Topaz Video AI v6.0.3
System Information
OS: Windows v10.22
CPU: Intel(R) Core(TM) i7-7820X CPU @ 3.60GHz 63.706 GB
GPU: NVIDIA GeForce RTX 3080 9.8174 GB
Processing Settings
device: 0 vram: 0.74 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 16.06 fps 2X: 08.54 fps 4X: 02.68 fps
Iris 1X: 14.67 fps 2X: 08.51 fps 4X: 02.86 fps
Proteus 1X: 15.14 fps 2X: 10.83 fps 4X: 02.84 fps
Gaia 1X: 05.36 fps 2X: 03.89 fps 4X: 02.50 fps
Nyx 1X: 08.36 fps 2X: 06.76 fps
Nyx Fast 1X: 16.82 fps
Rhea 4X: 02.06 fps
RXL 4X: 02.01 fps
Hyperion HDR 1X: 22.95 fps
4X Slowmo Apollo: 25.30 fps APFast: 51.81 fps Chronos: 11.82 fps CHFast: 21.78 fps
16X Slowmo Aion: 31.71 fps
Old HW set, practically unusable, today GPU upgrade z GTX 1080 Ti to RTX 3080 Ti
Topaz Video AI v6.0.2
System Information
OS: Windows v10.22
CPU: Intel(R) Core⢠i7-7820X CPU @ 3.60GHz 63.706 GB
GPU: NVIDIA GeForce GTX 1080 Ti 10.846 GB
Processing Settings
device: 0 vram: 0.74 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: ERR fps 2X: ERR fps 4X: ERR fps
Iris 1X: ERR fps 2X: ERR fps 4X: ERR fps
Proteus 1X: 00.70 fps 2X: 00.49 fps 4X: 00.10 fps
Gaia 1X: ERR fps 2X: ERR fps 4X: ERR fps
Nyx 1X: ERR fps 2X: ERR fps
Nyx Fast 1X: ERR fps
Rhea 4X: ERR fps
RXL 4X: ERR fps
Hyperion HDR 1X: 01.42 fps
4X Slowmo Apollo: ERR fps APFast: 20.26 fps Chronos: 00.23 fps CHFast: ERR fps
16X Slowmo Aion: ERR fps
RTX 5080
Topaz Video AI Beta v6.0.4.0.b
System Information
OS: Windows v11.23
CPU: AMD Ryzen 7 7800X3D 8-Core Processor 31.604 GB
GPU: NVIDIA GeForce RTX 5080 15.517 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 26.38 fps 2X: 17.69 fps 4X: 04.50 fps
Iris 1X: 29.59 fps 2X: 17.37 fps 4X: 05.47 fps
Proteus 1X: 29.29 fps 2X: 18.18 fps 4X: 05.39 fps
Gaia 1X: 12.78 fps 2X: 08.49 fps 4X: 05.16 fps
Nyx 1X: 10.88 fps 2X: 08.96 fps
Nyx Fast 1X: 21.04 fps
Rhea 4X: 04.07 fps
RXL 4X: 03.94 fps
Hyperion HDR 1X: 32.34 fps
4X Slowmo Apollo: 30.81 fps APFast: 75.70 fps Chronos: 17.01 fps CHFast: 32.14 fps
16X Slowmo Aion: 44.33 fps
*Previous RTX 3080:
Topaz Video AI v5.5.1
System Information
OS: Windows v11.23
CPU: AMD Ryzen 7 7800X3D 8-Core Processor 31.604 GB
GPU: NVIDIA GeForce RTX 3080 9.8174 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 20.69 fps 2X: 14.50 fps 4X: 04.55 fps
Iris 1X: 20.21 fps 2X: 12.87 fps 4X: 03.91 fps
Proteus 1X: 20.05 fps 2X: 14.06 fps 4X: 05.28 fps
Gaia 1X: 06.76 fps 2X: 04.71 fps 4X: 03.21 fps
Nyx 1X: 08.66 fps 2X: 07.18 fps
Nyx Fast 1X: 16.87 fps
Rhea 4X: 02.76 fps
4X Slowmo Apollo: 24.47 fps APFast: 65.65 fps Chronos: 14.91 fps CHFast: 24.35 fps
16X Slowmo Aion: 41.31 fps
Here is a pre-blackwell implementation RTX 5080 result. Note* The post includes 2x benchmarks. 5080 is at the top (I usually donât see more than one result until I scroll, thus thought it was worth noting).
