NVIDIA's next-generation GeForce RTX 5090 and RTX 5080 “Blackwell” GPUs are rumored to launch in Q4 2024.
NVIDIA's next-generation “Blackwell” gaming GPUs, the GeForce RTX 5090 and RTX 5080, are rumored to launch later this year
This year began with the launch of the GeForce RTX 40 “SUPER” series, which introduced three brand new GPUs based on NVIDIA's Ada Lovelace architecture. These GPUs are designed to offer better performance and value than their non-SUPER variants and have been met with generally positive reception.
Now it seems that NVIDIA is already planning to launch the next generation of gaming GPUs based on the recently introduced Blackwell architecture. The NVIDIA Blackwell GPUs announced so far are designed entirely for the AI and HPC segment, but we know for a fact that there's already a Blackwell GPU IP that will power next-generation RTX graphics. Graphics cards are expected to use this branding for the RTX 50 series.
that it mentioned That NVIDIA will once again focus on launching high-end graphics cards like the GeForce RTX 5090 followed by the GeForce RTX 5080 in Q4 2024. The graphics cards will be beneficial to NVIDIA's board partners in expanding their shipments and the launch of these GPUs will also help drive the average selling price GPUs (ASPs) are on the rise as more gamers upgrade towards the higher end of the spectrum. MSI is among the major brands that expect double-digit gains with its GeForce RTX 50 GPUs and related AI PC platforms which will be a huge deal in 2024 and beyond.
As NVIDIA enters the AI business, graphics card-related products also continue to introduce new products. Board card manufacturers expect NVIDIA's latest RTX 50 series graphics cards to be unveiled in the fourth quarter. Initially, they will focus on the high-end, high-end 5090 and 5080 product lines. The legal person is optimistic that NVIDIA's biennial launch of a GPU refresh will help card manufacturers drive shipments this year and will push up the average unit price (ASP) of graphics card products.
Based on previous rumours, NVIDIA is expected to have two flagship templates under the GB202 and GB203 banner. The former will likely be used by the GeForce RTX 5090 while the latter is expected to be used by the RTX 5080. The GB203 is rumored to be half of the GB202 GPU and both GPUs are expected to use the latest GDDR7 memory interface with up to 512-bit bus interfaces. Here's what the memory configurations might look like in the various Blackwell GPU SKUs:
- GB202 – 512bit / 28Gbps / 32GB (maximum memory) / 1792GBps (maximum bandwidth)
- GB202 – 384 bits / 28 Gbps / 24 GB (maximum memory) / 1344 GB/s (max bandwidth)
- GB203 – 256 bit / 28 Gbps / 16 GB (max. memory) / 896.0 GB/s (max. bandwidth)
- GB205 – 192bit / 28Gbps / 12GB (maximum memory) / 672.0GB/s (maximum bandwidth)
- GB206 – 128 bit / 28 Gbps / 8 GB (maximum memory) / 448.0 GB/s (max bandwidth)
- GB207 – 128 bit / 28 Gbps / 8 GB (maximum memory) / 448.0 GB/s (max bandwidth)
The last two major launches, Ampere (GeForce RTX 3090) and Ada (GeForce RTX 4090) occurred in late Q3 or early Q4. Accordingly, the release date of the fourth generation flagship cards, GeForce RTX 5090 and RTX 5080, in Q4 2024, seems reasonable but one must remember that this is not a specific time frame and can change depending on market conditions.
NVIDIA GeForce GPU SKUs:
generation | Blackwell | Ada Lovelace | Ampere | Turing | Pascal |
---|---|---|---|---|---|
Process node | TSMC 3nm? | TSMC 5nm | Samsung 8nm | TSMC 12nm | TSMC 16nm |
Launch year | 2024 | 2022 | 2020 | 2018 | 2016 |
Super excited SKU | GB202 | M102 | GA102 | TU102 | GP102 |
Enthusiastic SKU | GB203 | M103 | GA102 | TU104 | GP104 |
High-end SKU | GB205 | M104 | GA104 | TU106 | GP104 |
Mainstream SKU | GB206 | M106 | GA106 | TU106 | GP106 |
SKU for beginners | GB207 | M107 | GA107 | TU116/117 | GP107 |
“Typical beer trailblazer. Hipster-friendly web buff. Certified alcohol fanatic. Internetaholic. Infuriatingly humble zombie lover.”