Direct RGBA receive: fixed HTTP send-buffer diagnostic

Configuration and method

NCPU is the parent native CPU diagnostic. NSB adds one fixed HTTP socket send-buffer request.

The native server requests 262,144 bytes after source and ticket authentication, before it submits response headers.

All eight NSB connections recorded a successful get, set and get sequence. Each read 146,988 bytes before the request and 262,144 bytes after it.

The request increased the observed value. The parent did not record that option, so its initial value is unknown.

Both arms use the same BUDGET2 browser and source code. The source limit is 16,588,928 bytes and receiver credit remains one.

Noisedeck A sends 1920 × 1080 RGBA packets through Sync to Noisedeck B. Native camera publication remains active. B uses the HTTP BYOB reader.

The native executables differ. The packet layout, write path, CPU recorder and deadline rules remain unchanged.

The tests use GLSL and synth.testPattern. Timing uses pattern 0 with gridSize 4. Pixel controls use pattern 5 with gridSize 16.

Retained host metadata identifies an Apple M2, eight logical CPUs, 24 GiB of memory and Darwin 23.6.0.

Observed power samples show AC power. They do not prove continuous AC power between reads. The tests collected no in-run thermal or frequency data.

Delivery observations

The report contains 16 valid diagnostic gates and one rejected parent attempt. Each arm has four pixel controls, two five-second observations and two three-minute observations.

A valid diagnostic does not imply a passed delivery checkpoint. All eight NSB diagnostic gates passed. Both long delivery checkpoints failed.

NSB web delivered 18.312851 unique pairs/s and desktop delivered 16.320944. Each had 174 complete seconds below 50.

The minimum complete second was five pairs on web and one on desktop. All low cells remain in the data.

The NCPU parent delivered 54.863649 pairs/s on web and 58.467740 on desktop. The measured spans are separate sequential runs.

The measured throughput does not meet release requirements for this fixed setting. This comparison does not establish exclusive causation.

All 720 complete long-run B seconds, 20 short-run seconds and eight partial tails remain available. Input and final markers must match in the same render.

Rates use the full B measurement span, including the partial tail. The diagrams use a shared 0–66 scale.

Native submission and completion

The submission elapsed interval surrounds the synchronous uv_write call. The outer CPU interval includes calling-thread CPU clock reads around that interval.

The CPU difference is not the exact CPU time of the inner write interval. The diagnostic subtracts no overhead.

The callback elapsed interval includes submission and continues to the actual callback. It includes work after the call returns, waiting and scheduling.

These intervals are not additive. Neither callback elapsed nor elapsed minus CPU identifies a network, GPU or scheduler cost.

NSB had shorter mean submission time and much longer mean callback elapsed time than the parent. The socket setting can defer work into later completion.

That mechanism does not establish an exclusive cause for the delivery difference. The report retains both native measurements and actual delivery.

All native values cover setup through cleanup. They have no measured alignment to either page clock.

Retained records and event loss

The separate CPU recorder retained all slow calls. The parent recorded seven web and two desktop calls. NSB recorded five web and two desktop calls.

The older six-stage ring retained 256 records per long run. It overwrote 785 parent web records and 92 parent desktop records.

It overwrote 5,845 of 6,101 NSB web records and 4,449 of 4,705 NSB desktop records. These counts include all six stages.

Aggregate stage counts, totals and maxima remain available. They cannot reconstruct lost individual callbacks.

The slow-record diagram shows complete CPU slow records and only retained slow callbacks. The callback sample is a tail, not a complete distribution.

Exact sequence, generation and native start timestamps can join retained write records. A missing match remains unknown.

Source intervals and diagnostic cost

Measured A source intervals recorded 541 parent web drops and 152 parent desktop drops. NSB recorded 3,213 web drops and 3,098 desktop drops.

The source recorder retains positive pressure checks and actual drops. It does not measure continuous socket occupancy.

Positive checks and their actual drops are separate events. Drop totals count only actual drop records.

All pressure phases, thresholds, raw points and startup history remain in the numeric data. Source intervals use their own A clocks.

The pressure recorder, page observers, journal writes and native CPU reads can affect execution. There is no uninstrumented comparison with identical scheduling conditions.

The CPU recorder retains 22,744 bytes. The report lists its elapsed clock-read brackets separately. They do not measure total diagnostic overhead.

Rejected parent attempt

NCPU web-observe5-ncpuv2-a1 stopped during measurement. Its source renderer was not running before cleanup. The stop caller was not recorded.

It has no accepted rate or accepted B cell journal. The unchanged five-second repeat passed and remains a separate result.

