NVIDIA Blackwell Graphics Architecture GPU Codenames Revealed, AD104 Has No Successor

[btarunr]

The next-generation GeForce RTX 50-series graphics cards will be powered by the Blackwell graphics architecture, named after American mathematician David Blackwell. kopite7kimi, a reliable source with NVIDIA leaks revealed what the lineup of GPUs behind the series could look like. It reportedly will be led by the GB202, followed by the GB203, and then the GB205 and GB206, followed by the GB207 at the entry level. What's surprising here, is the lack of a "GB204" succeeding the AD104, GA104, TU104, and a long line of successful performance-segment GPUs by NVIDIA.

The GeForce Blackwell ASIC series begins with "GB" (GeForce Blackwell) followed by a 200-series number. The last time NVIDIA used a 200-series ASIC number for GeForce GPUs was with "Maxwell," as the GPUs ended up being built on a more advanced node, and with a few more advanced features, than what the architecture was originally conceived for. For "Blackwell," the GB202 logically succeeds the AD102, GA102, TU102, and a long line of "big chips" that have powered the company's flagship client graphics cards. The GB103 succeeds AD103, as a high SIMD count GPU with a narrower memory bus than the GB202, powering the #2 and #3 SKUs in the series. There is curiously the lack of a "GB104."



NVIDIA's xx04 ASICs have powered a long line of successful performance-thru-high end SKUs, such as the TU104 powering the RTX 2080, and the GP104 powering the immensely popular GTX 1080 and GTX 1070 series. The denominator has been missing the mark for the past two generations. The "Ampere" based GA104 powering the RTX 3070 may have sold in volumes, but a its maxed out RTX 3070 Ti hasn't quite sold in numbers, and missed the mark against the Radeon RX 6800 (similar price). Even with Ada, while the AD104 powering the RTX 4070 may be selling in numbers, the maxed out chip powering the RTX 4070 Ti, misses the mark against the RX 7900 XT with a similar price. This has caused NVIDIA to introduce the AD103 in the desktop segment—a high CUDA core-count silicon with a mainstream memory bus width of 256-bit—out to justify high-end pricing, which will continue in the GeForce Blackwell generation with the GB203.

As with AD103, NVIDIA will leverage the high SIMD power of GB203 to power high-end mobile SKUs. The introduction of the GB205 ASIC could be an indication that NVIDIA's performance-segment GPU will come with a feature-set that would avoid the kind of controversy NVIDIA faced when trying to carve out the original "RTX 4080 12 GB" using the AD104 and its narrow 192-bit memory interface.

Given NVIDIA's 2-year cadence for new client graphics architectures, one can expect Blackwell to debut toward Q4-2024, to align with mass-production availability of the 3 nm foundry node.

View at TechPowerUp Main Site | Source

 

[Camm]

This screams to me that whilst Nvidia couldn't quite get away with the shift across the entire product stack it tried to do this generation, it can get away with selling 60 series cards as 05 class cards rather than the 04 they've traditionally been.

 

[N/A]

102 is transitioned to 203 and 104 to 205 this means there is no optic shrink of the 10240 CUDAs 103 that would have been the 204.

we are left with 18432/ 256 bit and 7680/ 128 bit crystal. that way they can carve 5070 and 5080 and even the Ti models out of the 203 die. and have it coupled with 16 GB,
as for the 128 bit 12 Gb 3 gbit memory must be used similar to the 24 GB modules of RAM.

 

[usiname]

This screams to me that whilst Nvidia couldn't quite get away with the shift across the entire product stack it tried to do this generation, it can get away with selling 60 series cards as 05 class cards rather than the 04 they've traditionally been.

No my friend, the GB202 is for xx90 class cards, the GB203 is for xx80 and the GB205 is for xx70 class cards. xx60 will be on GB206. Nvidia took to much damage to shift all GPUs bellow xx80 to lower chips, they won't give you bigger chips now

 

[ir_cow]

With AMD dropping out and Intel not anywhere close to NVIDIA, I don't think we will see the 50 series until 2025 at the earliest. NVIDIA even cancelled the 4090 Ti because it has no competition.

