Page 5 of 15

Re: CUDASynth

Posted: Fri Dec 14, 2018 6:40 am
by Guest 2
admin wrote:
Fri Dec 14, 2018 6:32 am
I can't read Japanese so can't make much out of it.
It's a fork of AviSynth+ with CUDA support. I read it thru Google Translate service.

Re: CUDASynth

Posted: Fri Dec 14, 2018 11:02 am
by admin
OK, but what does "with CUDA support" actually mean?

Re: CUDASynth

Posted: Wed Mar 06, 2019 2:31 am
by hydra3333
OK, now I'm an old guy who is officially lost and asking for directions.

I'm looking for latest CUDAsynth downloads and I guess AvsCompat.dll as well, per these,
viewtopic.php?f=14&t=506&p=9119&hilit=AvsCompat#p9119
gonca wrote:
Thu Nov 08, 2018 4:49 am
hydra3333 wrote:
Wed Nov 07, 2018 11:30 pm
I seem to recall something about a custom avscompat.dll for vapoursynth ... is one needed ?
For the normal DGIndexNV and DGDecodeNV no
For the CudaSynth versions one will be needed
admin wrote:
Wed Oct 17, 2018 4:43 pm
Meanwhile, on the CUDASynth front, I have discovered some very interesting things focused around what Prefetch() does in Avisynth+, and what the PREFETCH macro does in Vapoursynth, and how they interact with CUDASynth. It opens a possible avenue for further performance gains. I'll write a post about it when I get an energy burst.
A (vapoursynth) patch would likely not be accepted as similar ideas have been rejected. No problem; it's easy to replace AvisynthCompat.dll.
This is the latest stuff I could find from circa 9/10/2018:
viewtopic.php?f=14&t=671&hilit=AvsCompat&start=60#p9016
http://rationalqm.us/misc/CUDASynth_0.2.rar
which had it's own (older?) DLL versions
DGDecodeNV
DGHDRtoSDR
DGPQtoHLG
DGSharpen

Other downloads ?which seem to not be "CUDASynth enabled"? include
dgdecnv2053.zip
http://rationalqm.us/hdr/DGHDRtoSDR_1.11.rar
http://rationalqm.us/misc/DGPQtoHLG_1.0.rar

Hence, may I enquire about the latest status and plans ?

Re: CUDASynth

Posted: Wed Mar 06, 2019 8:30 am
by admin
Thank you for your interest. I'll come back to this after dealing with a couple issues I have in DGDecNV.

Re: CUDASynth

Posted: Wed Mar 06, 2019 12:40 pm
by hydra3333
Sounds like a plan, first things first.

Like, I thought I'd retired but now I'm back at work for a while ... one does what one has to ;) :? :wow: :o :shock: :facepalm:

Re: CUDASynth

Posted: Wed Mar 06, 2019 12:47 pm
by admin
hydra3333 wrote:
Wed Mar 06, 2019 12:40 pm
one does what one has to
Definition of a responsible person. Congratulations!

Re: CUDASynth

Posted: Wed May 15, 2019 10:56 pm
by hydra3333
:) Just wondering, is there perhaps enough customer interest in CUDAsynth to consider taking it further ?

Re: CUDASynth

Posted: Thu May 16, 2019 5:29 am
by Guest
hydra3333 wrote:
Wed May 15, 2019 10:56 pm
:) Just wondering, is there perhaps enough customer interest in CUDAsynth to consider taking it further ?
+1

Re: CUDASynth

Posted: Thu May 16, 2019 8:48 pm
by admin
"customer"? Don't you mean "donor"? ;)

Not too much traction for the idea, I'm afraid. Nobody cares about performance anymore?

I'll come back to it, but right now I have a physics paper in the works.

Speaking of traction, see here. I've been cited in an influential place. Arthur Fine is a monumental figure in quantum foundations.

https://plato.stanford.edu/entries/qt-epr/

Re: CUDASynth

Posted: Thu May 16, 2019 10:04 pm
by admin
Bullwinkle fell asleep. Everyone needs a rest. Wheee!

Re: CUDASynth

Posted: Tue May 21, 2019 10:00 pm
by hydra3333
Graft, D. A., 2016, “ Clauser-Horne/Eberhard inequality violation by a local model”, Advanced Science, Engineering and Medicine, 8: 496–502.
Nice work, Sir.
(I cited without reading, on the basis of not smart enough to comprehend it :))

Re: CUDASynth

Posted: Tue May 21, 2019 10:49 pm
by admin
I'm sure you're smart enough; it's just a matter of studying the field first to avoid common errors.

That paper is probably my favorite. The idea hit me in the middle of coaching a swimming workout, and I developed the math on the back of the whiteboard while people were swimming a set. They asked me why I was suddenly in such a good mood! Just refuted a large number of experiments purporting to prove quantum nonlocality.

I'm going to do something about TrueHD demuxing then I will come back to CUDASynth. I can do it in parallel with my current physics paper. It's about EPR steering and Luders projection. Whee!

Re: CUDASynth

Posted: Sat Aug 03, 2019 12:55 pm
by admin
I'm back to CUDASynth for a while. I had to update everything because the CUDASynth DGDecodeNV was at 253.0.0.158 and current DGDecNV is 2053.0.0.179. Note that the DGHDRtoSDR used below is 1.11 and needs to be upgraded to 1.12.

