kiaspid Posted August 1, 2017 Share Posted August 1, 2017 (edited) hi guys, sorry can't speak English well. I planning to buy new hardware for OpenCL simulations on GPU. I study a little about pro-GPUs computing performance and find some information in https://en.wikipedia.org/wiki/FLOPS "Built using commercially available parts. AMD Ryzen 7 1700 CPU combined with AMD Radeon Vega FE cards in CrossFire tops out at over 50 TFLOPS (OMGOD!!!) at just under US$3,000 for the complete system." is it right? Can combine two AMD Radeon Vega FE cards in CrossFire with CPU? Does it help to speed up simulations in Houdini? Which Single-precision or Double-precision floating-point is more important in Houdini? is it depend on 32bit or 64bit process in software? https://pro.radeon.com/en-us/product/radeon-vega-frontier-edition/ OpenCL™ 2.0 Support Enables professionals to tap into the parallel computing power of modern GPUs and multi-core CPUs to accelerate compute-intensive tasks in leading CAD/CAM/CAE and Media & Entertainment applications that support OpenCL. Does Houdini support OpenCL 2ver? Edited August 1, 2017 by kiaspid Quote Link to comment Share on other sites More sharing options...
Atom Posted August 1, 2017 Share Posted August 1, 2017 Currently Houdini 16 can only use 1 OpenCL device at a time. This may change in the future. 1 Quote Link to comment Share on other sites More sharing options...
malexander Posted August 1, 2017 Share Posted August 1, 2017 Houdini either uses OpenCL 1.1 or 1.2. OpenCL 2.0 doesn't expose any features which Houdini needs right now. Single precision is also more important at the moment, and as Atom says, only 1 compute device is currently supported. 1 Quote Link to comment Share on other sites More sharing options...
Guest tar Posted August 1, 2017 Share Posted August 1, 2017 12 hours ago, kiaspid said: AMD Radeon Vega FE cards in CrossFire tops out at over 50 TFLOPS (OMGOD!!!) that's only possible for fp16, Houdini uses more precision with fp32+ Quote Link to comment Share on other sites More sharing options...
dudenay Posted August 2, 2017 Share Posted August 2, 2017 (edited) 21 hours ago, kiaspid said: Which Single-precision or Double-precision floating-point is more important in Houdini? is it depend on 32bit or 64bit process in software? Since almost five years, Houdini only support 64 bit platforms: https://www.sidefx.com/forum/topic/27117/ About Single Precision or Double Precision, it depends on what kind of data you want to store. For example, in the FBX import doc, it is specified that some vex datas can be single or double precision. http://www.sidefx.com/docs/houdini/commands/fbximport Maybe this other page of the doc will help you, it explains that you can changes the precision and amount of memory used to store an attribute. Warning: Not all operations in Houdini maintain 64-bit precision. http://www.sidefx.com/docs/houdini/nodes/sop/attribcast Edited August 2, 2017 by dudenay 1 Quote Link to comment Share on other sites More sharing options...
Guest tar Posted August 2, 2017 Share Posted August 2, 2017 On 01/08/2017 at 10:20 PM, kiaspid said: Which Single-precision or Double-precision floating-point is more important in Houdini? is it depend on 32bit or 64bit process in software? For OpenCL sims, single-precision is most important - double is used as an accumulation buffer to help avoid rounding errors afaik. This may be different for the OpenCL SOP. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.