OTOY Forums

are there any plans to get the latest Octane X on AMD Mac Pro 2019 that is capable to run on macOS 13?
Even though in a few months we will get macOS 14, I am still suck with macOS 12 Monterey because Octane isn't supported on macOS 13

I seem to always end up here complaining about how I love Octane, Path tracing is hands down the best look of many renderers with minimal effort, and yet I cannot use it in C4D on Mac. It crashes, or hangs consistently. I have a kind of workflow which involves animation in C4D then rough materials, then export as soon as possible to ORBX and finish and finesse in Standalone. It still fills me with dread using it, because I don't know when it will hang. Compare this to Blender which you can throw pretty much anything at and I feel confident that it won't die on me. But it doesn't render as nicely as Octane. This is such a shame.

I have it on iPad too and it's incredible. I recently did the same workflow ( rough in C4D and then export ORBX ) and finessed on the iPad with great success. It baffles me why it can't do the same in C4D.

RegSquires wrote:I seem to always end up here complaining about how I love Octane, Path tracing is hands down the best look of many renderers with minimal effort, and yet I cannot use it in C4D on Mac. It crashes, or hangs consistently. I have a kind of workflow which involves animation in C4D then rough materials, then export as soon as possible to ORBX and finish and finesse in Standalone. It still fills me with dread using it, because I don't know when it will hang. Compare this to Blender which you can throw pretty much anything at and I feel confident that it won't die on me. But it doesn't render as nicely as Octane. This is such a shame.

I have it on iPad too and it's incredible. I recently did the same workflow ( rough in C4D and then export ORBX ) and finessed on the iPad with great success. It baffles me why it can't do the same in C4D.

Please make sure to upgrade macOS to 13.3, it fixes an issue in Apple Metal Driver, and use c4doctaneX 2022.1.1-R2.

If not enough, please PM me with a link to a scene that is giving you issues, thanks.

ciao,
Beppe

bepeg4d wrote:Hi northalex78,

we are not ready for an unified OctaneBench CUDA/Metal version, maybe when Apple Silicon M3 and OctaneRender 2023.1 stable will be out

Apple is relatively new to GPU rendering, and I still think that it is like comparing artichokes with strawberries

Nvidia GeForce GPUs are designed to get the best rendering speed whatever it costs, while Apple silicon GPU are designed to give the best rendering power per watt.

Probably we should need to compare only with Laptop or Quadro Nvidia models to be more fair

Anyway, as said, rendering speed is not everything, especially if Apple Silicon is faster in all the other tasks, and you have PCs for final rendering.

Probably the best test you can do is to rent a MacBook Pro M2 Max or Apple Studio M1 Ultra with macOS 13.3, and try to work with your scenes in OctaneX 2022.1.1.

ciao,
Beppe

This is a bit of a cop-out answer.
"we are not ready for an unified OctaneBench CUDA/Metal version"
Can you say why?

Can we create Octane scenes on a Mac, save them to a network drive, then open those scenes on a PC and render?
Ok cool so why can't you just send it from the Mac to the PC slave?

"it is like comparing artichokes with strawberries"
Be that as it may, if we were going to buy new Mac hardware, why can't we get an idea of what kind of performance we're looking at?
Is an M1 or M2 fast enough to build a scene and look dev, then move to PC for rendering? You said in an earlier reply that you think that's the best workflow. Cool. But how does the experience of look dev on the Mac feel? How does it compare to lookdev on single 3080? Or some 2080s? Or anything??? We don't know because the cowards at Otoy won't release any kind of comparison data, but we KNOW they have it.

"Probably we should need to compare only with Laptop or Quadro Nvidia models to be more fair"
Silly comment, it's not about fair, it's simply about where do the Macs sit in the range of available options? "Oh it's not FAIR to compare to a 4090". I don't care if it's fair or not, we want to know.

What are all these eye-roll emojis you're using for anyway? Do they mean something else to you, than they do to me? Literally rolling your eyes at my questions, which I would think are quite reasonable. What's the rationale (other than "wah, wah, it's not fair") for hiding any sort of comparison or releasing a Mac benchmark tool?

I realise many people get emotional about Mac vs PC but for us, we are simply trying to decide if we need to leave Macs once and for all. We do not want to, we enjoy the Mac UX/UI etc. Apple Silicon is very promising. We simply seek data, cold hard data, to make better informed choices. But Otoy hides it, or is vague about upcoming releases. No netrender until M3? Maybe? Who knows??? Maybe the same time as Brigade, promised for 10 years already.

hey guys,
if it can help, I've been working on a project recently, and mainly working on the PC,
but I bought a M1 Max (64gb ram) and wanted to check how it would compare.

PC is 5950x + 128gb ram + 1x4090 + 1x3090 (in an external TB enclosure), and I got another PC with 1x4090 on the network
Mac is M1Max Macbook Pro with 64gb unified memory
Both run C4D 2023.2 and Octane 2022.1.1 R3

Here is the test Material has textures for roughness and bump, noise for displacement and randomwalk SSS Here are the settings, I've played a bit with adaptive sampling,
which is IMO one of the best way to save render time, depending on how much grain you tolerate Both machines rendered exactly the same image, which already is a good thing.

