NVIDIA RTX PRO 6000 Blackwell Benchmarks & Tear-Down | Thermals, Gaming, LLM, & Acoustic Tests

NVIDIA RTX PRO 6000 Blackwell Benchmarks & Tear-Down | Thermals, Gaming, LLM, & Acoustic Checks

YouTube video

The RTX PRO 6000 is the very best advertising for AMD’s Ryzen 9800X3D (learn our evaluation) that we use in our GPU take a look at benches, largely as a result of we’re seeing 5% to 14% scaling in gaming benchmarks versus the RTX 5090 (learn our evaluation). We additionally tore it down and located essentially the most insanely dense PCB we’ve ever seen with a reversion away from liquid metallic.

However this doesn’t have 96GB of VRAM simply because that’s the place all of it went after they misplaced it on the RTX 5080 (learn our evaluation), 5070 Ti (learn our evaluation), 5070 (learn our evaluation), 5060 Ti 8GB (learn our evaluation), and 5060 (learn our evaluation)– it’s additionally there as a result of it’s higher for… AI.

Right now, we’re benchmarking a few of these in a spherical of experimental benchmarks. We haven’t completed machine studying and LLM assessments earlier than and we don’t have a full methodology outlined but, however we’ve begun experimenting.

NVIDIA RTX 6000 Overview

This card will not be for gaming. That a lot is apparent. We’re nonetheless testing it in video games, and that’s largely for our personal functions: We need to understand how a lot room the 9800X3D has in it to scale for GPUs, and happily, it seems to be like there’s nonetheless lots for one more era of opinions. The 9800X3D retains up with the RTX PRO 6000 even when at 1080p in some conditions. 

The most important distinction between the PRO 6000 and the 5090 after the 96GB vs. 32GB reminiscence capability is the GPU itself: The PRO 6000 has 24,064 CUDA cores to the 21,760 of the 5090, practically an 11% improve. That’s an enormous distinction.

There are a number of variations of the RTX 6000. The one we’re taking a look at particularly is known as the Blackwell Workstation Version, however NVIDIA additionally has so-called “server editions” of the playing cards, and people are meant to enter servers and racks the place they’re getting power fed air from the entrance of the chassis moderately than utilizing an FE model cooler, which is the one we’re taking a look at in the present day. So, the one we’re taking a look at is extra just like what we’d see in form of the buyer to prosumer workstation class.

NVIDIA RTX 6000 Specs, Structure Fundamentals, and Worth

Entering into the specs:

The RTX PRO 6000 Blackwell GPU is constructed on the GB202 Blackwell die. The total GB202 die has assist for as much as 24,576 CUDA cores, 768 TMUs, 192 ROPs that may even be current, and can be used for the RTX 5090.

The PRO 6000 variation on this GPU has 24,064 CUDA cores, so it’s down by 512 and isn’t an ideal die, with 752 TMUs (down from 768) and nonetheless 192 ROPs, they usually’d higher all be there for the worth. The cardboard can be working 188 RT cores and helps PCIe 5.0 x16. Marketed clocks are 1590 MHz base and 2617 MHz enhance. The actual purpose folks purchase this card is its reminiscence, although, at 96GB of GDDR7.

The closest client class card to check in opposition to is the RTX 5090. It’s constructed on the identical die, however with 176 ROPs, 680 TMUs, 170 RT cores, 21,760 CUDA cores, 32GB of GDDR7 reminiscence, a 512-bit bus width, and a far lower cost at $2,000 to, realistically, $3,000 moderately than $8,000 to $11,000.

For some notable variations: The 5090 solely has a 3rd of the reminiscence capability of the PRO card, and the RTX PRO 6000’s die is nearly 98% current whereas the RTX 5090’s die is nearer to 88.5% current, with the rest both being fused off or faulty fallout and fused off. 

You would additionally technically purchase 4 5090s for the worth of 1 RTX PRO 6000, so the use case actually does come all the way down to having as a lot reminiscence as potential hooked up to a single GPU and PCIe slot.

NVIDIA RTX 6000 Tear-Down

The RTX Professional 6000 we examined is a 2-slot card with 2 giant followers on the entrance. The PCB is within the middle of the cardboard and you may see the place the fins begin to get taller. This means the place the PCB plate ends. 

The cardboard’s PCIe slot is related by way of a personalized pinout adapter. 

When you have a look at the bottom of the cardboard the place the followers are, you’ll discover a slight melancholy down in direction of the center of the fins. That is supposed to assist with strain drop. The design is full flow-through, which pushes air straight by way of the cardboard moderately than right into a PCB. 

The opposite aspect of the cardboard seems to be similar to a 5090. 

The highest of the RTX Professional 6000 retains the angle on the cardboard’s vented slats. This initiatives the air out at an angle that’s up and away from the cardboard. 

Just like the 5090, the cardboard makes use of a 12VHPWR cable.