Agreed.
Is this a fresh install of Windows?
What is your RAM frequency?
How do your results compare to other 5950x / 4090 setups?
You are likely CPU bound btw.
I am not CPU bound, and my render rate seems nearly 40% slower than other 5950x / 4090 setups. And yes, I have DOCP (XMP) enabled.
@20rushtonj results, same CPU + GPU. Where exactly are you 40% slower as claimed?
Topaz Video AI v4.0.0
System Information
OS: Windows v11.22
CPU: AMD Ryzen 9 5950X 16-Core Processor 63.928 GB
GPU: NVIDIA GeForce RTX 4090 23.59 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 31.33 fps 2X: 12.31 fps 4X: 02.88 fps
Iris 1X: 35.17 fps 2X: 13.61 fps 4X: 03.59 fps
Proteus 1X: 20.55 fps 2X: 08.78 fps 4X: 02.51 fps
Gaia 1X: 15.55 fps 2X: 10.35 fps 4X: 03.30 fps
Nyx 1X: 17.50 fps
4X Slowmo Apollo: 35.95 fps APFast: 60.74 fps Chronos: 30.44 fps CHFast: 30.35 fps
Topaz Video AI v4.0.8
System Information
OS: Windows v11.23
CPU: Intel(R) Xeon(R) w5-2465X 127.25 GB
GPU: NVIDIA GeForce RTX 4070 11.744 GB
GPU: NVIDIA RTX A4000 15.804 GB
Processing Settings
device: 0 vram: 1 instances: 0
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 18.34 fps 2X: 12.70 fps 4X: 03.93 fps
Iris 1X: 18.26 fps 2X: 10.16 fps 4X: 03.34 fps
Proteus 1X: 17.06 fps 2X: 11.96 fps 4X: 03.75 fps
Gaia 1X: 06.36 fps 2X: 04.39 fps 4X: 02.99 fps
Nyx 1X: 07.44 fps 2X: 06.14 fps
4X Slowmo Apollo: 27.48 fps APFast: 55.47 fps Chronos: 13.31 fps CHFast: 20.84 fps
```RTX 4070 75%TDPThen care to explain why my numbers are so much higher with a faster CPU and the same GPU?
UEFI settings / Overheating bc of old thermal paste and so on.
PCI-E link Speed.
Windows power settings.
Topaz Video AI v4.0.8
System Information
OS: Windows v11.23
CPU: 12th Gen Intel(R) Core(TM) i5-12500 63.745 GB
GPU: NVIDIA GeForce RTX 4060 7.7842 GB
Processing Settings
device: 0 vram: 0.95 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 10.21 fps 2X: 07.26 fps 4X: 02.33 fps
Iris 1X: 11.08 fps 2X: 05.81 fps 4X: 01.77 fps
Proteus 1X: 09.62 fps 2X: 06.90 fps 4X: 02.43 fps
Gaia 1X: 03.44 fps 2X: 02.32 fps 4X: 01.56 fps
Nyx 1X: 04.12 fps 2X: 03.55 fps
4X Slowmo Apollo: 15.68 fps APFast: 40.31 fps Chronos: 08.18 fps CHFast: 12.60 fps
Or maybe CPU speed actually matters. Anyway I don’t know why I even responded to them - going forward I’m not going to bother if their account doesn’t show they own the product.
Topaz Video AI v4.0.7
System Information
OS: Mac v14.0201
CPU: Apple M1 Ultra 64 GB
GPU: Apple M1 Ultra 48 GB
Processing Settings
device: 0 vram: 0.5 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 13.33 fps 2X: 09.10 fps 4X: 02.80 fps
Iris 1X: 11.77 fps 2X: 04.16 fps 4X: 01.97 fps
Proteus 1X: 13.21 fps 2X: 08.39 fps 4X: 02.68 fps
Gaia 1X: 04.18 fps 2X: 03.07 fps 4X: 02.38 fps
Nyx 1X: 04.97 fps 2X: 04.51 fps
4X Slowmo Apollo: 14.83 fps APFast: 55.40 fps Chronos: 04.45 fps CHFast: 06.64 fps
My system has the same CPU and a slower GPU. His numbers do seem low. (Although not 40%) My guess is slow RAM or possibly temperature issues with the CPU.
