OTOY Forums

bepeg4d wrote: Let's say that we have 5 Masters, we need 5 Combo and 10 Standalone license for a total of 100 GPUs.
C4D Team Render will spread your frame to 5 Master machines, while the Masters will spread their rendering to 20 GPUs.
Happy rendering,
ciao beppe

So 5 computers, 5 combo license. Buy why buy 10 standalone license?
I thought 5 computers with 20 gpu each will be covered by the 5 combo license.

@smicha: oh man, amazing, can't wait to see the new monster making of
@geo_n: as I said, it depends from the number of GPUs per node. In my previous example, we have 5 master with 2 slave each one, so we have 15 mascines, 5 Master + 10 Slaves with 7 GPUs in each node.
When the new smicha monster with 20 GPUs is ready, you need only one Combo license per node.
ciao beppe

Goldorak wrote:This is also a much more efficient way to use 100 GPUs for animation render jobs.

bepeg4d wrote:Hi geo_n,
it depends on how many GPUs there are on each slave.
Let's make an example:
With three of 7x GPUs smicha machines you can have 20 GPU for Octane and one for the Master display; for this setup you need one Combo Standalone + Plugin license and 2 Standalone license every three nodes.
Let's say that we have 5 Masters, we need 5 Combo and 10 Standalone license for a total of 100 GPUs.
C4D Team Render will spread your frame to 5 Master machines, while the Masters will spread their rendering to 20 GPUs.
Happy rendering,
ciao beppe

Maybe if you only use Octane (and not even C4Ds native renderers) for every single render job you ever run, or ever will run this approach might work.
In the real world however, it’s not an option. An example:

Say you have 3 jobs in your render queue. Job1 doesn’t use Octane, Job2 uses Octane, Job3 doesn’t use Octane.
With your setup it would require that I constantly have to monitor the queue and as soon as Job1 is finished I have to stop the queue, exclude/shut down any Team Render Client that’s not an Octane master, and reverse the whole process when Job2 is finished. But wait, even that will not work because as soon render clients becomes available they’re gonna start working on Job2. So I would have to run every job until all frames are finished before I can start the next one.
Not much of a render queue left.

So for my 30 slave machines setup we’re back at my earlier post.

Total cost for 1 master and 30 slaves:
Cycles4D - $1600
Thea - $800
Octane - $18000

OK. If you are looking to use Octane on 30 machines on our LAN does this mean you are running 30x high end GPUs? What is the power draw in your studio for this to work with cooling etc.?

Goldorak wrote:OK. If you are looking to use Octane on 30 machines on our LAN does this mean you are running 30x high end GPUs? What is the power draw in your studio for this to work with cooling etc.?

First of all you just admitted that you don't even read all the posts in this thread as I wrote about that on the previous page, like 2 days ago. And no, NOT with Octanes network render. Every client is independent and is assigned frames from C4Ds Team Render Server.

From my previous post:
"

Goldorak wrote:It was much worse in V2 when losing a single GPU killed the whole render, which V3 addressed. If you are going invest in machines to use all 20 GPUs w/ V3, then consider 2x 4U servers w/ 8x GPUs in each node, which will likely max out the power of most offices and provide ideal stability. Our DTLA office is in a former Pacbell building, with huge power supply and an A/C server room; but even, so an 2x4U 16- GPU config is about the limit of what we can reliably power

How is what works in your server room relevant?
Are you on 110 or 240 volt? Assuming you completely maxing your power draw it would mean something like 36 ampere if you’re an 110V but only 16 if you’re on 240. In my server room, which is nothing fancy, I have 4 separate 16A outlets and 4 10A and I’m on 240V so that’s 25 kW, enough for 135 GTX1080."

As of now I have 10 kW cooling, but the water pipes has capacity to hook up an additional 5-10 kW of cooling.

preciousillusion wrote: Total cost for 1 master and 30 slaves:
Cycles4D - $1600
Thea - $800
Octane - $18000

Does cycles4d come with some free render nodes? If not, how much are the render nodes?

