We've heard through rumours that Nvidia's next generation graphics architecture will be formally named as 'Blackwell'. This week at GTC, Nvidia officially announced exactly that, with the Blackwell platform promising to usher in "a new era" of computing.
Read More »Nvidia GB202 GPU expected to use the same TSMC 4N process as newly-announced B100
This week, Nvidia announced the launch of its next-generation Blackwell GPU architecture, designed for data centres and AI acceleration. The first Blackwell GPU, the B100, is going to be made available this year, with the likes of Microsoft, Amazon and Google already lining up to secure the GPU for their …
Read More »Nvidia’s upcoming GB202 GPU could feature a multi-chip package with 192 SMs and a 512-bit memory bus
Although Nvidia's Blackwell GPU line-up should be mostly similar to Ada in terms of memory specifications, the flagship seems to be the exception. Instead of a 384-bit memory bus like the one used on the AD102 GPU, the successor, the GB202, is expected to feature a 512-bit memory bus. Not …
Read More »Nvidia ‘Blackwell’ GPU leak points to significant memory speed upgrades
Contrary to previous rumours, it looks like the memory buses of the Nvidia RTX 50 series won't be much different from what we've seen in the RTX 30 and 40 series. Based on recent reports, the flagship GB202 GPU will feature a 384-bit memory bus, just like its predecessors. However, …
Read More »Nvidia’s next gen ‘Blackwell’ GPUs (RTX 50) will reportedly use GDDR7 memory
RTX 40 Super graphics cards are reportedly just around the corner but rumblings about Nvidia's true next-gen gaming graphics cards are starting to surface. The RTX 50 series is said to be based on the Blackwell architecture, and these GPUs are rumoured to make the jump to GDDR7 memory too. …
Read More »Nvidia GB202 GPU to feature 24,576 CUDA cores and a 512-bit memory bus
A recent report from a well-known Nvidia leaker suggests the upcoming GB202 GPU (Blackwell), which will presumably be used on the RTX 50 series flagship card, could be significantly faster than the AD102 GPU. According to kopite7kimi, the GB100 GPU (data centre use) will feature 8 GPC (graphics processing clusters), …
Read More »RTX 5090 rumoured be about 70% faster than the RTX 4090
While early reports have already claimed that Nvidia will wait until 2025 to release the RTX 5090, the first rumours about the upcoming GPU are already making the rounds online - including reports that it could be as much as 70% faster than the RTX 4090.
Read More »Nvidia ‘Blackwell’ architecture to succeed ‘Hopper’
It's common to see Nvidia naming new GPU architectures after renowned individuals. We've seen it multiple times, from Maxwell to Ampere and including the upcoming Ada (Lovelace) and Hopper. Based on a new report, it seems Nvidia will continue this trend, naming Hopper's successor Blackwell. The group behind a recent …
Read More »