OTOY Forums

Based on Metal, in theory, both CPU and GPU can participate in rendering calculations. In CUDA, only the GPU can participate in the rendering calculation.

In addition, Metal also supports ARM, so I don't know how to perform on iOS (iPadPro) and how to interact with iPhone.

These are what I expect. As for the pure GPU rendering speed, I can roughly guess that the rendering speed of the hardware of the same specification should be only 80% of the CUDA speed.

Good enough for me!

Weezer wrote:I doubt it will match a CUDA-based system, but the good news is that if it's not as fast as you'd like, just throw another GPU at it! Simples.

True but city living limit's me to one desktop tower ~ 4 gpu... I could see using a mac system with RNDR. I've been wanting to play with it but somehow my 20 tokens got lost... I hope to hear more about it, or maybe I'm not looking hard enough.

Otherwise if there's ANY integration with the XDR display that'll be one of the most persuasive reasons for me to switch back.

Yeah.. i think that the Metal platform will be interesting and also solves a bunch of issues..

1 - it unifies the mac and abstracts the underlaying architecture.
(i.e cpu gpu intel / amd / arm etc

2 - it lets octane use ALL the resources available instead of having to code separate CPU rendering options for "lowest common dominator situations"

(Remember V-ray guys difficult issues a few years back with GPU/CPU feature parity?) This lets Octane just let METAL handle it.

3 - This makes it useful in situations other than what we normally think of.. AKA Apple's ARKit and such.. for use with Brigade .. Could get really interesting on IOS as the brains for the AR glasses.

All these things are quite a nice side benefit of doing METAL.

I think this will be quite exciting going forward.

DOWNLOADING THE METAL VERSION RIGHT NOW. Its released!

All great points and very insightful CJ. I listen to a lot of Accidental Tech Podcast and they mention a big reason apple is years ahead of competitors in mobile chip development is because not only do they own the pipeline, so they can develop hardware in step with software, they also are able to scale things down as needed - for example the apple watch is years ahead of any other wearable tech. So a small market is the only reason they wouldn't scale up for "after burner" metal accelerators (maybe). And I think apple know's it's important to develop for their pro market even if it's small since they tend to be vocal and influence regular / prosumers.

They also mention that Apple's move to arm allows them to shed a lot of overhead that came with x86, and I can imagine CUDA has a lot of overhead for such an old platform now that it's 13 years old. But I'm not too educated in all this so I'm just excited to see all this happen

Octane X Release thread: viewtopic.php?f=15&t=75411

It’s Christmas!!!
Right I’m going to buy a Razer Core X.
What GPU / GPUs should I pack in there for working with Octane / C4D / Houdini / Da Vinci Resolve, Fusion / After Effects?

Here is the list tested from the announcement..

I notice that my intel gpu is listed but in the c4d plugin and stand alone I cannot select them.. also sadly looks like the RX580 I had is useless.. POUT
Looks like for now I'm stuck on my internal AMD gpu only for the moment.

Any chance the RX580 will be supported?


-----

I just want to know the OB score of all the AMD/Intel GPUs on 10.15 now

Just to be clear, these are still early numbers and have a +/- variance of 7% or so, but one thing that will likely hold is the relative performance in OB between these GPUs :

Radeon Pro Vega II Duo - 412 OB*
Radeon Pro Vega II - 206 OB
Radeon VII - 200 OB
5700 XT - 170 OB
Vega FE - ~161 OB
Vega 64 -~148 OB
Vega 56 - ~117 OB
Vega 48 - 108 OB
5600M - 108 OB
5500M - 77 OB
Vega 20 - ~50 OB
Intel iris 640 - ~17 OB
Intel iris 630 - ~15 OB
Apple A13 - ~15 OB

*dual ASICs (the 415 in the PR and 419 in the blog post are different runs that illustrate the variance noted above)

I can't get OctaneX for C4D on R20 to run i get an error
<<< Render failure detected!!! >>>
Please check render statistics to solve the problem.
MB:0/0 ST/MOV:0/1 Nodes:12 Tris:12 DispTris:0 Hairs:0 Meshes:1
Textures Grey8/16:0/0 Rgb32/64:0/0
>> ProcessUpdate(flags=CMOG cAsyn=1 dDelay=1
>> ProcessUpdate end res:0
>> ProcessUpdate(flags=CM cAsyn=0 dDelay=1
>> ProcessUpdate end res:0
>> ProcessUpdate(flags=CM cAsyn=0 dDelay=1
>> ProcessUpdate end res:0
and under R21 it asks for my login and password and when I enter it, it says it can't connect to the auth server.

Please help!