About a 50% increase compared to my 3080, but still 25% less than the average 4090. Will be interesting to see the results once blackwell has been implemented.
Edit* Adding my old 3080 results for comparison (bottom result - note* itâs from v5.5.1):
1 Like
https://nvidia.custhelp.com/app/answers/detail/a_id/5619
Topaz Video AI
Issue: Built-in benchmark does not work on Topaz Video AI 6.0
Fix: ETA February (as posted by NVIDIA)
Workaround: None
Other Information: Appears to be only an issue with the built-in benchmark, and not with the normal functionality in Topaz AI
1 Like
Uhm. What issue?
Look at my post just above yours.
Will optimization occur on earlier versions of topaz as well? I am using 5.5 because of recent frame rate and sound issues.
No, optimization will be released in an update to v6 very soon. VFR errors are fixed as of 6.0.3
1 Like
Windows 11 Pro 22h2 debloated
No Antivirus or cleaner programs
ASUS-PRIME Z790-A WIFI (Bios 1805)
i9-13900K + Kraken AIO
4090 drivers + 572.16 studio drivers
128 Gig DDR5-4800 UDIMM 1.1V CL40
Motherboard bios all auto no OC
Running Quick CPU
Core parking 100%
Frequency scaling 100%
Turbo boost 100%
Corsair HX1200 PS
4 onboard Nvme 4th gen drives
- Boot drive 2 TB
- Cache drive 2 TB
- Media drive 4 TB
- Mixdown drive 2 TB
Topaz Video AI v6.0.4
System Information
OS: Windows v11.22
CPU: 13th Gen Intel(R) Core(TM) i9-13900K 127.72 GB
GPU: NVIDIA GeForce RTX 4090 23.576 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 720x480
Benchmark Results
Artemis 1X: 177.44 fps 2X: 76.57 fps 4X: 22.62 fps
Iris 1X: 172.42 fps 2X: 100.12 fps 4X: 28.24 fps
Proteus 1X: 228.98 fps 2X: 113.21 fps 4X: 28.68 fps
Gaia 1X: 84.56 fps 2X: 61.76 fps 4X: 26.47 fps
Nyx 1X: 48.88 fps 2X: 44.66 fps
Nyx Fast 1X: 95.60 fps
Rhea 4X: 25.56 fps
RXL 4X: 23.95 fps
Hyperion HDR 1X: 98.44 fps
4X Slowmo Apollo: 217.39 fps APFast: 365.73 fps Chronos: 160.15 fps CHFast: 196.20 fps
16X Slowmo Aion: 127.32 fps
Topaz Video AI v6.0.4
System Information
OS: Windows v11.22
CPU: 13th Gen Intel(R) Core(TM) i9-13900K 127.72 GB
GPU: NVIDIA GeForce RTX 4090 23.576 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1280x720
Benchmark Results
Artemis 1X: 88.31 fps 2X: 31.61 fps 4X: 09.12 fps
Iris 1X: 76.49 fps 2X: 40.86 fps 4X: 10.46 fps
Proteus 1X: 87.94 fps 2X: 42.42 fps 4X: 10.82 fps
Gaia 1X: 35.25 fps 2X: 25.49 fps 4X: 10.23 fps
Nyx 1X: 31.64 fps 2X: 28.32 fps
Nyx Fast 1X: 58.92 fps
Rhea 4X: 09.42 fps
RXL 4X: 09.19 fps
Hyperion HDR 1X: 44.09 fps
4X Slowmo Apollo: 104.86 fps APFast: 169.36 fps Chronos: 73.27 fps CHFast: 87.48 fps
16X Slowmo Aion: 63.09 fps
Topaz Video AI v6.0.4
System Information
OS: Windows v11.22
CPU: 13th Gen Intel(R) Core(TM) i9-13900K 127.72 GB
GPU: NVIDIA GeForce RTX 4090 23.576 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 36.89 fps 2X: 16.04 fps 4X: 04.36 fps
