Video AI 6.1.X - User benchmarking Results

bbyers-0695 · March 30, 2025, 12:25am

Topaz Video AI  v6.1.3
System Information
OS: Windows v11.24
CPU: AMD Ryzen 7 9800X3D 8-Core Processor             31.151 GB
GPU: NVIDIA GeForce RTX 5080  15.517 GB
GPU: AMD Radeon(TM) Graphics  0.47438 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	27.54 fps 	2X: 	15.12 fps 	4X: 	03.88 fps 	
Iris		1X: 	28.47 fps 	2X: 	16.59 fps 	4X: 	04.61 fps 	
Proteus		1X: 	24.42 fps 	2X: 	16.77 fps 	4X: 	04.74 fps 	
Gaia		1X: 	11.96 fps 	2X: 	07.98 fps 	4X: 	04.46 fps 	
Nyx		1X: 	10.52 fps 	2X: 	08.62 fps 	
Nyx Fast		1X: 	19.90 fps 	
Rhea		4X: 	03.52 fps 	
RXL		4X: 	03.30 fps 	
Hyperion HDR		1X: 	32.46 fps 	
4X Slowmo		Apollo: 	35.31 fps 	APFast: 	71.70 fps 	Chronos: 	15.95 fps 	CHFast: 	30.87 fps 	
16X Slowmo		Aion: 	44.55 fps

Topaz Video AI  v6.1.3
System Information
OS: Windows v11.24
CPU: AMD Ryzen 7 9800X3D 8-Core Processor             31.151 GB
GPU: NVIDIA GeForce RTX 5080  15.517 GB
GPU: AMD Radeon(TM) Graphics  0.47438 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 720x480
Benchmark Results
Artemis		1X: 	143.24 fps 	2X: 	72.22 fps 	4X: 	19.31 fps 	
Iris		1X: 	130.91 fps 	2X: 	67.24 fps 	4X: 	27.90 fps 	
Proteus		1X: 	158.68 fps 	2X: 	74.76 fps 	4X: 	31.41 fps 	
Gaia		1X: 	65.37 fps 	2X: 	35.86 fps 	4X: 	22.89 fps 	
Nyx		1X: 	30.30 fps 	2X: 	29.04 fps 	
Nyx Fast		1X: 	78.08 fps 	
Rhea		4X: 	21.10 fps 	
RXL		4X: 	21.99 fps 	
Hyperion HDR		1X: 	100.85 fps 	
4X Slowmo		Apollo: 	187.70 fps 	APFast: 	371.39 fps 	Chronos: 	71.57 fps 	CHFast: 	136.62 fps 	
16X Slowmo		Aion: 	128.72 fps

memo90061 · April 1, 2025, 6:15am

This seems good for the price of the GPU!

brumult · April 1, 2025, 2:59pm

Yeah I’m very happy with it! I bet if Topaz spent some time they could get a little extra out of the GPU too.

memo90061 · April 1, 2025, 5:19pm

I will look into that GPU, but it seems like it’s out of stock everywhere.

ForSerious · April 1, 2025, 9:33pm

Yes indeed. I’m hoping to get one too, but it might be a couple of months before they come back in stock long enough to buy one.

wolfgang38 · April 2, 2025, 8:00pm

Topaz Video AI  v6.1.3
System Information
OS: Windows v11.23
CPU: AMD Ryzen 9 9950X3D 16-Core Processor            93.658 GB
GPU: AMD Radeon RX 9070 XT  15.349 GB
GPU: AMD Radeon(TM) Graphics  1.9646 GB
Processing Settings
device: -2 vram: 1 instances: 0
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	32.05 fps 	2X: 	12.72 fps 	4X: 	03.47 fps 	
Iris		1X: 	37.91 fps 	2X: 	17.33 fps 	4X: 	04.43 fps 	
Proteus		1X: 	35.03 fps 	2X: 	17.64 fps 	4X: 	04.45 fps 	
Gaia		1X: 	19.41 fps 	2X: 	11.63 fps 	4X: 	02.53 fps 	
Nyx		1X: 	14.79 fps 	2X: 	11.48 fps 	
Nyx Fast		1X: 	26.52 fps 	
Rhea		4X: 	03.02 fps 	
RXL		4X: 	02.42 fps 	
Hyperion HDR		1X: 	26.84 fps 	
4X Slowmo		Apollo: 	33.08 fps 	APFast: 	63.01 fps 	Chronos: 	23.20 fps 	CHFast: 	30.24 fps 	
16X Slowmo		Aion: 	44.28 fps

anton.y.kovalenko · April 4, 2025, 6:25pm

So yeah, TVAI most definitely needs DDR5 RAM. Made the cleanest comparison possible - upgraded to the very same MB but DDR5 variant, , all other components the same, and voila - almost 2x speed

