Nvidia’s RTX 4070 graphics card can run with more CUDA Cores and video memory than previously rumored, according to the latest GPU news.
The source for this is well-known hardware leaker Kopite7kimi, who shared the revised speculation for the RTX 4070 via Twitter.
There is an update from the RTX 4070. PG141-SKU3317680FP3212G 21 Gbps GDDR6X300WTSE >11000August 4, 2022
So the theory now – and this is still just a theoretical incarnation of the RTX 4070, of course, and we can’t be sure that Nvidia settled on this setup – is that Team Green decided to increase the CUDA Core count to 7,680 ( compared to the 7,168 rumors previously).
That’s a fair leap, and there’s another update supposedly applied where the graphics card is configured to use 12GB of VRAM (21Gbps GDDR6X) instead of the long-rumored 10GB loading.
Kopite7kimi further claims that the RTX 4070 will draw 300W in terms of power usage, a number that hasn’t actually changed from the previous leak we compared it to – good news in theory, as the rumored specs are more powerful now.
Analysis: Is it time for AMD to start getting really worried?
The RTX 4070 will be a more affordable model than the high-end Nvidia cards (4080 and 4090) that command the really big prizes, so seeing it get a sizable power boost is exciting news for the -to-be buyers of the future. While we are very aware of the dangers of rumored spec changes – they may not be right as the leaker may be misinformed. Also, as Nvidia goes through finalizing the RTX 4000 GPU’s specs, what might be true now could still change before launch.
All this to say as usual with rumors, resist the urge to get carried away. But if this is the final incarnation of the RTX 4070, many gamers will be breathing serious sighs of relief simply at the 12GB VRAM configuration. As the rumors that the graphics card was only 10GB was one of the biggest sources of frustration and disappointment in the entire RTX 4000 lineup, from what we’ve seen (well, that and the obvious power usage concerns, particularly in relation to the higher range).
Regarding the VRAM issue, if you remember, the theorizing was that Nvidia was saving the 12GB loadout for the RTX 4070 Ti, and what seems to have happened now is that the company basically turned the RTX 4070 in the 4070 Ti. Not only with this video memory being increased, but also with the CUDA core count being extended to use the full amount of cores available on the AD104 GPU.
This could mean that the RTX 4070 is set to outperform the current flagship RTX 3090 Ti, because, as Kopite7kimi tweets, the 3DMark TimeSpy Extreme score for the 4070 is now over 11,000. As the leaker said about the RTX 4070 Ti, it could easily match the 3090 Ti, and now this is apparently the case with the vanilla RTX 4070 – except it doesn’t have a predicted 400W power consumption like the 4070 Ti (it’s a 300W much smaller, supposedly).
Don’t underestimate how crucial this could be to the power of the RTX 4070 at the time – and what a handful that graphics card could become for AMD to handle – while also bearing in mind all the caveats we’ve already outlined.
There’s also another question: what can Nvidia do with the RTX 4070 Ti when it finally comes along? By which we mean, if the RTX 4070 is using the full version of the AD104 chip, then there’s nowhere to go, so presumably Nvidia would have to employ a scaled-down AD103 for the 4070 Ti (that’s the GPU that powers the RTX 4080). , or so the rumor believes).
Through VideoCardz (opens in new tab)