Question Advice needed - 3D graphics workstation build - $ 2,500 - $ 3,000 budget


Junior Member
Aug 15, 2023
Hi, everyone. Forum noob here, looking for advice.

Here’s my deal: I used to be involved in hardware, and know how to put a system together, but my knowledge is ~4 years out of date, and now that I need to put together a workstation, I’m overwhelmed AF with the options. I need it for my design business, which is thriving (which is why I need a better comp) but it’s killing my spare time (which is why I’m looking for advice and shortcuts).

It’s going to be a graphics workstation - Sketchup, Enscape, CorelDRAW, AutoCAD, some Revit. So the focus isn’t on frame rate but on processing power, reliability, and graphics performance.

I have $ 2,500 to spend, can push it up to $ 3,000 if needed.

Maybe allocate $ 500ish of that for a NAS box with RAID 0+1.
Have a monitor I’m happy with, so that’s not a concern.

Physical space is not a problem - I want a huge tower or desktop, with plenty of space. Don’t care about RGB lights, actually prefer the “A/V receiver” blackout look. Not married to any specific brand, so either Intel or AMD, either Nvidia or ATI, doesn’t matter. Whatever works better for the money.

The main dilemma I’m struggling with is “invest in a stronger CPU or max out videocards”? Like, should I spend $ 1,200 on a Threadripper or should I get more GPU power and RAM?

Then, Ryzen vs Threadripper vs Threadripper Pro? Or… what else is out there?

Videocard - according to Sketchup and Revit forums, MOAR VIDEO MEMORY FOR THE GOD OF VIDEO MEMORY is a thing. But where do you stop? 8GB? 10GB? 12GB? At what point does the price increase stop making sense? Again, it’s not for gaming, it’s for keeping track of complicated 3D models.

Should I spend $ 700 on RTX 3080 with 10GB, or get the 3060 8GB and invest another $300 into more (or faster) system RAM?

Should I do 2 less-expensive videocards or 1 maxed-out one?

Is there a way to start with a decent stack of SSD storage and add more later on? Adding more modules in an “array” rather than replacing them with higher-capacity modules? What do people do if they know they’ll have more money to invest in a system later on?

It’s all these kinds of decisions, and it’s overwhelming… any advice, whether general or specific items, would be much appreciated!