Topaz Video AI  v6.1.3
System Information
OS: Windows v11.24
CPU: 13th Gen Intel(R) Core(TM) i9-13900KF  63.851 GB
GPU: AMD Radeon RX 9070 XT  15.374 GB
Processing Settings
device: -2 vram: 0.99 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	33.25 fps 	2X: 	16.57 fps 	4X: 	04.26 fps 	
Iris		1X: 	39.99 fps 	2X: 	20.55 fps 	4X: 	05.50 fps 	
Proteus		1X: 	34.46 fps 	2X: 	20.26 fps 	4X: 	05.64 fps 	
Gaia		1X: 	18.44 fps 	2X: 	11.27 fps 	4X: 	02.99 fps 	
Nyx		1X: 	15.42 fps 	2X: 	12.96 fps 	
Nyx Fast		1X: 	30.43 fps 	
Rhea		4X: 	02.56 fps 	
RXL		4X: 	02.24 fps 	
Hyperion HDR		1X: 	23.24 fps 	
4X Slowmo		Apollo: 	26.80 fps 	APFast: 	64.06 fps 	Chronos: 	23.33 fps 	CHFast: 	27.49 fps 	
16X Slowmo		Aion: 	45.16 fps

runnermx · April 5, 2025, 12:08am

Topaz Video AI  v6.1.3
System Information
OS: Mac v15.04
CPU: Apple M4 Max  64 GB
GPU: Apple M4 Max  48 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	13.64 fps 	2X: 	08.31 fps 	4X: 	02.77 fps 	
Iris		1X: 	13.62 fps 	2X: 	05.41 fps 	4X: 	01.82 fps 	
Proteus		1X: 	14.50 fps 	2X: 	07.55 fps 	4X: 	02.46 fps 	
Gaia		1X: 	04.11 fps 	2X: 	03.12 fps 	4X: 	02.12 fps 	
Nyx		1X: 	05.32 fps 	2X: 	05.30 fps 	
Nyx Fast		1X: 	12.39 fps 	
Rhea		4X: 	01.12 fps 	
RXL		4X: 	01.21 fps 	
Hyperion HDR		1X: 	59.71 fps 	
4X Slowmo		Apollo: 	16.11 fps 	APFast: 	37.44 fps 	Chronos: 	05.52 fps 	CHFast: 	08.82 fps 	
16X Slowmo		Aion: 	16.07 fps

mft · April 5, 2025, 1:18pm

upgrade from RTX 5080 to RTX 5090

Topaz Video AI  v6.1.3
System Information
OS: Windows v11.24
CPU: Intel(R) Core(TM) i9-14900K  63.691 GB
GPU: NVIDIA GeForce RTX 5090  31.349 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	37.86 fps 	2X: 	16.68 fps 	4X: 	04.72 fps 	
Iris		1X: 	33.23 fps 	2X: 	18.16 fps 	4X: 	05.44 fps 	
Proteus		1X: 	40.11 fps 	2X: 	19.94 fps 	4X: 	05.75 fps 	
Gaia		1X: 	18.06 fps 	2X: 	12.77 fps 	4X: 	05.37 fps 	
Nyx		1X: 	16.58 fps 	2X: 	11.93 fps 	
Nyx Fast		1X: 	30.72 fps 	
Rhea		4X: 	04.44 fps 	
RXL		4X: 	04.88 fps 	
Hyperion HDR		1X: 	30.72 fps 	
4X Slowmo		Apollo: 	36.01 fps 	APFast: 	79.26 fps 	Chronos: 	21.08 fps 	CHFast: 	39.69 fps 	
16X Slowmo		Aion: 	39.64 fps

memo90061 · April 7, 2025, 4:14am

Now I really want the 9070XT

vincentabel42 · April 7, 2025, 6:03am

Hmm… those poor Rhea’s numbers, though!

