[Cin] Prepping HW upgrade options for AV1 de-/encoding

[Andrew Randrianasulu]

пт, 8 дек. 2023 г., 13:57 Andrew Randrianasulu <randrianasulu at gmail.com>:

>
>
> пт, 8 дек. 2023 г., 13:37 Terje J. Hanssen via Cin <
> cin at lists.cinelerra-gg.org>:
>
>>
>>
>> Den 03.12.2023 22:29, skrev Terje J. Hanssen:
>>
>> Already touched this topic barely in another thread
>> https://lists.cinelerra-gg.org/pipermail/cin/2023-December/007346.html
>>
>> But so many SW and HW pieces are mentioned around, it is almost a
>> full-time reading and study:
>> VAAPI, MESA, VULKAN, Intel Quick Sync Video etc......
>>
>> I realize my aging hardware which is fast enough for other tasks, needs
>> some "AV1 upgrade", if possible.
>>
>> But first I wonder, what is expected possible to do (obtain) with AV1
>> de-/encoding on my existing 64bit hardware:
>>
>> 1) laptop 2018: Dell XPS 13-9370: quad core i7-8550U CPU (8. gen
>> Kabylake) and Intel UHD Graphics
>>
>> 2) WS infinity:  MSI Z170A mobo:  quad core i7-6700K CPU (6. gen
>> Skylake), NVIDIA GeForce GT-730 graphics
>>
>>
>> A budget friendly first "AV1 HW upgrade" of the workstation 2) if
>> possible, would be to add a new GPU as Intel Arc A380.
>> But the question is if this will work at all on that much older (2015)
>> Skylake platform with i7-6700K CPU?
>> I've seen CPU bottlenecks has been mentioned and that Arc A380 is
>> targeted at newer generations CPU ...
>>
>>
>> Extracted from the first wikipedia reference below about Intel Alchemist
>> GPUs:
>>
>>    - Featuring 8 Xe-cores, the* A380 supports PCI Express 4.0* and has a
>>    total board power (TBP) of 75W. The graphics card is equipped with 6GB
>>    GDDR6 memory and a graphics memory interface of 96 bits, providing a memory
>>    bandwidth of 186GB/s.
>>    - Bus interface A380:         PCIe 4.0 x8 and for >=A580: PCIe 4.0
>>    x16
>>
>> That is, the keyword here seems to be PCIe 4.0 bus speed as a requirement
>> to utilize the Arc A380 GPU for HWA AV1 encoding (maybe also for other
>> GPUs?)
>>
>
> well, despite so much time spend looking at dev process for mesa3d I still
> do not know full details and media encoder process. But isn't it like
> putting uncompressed frame in vram (as long as you have enough of it - so
> probably n raw frames between keyframe ideally?) let media engine chw on
> it, pull resulting compressed bitstream out of vram via pci-express?
>
> So I speculate pcie bandwidth in itself will only matter if you compress
> both big frame size and long keyframes, so dma engine on card must
> constantly pump new raw frame data via bus.
>
> I saw some mention of big (resizeable) BAR as requirement for good
> performance, but opengl/vulkan IMO a bit different because they send often
> big amount of tiny objects (vertices) via bus for each frame. But may be
> default 256 Mb in size bar feels a bit small for sending like 1 second of
> 25 4k frames (300 mb/s)?
>

https://forums.gentoo.org/viewtopic-t-1165048.html?sid=1b9518b86afdcd1fa5e7710f16cd6893

not exactly about av1 in ffmpeg 6.1 but lists some components you need for
vaapi/qsv encoder on Arc 380 - also from dmesg resizeable bar support is
not essential ?



> gpus today use their own memory paging system, so may be this add
> additional restriction on how fast you can push frames to them. ....
>
> Also, windows and Linux drivers might differ ( I bet most reviews are from
> windows land).
>
> May be someone will post linux review of this particular aspect of Arc
> graphics, either video or text ...
>
> If this Suse Studio service still works may be you can compose your own
> live image with all components required to test that and walk to some
> offline place where you can testdrive new card .... { If weather feels
> favourable enough - we have around -14 C airtemp so I and my dog prefer
> short dashes around and back to warm place shortly}
>
>
> My existing Skylake WS 2) above has PCie 3.0 only.
>>
>> PCIe 4.0
>> https://no.wikipedia.org/wiki/PCI_Express
>> https://www.techreviewer.com/learn-about-tech/what-is-pcie-40/
>>
>> Motherboards Support PCIe 4.0
>> https://www.makeuseof.com/best-budget-pcie-4-motherboards/
>>
>> https://www.techreviewer.com/tech-answers/which-motherboards-support-pcie-40/
>>
>> Maybe also information of interest for WS building will be clarified
>> within a week, when Intel release their new "Meteor Lake, Core Ultra mobile
>> processors(?)
>> https://en.wikipedia.org/wiki/Meteor_Lake
>>
>>
>> References:
>> AV1 Encoding on a Budget: The Intel Arc A380 Approach
>>
>> https://medium.com/@contact_45426/av1-encoding-on-a-budget-the-intel-arc-a380-approach-d72367f2f349
>>
>> https://history-computer.com/intel-arc-a380-full-review-of-intels-entry-level-gpu/
>>
>> AV1 fixed-function hardware encoder is included in Alchemist GPUs as part
>> of the Intel Quick Sync Video core.
>> https://en.wikipedia.org/wiki/Intel_Arc#Alchemist
>> https://en.wikipedia.org/wiki/AV1#Hardware
>>
>> https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video#Hardware_decoding_and_encoding
>>
>>
>>
>>
>> --
>> Cin mailing list
>> Cin at lists.cinelerra-gg.org
>> https://lists.cinelerra-gg.org/mailman/listinfo/cin
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cinelerra-gg.org/pipermail/cin/attachments/20231208/80301f17/attachment.htm>