Iris 1X: 39.00 fps 2X: 19.08 fps 4X: 04.74 fps
Proteus 1X: 41.87 fps 2X: 18.38 fps 4X: 04.86 fps
Gaia 1X: 15.92 fps 2X: 10.62 fps 4X: 04.35 fps
Nyx 1X: 17.62 fps 2X: 14.44 fps
Nyx Fast 1X: 31.30 fps
Rhea 4X: 04.29 fps
RXL 4X: 04.28 fps
Hyperion HDR 1X: 27.79 fps
4X Slowmo Apollo: 40.93 fps APFast: 74.49 fps Chronos: 33.32 fps CHFast: 35.64 fps
16X Slowmo Aion: 31.46 fps
Topaz Video AI v6.0.4
System Information
OS: Windows v11.22
CPU: 13th Gen Intel(R) Core(TM) i9-13900K 127.72 GB
GPU: NVIDIA GeForce RTX 4090 23.576 GB
Processing Settings
device: 0 vram: 1 instances: 1
Input Resolution: 3840x2160
Benchmark Results
Artemis 1X: 08.21 fps 2X: 03.80 fps 4X: 01.04 fps
Iris 1X: 08.23 fps 2X: 04.16 fps 4X: 01.08 fps
Proteus 1X: 08.82 fps 2X: 03.94 fps 4X: 01.11 fps
Gaia 1X: 03.37 fps 2X: 02.29 fps 4X: 01.03 fps
Nyx 1X: 02.93 fps 2X: 03.57 fps
Nyx Fast 1X: 05.18 fps
Rhea 4X: 00.95 fps
RXL 4X: 00.96 fps
Hyperion HDR 1X: 12.33 fps
4X Slowmo Apollo: 14.09 fps APFast: 19.76 fps Chronos: 07.22 fps CHFast: 12.56 fps
16X Slowmo Aion: 15.33 fps
Topaz Video AI v6.0.4
System Information
OS: Windows v11.23
CPU: 12th Gen Intel(R) Core(TM) i5-12500 63.745 GB
GPU: NVIDIA GeForce RTX 4060 7.7705 GB
Processing Settings
device: 0 vram: 0.9 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 10.45 fps 2X: 07.34 fps 4X: 02.29 fps
Iris 1X: 10.77 fps 2X: 06.26 fps 4X: 01.77 fps
Proteus 1X: 10.17 fps 2X: 07.43 fps 4X: 02.39 fps
Gaia 1X: 03.41 fps 2X: 02.29 fps 4X: 01.54 fps
Nyx 1X: 04.12 fps 2X: 03.55 fps
Nyx Fast 1X: 07.54 fps
Rhea 4X: 01.34 fps
RXL 4X: 01.25 fps
Hyperion HDR 1X: 21.20 fps
4X Slowmo Apollo: 15.23 fps APFast: 40.63 fps Chronos: 08.16 fps CHFast: 12.59 fps
16X Slowmo Aion: 05.59 fps
Topaz Video AI v6.0.3
System Information
OS: Windows v10.22
CPU: AMD Ryzen 5 3600 6-Core Processor 31.928 GB
GPU: NVIDIA GeForce RTX 4070 SUPER 11.718 GB
Processing Settings
device: 0 vram: 1 instances: 0
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 19.96 fps 2X: 10.09 fps 4X: 02.57 fps
Iris 1X: 24.02 fps 2X: 10.05 fps 4X: 02.65 fps
Proteus 1X: 21.29 fps 2X: 10.40 fps 4X: 02.90 fps
Gaia 1X: 07.54 fps 2X: 05.25 fps 4X: 02.51 fps
Nyx 1X: 08.85 fps 2X: 07.80 fps
Nyx Fast 1X: 17.01 fps
Rhea 4X: 02.26 fps
RXL 4X: 02.12 fps
Hyperion HDR 1X: 20.52 fps
4X Slowmo Apollo: 26.72 fps APFast: 42.42 fps Chronos: 16.05 fps CHFast: 17.17 fps
16X Slowmo Aion: 25.27 fps
Topaz Video AI Beta v6.1.0.5.b
System Information
OS: Windows v10.22
CPU: Genuine Intel(R) CPU 0000%@ 318.66 GB
GPU: NVIDIA TITAN V 11.772 GB
Processing Settings
device: -2 vram: 1 instances: 0
Input Resolution: 1920x1080
Benchmark Results
| Column 1 | Column 2 | Column 3 | Column 4 | E | F | G | H | I | J | K |
|---|---|---|---|---|---|---|---|---|---|---|
| Artemis | 1X: | 09.69 fps | 2X: | 06.80 fps | 4X: | 01.19 fps | ||||
| Iris | 1X: | 09.25 fps | 2X: | 05.56 fps | 4X: | 02.14 fps | ||||
