Nvidia's AD102 Lovelace GPU reportedly offers gamers up to 18432 CUDA cores
Nvidia's next-generation architecture will be BIG
Published: 28th December 2020 | Source: 3DCenter.org via kopite7kimi |
Nvidia's AD102 Lovelace GPU reportedly offers gamers up to 18432 CUDA cores
Rumour has it that Nvidia's next-generation graphics architecture will deliver a huge performance leap over today's Ampere products, with 3DCenter.org and kopite7kimi claiming that the company's AD102 GPU core will feature 18,432 FP32 units.
Nvidia's next-generation GPU architecture will reportedly be called Lovelace after Ada Lovelace. This next-generation graphics architecture is also rumoured to use 5nm lithography and make up Nvidia's RTX 40 series of products.
If the specifications from kopite7kimi are accurate, Nvidia's AD102 GPU will feature 71.4% more CUDA cores than Nvidia's GA102 (Ampere) GPU. This factor alone should make Lovelace a huge generational leap over Ampere, assuming that Nvidia can offer the same (or higher) performance levels from each Lovelace FP32 unit.
At this time, it is unknown when Nvidia will release a post-Ampere graphics architecture. Right now, Nvidia hasn't released their entire Ampere lineup, and mobile Ampere products have not been released onto the consumer market. Even if we discount the possibility of an Ampere-based RTX 30 Super series refresh, Nvidia is unlikely to release a next-generation graphics architecture until mid-to-late 2021 at the earliest.
Rumour has it that Nvidia is working on RTX 3070 Ti and RTX 3080 Ti products, which are both due to release in early 2021. These products would negate the need for an Ampere RTX Super refresh, making it likely that Nvidia plans to release RTX 40 series products and a new GPU architecture sooner than expected.
Reports of Nvidia's accelerated product launches could be due to AMD's recent RDNA 2 products, which have proven to be very competitive with Nvidia's RTX 30 series offerings. Have AMD forced Nvidia to accelerate their roadmap?
Nvidia's next-generation GPU architecture will reportedly be called Lovelace after Ada Lovelace. This next-generation graphics architecture is also rumoured to use 5nm lithography and make up Nvidia's RTX 40 series of products.
If the specifications from kopite7kimi are accurate, Nvidia's AD102 GPU will feature 71.4% more CUDA cores than Nvidia's GA102 (Ampere) GPU. This factor alone should make Lovelace a huge generational leap over Ampere, assuming that Nvidia can offer the same (or higher) performance levels from each Lovelace FP32 unit.
At this time, it is unknown when Nvidia will release a post-Ampere graphics architecture. Right now, Nvidia hasn't released their entire Ampere lineup, and mobile Ampere products have not been released onto the consumer market. Even if we discount the possibility of an Ampere-based RTX 30 Super series refresh, Nvidia is unlikely to release a next-generation graphics architecture until mid-to-late 2021 at the earliest.
Rumour has it that Nvidia is working on RTX 3070 Ti and RTX 3080 Ti products, which are both due to release in early 2021. These products would negate the need for an Ampere RTX Super refresh, making it likely that Nvidia plans to release RTX 40 series products and a new GPU architecture sooner than expected.
Reports of Nvidia's accelerated product launches could be due to AMD's recent RDNA 2 products, which have proven to be very competitive with Nvidia's RTX 30 series offerings. Have AMD forced Nvidia to accelerate their roadmap?
You can join the discussion on Nvidia's rumoured AD102 GPU on the OC3D Forums.
So, nVidia's AD102 chip maybe is like:
— 3DCenter.org (@3DCenter_org) December 28, 2020
12 GPC
72 TPC
144 SM
18'432 FP32 units
~66 TFlops FP32 power (on 1.8 GHz) https://t.co/A8OnUktE1s
Most Recent Comments
Quote:
|
Originally Posted by AlienALX
Bah, so they didn't name it after Linda then
 I don't think AMD have accelerated it. Ampere has been a wash since its conception. Poor yields, wrong company making the dies and so on. |
I think it's more complicated than Samsung's process node just being terrible. Ampere is a not a terrible architectural leap. It's nowhere near as good as past generations, but in and of itself it's decent enough. Availability is poor, but so is everything using TSMC's 7nm, which has been available for ages now. Power consumption is pretty bad with Ampere, but there are loads of cores, and it's not exactly miles behind RDNA2. And that high power draw clearly hasn't made the cards impossible to cool.Quote
Meh... means nothing if only a minority of actual customers can get them without going to extraordinary lengths due to scalpers and retailers price gouging.Quote
Quote:
|
Originally Posted by AngryGoldfish
I've heard rumours that Nvidia will be using Samsung's 5nm process, not TSMC's.
I think it's more complicated than Samsung's process node just being terrible. Ampere is a not a terrible architectural leap. It's nowhere near as good as past generations, but in and of itself it's decent enough. Availability is poor, but so is everything using TSMC's 7nm, which has been available for ages now. Power consumption is pretty bad with Ampere, but there are loads of cores, and it's not exactly miles behind RDNA2. And that high power draw clearly hasn't made the cards impossible to cool. |
Will it be enough to kick AMD's ass? yeah, probably. However, if they remain with Samsung AMD will have a chance to drop with TSMC and stay in line.
Bit annoying if you think about "What if?". IE, what if Nvidia stuffed all of that core tech onto a 5nm TSMC die. Better clocks, better power consumption ETC.Quote
Quote:
|
Originally Posted by AlienALX
Then they will have the same issues. IE, Samsung's 5nm is more akin to TSMC's 7.
Will it be enough to kick AMD's ass? yeah, probably. However, if they remain with Samsung AMD will have a chance to drop with TSMC and stay in line. Bit annoying if you think about "What if?". IE, what if Nvidia stuffed all of that core tech onto a 5nm TSMC die. Better clocks, better power consumption ETC. |
CSS Valid
I don't think AMD have accelerated it. Ampere has been a wash since its conception. Poor yields, wrong company making the dies and so on.Quote