Edit. I see now its 185euro for single license and 60 for render nodes. What's the gpu limit per render node and the single master license?

Edit 2. The single license included 3 render nodes. Wow! Definitely going to check it out.
http://www.cgchannel.com/2016/11/insydi ... cycles-4d/

geo_n wrote:

preciousillusion wrote: Total cost for 1 master and 30 slaves:
Cycles4D - $1600
Thea - $800
Octane - $18000

Does cycles4d come with some free render nodes? If not, how much are the render nodes?

Edit. I see now its 185euro for single license and 60 for render nodes. What's the gpu limit per render node and the single master license?

Edit 2. The single license included 3 render nodes. Wow! Definitely going to check it out.
http://www.cgchannel.com/2016/11/insydi ... cycles-4d/

I dont think there's any limit on gpus, I haven't heard heard of anyone else imposing such limits.

preciousillusion wrote: Maybe if you only use Octane (and not even C4Ds native renderers) for every single render job you ever run, or ever will run this approach might work.
In the real world however, it’s not an option. An example:

Say you have 3 jobs in your render queue. Job1 doesn’t use Octane, Job2 uses Octane, Job3 doesn’t use Octane.
With your setup it would require that I constantly have to monitor the queue and as soon as Job1 is finished I have to stop the queue, exclude/shut down any Team Render Client that’s not an Octane master, and reverse the whole process when Job2 is finished. But wait, even that will not work because as soon render clients becomes available they’re gonna start working on Job2. So I would have to run every job until all frames are finished before I can start the next one.
Not much of a render queue left.

So for my 30 slave machines setup we’re back at my earlier post.

Total cost for 1 master and 30 slaves:
Cycles4D - $1600
Thea - $800
Octane - $18000

Hi preciousillusion,
if you insist to use only c4d Team Render, you will never take advantage of OctaneRender network rendering capabilities, and I wonder why you are so interest in the Octane Slave licence scheme if you will always need a combo license.
Since your slaves have both CPUs and GPUs, the beaty of OctaneRender network rendering is that you can use the GPUs of the slaves for Octane scenes, while rendering on CPUs via Team render your standard c4d scenes.
The point is not Team render against Octane net render, but to use them together.
All the best,
ciao beppe

bepeg4d wrote:

preciousillusion wrote: Maybe if you only use Octane (and not even C4Ds native renderers) for every single render job you ever run, or ever will run this approach might work.
In the real world however, it’s not an option. An example:

Say you have 3 jobs in your render queue. Job1 doesn’t use Octane, Job2 uses Octane, Job3 doesn’t use Octane.
With your setup it would require that I constantly have to monitor the queue and as soon as Job1 is finished I have to stop the queue, exclude/shut down any Team Render Client that’s not an Octane master, and reverse the whole process when Job2 is finished. But wait, even that will not work because as soon render clients becomes available they’re gonna start working on Job2. So I would have to run every job until all frames are finished before I can start the next one.
Not much of a render queue left.

So for my 30 slave machines setup we’re back at my earlier post.

Total cost for 1 master and 30 slaves:
Cycles4D - $1600
Thea - $800
Octane - $18000

Hi preciousillusion,
if you insist to use only c4d Team Render, you will never take advantage of OctaneRender network rendering capabilities, and I wonder why you are so interest in the Octane Slave licence scheme if you will always need a combo license.
Since your slaves have both CPUs and GPUs, the beaty of OctaneRender network rendering is that you can use the GPUs of the slaves for Octane scenes, while rendering on CPUs via Team render your standard c4d scenes.
The point is not Team render against Octane net render, but to use them together.
All the best,
ciao beppe

Octane network render is great for single frames. For animations, not so much.
Let's say you have a scene with 20 seconds preparation time and render 1000 frames.
Total render time with Octanes network render will be several hours longer vs distributed over 30 clients.

When it comes to slave licenses I of course assume they will work in a somewhat standard fashion, i.e. you only need one license per slave regardless of using Octanes network render, C4D Team Render, Backburner, Deadline or whatever.

Almost 6 days since I replied to you Goldorak.