Pro Graphics - 3XS Graphics Workstations for Redshift

What is Redshift?

Redshift is a biased rendering software application that artists and designers use to render scenes. Redshift supports a wide variety of modelling applications such as 3DS Max, Maya and Cinema 4D to name just a few.

What makes the best workstation for Redshift?

Redshift renders exclusively on NVIDIA GPUs, using the massively parallel nature of their architecture to speed up this process, as it would impractical to perform the same calculations on CPUs.

What CPU should I choose for Redshift?

As already mentioned Redshift performs best on NVIDIA GPUs, so CPU performance is less important than it used to be on a dedicated rendering system. The CPU does have a role to play in scene loading, but typically you should consider a CPU with as many PCIe lanes as possible, such as AMD Ryzen Threadripper PRO or Intel Xeon W in a workstation or AMD EPYC or Intel Xeon in a server, enabling the GPUs to communicate more efficiently. If you’re also planning on modelling on the same workstation then an AMD Ryzen 9 or Intel Core Ultra 9 is a cost-effective choice as they have high clock speeds which is important for these workflows.

What GPU do I need for Redshift?

As we’ve stated Redshift is a software application that relies on GPUs for the majority of tasks so the more powerful GPU you have the less time renders will take to complete. Like many dedicated rendering engines Redshift scales very well across multiple GPUs. This means you can scale the rendering ability in accordance with your budget. For example, a workstation with between one or four GPUs will provide cost effective performance for small to mid-size projects while one or more servers, each with up to eight GPUs, either hosted locally or provisioned in the cloud with NVIDIA vGPUs may be required for larger projects.

If you are considering a dedicated workstation for Redshift then we recommend consumer-grade NVIDIA GeForce RTX cards thanks to their outstanding value for money. However, if you’re also planning on modelling on the same workstation then it’s worth considering a professional-grade NVIDIA RTX GPU thanks to their larger memory buffer and certified drivers. For the same reasons we recommend professional grade GPUs in rendering servers.

How much RAM is recommended for Redshift?

The minimum RAM that we would recommend for Redshift is 32GB. As the application renders on GPUs, the amount of system RAM is not too important, but if you’re also planning on modelling on the same workstation then 64GB or more memory may be beneficial.

What storage does Redshift need?

We always recommend using an SSD for the boot drive that will host your OS, Redshift itself, and any active projects you are working on. The high speed of SSDs allows your system to boot, launch applications, and load files many times faster than any traditional hard drive. Alternatively, as your work projects increase you may want to consider a second SSD or higher capacity HDD - depending on your budget - so your OS, Redshift and live projects benefit from SSD speed, but your archived work uses the HDD where speed isn’t a necessity and costs can be lowered. Another archive solution could be to employ an external NAS device allowing sharing and RAID protection.

When it comes to deciding on drive capacity, this will very much depend on your number of files and the model sizes. You should also consider the space required by other applications and their associated files too.

Recommended Redshift Workstation Specifications

	High-End	Mid-Range	Entry-Level
CPU	AMD Ryzen 9 / Intel Core Ultra 9	AMD Ryzen 7 / Intel Core Ultra 7	AMD Ryzen 5 / Intel Core Ultra 5
RAM	64GB	32GB	32GB
GPU	NVIDIA RTX PRO 4000 Blackwell	NVIDIA RTX A1000	NVIDIA RTX A400