VSO Software Forum

Hello,

Motherboard: Asus Prime Z370-A
Processor: i7-8700K - Overclocked to its highest stable point
Video card: NVIDIA Geforce GTX 1080 FTW DT - Overclocked to its highest stable point

I just upgraded the video card and would like to make use of what I have to accelerate the VSO software to is best settings.

Under Hardware Optimizations I have the option enabled.
Maximum of hardware decoder(DXVA2) is selected as 6
Graphic adapter to use: NVIDIA Geforce GTX 1080

#______________Acceleration type___Max Bitrate___Nb
1 NVIDIA CUDA__NVIDIA NVENC_____16000000______2
2 NVIDIA CUDA__NVIDIA CUDA______14000000______2
3 OpenCL______Kronos OpenCL____14000000______2

4 Intel Quicksync soft (not available) 20000000
5 Intel Quicksync(not available) 14000000

I have not heard of Quicksync and reading up on it, it deals with integrated graphics. This MOBO does not have onboard graphics but I do have an Intel Processor and Quicksync looks to be an Intel product. it appears I have to enable this in the bios. I'm hesitant to do anything that will affect the speed of the video card. It does look like it offers the highest bitrate of all and perhaps I should enable it?

With this card, is there any setting that will help accelerate the process? I have several hundred DVDs to convert to mkv and any increase in speed is a great help.

Thanks!

You CPU has UHD 630 GFX chip and it is not powerful. You can enable it, but your 1080 is about 1200% faster.
Using that is not giving you any proper advantage, might drop few seconds from conversion time, might cause bottlenecks in CPU, might even crash.

Hi JJ, thanks for the reply.

Seeing the Quicksync max bitrate is much higher than the NVIDIA or OpenCL, it seems like this would be a good thing to have active. I assumed that the VSO software recruited all available sources of computational power and applied them as in a kind of threading to whatever task was being called on. If that's not how it works, then having #4 & #5 available doesn't make sense.

Last night I gave the GTX 1080 its first test and compared it to the GTX 1050 it replaced. I converted a ts file recorded from a TV special into mkv and the frame count never exceeded 45 Fps that I could see and it was much like using the older 1050. There was no great difference in speed of conversion, it took around 4 hours to do a conversion on a show that is two hours in duration. It seems like with the processing power this computer has between the CPU & GPU, I would see better Fps and a lower processing time.

With that said, I do get better results when converting DV9 files to mkv, often getting to the several hundred Fps. With such a low Fps with the ts file, I wonder then what I have not selected or, have selected poorly that I could correct? I have literally hundreds of video files to convert and would like to get the fastest conversion times without sacrificing quality.

I know the GTX 1080 has much greater processing abilities than with the GTX 1050; when not running VSO software I run BOINC and have the GPU dedicated to solely to SETI with the CPU devoted to Gene & Cancer studies. The GPU processes a SETI work unit in 7 minutes, much less time than with the 1050. It takes about 2 hours to do a SETI WU per CPU core @ 4,9 GHz, but there are 6 cores running simultaneously.

Thoughts as to what I might do to increase the Fps and reduce the conversion time with what I have available?

Thanks!

Video processing uses a lot of memory and does huge amount of reading and writing.
To get maximum speed you need to use fast drives, two SSDs is best bet, and set them to different buses so they don't steal bandwidth from each other.
Then use as many GPU tasks as possible, depends on system. Testing is best way to find what combination is fastest.
Quicksync might give better max bitrate in theory but awhen used with 1080 it just slows everything. Try without it.
Most likely best conversion speed comes with using Nvidia only, let CPU handle all file swapping, read/write etc.