After the catastrophic launch of the GeForce RTX 4080, the place one card ended up being “unlaunched” and the ultimate card had a ridiculously excessive price ticket caught on it, we’re hoping that Nvidia has discovered some classes for this GPU’s successor. The Nvidia GeForce RTX 5080 specs and efficiency are rumored to supply a number of enhancements over the RTX 4080, however with out rocking the boat an excessive amount of.
Whereas the RTX 5090 seems to be set to be the Nvidia flagship on this subsequent GPU technology, the place it’s more likely to take the perfect graphics card medal, the RTX 5080 ought to be less expensive. Plus, whereas Nvidia misjudged the pricing of the unique RTX 4080, it (nearly) returned to type when it launched the cheaper and quicker RTX 4080 Tremendous for $999 initially of 2024.
Nvidia GeForce RTX 5080 specs estimate
The most recent Nvidia GeForce RTX 5080 specs rumors level to the brand new GPU having round half the facility of the RTX 5090, with simply 10,752 CUDA cores and 16GB of GDDR7 VRAM.
| Nvidia GeForce RTX 5080 | |
| GPU | GB203-400-A1 |
| CUDA cores | 10,752 |
| SMs | 84 |
| RT cores | 84 |
| Interface | 16x PCIe 5.0 |
| VRAM | 16GB GDDR7 |
| Reminiscence interface | 256-bit |
| Reminiscence bandwidth | ~896MB/s |
| Energy draw | 400W |
| Energy connector | 1 x 16-pin |
The Nvidia GeForce RTX 5080 might be constructed on a brand new structure, utilizing one of many firm’s new Blackwell RTX 50 GPU designs. Whereas the RTX 5090 is rumored to make use of a big new chip, codenamed GB202, the newest RTX 5080 leaks say that this new graphics card will use a smaller chip, codenamed GB203.
It’s seemingly that this chip may also be utilized in different GPUs additional down the Nvidia stack, such because the RTX 5070 Ti, with some elements of the GPU disabled, however the RTX 5080 is rumored to make use of a variant codenamed GB203-400-A1.
In response to the newest rumor from tech leaker kopite7kimi, the RTX 5080 CUDA cores will stand at 10,750, unfold throughout 84 of Nvidia’s Streaming Multiprocessor (SM) items. This setup would additionally provide you with 84 RT cores, assuming Nvidia sticks to an analogous block design to its present Ada Lovelace GPU, the place every SM has an RT core.
These are related numbers to the ten,240 CUDA cores and 80 SMs discovered within the present RTX 4080 Tremendous, however they’re dramatically decrease than the rumored figures for the RTX 5090. The flagship GPU is rumored to have 21,760 CUDA cores, which is greater than double the ten,750 rumored for the RTX 5080.
Equally, in response to some leaks, the RTX 5080 has 16GB of GDDR7 reminiscence at launch, and whereas this might be quicker than the GDDR6X VRAM used within the RTX 4080, the quantity is staying the identical. Nevertheless, one other latest leak claimed {that a} 24GB RTX 5080 could be launched later, giving the cardboard a bit extra headroom for operating video games at excessive resolutions with a number of element.
Sadly, although, it seems to be as if the reminiscence interface goes to have the identical 256-bit width because the RTX 4080, which is disappointing if this finally ends up being an costly GPU. If the RTX 5090 does certainly have a 512-bit interface, then the RTX 5080 ought to arguably no less than have a 320-bit interface.
On the plus aspect, using GDDR7 VRAM ought to present a bit extra bandwidth. If the VRAM runs at 28GBps, for instance, that might imply the RTX 5080 reminiscence bandwidth is 896MB/s, which is faster than the 763MB/s of the RTX 4080 Tremendous.
Lastly, the RTX 5080 energy draw is rumored to face at 400W, which implies it is going to solely want one 16-pin energy connector. Some hypothesis has claimed the RTX 5090 wants two energy connectors, and whereas that is already unlikely, notably for Founders Version playing cards operating at inventory velocity, it’s out of the query for the RTX 5080 if it attracts 400W.
Nvidia GeForce RTX 5080 launch date estimate
Our greatest estimate for the RTX 5080 launch date window is January 2025, primarily based on rumor and hypothesis, as Nvidia hasn’t formally introduced a date.
Traditionally, the 80-series GPU has been the primary to launch, with the RTX 1080, 2080, and 3080 all arriving earlier than the dearer playing cards. Nevertheless, Nvidia threw us a curveball when it launched the RTX 4090 earlier than the 4080 again in 2022, and there are rumors that Nvidia might do that once more with its new Blackwell lineup.
Stories are blended right here, although. Early leaks pointed to the RTX 5090 launching in 2024, and being the one GPU from the brand new lineup to take action. Nevertheless, tech leaker kopite7kimi later claimed that the RTX 5080 would launch first, with the RTX 5090 coming later.
The identical leaker has now mentioned that no new RTX 5000-series Blackwell GPUs are anticipated to launch in 2024, with the expectation that each the RTX 5090 and RTX 5080 will launch in January 2025, presumably with an announcement on the CES tradeshow.
Nvidia GeForce RTX 5080 worth estimate
Our greatest estimate for the RTX 5090 worth is $999, primarily based on the present worth of the RTX 4080 Tremendous, which was considerably cheaper than the unique RTX 4080 when it launched.
We’re actually hoping that Nvidia discovered a lesson from the worth of the unique RTX 4080. Launching at $1,199, it was solely $400 cheaper than the RTX 4090, however considerably much less highly effective. What’s extra, its predecessor the RTX 3080 launched at an MSRP of simply $699, even when the next GPU scarcity meant it was virtually unattainable to truly purchase at that worth.
Sadly, we don’t suppose we’ll see the times of $699 80-series GPUs once more, no less than not until AMD or Intel steps as much as supply some critical competitors, however we don’t suppose Nvidia goes to shoot itself within the foot with one other $1,199 80-series GPU launch. $999 seems to be like the worth persons are keen to pay for this degree of energy, and that’s the worth we’re anticipating for the RTX 5080.
Nvidia GeForce RTX 5080 benchmarks estimates
Our Nvidia GeForce RTX 5080 benchmarks estimate places the cardboard as barely quicker than the RTX 4080 Tremendous in rasterization, however considerably faster in ray tracing.
There haven’t been many leaks about RTX 5080 efficiency but, however one rumor is that the RTX 5080 ray tracing velocity might be even faster than that of the RTX 4090. This comes from common tech leaker XpeaGPU, who claimed that the efficiency of the GB203 GPU rumored for use within the RTX 5080 is “near AD102 in raster and quicker in RT.”
Over the previous couple of generations, Nvidia’s new GPU architectures have made massive strides in ray tracing efficiency over earlier generations, however non-ray traced sport efficiency (rasterization) has improved by a smaller diploma. It seems to be as if this may also be the case with the RTX 5080.
Even with its purportedly low specs, this leak suggests the RTX 5080 will nonetheless be quicker than the RTX 4090, because of the enhancements made to the Blackwell structure. We’ll solely know for positive after we’ve had the playing cards in our arms to check for ourselves, nevertheless.
Within the meantime, whereas we anticipate the RTX 5080 launch, you possibly can learn our full information to Nvidia DLSS, the place we discover what makes Nvidia’s upscaling and frame-generation tech tick.
