Gigapixel AI Performance Issues

[joe.robinson]

Greetings, and thank you for taking the time to read this.

I’m a new gigapixel user. I started testing the software on my laptop, which is an older laptop. It’s running a Core i7-4710HQ. This is a 4th Gen intel proc running at 2.5Ghz/~3.5 Ghz turbo. It has a dedicate Nvidai GTX 970M.

After reviewing the forums and specs, I decided to build a new machine for optimal performance. I put together an i7-9700k. It’s a 9th gen intel proc running at 3.6Ghz/4.9Ghz turbo. It has 16GB of ram and and all SSD storage (windows volum is NVME). I opted for this processor as, according to the support forums, openvino supposedly works as well as a dedicated GPU.

I have found that the performance of gigapixel on the new machine is considerably slower. I started processing 5,000 images (on each machine) this morning before work. When I got home, my 5+ year old laptop had completed the task, but the new machine was less than half way through.

I’m looking to determine if I do indeed have a performance problem with this new rig, or if I’m incorrect in my expectations for the performance of OpenVino.

Any information is greatly appreciated!

[AiDon]

@joe.robinson

I opted for this processor as, according to the support forums, openvino supposedly works as well as a dedicated GPU.

I haven’t seen that quoted anywhere.

@joe.robinson

I’m looking to determine if I do indeed have a performance problem with this new rig, or if I’m incorrect in my expectations for the performance of OpenVino.

What GPU are you using and do you have dual GPUs? Go to Help-> Graphics info, press Copy and post the info here please.

[joe.robinson]

The machine that I believe is under-performing is using an Intel UHD 630 with OpenVino enabled. There is no dedicated GPU in the machine.

As I mentioned above, it’s my understanding that this combination was as good as a dedicated GPU. If I’ve been investigating GPUs, and if I need to add one, I’m looking at an RTX 2070… I just don’t want to spend the $500 if it won’t make a difference!

Application & Version: Topaz Gigapixel AI Version 4.4.5
Operating System: Windows 10 (10.0)
Graphics Hardware: Intel(R) UHD Graphics 630
OpenGL Driver: 3.3.0 - Build 26.20.100.7870
CPU RAM: 16038 MB
Video RAM: 1500 MB Total, 1200 MB In Use
Preview Limit: 3466 Pixels

[joe.robinson]

@AiDon

I opted for this processor as, according to the support forums, openvino supposedly works as well as a dedicated GPU.

I haven’t seen that quoted anywhere.

What kind of GPU is the key to speeding up Gigapixel AI?

I’m glad to see your reply. Allow me to continue my in-depth discussion here. First of all, I want to restrict the variables discussed so that we can get closer to the core of the problem more quickly and accurately. A. I’m not a gamer and I don’t care about professional CAD/CAE applications. B. Suppose my goal is just how to make Topaz’s GAI run faster! C. We only discuss desktop graphics cards Therefore, I quantified data once in 2018-2019 using graphics cards with GDDR6/HBM2 and VRAM no …

“In fact I have been just testing a AI app using OpenVINO and the combination of an i7/HD 630, with 16GB ram, is faster than using the GTX 1050/4GB on the same CPU.”

[Artisan-West]

@joe.robinson I think you have just proven that using the CPU is slower than a dedicated GPU but you should try processing 10 pictures on each to verify this. You don’t have to spend $500 on the GPU card. I would recommend the GTX 1660 Super which cost around $230. It is faster than my RX 580 which does pretty well on Gigapixel. Also, the memory on the 1660 Super (not the 1660) is much faster GDDR6 type and you will have 6GB instead of the 1.5 GB normal RAM you are using now.

[joe.robinson]

I do not believe I was actually experiencing an issue; I think my expectations were set too high from the research I did before my hardware purchase.

Adding a RTX 270 with 8GB, my frame processing time went from about 12 seconds per frame to about 3 seconds.