 

[Space Lynx]

With AMD dropping out and Intel not anywhere close to NVIDIA, I don't think we will see the 50 series until 2025 at the earliest. NVIDIA even cancelled the 4090 Ti because it has no competition.

Just a rumor AMD is dropping out. AMD is just delayed cause Apple bought 100% of the 3nm node from TSMC in 2024. AMD will be back in winter 2025 or spring 2026. I'd bet money on it. Apple is just whoring itself for the best node for the entirety of 2024 cause they rich as fuck and want to leave everyone else in the dust temporarily

 

[john_]

Well, here is my theory. I am saying for some time now that we are moving on a 4 years circle for gaming GPUs, with a refresh in between. I was basing this theory also on those rumors about a hi end Ge Force GPU with 512bit data bus.
So what we get? A 4090 Ti with 512bit data bus as an RTX 5090, an upgraded 4080 Ti as an RTX 5080 and under that a bunch of 5070s and 5060s and 5050s based on Blackwell GPUs. While this might look as a very very very long shot, meaning the high end parts being of an older architecture than the mid and low end models, it does makes sense if we consider the possibility Nvidia to be keeping ALL the big dies for AI. Using Ada for the high end models, means that Nvidia will keep it's big gaming dies on the 5nm process instead of using valuable and more expensive 3nm wafers for gaming purposes. The performance gap between the two top Ada models and everything under them, does give Nvidia the chance to play with new GPUs for the mid range that will offer higher performance than existing RTX 4070s and 4060s while remaining way under RTX 4080. A small price increase, like RTX 5070 Ti at $900 could also help to make GB105 financially valid if it is small enough for use in a gaming GPU.
We can also add here the rumor that there is no high end models in AMD's RDNA4 series. Maybe AMD also sees that there is no financial logic into selling mid size or bigger 3nm dies for gaming GPUs, even with their chiplet approach.

Then again there is the much more possible scenario of Blackwell to be fast enough that a GB203 will end up faster than an AD102, so GB203 could be used for a few ultra expensive $1500+ graphics cards and we can forget anything bigger for the gaming market.

 

[kondamin]

Just a rumor AMD is dropping out. AMD is just delayed cause Apple bought 100% of the 3nm node from TSMC in 2024. AMD will be back in winter 2025 or spring 2026. I'd bet money on it. Apple is just whoring itself for the best node for the entirety of 2024 cause they rich as fuck and want to leave everyone else in the dust temporarily

Or tsmc is having so many issues with 3nm and have such low yields they can only service apples demand.

wouldt be the first time products were relaunched under a new name because tsmc was stuck tinkering with a node.

 

[AusWolf]

Chip codenames are arbitrary. Whether you call something xx104 or xx203 doesn't matter. What the end product is capable of and how much it is selling for does.

Although, this does suggest a step back from the "let's make everything more expensive" movement that we've seen in the last couple of years. *touch wood*

For the article: There actually was a GM107 chip, powering the 750 (Ti). The 950 and 960's GPUs were called GM206.

 

[Space Lynx]

Or tsmc is having so many issues with 3nm and have such low yields they can only service apples demand.

wouldt be the first time products were relaunched under a new name because tsmc was stuck tinkering with a node.

its this partly and Apple by making the year long deal, it came with some perks, like they won't be charged for any bad yields. i read an article about the deal last week

 

[evernessince]

Just a rumor AMD is dropping out. AMD is just delayed cause Apple bought 100% of the 3nm node from TSMC in 2024. AMD will be back in winter 2025 or spring 2026. I'd bet money on it. Apple is just whoring itself for the best node for the entirety of 2024 cause they rich as fuck and want to leave everyone else in the dust temporarily

The idea that a foundry can suddenly not have enough capacity to supply partners their contractually obligated wafer allotments simply runs contrary to how the industry works. Apple and AMD book capacity years in advance and TSMC builds out capacity to meet those contracts. TSMC is required to give each company the wafer allotment agreed upon. Lack of such an obligation on TSMC's end would mean that companies could be holding the bag spending billions of dollars on a CPU or GPU designed with no wafers to actually make the product and thus why such contracts are standard practice and necessary.

