Jump to content

Hey everyone,

 

I would like to ask you guys for some tips on my pc-build. The rig will primarily be used for 3D architectural rendering and animation. Besides that, it will function as workstation and renderer at the same time.

At the moment I do most of the rendering CPU-based, but I want to build a rig that also has enough GPU-rendering power.

 

Software I currently use for CPU-based rendering are: Sketchup, Cinema4D and 3Ds MAx in combination with Vray.

Software I use for GPU-based rendering and gaming engines are: Octane in Cinema4D, Lumion and Unreal Engine 4.

My budget for the whole system is between $2500 and $3000.

 

I set up the following build mostly based on information I got through linustechtip video's. (Many Thanks)
http://pcpartpicker.com/list/H6WCD8

Dual CPU - Intel® Xeon® Processor E5-2620 v4
Dual HSF - Be Quit! Dark Rock Pro 3 
GPU - Asus Geforce GTX 1070
Motherboard - Asus Z10PE-D16 WS SSI EEB Dual-CPU LGA2011-3 
RAM - Curcial 32 GB (2x16GB) DDR4

SSD - 512GB Samsung
HHD - 4TB Hitachi Deskstar


Powersupply - 700W 80+ Titanium fully modular

Case- Phanteks Enthoo Pro Full tower

 

Please let me know if you see improvement in this build.

Thanks!

 

Link to comment
https://linustechtips.com/topic/695559-3d-architectural-rendering-rig/
Share on other sites

Link to post
Share on other sites

5 minutes ago, Dawson Wehage said:

There going to probably get really hot, I don't think air cooling is worth it

Air Cooling beyond the stock cooler will be more than sufficient. 

For the Best builds and Price lists here is a world where many points of the price have been predefined already for your convenience!

The Xeon E3 1231 V3 IS BETTER Than the Core i5 4690K and a Significantly better value for the non-overclockers or value shoppers.

The OS is like a kind food, Try it before saying if you like it or don't.

Link to post
Share on other sites

Id get a 6900k. It would be about the same speed in rendering(core scaling isn't perfect), but much faster in the view port, which is single threaded, 

 

Id only go xeons if you willing to spend a bit more for something like a 2650 v4

Link to post
Share on other sites

14 minutes ago, Electronics Wizardy said:

Id get a 6900k. It would be about the same speed in rendering(core scaling isn't perfect), but much faster in the view port, which is single threaded, 

 

Id only go xeons if you willing to spend a bit more for something like a 2650 v4

As far as I know the most important for CPU rendering is more cores, base frequency isn't so essential. Apart from that i found that xeon's are server processors that can handle many heavy workloads at the same time because its a server processor. So going for an 6900K or 2650 v4 isnt efficient for my type of workload i think..

Link to post
Share on other sites

Just now, JL-render said:

As far as I know the most important for CPU rendering is more cores, base frequency isn't so essential. Apart from that i found that xeon's are server processors that can handle many heavy workloads at the same time because its a server processor. So going for an 2650 v4 isnt efficient for my type of workload i think..

Its not cores times clockspeed. The i7's are much higher clocked.

 

The i7 and the xeon's are the same die, so they won't handle the workloads any differently.

Link to post
Share on other sites

49 minutes ago, Electronics Wizardy said:

Its not cores times clockspeed. The i7's are much higher clocked.

 

The i7 and the xeon's are the same die, so they won't handle the workloads any differently.

Okay okay, but why would you prefer a single CPU (more expensive) with less cores over dual CPU with more cores?

Link to post
Share on other sites

26 minutes ago, JL-render said:

Okay okay, but why would you prefer a single CPU (more expensive) with less cores over dual CPU with more cores?

There is much more than core count, that i7 is clocked much higher. You also have the problem of higher system latency and numa nodes.

 

Many tasks like single threaded performance. Most all of the viewport is single threaded, Web browsers are single threaded. 

 

That i7 is much higher clocked, and those programs won't use all those cores perfectly. Doubling the cores won't double the performance.

Link to post
Share on other sites

I am an architect and I have read that the ideal PC (in this case) is one where the CPU and the graphics card are perfectly matched, so that neither of them becomes a bottleneck for the other, what matters is, that the processor is fast enough To carry data to the graphics card. Generally the configuration you have is more oriented to different uses within the world of design as the modeling of characters, industrial objects, etc, in which case your pc would make more use of multi-processors because they are tasks with more complex calculations, but For rendering in applications like lumion and the vray rendering engine, it take advantage of the GPU and its CUDA Cores, in addition the CPU has a smaller impact on render times as it's mainly used for compressing videos and images

 

Nowadays it is more feasible for architects  build a gaming pc, but with a somewhat higher configuration such as a good i7, such as the Intel Core i7-6800K, a good GTX like the 1070 or even a 1080 that you can afford. If you want it just for render and animation I would go for I7, but in case you´re going to use other applications that really take advantage of the multiprocessors and HyperThreading would go for a wokstation with Xeon with 8 cores like the one you propose.

Link to post
Share on other sites

12 hours ago, Luien said:

I am an architect and I have read that the ideal PC (in this case) is one where the CPU and the graphics card are perfectly matched, so that neither of them becomes a bottleneck for the other, what matters is, that the processor is fast enough To carry data to the graphics card. Generally the configuration you have is more oriented to different uses within the world of design as the modeling of characters, industrial objects, etc, in which case your pc would make more use of multi-processors because they are tasks with more complex calculations, but For rendering in applications like lumion and the vray rendering engine, it take advantage of the GPU and its CUDA Cores, in addition the CPU has a smaller impact on render times as it's mainly used for compressing videos and images

 

Nowadays it is more feasible for architects  build a gaming pc, but with a somewhat higher configuration such as a good i7, such as the Intel Core i7-6800K, a good GTX like the 1070 or even a 1080 that you can afford. If you want it just for render and animation I would go for I7, but in case you´re going to use other applications that really take advantage of the multiprocessors and HyperThreading would go for a wokstation with Xeon with 8 cores like the one you propose.

Thank you very much for your input!

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×