Nvidia’s RTX 4000 gets new specs, and it’s not all good news

Nvidia’s upcoming Ada Lovelace graphics cards have received a new batch of rumored specs, and this time around, it’s a bit of a mixed bag.

While the news is good for one of the GPUs, the RTX 4070 has already received a cut when it comes to its specs – but the leaker says that won’t translate into a cheaper price tag.

And TBP, 450/420? / 300 watts.

And the[مدش]. kopite7kimi (@kopite7kimi) June 23, 2022

The information comes from kopite7kimi, a well-known name when it comes to PC leaks, who just revealed an update to the specifications of the RTX 4090, RTX 4080, and RTX 4070. I’ve heard previous whispers about specs From the RTX 4090 and RTX 4070, this is the first time we get predictions about the specifications of the RTX 4080.

Let’s start with the good news. If this rumor is true, it appears that the flagship RTX 4090 has received a slight bump in the core count. The number previously reported was 16128 CUDA cores, and that number has now increased to 16,384 cores, which translates into an upgrade from 126 Multiple Stream Processors (SMs) to 128. As for the rest of the specs, they remain unchanged – current expectations are That GPU will get 24GB of GDDR6X memory over the 384-bit memory bus, as well as 21Gbps of bandwidth.

The RTX 4090 includes the AD102 GPU, which reaches a maximum of 144 SMs, but it seems unlikely that the RTX 4090 itself will reach such heights. The full version of the AD102 GPU will likely be found in a better graphics card, whether it’s a Titan or an RTX 4090 Ti. It is also rumored to be monstrous energy requirements. This time, kopite7kimi hasn’t revealed anything new about that card, and as of now, we still don’t know the truth about its existence.

Moving on to the RTX 4080 with AD103 GPU, it is said to come with 10240 CUDA cores and 16GB of memory. However, according to kopite7kimi, it will be based on GDDR6 memory instead of GDDR6X. Since the leaker expected it to be 18Gbps, that would actually make it slower than the RTX 3080 with 19Gbps ​​of memory. The number of cores is exactly the same as in the RTX 3080 Ti. So far, this GPU doesn’t look that impressive, but it is said to come with a much larger L2 cache which is likely to provide an upgrade in gaming performance over its predecessors.

RTX 3080 graphics card on a pink background.
Jacob Roach / Digital Trends

When it comes to the RTX 4070, the GPU was previously rumored to come with 12GB of memory, but now, kopite7kimi expects only 10GB over the 160-bit memory bus. It is said to offer 7168 CUDA cores. While it’s certainly an upgrade over the RTX 3070, it may not be the generational leap that some users are hoping for. It’s also assumed that he won’t receive a price discount based on the spec drop, but we still don’t know the MSRP of this GPU, so it’s hard to judge its value.

Finally, the leaker introduced an update to the power requirements for the GPUs, which has certainly been the subject of a lot of speculation over the past few months. The expected TBP of the RTX 4090 is 450 watts. It’s 420 watts for the RTX 4080 and 300 watts for the RTX 4070. Those numbers are a lot more conservative than the 600 watts (and above) we’ve seen floating around.

What does all this mean for us – end users of the upcoming RTX 40-series GPUs? Not much yet. Specifications may change yet, and while kopite7kimi has a proven track record, it may be wrong about the specs as well. However, as it stands now, only the RTX 4090 appears to be a huge upgrade over its predecessor while the other two are a more modest change. It remains to be seen if prices will reflect that or not.

Editors’ Recommendations