Timjitsu · April 8, 2025, 10:59pm

Topaz Video AI  v6.1.3
System Information
OS: Windows v10.22
CPU: Intel(R) Core(TM) i9-14900K  31.747 GB
GPU: NVIDIA GeForce RTX 5070  11.576 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	17.85 fps 	2X: 	11.27 fps 	4X: 	03.63 fps 	
Iris		1X: 	19.17 fps 	2X: 	10.88 fps 	4X: 	03.45 fps 	
Proteus		1X: 	19.09 fps 	2X: 	13.15 fps 	4X: 	04.29 fps 	
Gaia		1X: 	08.19 fps 	2X: 	05.46 fps 	4X: 	03.82 fps 	
Nyx		1X: 	07.15 fps 	2X: 	05.96 fps 	
Nyx Fast		1X: 	13.18 fps 	
Rhea		4X: 	02.58 fps 	
RXL		4X: 	02.34 fps 	
Hyperion HDR		1X: 	23.36 fps 	
4X Slowmo		Apollo: 	25.99 fps 	APFast: 	56.61 fps 	Chronos: 	11.16 fps 	CHFast: 	19.79 fps 	
16X Slowmo		Aion: 	44.07 fps

Timjitsu · April 8, 2025, 11:01pm

Holy crap I was not expecting the 9070xt to perform that well! That’s shocking!

juraj.grossmann · April 9, 2025, 10:00am

Topaz Video AI  v6.1.3
System Information
OS: Windows v11.24
CPU: 12th Gen Intel(R) Core(TM) i7-12700K  31.779 GB
GPU: NVIDIA GeForce RTX 3080  9.8174 GB
GPU: Intel(R) UHD Graphics 770  0.125 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	19.74 fps 	2X: 	10.26 fps 	4X: 	03.27 fps 	
Iris		1X: 	19.03 fps 	2X: 	11.94 fps 	4X: 	03.56 fps 	
Proteus		1X: 	19.21 fps 	2X: 	12.81 fps 	4X: 	03.80 fps 	
Gaia		1X: 	06.58 fps 	2X: 	04.62 fps 	4X: 	02.98 fps 	
Nyx		1X: 	08.40 fps 	2X: 	06.85 fps 	
Nyx Fast		1X: 	16.25 fps 	
Rhea		4X: 	02.65 fps 	
RXL		4X: 	02.49 fps 	
Hyperion HDR		1X: 	30.82 fps 	
4X Slowmo		Apollo: 	23.95 fps 	APFast: 	61.62 fps 	Chronos: 	14.78 fps 	CHFast: 	24.09 fps 	
16X Slowmo		Aion: 	37.86 fps

memo90061 · April 9, 2025, 4:56pm

Same! I really want to buy it now!

Timjitsu · April 9, 2025, 10:08pm

I wonder if those numbers are real. How is possible almost be on par with a 5090 that has double the vRam and is DDR7 instead of DDR6, and double the bandwidth

z1nonly · April 10, 2025, 1:38am

Topaz needs to “optimize” their software just for the 5000 series and their efforts can be seen in some of the beta threads. It does seem like it’s not easy to improve 5000 series performance while also maintaining previous generations’ performance.

AMD’s 9000 series seems to function well in TVAI without any specific software changes just for AMD’s new GPU’s.

kingdom-3809 · April 10, 2025, 9:30am

It would be extremely important that the respective processor load (CPU) is also recorded in the benchmark.
This way you can see whether the CPU or the GPU / GPUs are the bottleneck.

topaz257 · April 10, 2025, 9:47am

Good idea. Although RAM speed influence can’t be properly listed. In 2X and 4X both GPU and CPU will be underutilized.

As a general rule. Everything with scaling 2X, 4X is limited by RAM/CPU speed (except when GPU is very underpowered compared to RAM/CPU).

Yuri · April 10, 2025, 5:29pm

Topaz Video AI  v6.1.3
System Information
OS: Windows v10.22
CPU: AMD Ryzen 7 6800H with Radeon Graphics           63.192 GB
GPU: NVIDIA GeForce RTX 3060 Laptop GPU  5.8555 GB
GPU: AMD Radeon(TM) Graphics  0.4744 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis		1X: 	09.35 fps 	2X: 	07.07 fps 	4X: 	02.40 fps 	
Iris		1X: 	08.73 fps 	2X: 	05.78 fps 	4X: 	01.81 fps 	
Proteus		1X: 	08.92 fps 	2X: 	06.82 fps 	4X: 	02.45 fps 	
Gaia		1X: 	02.95 fps 	2X: 	02.04 fps 	4X: 	01.40 fps 	
Nyx		1X: 	03.64 fps 	2X: 	03.22 fps 	
Nyx Fast		1X: 	07.36 fps 	
Rhea		4X: 	01.24 fps 	
RXL		4X: 	01.15 fps 	
Hyperion HDR		1X: 	23.79 fps 	
4X Slowmo		Apollo: 	13.11 fps 	APFast: 	34.59 fps 	Chronos: 	07.26 fps 	CHFast: 	11.94 fps 	
16X Slowmo		Aion: 	19.35 fps