Nvidia’s RTX 4080 graphics card might be a little less beefy than previously rumored, which quickly follows news that the RTX 4070 might be a livelier GPU than anticipated.
This new speculation from Lovelace again comes from Kopite7kimi, a regular hardware leaker on Twitter who’s been providing rumors at a fairly recent pace (although that’s to be expected as the RTX 4000 line approaches its release deadline, with an expected debut in October. or around , assuming there are no delays).
I’m not a chatterbox, but I need to make some updates. I hope you do not mind. a possible RTX 4080, PG136/139-SKU360AD103-300-A19728FP32256bit 16G 21Gbps GDDR6XFull power ~420WTSE ~15000Now I have completed the latest update for 4090, 4080 and 4070.August 5, 2022
The theory is that the ‘possible’ RTX 4080 – a reminder that nothing is necessarily set, even if this is Nvidia’s current thinking with GPU specs – could run with 9,728 CUDA cores, which is less than the 10,240 rumors earlier.
The leaker believes that the RTX 4080 will come equipped with 16GB of GDDR6X (at 21Gbps) and a power consumption of around 420W, all in line with what was previously reported – the only change is the lower core count. at this point.
Analysis: Nvidia’s positioning of the RTX 4000 models
The first thing to note here is that this is not a big change. Dropping 512 CUDA cores – which means reducing the on-board streaming multiprocessors by four, from 80 to 76 – isn’t a huge change, but it still represents a slight reduction in raw grunt for the GPU.
Let’s keep in mind that this is just a rumor, and the leak landscape surrounding Lovelace is apparently changing quite regularly these days – which could be a reflection of Nvidia tweaking and tweaking the relative specs as the company progresses. advances towards a bunch of finished products.
As mentioned at the beginning, Kopite7kimi also recently brought us news about the RTX 4070, with the apparent idea that the GPU was going to step up its performance level at this stage – to the specs that were previously leaked for the RTX 4070 Ti, in fact. A big step up for sure. So, assuming all of this is correct, what might Nvidia’s reasoning be here?
Decreasing the RTX 4080’s CUDA core count a bit, while seriously increasing the RTX 4070 – not just for CUDA cores, but VRAM loading as well – obviously means that the relative performance of the RTX 4070 and RTX 4080 is theoretically on par. being approached more closely.
With less differentiation between the two, and Nvidia’s xx80 model generally being a good chunk more expensive than the more conventional xx70 card – the relatively affordable high-end GPU, if you will – with Lovelace, we’ll be looking at an RTX 4070 which is even more in the sweet spot for most buyers?
Or put another way, if the RTX 4080 is not all much faster, who will shell out for one instead of a 4070? Unless stock is an issue for the latter, of course; that could come into play, especially if it’s really popular.
Also, we need to remember that CUDA Cores are not the complete story for performance, with many other factors coming into play such as clock speed. This is where there could be a clue in the RTX 4070 supposedly going around 300W for power consumption, while the RTX 4080 is supposedly going to hit 420W or more – this could point to Nvidia seriously increasing clock speeds (and therefore performance) with the high-end GPU compared to the 4070.
Also, Kopite7kimi kept the previous guestimate to a 3DMark TimeSpy Extreme score for the RTX 4080 (around 15,000), despite these CUDA cores being shaved, quite interestingly, this could also suggest that Nvidia might be putting more effort with the clocks here for the 4080 Although the leaker doesn’t specifically mention clock speed.
It certainly doesn’t make sense to have the RTX 4080 as a relatively unattractive proposition compared to the 4070 – much higher power usage won’t help either, in terms of forcing PSU upgrades and power costs in general – so for now, let’s go. suppose that if this rumor is true, we still can’t see the full picture of what Nvidia is trying to do with the placement of Lovelace GPUs. Or that there will be another spec revamp yet to come…
Relative price may also be a factor here – we haven’t heard much about it from the rumor yet, but we can’t imagine that the RTX 4000 series will launch with affordability in mind. Particularly not if Nvidia has a lot of excess RTX 3000 stock to deal with, as the vine claims, and change everything that overlaps with the Lovelace release (which is likely only a few months away).
Whatever the case, you’d imagine Team Green would have learned from their mistakes in terms of the RTX 3080 starting off on the wrong foot, coming with just 10GB of VRAM in its early incarnation – an unpopular move made up by the later release of a variant. of 12 GB. The latter was beefier not just in terms of additional video memory, leaving some buyers of the original RTX 3080 with a serious case of regret.
Through VideoCardz (opens in new tab)