Only wafers above those contractually obligated are not guaranteed, much like what we saw during the pandemic. The silicon foundry business does not turn on a dime.

I'm not even sure Apple is taking all of the excess capacity, last I checked TSMC has spare capacity. In addition, Apple has saw a drop in sales over the last 3 quarters consecutively and has cut product shipments. TSMC also offers 2 3nm products, N3E and N3P. N3E being the mobile geared node, Apple is likely using that. It's how to tell how much that would impact AMD, who will almost certainly be using N3P for it's graphics cards and CPUs.

If AMD is late to 3nm, it's extremely likely that it down to AMD's own choice and not the fault of Apple. TSMC will have a glut of wafers in 2025 as that's when production volume that was pushed for during the pandemic starts coming online. On top of that Intel has been seeing success in it's foundry for hire business as well, which can only be good for the industry as a whole. The way I see it, capacity is set to increase while sales are set to decrease. Unless some large demand comes around again I see the potential for too much supply.

 

[Space Lynx]

The idea that a foundry can suddenly not have enough capacity to supply partners their contractually obligated wafer allotments simply runs contrary to how the industry works. Apple and AMD book capacity years in advance and TSMC builds out capacity to meet those contracts. TSMC is required to give each company the wafer allotment agreed upon. Lack of such an obligation on TSMC's end would mean that companies could be holding the bag spending billions of dollars on a CPU or GPU designed with no wafers to actually make the product and thus why such contracts are standard practice and necessary.

Only wafers above those contractually obligated are not guaranteed, much like what we saw during the pandemic. The silicon foundry business does not turn on a dime.

I'm not even sure Apple is taking all of the excess capacity, last I checked TSMC has spare capacity. In addition, Apple has saw a drop in sales over the last 3 quarters consecutively and has cut product shipments. TSMC also offers 2 3nm products, N3E and N3P. N3E being the mobile geared node, Apple is likely using that. It's how to tell how much that would impact AMD, who will almost certainly be using N3P for it's graphics cards and CPUs.

If AMD is late to 3nm, it's extremely likely that it down to AMD's own choice and not the fault of Apple. TSMC will have a glut of wafers in 2025 as that's when production volume that was pushed for during the pandemic starts coming online. On top of that Intel has been seeing success in it's foundry for hire business as well, which can only be good for the industry as a whole. The way I see it, capacity is set to increase while sales are set to decrease. Unless some large demand comes around again I see the potential for too much supply.

Apple procures TSMC's entire supply of 3nm chips, insiders say

www.techspot.com

high end gpu's need 3nm to truly shine. Apple owns them all.

yep, I already said AMD will make a come back in late 2025 or early 2026.

 

[N/A]

Maybe backported to N4, GB202 ~~ 800mm² 512bit GDDR6X, GB203 600mm² 384 bit, 400,300,200, hence the 2 prefix

 

[evernessince]

Apple procures TSMC's entire supply of 3nm chips, insiders say

www.techspot.com

high end gpu's need 3nm to truly shine. Apple owns them all.

yep, I already said AMD will make a come back in late 2025 or early 2026.

If you look at the article's source, it specifically states Apple secured the initial batch of 3nm wafers ala 2H 2023. It does not touch upon 2024's production.

 

[BorisDG]

1. We already knew it's Blackwell
2. It's said from numerous sources, that it comes at earliest 2025.

 

[Gucky]

Those are just naming schemes, as long as we don't know the sizes, we can't talk about successors.

 

[Dr. Dro]

Just a rumor AMD is dropping out. AMD is just delayed cause Apple bought 100% of the 3nm node from TSMC in 2024. AMD will be back in winter 2025 or spring 2026. I'd bet money on it. Apple is just whoring itself for the best node for the entirety of 2024 cause they rich as fuck and want to leave everyone else in the dust temporarily

Even though AMD should. Take some time to recollect and develop the software ecosystem, defend the home front against Intel's onslaught, iterate and then come back to the high end.