| Proteus | 1X: | 10.01 fps | 2X: | 07.28 fps | 4X: | 02.52 fps | ||||
| Gaia | 1X: | 04.49 fps | 2X: | 01.85 fps | 4X: | 00.15 fps | ||||
| Nyx | 1X: | 04.86 fps | 2X: | 03.57 fps | ||||||
| Nyx Fast | 1X: | 08.53 fps | ||||||||
| Rhea | 4X: | 00.15 fps | ||||||||
| RXL | 4X: | 00.15 fps | ||||||||
| Hyperion HDR | 1X: | 13.28 fps | ||||||||
| 4X Slowmo | Apollo: | 15.82 fps | APFast: | 37.99 fps | Chronos: | 06.53 fps | CHFast: | 11.81 fps | ||
| 16X Slowmo | Aion: | 24.84 fps |
For EVERYONE who (like me) had no idea how to do color formatting, do not remove the upper and lower characters, such as: _ ```Topaz Video AI Beta v6.1.0.5.b
System Information
OS: Windows v10.22
CPU: Genuine Intel(R) CPU 0000%@ 318.66 GB
GPU: NVIDIA TITAN V 11.772 GB
GPU: NVIDIA GeForce RTX 3060 11.831 GB
Processing Settings
device: -2 vram: 1 instances: 0
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 09.65 fps 2X: 07.20 fps 4X: 02.38 fps
Iris 1X: 09.36 fps 2X: 05.77 fps 4X: 01.83 fps
Proteus 1X: 09.38 fps 2X: 06.95 fps 4X: 02.51 fps
Gaia 1X: 03.13 fps 2X: 02.17 fps 4X: 01.47 fps
Nyx 1X: 03.86 fps 2X: 03.33 fps
Nyx Fast 1X: 07.65 fps
Rhea 4X: 01.22 fps
RXL 4X: 01.16 fps
Hyperion HDR 1X: 11.51 fps
4X Slowmo Apollo: 14.00 fps APFast: 42.47 fps Chronos: 07.46 fps CHFast: 12.44 fps
16X Slowmo Aion: 21.73 fps
Topaz Video AI Beta v6.1.0.5.b
System Information
OS: Windows v10.22
CPU: Genuine Intel(R) CPU 0000%@ 318.66 GB
GPU: NVIDIA GeForce RTX 3060 11.831 GB
Processing Settings
device: -2 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 09.51 fps 2X: 07.09 fps 4X: 02.35 fps
Iris 1X: 09.31 fps 2X: 05.85 fps 4X: 01.82 fps
Proteus 1X: 09.33 fps 2X: 06.93 fps 4X: 02.50 fps
Gaia 1X: 03.11 fps 2X: 02.16 fps 4X: 01.45 fps
Nyx 1X: 03.85 fps 2X: 03.31 fps
Nyx Fast 1X: 07.30 fps
Rhea 4X: 01.20 fps
RXL 4X: 01.15 fps
Hyperion HDR 1X: 11.25 fps
4X Slowmo Apollo: 13.88 fps APFast: 36.37 fps Chronos: 07.46 fps CHFast: 12.38 fps
16X Slowmo Aion: 21.32 fps
Topaz Video AI Beta v6.1.0.5.b
System Information
OS: Windows v10.22
CPU: Genuine Intel(R) CPU 0000%@ 318.66 GB
GPU: NVIDIA GeForce RTX 3080 11.81 GB
Processing Settings
device: -2 vram: 1 instances: 1
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 20.10 fps 2X: 09.98 fps 4X: 02.85 fps
Iris 1X: 19.00 fps 2X: 12.42 fps 4X: 03.14 fps
Proteus 1X: 19.32 fps 2X: 12.52 fps 4X: 03.20 fps
Gaia 1X: 06.10 fps 2X: 04.40 fps 4X: 02.82 fps
Nyx 1X: 08.16 fps 2X: 06.45 fps
Nyx Fast 1X: 16.89 fps
Rhea 4X: 02.38 fps
RXL 4X: 02.38 fps
Hyperion HDR 1X: 14.63 fps
4X Slowmo Apollo: 21.08 fps APFast: 44.28 fps Chronos: 14.82 fps CHFast: 19.78 fps
16X Slowmo Aion: 26.32 fps
Thus, we can see that when using two GPUs in the program and benchmark, where (just to emphasize) only ONE video file is processed, we see the ABSENCE of GPU performance SUMMATION.
