The GeForce RTX 30 collection is already tremendous quick. The upper-end playing cards in that household chew up and spit out video games at excessive resolutions the way in which playing cards from 5 years in the past have been simply beginning to grasp 1080p. It will get somewhat bit foolish at instances after we’re benchmarking these graphics playing cards as a result of it’s generally troublesome to search out benchmarks that may actually put the harm on one thing like a GeForce RTX 3080 Ti—to say nothing of the 3090 or 3090 Ti.
That is why speaking in regards to the GeForce RTX 4000 collection can get our heads spinning, as a result of these playing cards are sounding like a monumental leapp forward of the Ampere era. Most of what we predict we all know in regards to the upcoming GeForce era comes from leakers. We have heard tales of huge GPUs pulling unbelievable energy numbers, and tall tales of as much as 48GB of reminiscence, however the actuality might be somewhat extra modest.
The newest leaks from kopite7kimi sound much more reasonable than a few of the stuff we have heard earlier than. In a tweet in the present day, he supplied some “updates” on the anticipated specs of the GeForce RTX 4090, 4080, and 4070. The tweet contains vital specs for the upcoming GPUs, and in a follow-up tweet, he supplied his expectations for the playing cards’ whole board energy.
Ranging from the highest, the leaker says that the GeForce RTX 4090 will use the AD102-300 GPU, and have absolutely 16,384 FP32 ALUs, or “CUDA cores” in NVIDIA nomenclature. That may be 128 Ada Lovelace Streaming Multiprocessors (SMs), which is an improve from kopite7kimi’s final leak. Earlier leaks have indicated that the AD102 chip has 144 SMs in its full-fat configuration, so this leaves some room for a RTX 4090 Ti or “Titan” SKU.
The reminiscence configuration on the RTX 4090 will supposedly be some 24GB of GDDR6X reminiscence operating at a blistering 21Gbps on a 384-bit reminiscence bus. That is, notably, the very same configuration because the GeForce RTX 3090 Ti, which lends some credence to the concept that the 2 playing cards (4090 and 3090 Ti) use closely-related PCBs, as has been rumored prior to now.
Matching up with earlier rumors, kopite7kimi additionally places the RTX 4090 down for a 450W whole board energy. This coincides with the utmost quantity of energy that the 12VHPWR connector is allowed to hold with out having the additional 4 sense pins. Nearly no one has an influence provide with that functionality but—there’s just one available on the market—so it appears unlikely we’ll see playing cards exceed 450W nominal energy within the short-term.
Stepping all the way down to the GeForce RTX 4080’s rumored specs, we see a a lot bigger leap than within the 3000 collection between x80 and x90 playing cards. The GeForce RTX 4080 is supposedly going to ship with a distinct processor altogether in comparison with the top-end card, referred to as AD103-300. Now we have seen a “103” GPU earlier than; there is a GA103 that’s used within the top-end Ampere cellular GPUs. The Ada Lovelace model will allegedly have 10240 CUDA cores, giving it 80 SMs of a doable 84.
This newest leak pegs the GeForce RTX 4080 with a narrower 256-bit reminiscence interface utilizing 16GB of GDDR6 reminiscence, presumably at 18Gbps. This appears conservative to us; we would anticipate the x80 card to ship with GDDR6X, not less than, and certain on the greater 21Gbps clock charge. The rumored configuration would find yourself with much less reminiscence bandwidth than even the RTX 3070 Ti, though even 256-bit GDDR6X at 21 GT/s could be behind the RTX 3080 and its a lot wider bus. It is an fascinating rumor for certain; maybe the a lot bigger L2 cache on Ada Lovelace could make up the distinction, or NVIDIA has devised a method to successfully enhance reminiscence bandwidth via a brand new compression know-how or Infinity Cache-like function.
The leaker appears much less assured in his energy estimate for the GeForce RTX 4080, itemizing it with a query mark. The estimate is 420W, which is not all that removed from the GeForce RTX 4090. Given the large distinction in allotment of purposeful models, it is doable that the GeForce RTX 4080 is clocked a lot greater than its bigger sibling to maintain them from being too far aside in remaining efficiency. That may push it out of the optimum space on the efficiency/watt curve, probably making the 4090 a extra environment friendly card.
Lastly, for the RTX 4070, the leak predicts that it’s going to use AD104-275 with 7168 CUDA cores in 56 SMs. Based mostly on the leaked specs for the Ada Lovelace GPUs, that might be the identical 4-SM lower from the full-fat AD104 GPU as we noticed in AD103 for the RTX 4080. He predicts 10GB of 18Gbps GDDR6 reminiscence on a 160-bit bus for the RTX 4070, one other brief reminiscence bandwidth allocation placing it on par with the RTX 3060. Appears doubtful to us, however once more, Ada Lovelace is drastically rising the on-die GPU cache in comparison with Ampere. Possibly that makes up the distinction.
Unsurprisingly, provided that it is a a lot smaller GPU, the RTX 4070 will get put down for a downright-reasonable 300W TBP. Keep in mind when a 300W graphics card appeared absurd? It wasn’t all that way back that folks have been making memes evaluating the R9 290X’s blower to a jet engine.
One sore level of the GeForce RTX household for a lot of PC lovers has been the sharp rise in value at each tier. The GeForce GTX 1080 shipped with a nominal MSRP of $599, which dropped to $499 after 9 months. The GeForce RTX 2080 debuted at $699, and the RTX 3080 piled one other $100 on prime of that.
Sadly, it appears like that pricing development will proceed. Because the final a part of the unique tweet states, “do NOT anticipate a decrease MSRP.” Hopefully that signifies that we not less than will not see even greater pricing for discrete GPUs, nevertheless it does current a chance for a competitor (like AMD or even perhaps Intel) to slide in with decrease pricing and steal some market share. We’ll see what occurs later this yr when the higher-end Arc playing cards and AMD’s Radeon RX 7000 collection hit the market.