Intel is after AMD's lunch and they poached several former AMD employees who understand how AMD thinks and the limitations of their MO.

Arc cannot be ignored for much longer, especially with it providing the same Linux friendliness of Radeon, achieving a decent amount of performance and with ultra high end GeForce cards being priced ever out of reach.

It'd be a safe bet that the 5090 will be 2K, and we don't even have software that can take full advantage of Ada features yet.

 

[Assimilator]

especially with it providing the same Linux friendliness of Radeon

Nobody cares about Linux.

 

[Dr. Dro]

Nobody cares about Linux.

In absolute numbers perhaps, but Linux support is one of the major selling points of Radeon. It's a faithful, steady slice of the market share. The driver support for it is almost as good as Nvidia on Windows.

 

[Space Lynx]

Nobody cares about Linux.

I do, Linux mint is awesome and so is steam OS. They rock my socks and I am 100% buying steam deck 2 on launch day in a few years.

 

[ARF]

No my friend, the GB202 is for xx90 class cards, the GB203 is for xx80 and the GB205 is for xx70 class cards. xx60 will be on GB206. Nvidia took to much damage to shift all GPUs bellow xx80 to lower chips, they won't give you bigger chips now

This means the *60 tier will be the new entry level / low-tier garbage parts. Historically it had never been so, and at that point it would have been the first time this actually happened.

Just a rumor AMD is dropping out. AMD is just delayed cause Apple bought 100% of the 3nm node from TSMC in 2024. AMD will be back in winter 2025 or spring 2026. I'd bet money on it. Apple is just whoring itself for the best node for the entirety of 2024 cause they rich as fuck and want to leave everyone else in the dust temporarily

There are still Samsung 3N nodes as alternative solutions.

3 nm process - Wikipedia

en.wikipedia.org

I think there is too much philosophy in the thread, negative thinking and lack of actual facts and solutions.

If indeed the new gens both from AMD and nvidia are delayed till 2025 | 2026, then you can safely assume that's the end of the graphics market as we know it.
2018 - RTX 2000
2020 - RTX 3000
2022 - RTX 4000
2025 | 2026 - RTX 5000

3 | 4 years to launch a new generation ?

 

[TheinsanegamerN]

Chip codenames are arbitrary. Whether you call something xx104 or xx203 doesn't matter. What the end product is capable of and how much it is selling for does.

Although, this does suggest a step back from the "let's make everything more expensive" movement that we've seen in the last couple of years. *touch wood*

For the article: There actually was a GM107 chip, powering the 750 (Ti). The 950 and 960's GPUs were called GM206.

When you take a chip typically put on a 104 die, move it to a 106 die, but sell it for 104 die prices, you deserve to be called out for it. "but muh chip names dont matter" is a very poor excuse for this behavior.

Even though AMD should. Take some time to recollect and develop the software ecosystem, defend the home front against Intel's onslaught, iterate and then come back to the high end.

HARD disagree. The last time AMD tried this, it gave nvidia the upper mid range and high end on a silver platter, leading not only to price increases, but nvidia getting more money from sweet high end margins to continue pulling ahead. When you drop out of a market, its HARD to break back in, as everyone will have moved on.

 

[ViperXZ]

The "2xx" moniker lets someone guess this is just a rebranded ADA, basically a refresh of ADA which was renamed, but the "2xx" remained. Comparable to Maxwell back then, but they changed the arch name.

 

[ARF]

The "2xx" moniker lets someone guess this is just a rebranded ADA, basically a refresh of ADA which was renamed, but the "2xx" remained. Comparable to Maxwell back then, but they changed the arch name.

A refresh that needs 3 or 4 years to execute, and then someone says it is backported to the current 4nm process?
Just wow

 

[ViperXZ]

A refresh that needs 3 or 4 years to execute, and then someone says it is backported to the current 4nm process?
Just wow

A refresh because why else name it "2xx" if it's a brand new arch? Ofc I'm just speculating like everyone else here. But it's "just" 3 years.