What are typical scores for benchmark in Contextcapture for different GPU cards? Does GPU RAM size influence much?
Benchmark scores can be checked in Contextcapture Settings->Configuration->Start Benchmark
2x AMD Radeon VII - 226[Vulkan API]
RTX 2080 - 215/100
Titan V (Volta) -?Quadro Pxxx-?
TITAN Xp -?
TITAN X (Pascal) (recommended)- ?
GTX TITAN X (GTX 1080 Ti) Maxwell - ?
GTX 1080 (now recommended) - (191/100) or in latest versions - 215/100
GTX 1070 - ?
GTX 980 Ti (Titan) (from CC recommended list) - exceeds it by far (150/100)
GTX 980 - just exceeds 100 (117/100)
GTX 1060 (recommended) - 99/100
NVIDIA GeForce GTX 980M - (75/100)
NVIDIA GeForce GTX 960M - (45/100)
GTX 780 Ti (from CC recommended list) - ??
Nvidia Tesla K40m - 57/100
NVIDIA Quadro K4000 - 24/100
NVIDIA Quadro K2000M - (20.3/100)
GTX 650 Ti - 24/100
GTX 260 - 12.6/100
GTX 730 - 10/100
Also there seems there is no linear correlation with benchmark scores regarding processing time. For example on same machine GTX 650 Ti processes sample model 5 hours while GTX 980 Ti processes it at 3 hours while benchmark scores differ 7 times performance is only 2 times. General Contextcapture System Requirements:Minimum HardwareAt least 8 GB of RAM and NVIDIA or AMD graphics card, or Intel integrated graphics processor compatible with OpenGL 3.2 with at least 1 GB of dedicated memory.
Recommended HardwareMicrosoft Windows 7/8/10 Professional 64-bit running on a PC with at least 64 GB of RAM, and Intel I7 (4+ cores), 4.0+ Ghz, Hyper threading should be enabled, and a NVIDIA GeForce GTX 1080 Ti graphics card (or Titan X, GTX 1080, GTX 980ti) and data should preferably be stored on fast storage devices (fast HDD, SSD, or SAN).P.S. ContextCapture also does not support NVidia Tesla cards yet.
Please post your scores!
I dont suppose anyone has updated this list with the Pascal Quadro cards ?
Doesn't seem to make a licking bit of difference .. I don't know what I've got wrong or what is going on - got a 1080 and M60's - and none of my instances use the GPU more than about 2 - 3% of the time during reconstruction ..
The 1080 scaores about 195 .. per above .. got a 1080Ti sitting on my desk too .. but I can't keep any gpu I've got even moderately busy ..
Can anyone running a reconstruction fire up either Windows 10 task manager with the GPU perfmon or MSI Afterburner (say) and monitor their GPU usage and tell me if they see anything different? I get the odd series of spikes every few mins or so, flat line the rest of the time .. CPU maxed.
Yes this is typical that reconstruction is CPU/GPU dependent some tasks ar CPU intensive and some are GPU. So you will see spikes. In latest Windows 10 you can see GPU load in Task Manager so no need for MSI.
But you should also see by some load test that GPU is not throttled by overheating or other issues. There is more discussion about it, from my observations avarage GPU load is only 25% so there is room for improvement - https://communities.bentley.com/products/3d_imaging_and_point_cloud_software/f/contextcapture-descartes-pointools-forum/126763/contextcapture-windows-timeout-detection-and-recovery-tdr-errorIt would be better to start a new discussion regarding optimizing GPU for ContextCapture.
There is very low activity from community for posting scores. I bet that Acute3D have all the scores collected from customers from received logs just would like to ask them to share. Because it not always cost effective to by latest card just to get only small increase in scores.
Microsoft Windows 7/8/10 Professional 64-bit running on a PC with at least 64 GB of RAM, and Intel I7 (4+ cores), 4.0+ Ghz, Hyper threading should be enabled, and Kickass Proxy Unblocked List Working a NVIDIA GeForce GTX 1080 Ti graphics card (or Titan X, GTX 1080, GTX 980ti) and data should preferably be stored on fast storage devices (fast HDD, SSD, or SAN).