Yup, it usually is. Great job, btw!
1 Like
Intel will not call already sold CPUs for RMA.
Wow! Iām sorry to know that, Imo.
phew I got dehydrated just thinking about those thermals. 
j/k
Wow! Interesting. That does seem weird.
Indeed, I returned to 5.0.4, because of enable to adress the issue 
The load is absolutely not what advertised is but I will give a try with Nyx and exact Topaz conditions.
Here I am on Proteus(x4) + Artemis.
Thereās a bottleneck, but where ā¦ I would be very very happy thatās on my side, but did so much , os so many platform, GPUs and mobos tests, always the same issue.
So we are very far from allmost perfect scalability
5.04 was less prone to crashing on my system with multi-GPU, but my multi-GPU processing speed was actually slower than it was with a single GPU.
Iāve got the same problem here 1 RTX 4090 >> 1 RTX 4090 + 2 RTX 3070 ā¦
Multi-GPU usually doesnāt work well unless the processors are equivalent in speed. Mismatched processors usually produce a combined speed somewhere between the fastest and slowest processor. Which pretty much negates the whole point of multi-GPU.
With a well handled queue this shouldānt occur, moreover using tiles or GOP,
Above is 4x MSI RTX 3070 GAMING Z TRIO LHR
One of it runs well, 99% load
Two of it, problems start. One is allmost fully loaded the second jerky
Three to four, they run at 30-60% very jerky, unable to get a full load.
It was said elsewhere the CPU is unable to feed them. Well ā¦ if so the CPU should be @ 100% load and agonizing, it is not the case, whatever the CPU I use 9950X / 2990WX which are not on the paper pure bullshit.
Found it back
8K input ā¦ ok ā¦ building an 8K sample test file with massive noise
Got it ! It FULLY depends on the model 
Gaia full 100% everywhere (the useless model 
)
Proteus : no 100%, Artemis no 100%, so going to test it all and publish true results
I edited my previous post with a long term estimation, not the 5 first minutes
EDIT : Chances that advert is TRUE, got a 100% mega full load over all GPU, CPU is sleeping or almost, so Nyx IS multi optimized, others ā¦ benchs soon
So nothing to hunt on my hardware, runs fine, it is mosly very model dependant. and scale you use it up.
I missed this detail when I first replied, but CEP doesnāt like it when AC and DC LLC are not 1 to 1. Before 0x129, my Asus board was defaulting to 0.4 and 1.1. I had to manually match them. 0x129 defalts to 0.7 something for both. (On my board at least)
Main bottleneck when doing any CPU related task (e.g. upscaling) is usually the RAM speed. Run for example Proteus with manual settings and no scaling and u will get much better GPU utilization compared when using auto-values and upscaling.
I never use automatic values.
But you are right about one thing, scaling SHOULD NOT be CPU-specific, and that seems to be the case.
I donāt know if bilinear or bicubic is used, but GPUs should handle it like at least 100x faster for the same quality.
My RAM is top on the market, did not find better for Zen5 arch. 30-36-36-72 @ 6000MT/s
84K read / 81K write / 78K copy, isnāt enough ?!? Or the algo. is crap ā¦
There is just no CPU+RAM combination which can upscale the image fast enough for a 4090 it seems. The 4090 can process a single frame just faster than any CPU RAM combo out there can upscale that frame. At least not with the current code TVAI is using for upscaling. Maybe better utilization of many CPU cores by upscaling multiple frames in parallel or sth similar could help.
1 Like
:S You can easily pipe a Tensor job and whatever in anotehr side.
Also, as said a frame upscaled + AI rendered will be much more fatser than giving job to CPU.
The VRAM is vastly unused, so no excuse for not optimizing and using it.
If not perfect at least DirectStroage exists, it is allready used in supercomputing, so why not here in a kick ass brand famous AI upscaler ?
So with such latency I can get the code is DirectML, if CUDA we wouldnāt get this
Started to show artifacting after the recent update, can anybody help?
Topaz Video AI v5.3.1
System Information
OS: Windows v10.22
CPU: Intel(R) Coreā¢ i9-7900X CPU @ 3.30GHz 63.702 GB
GPU: NVIDIA GeForce GTX 1080 Ti 10.846 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 04.37 fps 2X: 03.02 fps 4X: 01.24 fps
Iris 1X: 04.48 fps 2X: 02.54 fps 4X: 00.93 fps
Proteus 1X: 04.08 fps 2X: 02.72 fps 4X: 01.30 fps
Gaia 1X: 01.63 fps 2X: 01.10 fps 4X: ERR fps
Nyx 1X: 01.95 fps 2X: 01.67 fps
Nyx Fast 1X: 04.38 fps
Rhea 4X: ERR fps
4X Slowmo Apollo: 05.74 fps APFast: 22.28 fps Chronos: 02.67 fps CHFast: 04.27 fps
16X Slowmo Aion: 08.92 fps
supposedly this artifacting is from GPU failure, can anyone recommend an ideal replacement for my sytem to efficiently run Topaz Video AI that is compatible with my system? I have no idea what type of card I should be looking for an what is compatible with my system. I have an X299 Designaire motherboard and an Intel(R) Coreā¢ i9-7900X CPU @ 3.30GHz 63.702 GB
4070 Super or, if your budget allows, a 4070-Ti-Super with 16gb of Vram.
Thank you for the recommendation. Does it make sense to upgrade my GPU to the 4070 Super or Ti if my intel processor / motherboard are still from 2018? I donāt want to abandon my system entirely as it was fairly expensive and Iām not ready to write it off yet. Would this be a worthy upgrade or am I putting too nice of a GPU in too old of a system?
Whether u can actually utilize the power of a 4070 depends a lot on your workload. If u upscale x2 or especiall x4, a 4070 will most likely be under utilized because the CPU+RAM are too slow to upscale the image for it.
So for more specific recommendations, it would be interesting to know what your typical processing tasks are.
Iām a doc filmmaker, and I use it professionally to clean up archival footage. Most work is 1080p enhancement with Iris, though I do occasionally upres to 4K for individual shots for the purposes of reframing. Is the 4070 compatible with my motherboard?
if its under utilized, does that just mean im throwing money away, or does it mean I can get away with using a lesser card? I had no issues with my previous GeForce 1080 ti SC2 (until the recent artifacting), the render times never bothered me because I usually handle those tasks overnight anyways. I guess iām wondering if Topaz Video AI burned out by GPU by doing overnight renders, or if its just an age issue and not the fault of enhancement software.
Youād be fine with an RTX 3060 TI.
1 Like
