Don’t worry, Nvidia’s RTX 6000 workstation GPU isn’t designed for gamers – so we’re not dealing with a new and particularly frightening chapter in the history of soaring gaming graphics card prices. But just under 8,000 dollars according to a first dealer listing is also a real announcement for a professional graphics card, which is flanked by extreme technical data.
The RTX 6000 is based on the Ada Lovelace architecture, which is also the official product page prominently emphasized. The same architecture is already used in the new gaming GPUs of the RTX 4000 generation, which also includes the much-discussed RTX 4080. You can find out more about this here:
Column on the RTX 4080
In the end, the customer has the upper hand
The fact that the RTX 6000 relies on the identical architecture as the RTX 4000 GPUs enables a direct and technically quite interesting comparison with Nvidia’s current Geforce top model, the RTX 4090. From our point of view, it is particularly noteworthy in three categories:
- The RTX 6000 relies almost entirely on the AD102 chip and is proud of it 18,176 units of account (RTX 4090: 16,384 arithmetic units).
- The memory is saturated with 48 GB very large (RTX 4090: 24 GB). At the same time, it works a bit slower at 960 GB/s (RTX 4090: 1,008 GByte/s), but it offers ECC error correction, which is important for professionals.
- The TDP falls with it 300 Watt comparatively low from (RTX 4090: 450 watts). This suggests that the clock rates are a little lower than in the gaming flagship, which is not unusual in the professional segment.
Some links included on this page are affiliate links. Depending on the provider, GameStar receives a small commission for purchases made via these links without affecting the price.
More info.
RTX 6000 vs. RTX 4090: Technical data at a glance
| RTX 6000 | RTX 4090 | |
|---|---|---|
| units of account | 18.176 | 16.384 |
| Tensor-Kernel | 568 | 512 |
| Raytracing Core | 142 | 128 |
| Storage | 48 GB GDDR6X with EEC | 24 GB GDDR6X |
| memory bandwidth | 960 GByte/s | 1.008 GByte/s |
| TDP | 300 Watt | 450 Watt |
| Theoretical computing power (FP32) | 91.1 Teraflops | 82.6 Teraflops |
| approx. price | that. 7,900 dollars | that. 2,300 dollars |
The Europe dealer Delta-Computer currently sets a price of 7,895.71 dollars for the RTX 6000, as reported by Heise.de. However, the naming can lead to confusion.
This is how the predecessor model with Ampere architecture is named RTX A6000
at a price of about 5,000 dollars, while also their predecessors with Turing architecture and the name Quadro RTX 6000
can still be found in stores for around 4,300 dollars.
Fortunately, since these are professional cards, none of this plays a major role for us players, but it can still be interesting to take a look at this segment, both technically and in terms of price. By the way, you can find out how such graphics cards perform in gaming in this article:
How does Cyberpunk 2077 play with a graphics card for almost 6,000 dollars?
What do you think of the professional graphics cards from Nvidia? Technically interesting GPUs with fully developed chips, completely irrelevant for gamers or something in between? Feel free to write it in the comments!