Here is a test script:

loadplugin("D:\Don\Programming\C++\Avisynth filters\CUDASynth\DGDecodeNV\x64\release\dgdecodenv.dll")
loadplugin("d:\don\Programming\C++\avisynth filters\CUDASynth\DGHDRtoSDR\x64\release\dghdrtosdr.dll")
dgsource("LG Chess 4K Demo.dgi",fulldepth=true,fdst="gpu0")
dghdrtosdr(impl="255",light=250,fsrc="gpu0",fdst="gpu1",fulldepth=true)
dgdenoise(fsrc="gpu1",fdst="gpu0",chroma=true)
dgsharpen(fsrc="gpu0")
trim(0,999)

For non-CUDASynth operation all fsrc and fdst are replaced with "cpu".

Here are the test results showing a very healthy FPS improvement of x3.6 , enough to make the difference between non-real-time and real-time playback:

-----
D:\Don\Programming\C++\Avisynth filters\CUDASynth\CUDASynth Test 1>avsmeter64 "LG Chess 4K Demo - No CudaSynth.avs"

AviSynth+ 0.1 (r2728, MT, x86_64) (0.1.0.0)

Number of frames: 1000
Length (hh:mm:ss.ms): 00:00:16.683
Frame width: 3840
Frame height: 2160
Framerate: 59.940 (60000/1001)
Colorspace: YUV420P16

Frames processed: 1000 (0 - 999)
FPS (min | max | average): 8.496 | 30.45 | 27.05
Memory usage (phys | virt): 316 | 1362 MiB
Thread count: 19
CPU usage (average): 12%

Time (elapsed): 00:00:36.974

D:\Don\Programming\C++\Avisynth filters\CUDASynth\CUDASynth Test 1>avsmeter64 "LG Chess 4K Demo.avs"

AviSynth+ 0.1 (r2728, MT, x86_64) (0.1.0.0)

Number of frames: 1000
Length (hh:mm:ss.ms): 00:00:16.683
Frame width: 3840
Frame height: 2160
Framerate: 59.940 (60000/1001)
Colorspace: YUV420P16

Frames processed: 1000 (0 - 999)
FPS (min | max | average): 54.09 | 103.0 | 97.40
Memory usage (phys | virt): 316 | 1356 MiB
Thread count: 19
CPU usage (average): 12%

Time (elapsed): 00:00:10.267
-----

Re: CUDASynth

Posted: Sat Aug 03, 2019 3:12 pm
by Guest
Do you still have your 1080ti lying around?
Would be interesting to see two card performance if possible

Re: CUDASynth

Posted: Sun Aug 04, 2019 5:53 am
by admin
I don't have a system that properly supports two cards.

Re: CUDASynth

Posted: Sun Aug 04, 2019 6:32 am
by Guest
It was just an idea.
I am sure the average user doesn't have two cards installed either

Re: CUDASynth

Posted: Sun Aug 04, 2019 10:40 am
by admin
Need all that extra power and power cables, extra PCIe x16 slot, and extra lanes on the CPU. I hope my next system will have all that. Anyway, expect about a x1.5 to x2.0 boost. ;)

Re: CUDASynth

Posted: Sun Aug 04, 2019 10:57 am
by Guest
I am sure the average user doesn't have two cards installed either
I am a pretty average user, don't have two cards and don't know if my system could handle it for the reasons you mentioned
Just curiosity

Re: CUDASynth

Posted: Sun Aug 04, 2019 11:06 am
by admin
Curiosity killed the cat. :lol:

Re: CUDASynth

Posted: Wed Aug 07, 2019 2:06 am
by Guest 2
I'd really love to see MVTools2 and KNLMeans ported to CudaSynth :bow:

(My mostly used filter is SMDegrain ;) )

Re: CUDASynth

Posted: Wed Aug 07, 2019 5:12 am
by admin
DGDenoise is my CUDASynth version of KNLMeans. It's included in DGDecodeNV.dll (standard and CUDASynth versions). Have you looked at it? It's already CUDASynth enabled. I will release the latest CUDASynth today.

Re: CUDASynth

Posted: Wed Aug 07, 2019 8:29 am
by admin
CUDASynth 0.3:

http://rationalqm.us/misc/CUDASynth_0.3.rar

Make your DGI files with DGIndexNV 2053.0.0.179.

Re: CUDASynth

Posted: Wed Aug 07, 2019 9:28 am
by hydra3333
:bravo:

Thank you ! :D

Re: CUDASynth

Posted: Wed Aug 07, 2019 12:14 pm
by Guest 2
admin wrote:
Wed Aug 07, 2019 5:12 am
DGDenoise is my CUDASynth version of KNLMeans. It's included in DGDecodeNV.dll (standard and CUDASynth versions). Have you looked at it?
Yes, thanks and it has great results when used on its own. :salute:

I have tried to use it as prefilter for SMDegrain but I can't replicate the same results of KNLMeans as the latest is more "integrated" in the script and has some optimizations.

Re: CUDASynth

Posted: Wed Aug 07, 2019 12:27 pm
by admin
Guest 2 wrote:
Wed Aug 07, 2019 12:14 pm
I have tried to use it as prefilter for SMDegrain but I can't replicate the same results of KNLMeans as the latest is more "integrated" in the script and has some optimizations.
Lack of relevant explanation and details will guarantee that you are ignored. Tough love!