So, the Mac was 2:17min, which is not too bad in my opinion.

PC with 3090 is 47s
PC with 4090 is 25s
PC with both 3090 and 4090 is 20s
PC with both cards + 4090 on the network is 17s

Just a simple unscientific test, but frankly I've started this scene with the Mac,
until I really wanted to switch to the PC to precisely work on the material, I could almost instantly see what I was doing.

I think it's a decent machine to work on, it's extremely quick with everything else, almost quicker than my PC on certain tasks,
and certainly blows my PC out when working with Premiere and After Effects.
Also you have to consider the portability, the power consumption which I think is stupid for the PC.

Potentially with a Mac Studio M2 Ultra maxed up we'd almost have a decent CGI machine... but at what price ?

Also, Bepe, may I ask you guys to update the MacOS C4D plugin link ? It still points to an older version,
and you have to go on the forum to download 2022.1.1 R3 when the PC version.

Hope that helps

It's been a while since I've used Octane but I'm giving it another shot on my Mac Pro running 12.6.3 on an AMD 6900XT GPU.

In Houdini, I can't get any image to appear in the IPR, and I'm sure it's user error but I was wondering if anyone had a clue as to what I'm doing wrong?

I have followed the instructions in the manual for the basic scene, created a ROP in the OUT context, created a Render Target in the MAT context with a couple of Octane Materials and assigned the materials.

I hit IPR and I feel like I should be seeing something, but it's completely black.

Thanks for any help you guys can provide.

PS

FWIW - the Octane standalone works fine, so I know it's not something with the GPU or my machine.

Midphase wrote:It's been a while since I've used Octane but I'm giving it another shot on my Mac Pro running 12.6.3 on an AMD 6900XT GPU.

In Houdini, I can't get any image to appear in the IPR, and I'm sure it's user error but I was wondering if anyone had a clue as to what I'm doing wrong?

I have followed the instructions in the manual for the basic scene, created a ROP in the OUT context, created a Render Target in the MAT context with a couple of Octane Materials and assigned the materials.

I hit IPR and I feel like I should be seeing something, but it's completely black.

Thanks for any help you guys can provide.

PS

FWIW - the Octane standalone works fine, so I know it's not something with the GPU or my machine.

Hi,
it seems you have installed the new version for M1/M2 Apple Silicon only.

For AMD GPUs under macOS Monterey, you need to install old PR14 version:
viewtopic.php?f=102&t=79383

ciao,
Beppe

But why is there a specific Intel version included in the file then?

This doesn't make any sense. Also PR14 doesn't even work with Houdini 19.5.

Midphase wrote:But why is there a specific Intel version included in the file then?

This doesn't make any sense. Also PR14 doesn't even work with Houdini 19.5.

Hi Midphase,

Thanks for the post.
The Intel version still works on Silicon chips.

Yeah, the last version we supported with PR14 is for Houdini_19.0.531.



cheers

@studiofy

"So, the Mac was 2:17min, which is not too bad in my opinion.
PC with 3090 is 47s
PC with 4090 is 25s
PC with both 3090 and 4090 is 20s
PC with both cards + 4090 on the network is 17s

Just a simple unscientific test, but frankly I've started this scene with the Mac,
until I really wanted to switch to the PC to precisely work on the material, I could almost instantly see what I was doing.

Potentially with a Mac Studio M2 Ultra maxed up we'd almost have a decent CGI machine... but at what price ?"

Thanks for this. Doing what Otoy will not!
We need more of these comparisons.

It does sound like we'd need a Mac Studio Ultra - or better - to be able to work effectively on Mac. And this is what we've been asking Otoy about for months, to which they keep saying "why do you want to compare render speed". Ugh, so frustrating.

What I am interested in, is the look dev experience. Setting up textures, lighting, getting a clean image preview from the live viewer.
Can we work effectively on a Mac, to develop scenes, before sending over to the PC to render?
And Otoy just say "Macs are fast to work with, why do you care about render speed? This is not what should be compared". Seriously said to me in PM.
Uh, maybe because our job is to make nice images, and we need to SEE the image, and WAIT for that image to resolve in the LV?

So we just want to be able to benchmark different Macs.
If 2022.1 is functionally identical on Mac and PC, ie, you can open the same scene on either and render... why can't we have an updated Octane Bench, one for PCs and one for Macs?
As we want to know
- which Mac is better?
- M1? M1 Max? M1 Ultra in the Mac Studio? Vs M2? M2 Pro?
Who knows how they all stack up? Which one should we buy? Any? None?
- how much of an improvement (if any) could we see, moving from, I don't know, a PC with 4x2080s? Probably none. Or a Mac with 2x1080Tis? Should we just keep using that Mac? Should we just say f**k mac, go to PC? AGAIN WHO KNOWS

Because for some reason Otoy won't publish any kind of bench. Otoy, do you have some agreement with apple or something that you won't show this? It makes ZERO sense to not be interested in the factor that is an ENORMOUS part of every 3D artist's day - improving images, which relies on speed of iteration, which relies in turn on speed of render in the LV to see enough detail, or low enough noise, to evaluate our decisions.