All actual owned processes exited. Source, pressure and cell journals are incomplete. A and B telemetry journals are complete.

The rejected native lifetime independently recorded 358 writes and one slow write. Its native evidence remains separate from delivery comparisons.

The slow frame sequence was 52 and the source start-cut sequence was 312. These values do not align native and page clocks.

NSB completed all eight gates on their first attempts. Low delivery rates did not cause those valid diagnostics to be discarded.

Limitations and references

GPU contention on the busy test machine is a working hypothesis for intermittent source-entry gaps. These tests did not measure GPU scheduling or contention.

This hypothesis is not an established explanation for the poor throughput of the socket arm.

These are sequential macOS measurements, not a cross-platform certification, a causal ranking or an hours-long performance claim.

The data does not establish an end-to-end latency change, a garbage collection cause or external camera delivery.

The earlier reports remain unchanged. Numeric data contains exact result, source, receipt and journal digests. Private logs and credentials remain outside this report.

Configuration and method
Arm | Context | Gate | Diagnostic | Native calls | Result SHA256
NCPU | web | positive | Pass | 577 | 238479620f001af4b2098bf9880fb3cb53df8a29c4b286a95174a5ad871817b2
NCPU | web | negative | Pass | 527 | 0b4c57e3f0b3d94b365a7a726a9273c3776a7a1492d7cc772732f7568564b220
NCPU | web | observe5 | Pass | 581 | 865997ff65101c84485c26340e1d40c0acaece6dfb6a932fd71a43f0924ec5b9
NCPU | web | observe180 | Pass | 10176 | 4dade0ebba66bfd51e69e3a7c4f4a2cc2513224be3a44321264ba093d884644c
NCPU | desktop | positive | Pass | 580 | e79194134af9a8631d4c1d844959647bc74a2cc45a0d15863368ce74fb5f868b
NCPU | desktop | negative | Pass | 582 | bb789d0969bacbc405dc187482352dc086dc1538fad7f3e92d5cd78d0e1a3d3c
NCPU | desktop | observe5 | Pass | 596 | 1527b56d0b8a8ac1a137c92982c40f66597f61c58c5394e213c321c514907f30
NCPU | desktop | observe180 | Pass | 10827 | 06eb8c2d7725d438d1cc1ae27f85d005c1f69883b6b732079d64d55410df5d84
NSB | web | positive | Pass | 456 | f5ce7a923088932f049f71ec26933840db93e207c3624aa4231e537ec43d749d
NSB | web | negative | Pass | 492 | 93a90b6cc6845e78d3e2d6682b6775ab5a54d210eabc0eb06e8499bd3c7877fd
NSB | web | observe5 | Pass | 89 | 279c14102477c1306fc8b54b9cc120a8c0bbd8a2b898bf99c2a078fc261dd33e
NSB | web | observe180 | Pass | 3390 | 9c75b712c6d40d58e41b8fecc1daf971d638146988c0b19fea5943fb2a38e73b
NSB | desktop | positive | Pass | 444 | 4fe2c03c2e3ae21fa880a1e95b7317527e0065645a882d7bea491b59e0f52803
NSB | desktop | negative | Pass | 418 | 4dce206fa9892bdf19e3b23a239b89b9228abdea06420260158c517f98381d11
NSB | desktop | observe5 | Pass | 582 | a66c481af1e4e68c4aa08af66a0b204a97b2a6abdd860a5c1ea699460321a0e8
NSB | desktop | observe180 | Pass | 3105 | 9d0f9fcb154ab8865ce432d25d6c9b2e97f4fed21afac2f4912fb6f7c23ec7d2
Configuration and method
NSB context | Gate | Request bytes | Before bytes | After bytes | Get status | Set status | Get-after status | Generation
web | positive | 262144 | 146988 | 262144 | 0 | 0 | 0 | 3445202994453119486
web | negative | 262144 | 146988 | 262144 | 0 | 0 | 0 | 15911658661492902753
web | observe5 | 262144 | 146988 | 262144 | 0 | 0 | 0 | 13117177670269749844
web | observe180 | 262144 | 146988 | 262144 | 0 | 0 | 0 | 10380259390602513665
