Possibly the bandwidth of those expansion slots is to low and is limiting the cards. Splitting up a PCI-E 3.0 x 16 Lane in 2 8x or 4 4x Lanes should work just fine without much speed loss. Though if you are running 4 cards on a 2.0 x 16 Lane that could be the root of your problem.

You should really look into the Standalone GUI first, the C4D plugin is still in development and gets lots of improvements on a daily basis. Learning how Octane was intended to be used will help you a lot and you should be able to understand all the important basics within a few days.
The C4D plugin ...

Is someone using Octane on Linux?
I'm curious if the performance with the latest proprietary Nvidia Linux drivers is worse than with the latest Windows drivers.

+1
saves a lot of time

Is shared GPU memory still not possible with the latest CUDA and Compute Model version?
PCI-E 3.0 has a max bandwidth of about 64GB/s, that would definitely be enough to have apropriate render times over multiples cards.
Slower than having everything on every GPU but it would enable larger scenes ...

interesting to say a least! though somehow I believe it will not be available for mases but rather for big companies only..'cos of steep price tag for renting a service like that..so the future might be here..for some.. /sorry for the sad thoughts/

cheers

This is a Javascript based video codec ...

I think he wants a procedural texture like the noise filters that are available, not how to achieve a fabric material.
+1 to that

I gotta say the render priority isn't helping at all right now. The rest of the system is still lagging like hell :/

The render prioity setting is great thing, though it seems this only affects Octanes UI. Would it be possible to set the priority for the whole executable so other programs don't slow down so much? This is really one of the most annoying things yet with Octane. Most of the time you can't use your ...

If its not already planned. I'd like to request a subscription plan for daily usage. Even with just one GPU it would be quite expensive with 1$ per GPU to use 8h per day.