[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k] [s4s] [vip] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Home]
Board
Settings Home
/3/ - 3DCG



Thread archived.
You cannot reply anymore.



File: AMD-Threadripper.png (617 KB, 1543x824)
617 KB
617 KB PNG
Curious about how AMD CPUs and GPUs perform with 3D software (more specifically, Blender, C4D, Maya, Arnold, Cycles, Octane), as well as Adobe products. Currently looking at the $500, 8core, Ryden 7 1800x and it aparently peforms similar to a $1k Intel i7. Not only that but their new Ryzen Threadripper will aparently cost less than $1k and will have double the cores. There must be some kind of catch, right?

What about its GPUs? Aparently GPU rendering on OpenGL/CL is very slow compared to CUDA, is that true?
>>
New Ryzen CPU's perform just fine, but you're better off with nVidia GPU, both for Maya and Adobe packages. You don't have to pair AMD CPU and GPU.
>>
i have blender and i use OpenCL (AMD cores)
yes it does work, but make sure its supported in the documentation. personally i have 7870HD which is old but blender seem to support it.

the downside is that it takes a few seconds to load the render kernals but the rendering itself is fuckin sweet.
yeah openCL does work but people rrather use cuda because they have more products supports.
people also say OpenCL is buggy because of shitty render support (from the third party standpoint, not AMD)
>>
>>567690
and to answer your question. my openCL renders like 50% faster than CPU rendering so im happy. i don't know about cuda but my CPU render as noisy as hell and with the GPU sort of takes less time to get a good render
>>
>>567688
Don't even bother with the 1800x, all of the 8-core chips are the same silicon, just of different quality. You can get a 1700x or 1700 and clock them up to the same speeds, and save a bunch of money in the process.
There's no real catch, AMD simply has the superior scale-up architecture. TR will in theory be able to clock as high as Ryzen quite simply because it's two of the same chips.

For GPU I would err on the side of CUDA simply because it's the more mature platform.
>>
>>567688
the older the platform the more stable its gonna be thats why studios dont buy the latest shit on day one or even on the same year

so yeah CUDA is better




Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.