Hi Guys,
I would like to buy a render manager to manage a small farm (max 10 machines).
I’m on Windows 10 and my candidate are :
I would be glad to hear some advise of people who have to make the same choice.
Price / simplicity / reliability are the 3 most important factor of the equation.
Royal Render and Muster looks to give a good balance for price / perf at the moment ?
What do you think ? 
Cheers
E
I only have experience with Deadline (as a user, no idea about price.) Works out of the box for simple rendering, and extensible to do anything you want if you have scriptors/programmers dive into it.
You might want to Give Puli a try. It’s open source from Mikros Image. It has it’s quirks but it get’s the job done.
For paid solutions i like Rush , small, fast and easy to install and manage, writen in perl and rock solid.
I’ve used Royal Render in farms of 10 nodes, 40 nodes and 150 nodes and it’s always been solid, with a great feature set.
That said have also been wanting to take a look at deadline as they’ve had per gpu job rendering for a while now.
Tried Muster and Deadline, and Deadline is a lot better. In fact back when I tried it Muster was a pain in the ass to get working.
Deadline worked out of the box with minimum effort.
if you’re on a single platform (windows) and just a few nodes, any of them should work fine.
Where it really starts making a difference is working on a lot more nodes and on multiple platforms.
From what I understand and have seen, Deadline is the biggest, heaviest of the bunch in terms of how many resources, cpu cycles, and network traffic it uses to run and manage all its database and statistics stuff.
We’re using Smedge here on windows workstations and linux render farm. Not that it means I’m endorsing it, because it can be finicky at times. It’s lightweight and fast with far fewer features, though we’ve ever really needed any databases or stats. The things we’ve wanted have actually made more sense to script linux to do in conjunction with our FTP rather than relying on the render manager to do it by putting more strain on the network and taking away from rendering CPU cycles.
Smedge is almost half the price as Deadline not to mention MUCH cheaper to upgrade (less than 27%) which becomes important when you start adding more nodes down the road.
Many thanks for all your input guys.
Deadline looks great , but too expensive for what i need.
Smedge looks nice but i don’t like the UI so much.
Muster doesn’t mention explicitly houdini.
At the moment Royal Render attract me the most
I’m using HQUEUE at the moment but i am not that impress by the tool …
More feedback on Royal Render in real production scenario would be cool !
Cheers
E
Royal Render seems pretty solid and has reasonable pricing
I agree about Smedge’s look. I can’t stand Smedge’s candy-looking icons and turn them off ASAP. All their functions are right there in the menus anyway and I’d rather have the additional screen space for actual job info.
I’ve tried all those except Royal Render and Deadline is the easiest. Smedge isn’t bad either.
Also I heard Butterfly Render is good also.
You can use for other app as long as they have commandline for rendering.
I used to render Fusion, Nuke and Terrangen with it.
We had 20 ws and 50 render nodes and did not have much problem.
Royal Render is great, I used it at some studios. Smedge was nice too, but RR had some features that made it a bit more comfortable.
We use Renderpal here. It’s pretty solid and the pricing scheme is very attractive.
I can’t say much about Muster, but when we used it the last time (YEARS ago) it was a pain.
for me GPU rendering is an important part…
Deadline GPU AFFINITY SETTINGS FOR SLAVES
GPU affinity can now be set for Slaves, in the same way that CPU affinity can be set. These GPU affinity settings can be used by Deadlines application plugins to set the GPUs to use for the render if the jobs renderer supports it. Currently, the GPU affinity settings are only used by the Maya and Softimage plugins when the RedShift renderer is being used.
For example, if you have two Slaves running on a machine with 4 GPUs, you could set their affinity so that one Slave uses GPUs 0 and 1, and the other Slave uses GPUs 2 and 3. When these Slaves render RedShift jobs, they will pass the GPUs they want to use to RedShift so that each render only uses 2 GPUs.
generally i would say it depends on the criteria you have. Regarding my criteria…
…i changed from royal render to deadline, because documentation and premium support was way better from thinkbox (for me its very important, because having a showstopper in my pipeline would keep me away from sleeping well). Also customization with configfiles and scripting in python is currently better in my opinion.
If this isn’t important to you, i would go with royal render first. In terms of scalability both are good enough!
Hope this helps a little bit!
Thanks for all your input guys !
for the moment i think i will try royal render as feedback are globaly very good.
deadline looks more advanced but i’m not sure i’m ready to pay higher price for just a small farm.
Cheers
E