My other system has a slower CPU (slower for productivity) with a 4090 and also does better:
2 Likes
iMac 2017 Quadcore Ventura
Topaz Video AI v4.0.8
System Information
OS: Mac v13.0601
CPU: Intel(R) Core(TM) i5-7600 CPU @ 3.50GHz 16 GB
GPU: AMD Radeon Pro 575 4 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 01.41 fps 2X: 00.93 fps 4X: 00.37 fps
Iris 1X: 00.96 fps 2X: 00.57 fps 4X: 00.19 fps
Proteus 1X: 01.41 fps 2X: 00.97 fps 4X: 00.43 fps
Gaia 1X: 00.47 fps 2X: 00.33 fps 4X: 00.25 fps
Nyx 1X: 00.62 fps 2X: 00.57 fps
4X Slowmo Apollo: 01.24 fps APFast: 04.91 fps Chronos: 00.27 fps CHFast: 00.56 fps
iMac 2017
Topaz Video AI v4.0.8
System Information
OS: Mac v13.0601
CPU: Intel(R) Core(TM) i5-7600 CPU @ 3.50GHz 16 GB
GPU: AMD Radeon Pro 575 4 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 720x480
Benchmark Results
Artemis 1X: 08.06 fps 2X: 05.48 fps 4X: 02.39 fps
Iris 1X: 06.42 fps 2X: 03.99 fps 4X: 01.10 fps
Proteus 1X: 08.60 fps 2X: 05.18 fps 4X: 01.78 fps
Gaia 1X: 02.87 fps 2X: 02.04 fps 4X: 01.59 fps
Nyx 1X: 02.32 fps 2X: 02.26 fps
4X Slowmo Apollo: 07.37 fps APFast: 26.42 fps Chronos: 01.92 fps CHFast: 03.56 fps
Topaz Video AI v4.0.8
System Information
OS: Windows v11.23
CPU: AMD Ryzen 9 5900X 12-Core Processor 31.909 GB
GPU: NVIDIA GeForce RTX 4070 11.743 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 17.30 fps 2X: 09.04 fps 4X: 02.53 fps
Iris 1X: 18.14 fps 2X: 09.13 fps 4X: 02.60 fps
Proteus 1X: 15.13 fps 2X: 08.53 fps 4X: 02.61 fps
Gaia 1X: 06.19 fps 2X: 04.22 fps 4X: 02.63 fps
Nyx 1X: 07.33 fps 2X: 06.17 fps
4X Slowmo Apollo: 26.83 fps APFast: 44.30 fps Chronos: 13.52 fps CHFast: 21.32 fps
Topaz Video AI v4.0.8
System Information
OS: Windows v11.23
CPU: AMD Ryzen 9 5900X 12-Core Processor 31.909 GB
GPU: NVIDIA GeForce RTX 4070 11.743 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 3840x2160
Benchmark Results
Artemis 1X: 03.89 fps 2X: 02.45 fps 4X: 00.51 fps
Iris 1X: 04.14 fps 2X: 01.97 fps 4X: 00.66 fps
Proteus 1X: 03.58 fps 2X: 02.30 fps 4X: 00.51 fps
Gaia 1X: 01.36 fps 2X: 00.93 fps 4X: 00.60 fps
Nyx 1X: 01.24 fps 2X: 01.59 fps
4X Slowmo Apollo: 06.99 fps APFast: 14.91 fps Chronos: 02.92 fps CHFast: 05.78 fps
Funny you should say that, as my results with an rtx 4080 and R7 5800x have improved since I removed the case side and now it’s close to or even better than some of the 4090 results posted for Artemis, Iris and some of the slomos. Video AI 4.0.X - User Benchmarking Results - #196 by DaveL
This is in spite of many internet claims that you should always keep your case sides on. Well clearly, that depends! Cheaper for me than upgrading case fans and cooling (and my Gigabyte 4080 is almost silent). But my results are nowhere near the faster 4090 results using much faster CPUs, presumably better cooling and faster memory.
So as well as CPUs, GPUs and memory speed (not forgetting to set the 2x XMP profile!), it’s about cooling too IMO - ESPECIALLY for folks with a 4090.
One of my workstations is a typical big, open box with fans in the front and vents in the rear. Removing the cover makes everything run cooler.
The other has two inner panels inside the cover that duct air from the front fans so that each of the two CPUs gets its own independent cool air inlet and the CPU, RAM and GPU sections of the MB are isolated from one another. Remove these inner panels, and all the temps go up even with the cover off.
Topaz Video AI v4.0.8
System Information
OS: Windows v10.22
CPU: AMD Ryzen 5 2600 Six-Core Processor 15.927 GB
GPU: NVIDIA GeForce RTX 2060 SUPER 7.8311 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 09.94 fps 2X: 07.36 fps 4X: 02.11 fps
Iris 1X: 10.22 fps 2X: 05.72 fps 4X: 01.84 fps
Proteus 1X: 09.68 fps 2X: 07.07 fps 4X: 02.08 fps
Gaia 1X: 03.06 fps 2X: 02.04 fps 4X: 01.41 fps
Nyx 1X: 03.73 fps 2X: 03.35 fps
4X Slowmo Apollo: 04.29 fps APFast: 38.38 fps Chronos: 07.25 fps CHFast: 11.54 fps
Topaz Video AI v4.0.8
System Information
OS: Windows v11.23
CPU: AMD Ryzen 7 7800X3D 8-Core Processor 31.713 GB
GPU: NVIDIA GeForce RTX 4070 11.744 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 18.81 fps 2X: 13.30 fps 4X: 04.00 fps
Iris 1X: 20.21 fps 2X: 10.85 fps 4X: 03.40 fps
Proteus 1X: 17.73 fps 2X: 12.44 fps 4X: 04.25 fps
Gaia 1X: 06.35 fps 2X: 04.43 fps 4X: 03.01 fps
Nyx 1X: 07.53 fps 2X: 06.50 fps
4X Slowmo Apollo: 28.25 fps APFast: 67.88 fps Chronos: 13.83 fps CHFast: 22.33 fps
What do you think? are they in line with the performance the program should have with my PC components?
1 Like
Looks to be similar in results to my RTX 3080 system (albeit it has a 5950X Ryzen 9). So I think this is normal since the 4070 is similar in performance to a 3080. You can check my result @ Reply #175.
1 Like
Thank you for your post here. I was wondering how 7800X3D run with 4070 and it is the same as my old 5900X, so no need to upgarde for me 
Mind that 5900X is 12-core CPU so 7800X3D does pretty good job.