desktop | positive | 262144 | 146988 | 262144 | 0 | 0 | 0 | 2608385300020657393
desktop | negative | 262144 | 146988 | 262144 | 0 | 0 | 0 | 9572658502625154086
desktop | observe5 | 262144 | 146988 | 262144 | 0 | 0 | 0 | 9750754913233795095
desktop | observe180 | 262144 | 146988 | 262144 | 0 | 0 | 0 | 14623024790689434535
Configuration and method
Arm | Context | A browser | B browser | Start UTC | End UTC | Observed AC samples
NCPU | web | Chrome/152.0.7977.83 | Chrome/152.0.7977.83 | 2026-09-09T09:33:51.967Z | 2026-09-09T09:37:21.155Z | 5
NCPU | desktop | Chrome/148.0.7778.180 | Chrome/148.0.7778.180 | 2026-09-09T09:37:21.389Z | 2026-09-09T09:40:54.268Z | 5
NSB | web | Chrome/152.0.7977.83 | Chrome/152.0.7977.83 | 2026-09-09T10:43:59.656Z | 2026-09-09T10:47:46.143Z | 5
NSB | desktop | Chrome/148.0.7778.180 | Chrome/148.0.7778.180 | 2026-09-09T10:47:46.515Z | 2026-09-09T10:51:30.683Z | 5
Delivery observations
Arm | Context | Run | Unique pairs | Elapsed ms | Pairs/s | Input min | Final min | Input below50 | Final below50 | Average50 checkpoint | All seconds50
NCPU | web | observe5 | 293 | 5024.799999237061 | 58.310779 | 56 | 56 | 0 | 0 | True | True
NCPU | web | observe180 | 9876 | 180009.90000152588 | 54.863649 | 32 | 32 | 41 | 41 | True | False
NCPU | desktop | observe5 | 295 | 5006.900001525879 | 58.918692 | 54 | 54 | 0 | 0 | True | True
NCPU | desktop | observe180 | 10525 | 180013.8000011444 | 58.467740 | 36 | 36 | 4 | 4 | True | False
NSB | web | observe5 | 41 | 5012.900001525879 | 8.178898 | 6 | 6 | 5 | 5 | False | False
NSB | web | observe180 | 3297 | 180037.5 | 18.312851 | 5 | 5 | 174 | 174 | False | False
NSB | desktop | observe5 | 295 | 5015.5 | 58.817665 | 56 | 56 | 0 | 0 | True | True
NSB | desktop | observe180 | 2938 | 180014.10000038147 | 16.320944 | 1 | 1 | 174 | 174 | False | False
Delivery observations
Arm | Context | Run | Tail ms | Input tail | Final tail
NCPU | web | observe5 | 24.799999237060547 | 2 | 2
NCPU | web | observe180 | 9.900001525878906 | 0 | 0
NCPU | desktop | observe5 | 6.900001525878906 | 1 | 1
NCPU | desktop | observe180 | 13.80000114440918 | 0 | 0
NSB | web | observe5 | 12.900001525878906 | 0 | 0
NSB | web | observe180 | 37.5 | 0 | 0
NSB | desktop | observe5 | 15.5 | 1 | 1
NSB | desktop | observe180 | 14.100000381469727 | 1 | 1
Native submission and completion
Arm | Context | Lifetime s | Calls | Mean submit ms | Mean outer CPU ms | Mean callback ms | Max submit ms | Outer CPU at max submit ms | Max callback ms
NCPU | web | 188.336557166 | 10176 | 4.922948 | 3.467390 | 5.049630 | 227.731292 | 4.019 | 227.751
NCPU | desktop | 188.203864333 | 10827 | 5.119771 | 3.496210 | 5.231919 | 42.512834 | 3.087167 | 42.519
NSB | web | 189.33030604200002 | 3390 | 1.370095 | 0.553356 | 38.587391 | 52.688542 | 0.571833 | 568.584
NSB | desktop | 189.426996708 | 3105 | 0.468272 | 0.178995 | 46.132891 | 56.977709 | 0.127375 | 768.76
Retained records and event loss
Arm | Context | All-stage slow | Retained | Overwritten | CPU slow | CPU overwritten
NCPU | web | 1041 | 256 | 785 | 7 | 0
NCPU | desktop | 348 | 256 | 92 | 2 | 0
NSB | web | 6101 | 256 | 5845 | 5 | 0
NSB | desktop | 4705 | 256 | 4449 | 2 | 0
Retained records and event loss
Arm | Context | Call index | Sequence | Native start s | Submit wall ms | Outer CPU ms | Retained matching callback ms
NCPU | web | 2232 | 2250 | 37.577803 | 227.731292 | 4.019000 | None
NCPU | web | 7018 | 7447 | 124.174078 | 26.480417 | 3.303459 | None
