opens, starting ft-camd if
nothing runs it (and stopping it on quit), instead of saying the cameras
aren't running. The round's lighting defaults to what the cameras measure:
daylight or indoor, from the mono cameras' ambient infrared. Lamps give off
little infrared, so dim and normal rooms read alike; picking dim, room or
daylight still overrides it. session.json gets source, measured and
ambient_ir.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
vrserver sends a controller's state only when something on it changes, and one lying still
or asleep may not even send its first one. The check subscribed to one controller and failed
after 3 s of silence. It now subscribes to all, and silence after a good handshake is a skip.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The gaze service ran ft-gaze and our own eye tracker all the time: with gaze mode off, ft-eyes
still took about 60% of a core, and ft-eyegrab, ft-gaze and ft-gazed 3 to 4% each. Now ft-gaze
and our tracker run only while gaze mode is on and someone wears the headset, while a check or
the calibration is open or asked for, or under a "wake" lease, which the Gaze page of Frametop
Input Settings renews while it's open. 30 s after the last use they stop, and the frame grabber
idles with our tracker.
- The pointer helper answers "gaze ? headset" with worn|away (SteamVR's activity level for the
headset); an older helper answers it as before, and the service then goes by gaze mode alone.
- A quick check, calibration, or fit check asked for while idle wakes the tracker and opens once
it sends; the automatic calibration waits quietly while it starts.
- Status has "awake" and "idle" (why), and the Gaze page shows it.
- gaze/test/idle-test.py runs the service with a fake helper and ft-gaze, offline.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ft-camd tells the side cameras apart by XRService's buffer allocation order,
which some XRService starts reverse; both of 2026-10-02's starts did, so the
cutouts missed the hands. HANDS_SWAP_SIDES=auto (the default) has ft-hands
vote from hands seen in both side cameras: the landmark rays meet in front
of both cameras only under the right naming. While undecided it probes the
exchanged naming with the landmark model. It decides in about 2 s of hands
(right on all 7 recordings replayed), swaps the views in place, and publishes
sides.json. 0 and 1 still force it, with a warning when the hands disagree.
Recordings carry each part's naming and the session's decision; review,
export, validate and ft-handreplay put the names right, and takes.py sides
records a decision by hand.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Frametop kept using the headset during games: with gaze mode off, our eye tracker still took
about 60% of a core, remote desktop about 2 cores while on, and KWin kept drawing hidden
screens because ft-screens sent their frame callbacks at 90 Hz. Pausing gives that back, and
resuming brings back only what pausing stopped. It's also a way to keep the gaze service and
our eye tracker off during games, which PR #13 asked for.
Paused (input/game_pause.py, run by the input relay):
- frametop-gaze stops (ft-eyegrab then idles by itself), and hand tracking and remote desktop
stop if they run
- the desktop hides and slows down: ft-screens "pause on" hides every panel and sends KWin a
frame callback once a second; or, with pause_desktop "close", the desktop closes and starts
again on resume
- the relay lets go of the 3D mouse, typing goes to Steam, and mapped buttons and key
combinations do only pause_toggle, steam_menu and commands
Toggled by both thumbsticks clicked together twice (configurable), read passively from
vrserver's web socket (input/vrws.py) so it works in games and takes nothing from them; by the
new pause_toggle action; by input/ft-pause; and, with pause_auto (default on), by a VR game
starting and ending, which the pointer helper now reports ("vrgame 1|0"). Frametop Input
Settings has a Games page for it. update-check.py checks the web socket, and doesn't count a
paused gaze service as failed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Since the relay leaves a node it can't read yet for the next scan (12f2e84, PR #12), it
checks os.access first, and the test's fake /dev/input paths don't exist, so the relay never
opened them and every key check failed. The fake os now says they're readable.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The second real session took 16 minutes, half of it 36 still poses. Labels
come from the auto-labeller, so what matters is variety, not clean holds.
A sweep step shows a strip of pose pictures and lights one every 4 s while
the hands move slowly near and far; each cue is a prompt event with
"cue": true. The core session is now 15 steps, about 5 minutes recorded.
The pose groups, the one-hand sweeps' groups and the cue order are shuffled
per session, seeded from its id and saved in session.json. A quick round
(--quick, or the checklist's choice) is about 2 minutes for extra lighting.
