NVIDIA's upcoming Blackwell "GeForce RTX 50" GPUs might be affected by changes in the launch schedule due to data center prioritization.
NVIDIA GeForce RTX 50 "Blackwell" GPUs Could Be Affected By Data Center Chip Design Changes
NVIDIA's GeForce RTX 50 "Blackwell" GPUs are going to be the next major launch for the graphics market on PC platforms. These chips are anticipated by a lot of gamers and enthusiasts who are planning to upgrade from their older RTX GPUs. However, certain roadblocks might affect launch schedule for gaming Blackwell chips.
We recently reported that it made specific changes to the Data Center Blackwell AI chips associated with the mask of the GPU. This change has now been completed & Blackwell is on track for volume production in Q4 2024 which is also when the chips will be generating billions of dollars in revenue for the green team.
However, recent reports from industry sources (via MyDrivers) suggest that these changes might also have affected the consumer-tier GPUs. The main issues that prompted the redesign of the top-mask layer were reportedly thermal expansion (CTE) mismatch between the two GPU dies, LSI bridge, RDL interposer, and the main substrate which led to chip warping and system failure. The solution was a redesign of the top metal layer and bumps.
While Blackwell's Data Center and Consumer tier chips will be vastly different than one another, the fundamental architecture will be very similar so it is being reported that the upcoming GeForce RTX 50 series including GeForce RTX 5090 might have been affected by this too however, NVIDIA has already cleared the dust that the samples they sent out to customers were the fixed units and the volume production they are heading towards also have the updated design, NVIDIA is prioritizing Data Center "Blackwell" GPUs for now since it makes up the bulk of its revenue and the chips are in huge demand with the supply being bottlenecked only due to production capabilities of TSMC.
NVIDIA GeForce RTX 50 "Blackwell" GPU Lineup (Preliminary) Specs:
| Graphics Card | GPU | Max GPU SMs / Cores | Memory Configuration | TDP |
|---|---|---|---|---|
| GeForce RTX 5090 Ti? | GB202 | 192 / TBD | 512-bit (GDDR7) | 450W? |
| GeForce RTX 5090? | GB202 | TBD / TBD | 448-bit (GDDR7) | 400W? |
| GeForce RTX 5080 Ti? | GB202 | TBD / TBD | 384-bit (GDDR7) | 350W? |
| GeForce RTX 5080? | GB203 | 96 / TBD | 256-bit (GDDR7) | 320W? |
| GeForce RTX 5070? | GB205 | TBD | 192-bit (GDDR7) | 225W? |
| GeForce RTX 5060 Ti? | GB206 | TBD | 128-bit (GDDR6?) | 160W? |
| GeForce RTX 5060 | GB207 | TBD | 128-bit (GDDR6?) | 120W? |
Previously, it was mentioned that NVIDIA's GeForce RTX 50 "Blackwell" GPUs might move from a Q4 launch to CES 2025. Furthermore, Kopite7kimi has also shared some details on the power input of the chips and according to him, the upcoming GPUs will see some increase in power consumption while the higher-end SKUs seeing bigger increases though it isn't known if this power increase is only related to the prototypes/ES chips or the final retail SKUs.
I know you someone got the details of GeForce of Blackwell recently.
Both of them all have some increase in power consumption, with higher SKUs increasing more.— kopite7kimi (@kopite7kimi) September 3, 2024
Regardless, the hype around RTX 50 series is quite evident and we can't wait to see what NVIDIA cooks up for the gaming market.
About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.
Follow Wccftech on Google to get more of our news coverage in your feeds.
