You are not logged in. Please login or register.
SmoothVideo Project → Posts by Chainik
> when I try to do this, access violation error appears and the player close
SVP menu -> Utilities -> Set environment variables for Vapoursynth
constasmile
> please try implement new ai scene detection
dunno, I don't see any real problem with the current simple and fast method
another neural network just for a scene change detection looks like an overkill to me
24 to 48 is _always_ 1n2n3n4n..., this is exactly why I said "not sure "1m"/"2m" do anything in 24->48 conversion"
24 to 60
"uniform": 1nnnn3nnnn5....
"2m": 11n22n33...
"1m": 1nn2n3nn4n...
something like this
"n" - interpolated frames, numbers - source frames
what exactly you'll get after enabling 1m/2m in any other XX fps to YY fps conversion - is a kind of mystery 
not sure "1m"/"2m" do anything in 24->48 conversion
> Can someone integrate the script in MPC Player to add MadVR in the pipeline ?
what do you mean? just run MPC-HC with 'Vapoursynth Filter' instead of 'Avisynth Filter'
it doesn't matter in this case that others have same video card
it's about system's memory bandwidth
because others have different hardware?
DragonicPrime
2 identical videos, 1 in SDR the other in HDR, the SDR 4k runs perfectly, HDR lags.
it's still 10-bit from the video decoder and to the point in Vapoursynth script where it got converted to 8-bit...
try MPC-HC + Vapoursynth Filter, and disable P010/P016 right in LAV Video decoder properties
aloola
btw:also maybe it a bug? every time black bar detection was on, the TRT CMD pop out again.
with black bars removed the frame size changes, obviously, and it'll need to generate cache for the new size
that's the problem, 3-5 mins waiting is not exactly a smooth experience
BTW it can generate a "multi-resolution" cache, for example one cache for "1920*1080 and below" but then the speed drops by ~30% 
ALL
FYI, 10-bit color depth needs _twice_ the memory bandwidth compared to 8-bit
HDR doesn't strictly require 10-bit processing, if you play HDR without "allow 10-bit" in SVP then nothing bad will happen
edVL
and you're still capped by the video encoder
> Grobalt said 1080p real-time should be ok on RTX 3070 laptop GPU
not sure what you're doing, but it's OK even on a 2060 laptop now
GOOD NEWS EVERYONE!
updated instructions
should improve FPS on 4080-and-better (probably 4070/3080 too, dunno), when performance is bound by the system's RAM bandwidth, not GPU power
i.e. for 4K playback
grobalt
> P.S. i ordered new hardware to test .. Ryzen 7700x, DDR5 7200 Memory etc.
cancel your order
don't do "fixed 60 fps" with RIFE
UHD
> We are now testing
you are not
if it creates cache files - then it's working
aloola
> I just want to report that I got 30fps on 4k transcoding
and you're capped by the encoder speed
try h264 instead
or set preset=fast in 'advanced' mode, cause it's set to "slow" by default
I suppose "two units" doesn't mean it can encode ONE stream using two units, it's more like it can encode several video streams at the same time.
for AV1 you can just replace SVP 4\mpv64\mpv.exe with the latest shinchiro build
https://sourceforge.net/projects/mpv-pl … les/64bit/
---> DON'T FORGET TO RESTART SVP after .js/.py file replacing, it may not reload these files automatically in some cases <---
> при 8к 60фпс
вот проблемы-то у людей...
а если будет работать 8к 60фпс, тогда будем жаловаться что 16к 120фпс плохо работает?
here's a full script you can use for testing
run with
> vspipe test.py .
variables:
- frame size on line 16, "length" is the number of source frames
- CPU threads on line 6
- GPU streams on line 27
- SVP's add. functions on/off - comment line 69
I thought the problem was in a CPU threads number, which was 4 before, and it's not enough to feed the GPU. Now it's 8, but you're saying there's no difference. Which is strange.
DON'T FORGET TO RESTART SVP after .js/.py file replacing, it may not reload these files automatically in some cases
---
BTW use_cuda_graph=True gives only like 2% difference in my case
try this
> 115 fps in 1080p.
just to be sure - what is the encoding speed with the generic SVP video profile?
I hope you're using H.264/NVENC, not "software" mode as on the screenshot above
> Based on the benchmark
benchmark is for 4.4 model, and I set 4.6 as a default one here
plus note the "GPU threads" option in the video profile, it directly translates into "TRT streams" count
> you need put the *.onnx files here C:\Program Files (x86)\SVP 4\rife\models\rife
yeah, my bad... fixed now
That one - https://github.com/AmusementClub/vs-mlrt - is MUUUUCH better
-------------
Post updates:
15/01: instructions removed, just install the "TensorRT" package via SVP's "Additional programs and features"
09/01: updated both libs and scripts, should improve 4K performance on 4080+
SmoothVideo Project → Posts by Chainik
Powered by PunBB, supported by Informer Technologies, Inc.