SharpCap Forums

Hi,
I noticed that weiner deconvo is a function that improves the live image but it also has a strong impact on performances. When it is activated, there is a lag of about 10 sec at least for each action on the histogram mainly.

Is that an issue related to the performance of the computer or its configurations ?
Can we improve that by adding some components ? improved processor, memory, graphic card, etc...

My configuration is a, AMD Ryzen 5 1600 six core 3.2 Ghz, 12 Gb of RAM, very basic graphic card, and SSD Samsung 860 Evo 1Tb, running Windows 10 last level, Sharpcap 3.2 64bits last version

Hi,

Currently, the most important thing for the performance of the deconvolution is your CPU speed – that is the raw gigahertz speed of the CPU since right now the deconvolution only uses one core.

It looks like most of the time is taken up in Fourier transform type functions, so there may be some scope to either use more than one core or offload some of the work to the GPU in a future version.

Cheers, Robin

Thanks Robin,
yes, using more core could be a good option, at least for me as i don't have a correct GPU..
So i wait for may be a new enhancement...
Bernard

Hi Bernard,

I had a play around with this this afternoon and enabling the GPU was relatively simple and doubled the speed for me (£150 GPU faster by a factor of two than £500 CPU). Pushing the processing of the three separate colour planes into separate threads doubled the speed again – a speed up that should also apply for those whose GPU doesn't have the power to help out. Watch out for the next version…

Cheers, Robin