NVIDIA's GeForce RTX 5090 & RTX 5080, the next generation of GPUs from the green team for gamer, have their specs revealed by Kopite7kimi.
NVIDIA GeForce RTX 5090 & RTX 5080 GPUs To Have A Huge Specs Gap According To Rumored Specs
NVIDIA is preparing its next-generation GeForce RTX 50 "Blackwell" GPU lineup which is expected to bring some big performance gains. So far, we have seen rumors regarding the GPU configurations, a list of possible SKUs, and the expected launch dates but Kopite7kimi, one of the best leakers in the tech scene, has now revealed the specifications of the top two cards, the GeForce RTX 5090 & the GeForce RTX 5080.
NVIDIA GeForce RTX 5090 Graphics Card
Starting with the details, we first have the NVIDIA GeForce RTX 5090 which is going to be the top GPU of the lineup. The graphics card is expected to feature the PG144/145-SKU30 PCB design & that's going to incorporate the GB202-300-A1 GPU core. The GPU will have 170 SMs enabled out of the total 192 SMs and will feature 21760 cores instead of the total 24,576 cores. This marks an 11.4% reduction which is slightly more than the -11.1% reduction of the RTX 4090 versus its full AD102 die.
GeForce RTX 5090
PG144/145-SKU30
GB202-300-A1
21760FP32
512-bit GDDR7 32G
600W— kopite7kimi (@kopite7kimi) September 26, 2024
On the memory side, the NVIDIA GeForce RTX 5090 is going to be a beast with 32 GB of GDDR7 VRAM capacity running across a 512-bit interface. The first generation of GDDR7 memory is likely to scale between 28-32 Gbps which would lead to 1.792 or 2.00 TB/s of total bandwidth. The card will also come with a 600W TBP rating though it should be mentioned that the TBP value shouldn't be compared with actual real-world or gaming power consumption as that ends up much lower. The leaker also states that despite the 600W design, the card will feature a 2-slot cooler in Founders Edition flavors.
NVIDIA GeForce RTX 5080 Graphics Card
Moving over to the second card, we have the NVIDIA GeForce RTX 5080. The RTX 5080 graphics card will be based on the PG144/147-SKU45 PCB and will incorporate the GB203-400-A1 GPU die. This card will utilize the full GB203 GPU die with 84 SMs and 10,752 cores but that's a big -51% reduction compared to the RTX 5090. For comparison, the RTX 4090 and RTX 4080 had a -40 percent difference in the number of cores so the overall performance is going to vary by a huge margin.
GeForce RTX 5080
PG144/147-SKU45
GB203-400-A1
10752FP32
256-bit GDDR7 16G
400W— kopite7kimi (@kopite7kimi) September 26, 2024
Other than that, the RTX 5080 is also going to feature half the VRAM configuration with a 16 GB capacity running across a 256-bit bus interface while utilizing GDDR7 modules. We can expect the memory bandwidth anywhere between 896 GB/s and up to 1024 GB/s. The card will also be rated at 400W TBP, similar to the power configurations that were revealed a while ago. This will be a 25% increase in the power wall but once again, the real-world figures should be very different.
Having such a huge difference between the 90 and 80 class graphics cards might not go in NVIDIA's favor as we saw with the RTX 4080. The RTX 4080 was among one of the poorest values for an 80-class product which was only slightly rectified with the launch of the "SUPER" variant. We can hope that the pricing of the 5080 will be much more reasonable than the $1200 US MSRP last time & even the RTX 4080 SUPER at $999 US is a tad bit more expensive than what we used to get in the past so fingers crossed that we can get a 5080 GPU for under $1000 US but given with zero competition in the high-end space, the possibility of that happening are very little.
While the specifications listed here might be possible, we should still be vary that early contig can change a lot before the launch as we have seen with previous launches. NVIDIA is said to have the final specs laid out but we can't be 100% certain and it's wise to wait for the official announcement from the green team themselves. With the launch closing in for later this year or early next year, we can expect more information in a few months.
NVIDIA GeForce RTX 5090 & RTX 5080 GPU Specs (Rumor):
| 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) |
