Everything we know about the new Nvidia GeForce RTX 4080 GPU

Everything we know about the new Nvidia GeForce RTX 4080 GPU ...

We''re just off to the start of Nvidia''s ''Lovelace'' RTX 4000 graphics cards, which will be on September 20 and will feature the Nvidia GeForce RTX 4080. As of today, three GPUs will be expected, although this does not include the RTX 4070.

The flagship RTX 4090, along with two different flavors of the RTX 4080, one with a 12GB of GDDR6X VRAM, and a full-fat version with a larger 16GB capacity, should be included.

The Nvidia GeForce RTX 3080 was such an incredible GPU, and it has introduced 4K gaming to the mainstream PC space. However, it has the potential to be one of the best Nvidia graphics cards on the market.

As you continue to scroll, much of what we''ve''know so far is actually based on leaks, speculations, and reasonable logic extrapolation. There''s no way to know until September 20, so we await with a breath of calm.

Cut to the chase

  • No official release date yet, should be revealed September 20
  • Current leaks put on-sale date in October or November
  • Unclear whether new cards will launch at the same time

Debate has simmered for weeks on when the RTX 4080 will be released, even when it comes to the Nvidia presentation where it will be revealed. The main one is from famous Twitter leaker @kopite7kimi, who claimed that Lovelace, the architecture used by the RTX 4080, would be launched "a little earlier."

This was based on previous claims that the next-generation graphics card would be introduced at the end of 2022. Coupled with a different leaker previously that the RTX 4080 will be released in October, with the 4080 expected shortly thereafter in November.

RTX 2080, for example, was launched in September 2018, about two years after the GTX 1080''s may launch in May 2016. And, on the other hand, the RTX 3080 was almost two years on the dot. So, if Nvidia is going to keep its regular cadence with its next GPUs, we can assume that current leaks are accurate and at least one RTX 4000 card will be launched in October 2022.

  • Very little info on pricing at this stage
  • Current-gen RTX 3080 originally retailed at $699
  • Two different models make guessing prices tricky

Back when we first reviewed the Nvidia GeForce RTX 3080, we applauded it for seeing a significant improvement in performance at a significantly lower price point than the RTX 2080 Super. Evidently, GPU prices were at some frankly outrageous prices, so that point is probably double-edged.

However, the crypto problem erupted, as the value of cryptocurrencies plummeted and Ethereum''s great ''Merge'' saw conventional mining become redundant. Most current-gen GPUs are now available at (or near) MSRP once more as the market stabilizes.

RTX 3080 had an MSRP of $699 (649) when it first came out, and generational pricing tends not to shift too much when Nvidia is concerned, so we may expect a price tag somewhere in the region. Pricing information has yet to be revealed (or leaked), so this is entirely pure speculation on our part.

While two versions of the RTX 4080 will be available, pricing will be difficult to accurately anticipate. While it''s a relatively safe bet that one of these two versions will stick around the 3080''s $699 retail price, which one will be complete guesswork at this stage.

Our personal prediction, based somewhat on the piggy-bank price rumors around the RTX 4090, is that the 12GB model will cost $699, while the 16GB version will be around the $800-900 mark. We''ve hoped the cards are cheaper than the one, but we''re still trying to be realistic with our expectations.

  • 16GB model will feature 9728 CUDA cores, 2,505MHz boost clock, 256-bit memory bus and memory bandwidth of 736GB/s
  • 12GB model will feature 7680 CUDA cores, 2,610MHz boost clock, 192-bit memory bus and memory bandwidth of 504GB/s
  • Power demands won''''t be as severe as originally feared

The 12GB version of the RTX 4080 was, in fact, used as the RTX 4070 or 4070 Ti, although it is thought to be a slightly different central GPU (though still manufactured on Nvidia''s 4nm Ada Lovelace process).

Given that it was previously mistaken for a lower-tier card, it unsurprisingly falls under the 16GB version in just about every area, not just VRAM. Besides, there isn''t less memory, it''s less powerful, with a smaller bus and a bandwidth of 504GB/s, as opposed to the 16GB model''s 736GB/s.

The 16GB 4080 also has faster memory than even the RTX 4090, with a speed of 23Gbps compared to 21Gbps for the other two cards. It''s worth noting that both the leaked versions of the RTX 4080 actually have a slightly reduced memory bandwidth than the original RTX 3080, but the memory is faster across the board, and that card only had a 10-GB of VRAM to begin with.

With Lovelace, the previous-generation Ampere architecture did not provide enough CUDA cores. This means that the 16GB model of the RTX 4080 is less expensive than the 3080, but the 12GB model does not. Nvidia''s name for graphics processor cores is, however, more.

When it comes to power requirements, there has been a lot of worry about these cards being serious power-hungry. Historically, leaks pointed to power requirements upwards of 650W, but it now appears like the 16GB RTX 4080 will have a much greater base TGP of 340W (285W for the 12GB model), but when it comes to overclocking, the ceiling appears to be 516W.

Overclocking is a problem for these two GPUs: 2,505MHz and 2,610MHz. Surprisingly, the faster card is actually the 12GB version, although this is because the beefier 16GB model will need to throttle sooner to avoid overheating. In either case, these are massively higher than the original 1710MHz boost clock of the RTX 3080.

  • Quad-slot cooling design with three fans as standard
  • No leaked images of Founder''''s Edition cards yet
  • Expected to be larger and heavier than anything we''''ve seen yet

It''s impossible to discern at this point which image leaks of previously assumed RTX 4000-series GPUs are genuine - and they so far have all been focused on the flagship RTX 4090. Our US computing editor was somewhat suspicious of the fonts used on some supposed Zotac Gaming RTX 4090 packaging.

Despite the fact that the RTX 4080, which will presumably incorporate the same new quad-slot design as in leaked RTX 4090 pictures, is said to be powered by four fans (a standard three-fan design, with an extra fan attached to the rear) and a meaty heat sink.

We''ve yet to see Founder''s Edition (FE) versions of these cards leak right, so the exact design of these is unclear at this point (although various renderings have been circulating).

If you think that Gigabyte GPUs would be superior to MSI ones, then proceed with your gut. Either way, these new GPUs will cause problems for PC manufacturers, since they''ll be bigger, heavier, and more complex than we have ever encountered previously.

Remember that everything we''ve discussed here is pure conjecture at this point; we''ve collated our most trusted and believable leaks to form this report, but there''s no way of confirming which information is correct and which is hogwash. We''ll be back to this page on September 20 once we have all of Nvidia''s news!