Anyone know a rough estimate of cost to setup back burner?
Or is there a way to use a local machine and render to AWS? Or is that daft?
B = the cost of Burn
L = labor, in hours
M = rate, in $USD
C = the cost of a computer
B = (L * M) + C
Where L = 1
Let’s be clear here… Backburner is just the render manager software, and that is free. What you are asking for is Burn, which is also free. If you can install flame, you can install burn. It just comes down to the cost of render node machines. And unless you have Flame in AWS, then no, you can’t offload render to there, unless you want to create a world of hate for yourself.
2 Likes
Ok. Thanks @ALan - a decent size tender farm for 7 flames to run burn on.
And thinking out loud, might it be more efficient to store media /renders on a server rather than local frame stores? I am wondering whether this would save export time to the farm. I have not programmed computers since using a commodore pet nigh on 40 years ago. Nor built a computer since pentium 3 days.
1 Like
Hi John,
I’ve made a few Burn render farms and they’re incredibly exciting to use. A little finicky as well. My long view on the Burn idea is that as the Flame hardware gets faster, the Burn nodes become less useful so it’s a short term solve to an immediate problem. Or it’s a long term commitment to keeping it current.
But I if you think you need the render horsepower and can afford to write it off on the CURRENT job you’re working on, then it’s a strong consider. But then again, you should be committed to troubleshooting or have someone that can jump in when things inevitably go sideways setting it up.
I made a decision graph that might help:
BurnRubric.pdf (26.9 KB)
1 Like
Burn in my experience scales very poorly. With the dual beefy gpus in the box where the images live it is very dissapointing sending things across the network and the more nodes you add the more the network hurts.
1 Like
True… Burn does not scale linearly. Also, several render inefficiencies within the Burn render engine that makes it a poor solution. Spend the money to get the most powerful workstation possible and render local.
2 Likes