NVIDIA GeForce RTX 5080 will be the first GPU to launch with 32 Gbps GDDR7 memory.
The NVIDIA GeForce RTX 5090 and RTX 5080 will be the flagship products of the new Blackwell generation, and the RTX 5080 is expected to arrive with the fastest DRAM solution on the market.
According to various sources, the NVIDIA GeForce RTX 5080, built on the GB203 GPU die, will carry 16 GB of GDDR7 memory running at 32 Gbps. That is 4 Gbps faster than the 28 Gbps chips on the GeForce RTX 5090. The card will use a 256-bit memory bus, so total memory bandwidth is expected to reach 1 TB/s.
That would be a big step up from the RTX 4080 series, which tops out at 736 GB/s of bandwidth with 23 Gbps GDDR6X. GDDR6X is impressive, but GDDR7 should change the game and give GPUs more bandwidth, especially those with narrow buses. Note that 32 Gbps will not be available across the whole lineup. Only the RTX 5080 will get it first. NVIDIA may well move the lineup to faster memory (G7) and larger capacities down the road, but for now, let's focus on what to expect from the first wave of these cards.
Based on the information provided, the NVIDIA GeForce RTX 5080 will use the PG144/147-SKU45 board and the GB203-400-A1 GPU chip. The card will use the full GB203 die, with 84 streaming multiprocessors (SMs) and 10,752 cores, which is 51% fewer than the RTX 5090. By comparison, the RTX 4090 and RTX 4080 differed in core count by 40%, so overall performance will be quite far apart. The card is also rated at 400W TBP, a 25% increase in power, though real-world figures could vary quite a bit.
NVIDIA is also expected to use 3 GB GDDR7 modules in the future, which would allow up to 24 GB of VRAM on the NVIDIA GeForce RTX 5080. That would benefit gamers and content creators, letting them use the extra memory for content creation, AI workloads, or to keep games running smoothly and stably at high resolutions with enhanced texture packs.
The NVIDIA GeForce RTX 5080 is expected to launch at CES 2025, so stay tuned for more information. The official unveiling is still a few months away.
NVIDIA GeForce RTX 5090 and RTX 5080 GPU Specifications
| Graphics Card Name | NVIDIA GeForce RTX 5090 | NVIDIA GeForce RTX 5080 | NVIDIA GeForce RTX 4090 | NVIDIA GeForce RTX 4080 |
|---|---|---|---|---|
| GPU Name | Blackwell GB202-300 | Blackwell GB203-400 | Ada Lovelace AD102-300 | Ada Lovelace AD103-300 |
| GPU SMs | 170 (192 Full) | 84 (84 Full) | 128 (144 Full) | 76 *80 Full) |
| GPU Cores | 21760 (+33%) | 10752 (+11%) | 16384 | 9728 |
| Clock Speeds | TBD | TBD | 2520 MHz | 2505 MHz |
| L2 Cache | TBD | TBD | 72 MB | 64 MB |
| Memory Capacity | 32 GB GDDR7 (+33%) | 16 GB GDDR7 (0%) | 24 GB GDDR6X | 16 GB GDDR6X |
| Memory Bus | 512-bit (+33%) | 256-bit (0%) | 384-bit | 256-bit |
| Memory Speed | 28-32 Gbps | 32 Gbps | 21.0 Gbps | 23.0 Gbps |
| Bandwidth | 1792-2048 GB/s | 1024 GB/s | 1008 GB/s | 736 GB/s |
| TBP | 600W (+33%) | 400W (+25%) | 450W | 320W |
| Power Interface | 1 12V-2x6 (16-Pin) | 1 12V-2x6 (16-Pin) | 1 12VHPWR (16-Pin) | 1 12VHPWR (16-Pin) |
Source: WCCFTECH
