Increasing Performance for 4K video

general questions about Neat Video
Post Reply
Posts: 3
Joined: Thu Oct 13, 2016 1:03 am

Increasing Performance for 4K video

Post by estarkey7 » Wed Jul 19, 2017 9:52 pm

I'd like to know what things can be improved in my system to increase noise reduction performance in 4K video?

Which has the greatest impact?

1. CPU Cores?
2. CPU Speed?
3. CPU Memory Bandwidth?
4. CPU Memory Amount?
5. CPU L1 cache size?
6. CPU L3 cache size?
7. CPU L3 cache size?
8. CPU Instruction Set (SSEx,AVX, FMA, etc.)?
9. GPU computational units (shading unit quantity)?
10. GPU Memory Bandwidth?
11. GPU Memory Amount?
12. GPU Instruction Set (OpenCL or CUDA version).
13. GPU Quantity?
13. Storage Speed in MB/Sec.

I am planning a system build, and after looking at performance gains when adding multiple GPU's, there is little performance between one and 4 GPUs in a system. System memory seems to be of little help, as does having dozens of CPU cores. To sum it up, I doesn't want to have a $6000 computer that runs Neat Video as fast as a $1000 PC.

Is this an architecture limitation? Is there a plan to increase the noise reduction scaling when multiple cores and multiple GPUs are available?

On a more technical note, does your algorithm have an O^2 or O^N operations, particularly when using temporal filtering. Are there dependencies in the calculations that prevent processing through parallelism?

Posts: 2261
Joined: Thu Sep 01, 2005 4:12 pm

Post by NVTeam » Thu Jul 20, 2017 11:41 am

Most of the parameters you have listed do have some impact on the overall speed of a render process that involved Neat Video.

When you consider a CPU, the most important parameters are the number of cores, clock speed, memory bandwidth, cache size.

When you consider a GPU, the most important parameters are the processing power (GFLOPS), memory bandwidth and size. Two fast GPUs will deliver higher speed than one, but the speed is usually not doubled (the speedup is typically around x1.5).

We continue to work on further optimizations to improve the performance of the filter and expect that the above parameters will remain relevant. Therefore it makes sense to invest into a machine with a good new CPU and fast GPU or GPUs (or possibility to add more GPUs).

As the first step please try to run Optimize tool in Neat Video Preferences to see how the speed of the filter scales with the number of cores used and with added power of the GPU. You can also run the standalone benchmark tool -- Neat Bench -- on different machines to evaluate their performance. It is a command line tool that you can control using command like keys. This way you can make it test 4K resolution too.

Thank you,
Image Image Neat Video team
noise reduction for video and photos

Post Reply