GPU simulations?

[cb LLC]

Hi,
any news when we can use GPUs to simulate?

Do you think this one AMD EPYC 7702 64-Core 2.0GHz (3.35 GHz Max Boost) or this one Intel Core i9-7980XE X-Series Extreme Edition 2.6 GHz 18-Core LGA 2066 is faster?



Currently I have 10 core i9 I got?



Thank you

[Svetlin.Nikolov]

Hey,

GPU simulations will be good to have. Would it be good to have them at the expense of bugfixes and new features though? Will ping you when there is something, and it's important to know that there is interest and hope for GPU simulations to turn things around. Two years ago, there was so much we could optimize on CPU for simulations, that it was obviously worth the effort, so we did it and now a Phoenix 3.99 simulation is significantly faster than a Phoenix 3.00.01 simulation on CPU. There are still at least 2 more large algorithms we could optimize significantly, so CPU sims can get faster even without changing the hardware.

I could not tell you any more about these two CPUs than what we've put on the docs site's Hardware Advice section, so if anyone has any first-hand experience with them, please do share

Cheers!

[cb LLC]

Well we changed 10 core 7900x with 18 core 9980xe and it is slow in simulations. It went from 1.32m vox/s to just 1. I am shocked. it has almost twice more cores.

we also tried limit cores from 18 to 10 and the speed is the same. And overclocked it to 4.2 ghz

I also noticed core speeds are very jumpy from 4200 to 1600. I attached the video. I remember old phoenix used to use 100% of cpu without any jumping. Maybe a bug?

all these tests were done with fire and smoke. We havent tried liquid yet

[Ivaylo Katev]

try ti swith off the CPU throttle if there is such possibility

[cb LLC]

try ti swith off the CPU throttle if there is such possibility

Thanks is there like a software for this or bios? I never heard about it

[cb LLC]

try ti swith off the CPU throttle if there is such possibility

Ok I found this article bellow. Now it stays at 4ghz but no improvement in speed. Plus I noticed it spikes to 100% sometimes but most of the time at stay at 50% and it takes 1min 30 sec per frame. Not sure why wouldnt phoenix use 100% all the time. Plus we got fast ram and nvme and each smoke sim is only 50mb


So if it stays at 100% for only 4 seconds and the rest of the frame at 50% that means it takes 4 sec to calculate smoke and the rest is bunch of data movement? Grid is only 50mil cells.






Go to Power Option
Click on Change plan settings
Click on Changed advanced power settings
Click on Processor Power Management and set both option for Maximum and Minimum up to 100 Percent.

[cb LLC]

I did another test it's just lots of smoke in the box using direct smooth. It keeps CPUs at 100% When I use buffered smoke it is way slower and cpu is not at 100% I know buffered is harder to calculate but why not 100%

[Ivaylo Katev]

The bottle neck of Phoenix is the RAM speed, not the calculations. Actually, this is true for big amount of contemporary software, just the working frequency is so big, that the memory appears like an "external device" to the processor. So, with this in mind, faster RAM and bigger cache is better than multiple cores. This also answers to you question why buffered is lower - as the name says, the buffered method requires a "buffer", one more piece of RAM involved in the calculations. The direct method is working directly with the velocity data, no buffer is involved, thus having lower buss usage.

[Svetlin.Nikolov]

Hey cb LLC , would it be possible to send over the Phoenix logs from these runs? This would give us more info.

[cb LLC]

Hey cb LLC , would it be possible to send over the Phoenix logs from these runs? This would give us more info.

HI I will soon.
I noticed when I am at 14mil cells cpu is 100% once smokes expands to around 20mil CPU power stays at 50% 14 and 20 not much difference but it drops by 50%

[cb LLC]

here is the scene after 100 frames of simulations you will see cpu drop.

[cb LLC]

another update:

I paused the simulation. Then I merged all sim objects into the other scene to send it to you, After I resumed the simulation CPU went back to 100%
it is a bug I think. Sometimes it is 100% sometimes half. Cpu 2 is usually always at 100% Usually buffered has this

I can actually do 2 simulations at the same time to keep cpu at 100% with almost the same calculations per frame

[urbanite]

I am on 9940 i9 and speed simulation vary from 20-50%. This is on default fire preset. Maybe it is how it suppose to be but am having similar results as I had from amd 1950x which is NUMA...

[Svetlin.Nikolov]

Hey guys, since we don't have these CPUs here it would be very useful if you can send over your Phoenix logs from these simulations, so we can see how they compare with our setups.

[cb LLC]

Hi Svetlin,
here it is. We have 5 phoenix grid in the scene I guess they all go to one log?