Yes, my examples have rather diverse GPUs.
Moreover, with joint use, for example, RTX 3080 + 3060 (both GPUs are loaded up to 100%), the final result is LIMITED by the performance of the weaker GPU.
Oops, I didnât post the RTX 3080 + 3060 result here (since I was discouraged of low perf. and conducted many tests with different hardware settings inside Topaz Video AI). I couldnât get any positive difference.
I note by reading these benchmark that the most powerful Apple M4 configurations are systematically less efficient, especially with the RHEA model. Is it normal? Please work on the macOS application and accelerate performance to make up for this immense delay. THANKS.
1 Like
Do you have the pro version?
No just The normal version
My question was addressed to sergyggu.
Uhm⌠How is software supposed to fix bad hardware?
I tried to google a combination of mac/apple vs amd/radeon vs nvidia/rtx gpu just now and saw no benchmarks or testing from trustworthy sources on the first page. Only thing i found was a test between the rtx 4060, which performed 150% better than the M4, and rtx 4060 is a budget gpu at around $400. The benchmarks you see here from people running rtx 3080/-90, 4080/-90 and 5080/-90 are playing in a different solarsystem than Apple M4.
Even for high end nvidia rtx GPUâs the Rhea model is slow. Processing at single digit fps. Itâs not only Appleâs gpu. The Rhea model is just large and require a lot of computes.
Topaz Video AI v6.1.0
System Information
OS: Windows v10.22
CPU: AMD Ryzen Threadripper 3970X 32-Core Processor 127.87 GB
GPU: NVIDIA GeForce RTX 3090 23.756 GB
Processing Settings
device: -2 vram: 1 instances: 0
Input Resolution: 1920x1080
Benchmark Results
Artemis 1X: 24.36 fps 2X: 11.74 fps 4X: 03.58 fps
Iris 1X: 22.37 fps 2X: 14.54 fps 4X: 03.76 fps
Proteus 1X: 23.51 fps 2X: 14.50 fps 4X: 04.27 fps
Gaia 1X: 08.27 fps 2X: 05.75 fps 4X: 03.33 fps
Nyx 1X: 10.01 fps 2X: 08.37 fps
Nyx Fast 1X: 20.64 fps
Rhea 4X: 03.06 fps
RXL 4X: 02.94 fps
Hyperion HDR 1X: 20.51 fps
4X Slowmo Apollo: 33.69 fps APFast: 59.37 fps Chronos: 18.94 fps CHFast: 28.16 fps
16X Slowmo Aion: 25.23 fps
Theory: Tensor RT will increase the Perfomance of 5XXX GPUs by around 30%.
So 5080 will land at 4090 Performace (Close to).
5080 Super will be at the same Performane or slightly higher than 4090.
The software is expected to make use the capabilities of the hardware present.
And at least on the Mac it doesnât at least in quite some scenarios. In general Iris is too slow (and we do know it could be faster as it was faster in the past). And, even only looking at the raw specs, an M2 ultra should be at least on par with a 4060 or somewherre in between a 4060 and a 4070, but sometimes is much slower than a 4060 (again, especially Iris and again, we already had speeds here that were about right in the past).
The whole performance on the Apple Silicon chips is a total mess with TVAI not really scaling with higher specced chips, and some quite wierd things, including that forcing RAM to 10% can lead to higher speeds than if it was at 80-100%, someone even had higher fps with energy saving mode enabled.
The last two announced âperformance optimisationsâ (in Gigapixel and PhotoAI) in fact lead to (significantly) longer rendering times here on my machine.
So, apparently Topaz have no good idea how to get the best performance out of those chips, especially on when to preferably use the GPU and when the NE.
P.S.: âBad hardwareâ is when the hardware fails or runs unstable due to e.g. issues with power-draw. So this would more fit to the high end NVidia cards.
I guess you meant slow hardware.
1 Like
