It is still unclear when exactly the successor to Nvidia’s extremely popular Geforce RTX 3080 will appear. However, according to the latest rumours, the assumed technical data sound more and more interesting.
The source is once again the one known for leaks Twitter user kopite7kimi, who keeps posting updates on the specs of Nvidia’s upcoming RTX 4000 GPUs. The fact that technical details change relatively shortly before release is nothing completely unusual. Nevertheless, these are only rumors – which in this case sound particularly promising.
Reports on a possible replacement for silicon, which is currently extremely important in the hardware sector, are also promising and could do some things better. You can find out more about this in the following article:
Significantly faster hardware thanks to new material? This is behind the discovery
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.
Which rumor is it exactly?
The most recent tweet from kopite7kimi stands out in two main points: the expected power consumption and the storage speed. Both are moving in a positive direction compared to previous rumors about the new GPU of the Ada Lovelace generation (RTX 4000):
link to Twitter content
For a long time it was assumed that the RTX 4080 has a significantly higher power consumption than its predecessor in the range of 100 watts more. The latest specification of 340 watts, on the other hand, is only slightly above the Founders Edition of the RTX 3080 (320 watts).
If the information in the tweet above is correct, the storage speed would also increase. Instead of 16.0 GB of GDDR6X memory with 21 Gbps, modules with 23 Gbps are now being discussed. For comparison: The RTX 3080 relies on 10.0 or 12.0 GB GDDR6X with a speed of 19 Gbps.
Both changes sound plausible. So it is conceivable that Nvidia, even in the course of the often negative reactions to the reports about the high power consumption, has set about improving at least the RTX 4080 (and slower models). Another possible explanation: the high power consumption was only planned for the top models RTX 4090 (Ti) from the start.
There are no differences compared to older rumors in the number of shader units. It is therefore still 9,728, while the RTX 3080 of the current Ampere generation has 8,704 of them. The RTX 3080 with 12.0 GB VRAM has 8,960 shader units.
When will the Geforce RTX 4080 be released?
There is still no consensus in the rumor mill. Recently, it was assumed that only the absolute high-end squad in the form of the Geforce RTX 4090 (Ti) would come onto the market this year. A release of the RTX 4080 in 2022 is not yet off the table.
However, it is becoming more and more likely that Nvidia will generally be later than the RTX 3000 release two years ago, which took place in mid-September 2020. So there were the first official teasers at the beginning of August, while Nvidia has been keeping a low profile more than two years later.
A possible reason should be the (too) full warehouse with RTX 3000 models, which first have to be sold in larger quantities. You can find out more about this in the following article:
Nvidia graphics cards
Price drops due to too many RTX 3000 models?
What do you think of the current rumors about the upcoming Geforce RTX 4080? Do you think the power consumption will really only be slightly higher than in the case of the RTX 3080 or do you rather assume a plus in the range of 100 watts? And when do you expect the RTX 4080 to be released – still in 2022 or next year? Feel free to write it in the comments!