NCPU | web | 7068 | 7520 | 125.465042 | 29.938292 | 3.039208 | None
NCPU | web | 7123 | 7618 | 127.072926 | 42.202167 | 3.130333 | None
NCPU | web | 9231 | 10022 | 167.154725 | 31.472959 | 3.215208 | 73.369209
NCPU | web | 9410 | 10241 | 170.736279 | 45.271917 | 2.077417 | 45.280167
NCPU | web | 9543 | 10409 | 173.630586 | 29.772000 | 1.637250 | 29.781125
NCPU | desktop | 3752 | 3889 | 64.927500 | 42.512834 | 3.087167 | 42.519959
NCPU | desktop | 10058 | 10314 | 172.189918 | 29.963500 | 2.845750 | 29.971958
NSB | web | 85 | 284 | 4.832020 | 35.887250 | 0.118625 | None
NSB | web | 404 | 1340 | 22.455274 | 52.688542 | 0.571833 | None
NSB | web | 566 | 1855 | 31.093714 | 31.244125 | 0.385959 | None
NSB | web | 2365 | 7873 | 131.354754 | 29.193208 | 0.537125 | None
NSB | web | 2582 | 8865 | 147.875039 | 31.711750 | 0.716875 | None
NSB | desktop | 803 | 1854 | 31.001877 | 56.977709 | 0.127375 | None
NSB | desktop | 807 | 1875 | 31.363471 | 25.915292 | 0.199750 | None
Retained records and event loss
Arm | Context | Stage | Count | Total us | Maximum us
NCPU | web | ingestGap | 10742 | 188191659 | 257426
NCPU | web | tapCopy | 10563 | 10345078 | 24321
NCPU | web | cameraPublish | 10742 | 13230885 | 42708
NCPU | web | writeSubmit | 10176 | 50091044 | 227731
NCPU | web | writeCallback | 10176 | 51385038 | 227751
NCPU | web | ackCallback | 10175 | 178368796 | 255867
NCPU | desktop | ingestGap | 11134 | 188076839 | 190010
NCPU | desktop | tapCopy | 10952 | 8268334 | 46864
NCPU | desktop | cameraPublish | 11134 | 10687531 | 63846
NCPU | desktop | writeSubmit | 10827 | 55426550 | 42512
NCPU | desktop | writeCallback | 10827 | 56645982 | 42519
NCPU | desktop | ackCallback | 10826 | 180932771 | 173252
NSB | web | ingestGap | 7958 | 189078036 | 332145
NSB | web | tapCopy | 7771 | 16061726 | 85660
NSB | web | cameraPublish | 7958 | 21562445 | 122626
NSB | web | writeSubmit | 3390 | 4642998 | 52688
NSB | web | writeCallback | 3390 | 130811255 | 568584
NSB | web | ackCallback | 3389 | 183815010 | 580163
NSB | desktop | ingestGap | 8079 | 188991305 | 409191
NSB | desktop | tapCopy | 8004 | 16760574 | 194716
NSB | desktop | cameraPublish | 8079 | 20664080 | 183440
NSB | desktop | writeSubmit | 3105 | 1452476 | 56977
NSB | desktop | writeCallback | 3105 | 143242628 | 768760
NSB | desktop | ackCallback | 3104 | 184719394 | 869997
Source intervals and diagnostic cost
Arm | Context | A elapsed s | Offers | Accepted | Sent | Admission drops | Pre-encode drops | Post-encode drops | All pressure drops
NCPU | web | 180.08710000038147 | 10806 | 10508 | 10265 | 298 | 243 | 0 | 541
NCPU | desktop | 180.09129999923707 | 10806 | 10713 | 10654 | 93 | 59 | 0 | 152
NSB | web | 180.2821999988556 | 10817 | 8754 | 7604 | 2063 | 1150 | 0 | 3213
NSB | desktop | 180.1848000011444 | 10811 | 8567 | 7713 | 2244 | 854 | 0 | 3098
Source intervals and diagnostic cost
Arm | Context | Mean before us | Max before us | Mean after us | Max after us
NCPU | web | 1.997889 | 643.167 | 3.730281 | 2919.833
NCPU | desktop | 2.202276 | 528.208 | 3.638133 | 467.791
NSB | web | 2.440437 | 520.958 | 5.913816 | 3729.583
NSB | desktop | 4.974886 | 4247.541 | 3.880577 | 999.292
Rejected parent attempt
Attempt | Accepted rate | Native calls | Slow wall ms | Outer CPU ms | Sequence | Result SHA256
web-observe5-ncpuv2-a1 | None | 358 | 32.55175 | 3.65325 | 52 | ca5fce9a08deb791e63fac2c42c0e10caedeeb4ac622971e73d8df3929f5fa81