Tuesday, December 22nd 2020
NVIDIA to Introduce an Architecture Named After Ada Lovelace, Hopper Delayed?
NVIDIA has launched its GeForce RTX 3000 series of graphics cards based on the Ampere architecture three months ago. However, we are already getting information about the next-generation that the company plans to introduce. In the past, the rumors made us believe that the architecture coming after Ampere is allegedly being called Hopper. Hopper architecture is supposed to bring multi-chip packaging technology and be introduced after Ampere. However, thanks to @kopite7kimi on Twitter, a reliable source of information, we have data that NVIDIA is reportedly working on a monolithic GPU architecture that the company internally refers to as "ADxxx" for its codenames.
The new monolithically-designed Lovelace architecture is going make a debut on the 5 nm semiconductor manufacturing process, a whole year earlier than Hopper. It is unknown which foundry will manufacture the GPUs, however, both of NVIDIA's partners, TSMC and Samsung, are capable of manufacturing it. The Hopper is expected to arrive sometime in 2023-2024 and utilize the MCM technology, while the Lovelace architecture will appear in 2021-2022. We are not sure if the Hopper architecture will be exclusive to data centers or extend to the gaming segment as well. The Ada Lovelace architecture is supposedly going to be a gaming GPU family. Ada Lovelace, a British mathematician, has appeared on NVIDIA's 2018 GTC t-shirt known as "Company of Heroes", so NVIDIA may have already been using the ADxxx codenames internally for a long time now.
Sources:
kopite7kimi, via VideoCardz
The new monolithically-designed Lovelace architecture is going make a debut on the 5 nm semiconductor manufacturing process, a whole year earlier than Hopper. It is unknown which foundry will manufacture the GPUs, however, both of NVIDIA's partners, TSMC and Samsung, are capable of manufacturing it. The Hopper is expected to arrive sometime in 2023-2024 and utilize the MCM technology, while the Lovelace architecture will appear in 2021-2022. We are not sure if the Hopper architecture will be exclusive to data centers or extend to the gaming segment as well. The Ada Lovelace architecture is supposedly going to be a gaming GPU family. Ada Lovelace, a British mathematician, has appeared on NVIDIA's 2018 GTC t-shirt known as "Company of Heroes", so NVIDIA may have already been using the ADxxx codenames internally for a long time now.
35 Comments on NVIDIA to Introduce an Architecture Named After Ada Lovelace, Hopper Delayed?
amd rdna3 5nm
i really hope this, then only engineer works both side shows clearly, which one has fastest and efficencies gpus.. i guess Q1-2/2022 out.
few things is sure.
1st,amd cant make same thing like its soing rx 6000 series, and i mean using 'old' rx 5000 series gpus and sqeezze them one gpu,aka 2x5700xt,i mean 3x5700xt are not possible
so, amd must create brand new one. let see them skills.
nvidia..well lovelace is something new what no1 knows except nvidia engineers and leaders,but, sure is that it is 5nm die tech...is it samsung or TSMC we will see, but not so long ago,
we heard that samsung and nvida make new deal...so is it mean samsung engineer has sure nvidia peoples that them die tech is better ..or is it cheaper.. or both?
we will see.
hope we see amd rdna3 and nvidia lovelace early 2022..Q1-2.... i cant belive earlier.
bfore that we seen nvidia super models and of coz rtx 3080 ti and others variants from nvidia also..... how about rtx 3070 sli...i say, NO.
and. also, where is intel gpu? it was nice to see 3rd gpu for battle line.
i just wondering, sure thouse both are remendes fast and return back under 300W can forget...questions is, where we need so much power,bcoz only few percent gamers use 4K monitor.
hmm,soon, only one gpu take same power than hole PC rig bfore.....no good.
I don't look at each and every generation as a must-have upgrade (because, like you noted, improvements can be pretty minute). I just look at it like "$200 bought you 100% performance last year, this year it buys you 105%".
All the lame excuses now, about "but node advantage", is just an attempt to address that acute green pain in the lower back of the respective green zealots.
While node advantage is indeed there, AMD also happens to use SMALLER chips that consume LESS power. That is all what one would expect from a node shrink to do. Compare transistor counts, AMD has matched green chips at perf/transistor, beaten at perf/watt AND is using slower VRAM.
What remains is the lame "but RT" argument, 2 years into ungodly RT goodness, we have "whopping" two dozen games, most of which exhibit barely any visual difference, with the most notable "feature" that RT brings being sharply dropping your framerate.
Oh, but to "compensate" there is "but if you run it at a lower resolution... from far above you wouldn't notice... especially if we apply that glorified TAA derivative we call 'DLSS 2'".
AT this pace, RDNA3 will at the very least exchange punches with the most obnoxiously priced crap team green will roll out, but we can also get into this:
Only for connoisseurs. ;)