Touch the dot has 6 dots, the push sections two heights.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
After the headset wakes, XRService sometimes fails to load the colour module's
VCINT FPGA image; then only the two side cameras run, without the IR light,
and the tracker finds no hands. hands/camcheck.py reads XRService's log, the
video nodes it holds and ft-camd's ring, and says ok, degraded or unknown.
The recorder won't start while degraded (--ignore-cameras overrides it), offers
a confirmed SteamVR restart, and stops the first hand-size step when the
tracker sees no hand at all. ft-camwatch (unit file only, not enabled) follows
the log, notifies, and with CAMWATCH_AUTO_RESTART=1 restarts SteamVR when the
headset isn't worn and nothing else uses VR.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An X11 window in a Flatpak can give KWin an app id with no desktop file:
RustDesk's says com.carriez.flutter_hbb (its GTK application id), and the
Flatpak's desktop file is com.rustdesk.RustDesk. A profile recorded that
id, so it couldn't relaunch the app (PR #16 keeps that from killing
ft-floatd's socket). And the window's pid is the sandbox's own, so a
launched window matched neither by process nor by app id, and didn't
float.
desktop_name() finds the desktop file whose StartupWMClass names the
window's class (or app id) when the app id has none. Capture records
that name, a profile claims open windows by it, and a launch's window
matches by it.
Profiles saved before this keep the old id; save them again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The headset's right-side click button (KEY_SELECT on gpio-keys, read without
a grab) now works the session: Next while a step waits, pause during a hold,
resume while paused. With no mouse connected the hints lead with it.
The push sections say plainly to push straight out from the headset and pull
back, with a side-view picture of the head, the headset and the arrow, and
the bar's ends read "At your chest" and "Arm out". The bar labels are sent
as one field, so labels with spaces no longer split.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The first real session moved on every 5 s with text only, too fast to follow.
Each step now waits for Next (Space or the window's button), counts down 3-2-1
while recording, then holds. P pauses, R redoes a step, S skips a section;
"Advance by itself" (--auto) keeps the old timed flow. Nothing records while
a step waits: each step is its own recording part.
The panel and the window show a picture of each pose (hands/rec/poses,
generated by make_poses.py from a parametric hand, MIT) and a diagram of where
to hold the hands and how far out. prompts.jsonl gains ready, wait and redo
events; session.json gains mode.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
hands/rec/validate.py (standard library; Linux and Windows, Python 3.12+)
checks an export before upload and when it's received: SHA256SUMS, an
allow-list of files, the manifest's schema and keys, the consent version,
a uuid4 contributor, no identifying fields in calibration.json or
device.json, every sets.bin.zst decompressed to its end as a stream with
each FHSET01 header checked against the manifest, jsonl lines, the total
size. It decompresses with compression.zstd, zstandard or the zstd
program. validate.py DIR [--json].
hands/rec/hub.py uploads an export with huggingface_hub, as a pull
request to contributions/<contributor>/<session>: validate first, refuse
a repeat of the same export, check the login (whoami) and access
(auth_check), upload_folder(create_pr=True) with the manifest summary as
the description, then record the PR under "uploads" in session.json.
Errors are explained (terms not accepted, not found, 401/403, network).
--dry-run makes no network calls. FT_HANDREC_DATASET overrides
HF_DATASET (DeeJanuz/frametop-hands); while the texts are drafts a real
upload needs FT_HANDREC_ALLOW_UPLOAD=1.
The Upload page shows the login with "Check again" and how to run
hf auth login in a terminal (the token never enters the window), then
Upload with a phase, progress and Cancel (hub.py as a child process), the
PR link, and a warning for an export uploaded before. The manual
command stays as the fallback. ft-handrec --hub-dry-run.
session.py also saves device.json: cv.cad_from_cal and head from
/persist/device_config.json, the labeller's shape, nothing identifying;
export copies it. Session ids with a -N suffix are accepted everywhere.
hands/rec/rehearse.sh runs it all without the headset: ft-ringplay plays
30 s of a capture into a ring, session.py records a short test script
with ft-handpanel --no-vr and a tracker, then export, validate and a
dry-run upload (--repo ID uploads for real). It runs in one frame-job
scope, deletes its data and stops its processes, also on Ctrl+C.
hands/rec/tests/test_validate.py covers good and broken exports and hub.py
without the network. The dev container gains python3-huggingface-hub.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
