Nvidia’s RTX 4070 might not be among the first batch of high-end GPUs released and, if the rumor is correct, we’ll have two different rounds on the RTX 4080 coming, as well as the flagship RTX 4090.
The Vine’s latest talk about Lovelace comes from several sources that VideoCardz (opens in new tab) spoke, combined with information pulled from the Kopite7kimi hardware leaker on Twitter (which recently summed up (opens in new tab) all your latest discoveries), and chipell (opens in new tab) (Chinese forums), which corroborates VideoCardz’s findings in some places.
There’s some new information about clock speeds and power usage, but the biggest reveal here is that Nvidia appears to be planning to release two versions of the RTX 4080, one with 16GB of VRAM and one with 12GB – but there will be other big differences with the spec (take it all with big piles of spice, naturally, as in any rumor).
The 16GB RTX 4080 is what is said to be built on the AD103 GPU with 9,728 CUDA cores, with the 12GB spin on the 4080 reportedly dropping to use the smaller AD104 GPU, and will have 7,680 CUDA cores – a big difference.
As VideoCardz notes, what appears to be the case is that the RTX 4080 12GB is what was previously identified in the rumor mill as the RTX 4070 (or 4070 Ti) – which explains why it looked so beefy, something we’ve talked about in recent times.
VideoCardz informs us that Nvidia has made a “last minute name change” to this card to turn it into a smaller spin of the RTX 4080, and right now, there’s no RTX 4070 – with graphics card manufacturers not expecting that. that will change (with the release of Lovelace now so close: it’s slated for September 20 at GTC).
Other spec nuggets shared in this new rumor include that the RTX 4080 12GB will have a power usage of 285W (TGP or Total Graphics Power), which can be increased to 366W. The boost clock will supposedly be 2610MHz.
The stronger version of the RTX 4080 with 16GB of VRAM is supposed to increase to 2505MHz, with a TGP of 340W (able to be increased to 516W maximum). GDDR6X memory is said to run at 23 Gbps remarkably, and this will be the only GPU to have that fast VRAM.
As for the RTX 4090, that increases to 2520MHz, and the rumor again points to the TGP of 450W, but at most it can be pushed to 660W. The 24GB of video RAM will be 21Gbps modules, the same as the smaller 12GB RTX 4080.
The tech site tells us that the RTX 4090 is due out in the first half of October, and the RTX 4080 models are apparently due to debut in the first two weeks of November.
Review: RTX 4070 ETA?
So what about the RTX 4070? Well, we heard a few weeks ago – again from Kopite7kimi on Twitter – that there were two possible specs for the 4070 graphics card. The high end possibility mentioned now is the RTX 4080 12GB, in theory, so that would leave the other configuration of an AD104 chip with 10GB of VRAM and 7,168 CUDA Cores to be the RTX 4070. Maybe. (Let’s face it, the Lovelace rumor space is getting more and more confusing.)
If that’s what’s happening, it’s going to be disappointing for a few reasons. Firstly, the RTX 4070 with only 10GB of video memory will not go down well, and secondly, the prospect of arriving late at the table is also off-putting. When will you show up? If there’s still no vision of that, is this a GPU that Nvidia plans to delay until early 2023, as some rumors have suggested?
Before we get too disheartened, what we don’t have yet is any price indication for Lovelace outside of the RTX 4090 (we’ve just heard some hints about the flagship’s pricing, and unfortunately the news isn’t what we wanted to hear).
If we’re not getting an RTX 4070 initially, but we’re getting an RTX 4080 12GB that is relatively well priced for its performance – considering it drops quite a bit from the 4080 16GB in terms of a smaller GPU and considerably fewer CUDA Cores – then maybe it doesn’t matter. . Which is to say, it’s all about the price/performance ratio of these GPUs, whatever they’re called.
That Nvidia has reportedly decided to up the name to an RTX 4080 from the 4070, however, leaves a worrying feeling about pricing in our gut. Particularly due to the aforementioned RTX 4090 price leak. Time will tell, as always, and with less than a week before Lovelace’s release, we won’t have to wait long to find out more about Team Green’s next-gen graphics cards.
All of this leaves us wondering, however, if the tactic might be to charge a premium for the new products – pretty much as expected for high-end GPUs fresh off the assembly lines – and drive down the prices of that excess stock RTX 3000 even further, as the alternative option for those who don’t want to squander also A lot of.