Tech Reviews & Gadgets

Nvidia GeForce RTX 5080 release date, price and specs

Jensen Huang now proudly shows off its new product NVIDIA GeForce RTX 5080 We were at the CES trade show in Las Vegas for gaming GPUs. Thankfully, it’s significantly cheaper than the RTX 4080 and has similar core specs, with 16GB of VRAM. We believe there won’t be any “launch” this time. In this guide, we’ll take you through the RTX 5090 release date, as well as the price and specs of the new graphics card.

While the RTX 5090 takes pride of place as Nvidia’s flagship in its new Blackwell lineup of gaming GPUs, and will likely win a Best Graphics Card medal, the RTX 5080 is half the price. Also, while Nvidia misjudged the pricing of the original RTX 4080, it (almost) got it right when it launched the cheaper, faster RTX 4080 Super in early 2024 for $999.

Nvidia GeForce RTX 5080 release date

RTX 5080 release date is Thursday, January 30, 2025This comes just weeks after Nvidia unveiled its new GPUs on January 6 at the CES trade show in Las Vegas.

Nvidia GeForce RTX 5080 price

RTX 5080 sells for $999 For MSRP, refers to Nvidia Founders Edition stock rate. That’s in line with the price of the RTX 4080 Super when it launched, and $200 cheaper than the original RTX 4080’s $1,199 price.

Nvidia GeForce RTX 5080 specifications

Nvidia GeForce RTX 5080 specs show it to have similar performance to the RTX 4080, with 16GB VRAM and a 256-bit memory interface, although it also has the advantage of using GDDR7 memory.

NVIDIA GeForce RTX 5080
CUDA Core 10,752
base clock 2.3GHz
boost clock 2.62GHz
RT core Fourth generation, 171 TFLOPS
tensor core Fifth generation, 1,801 AI TOPS
interface 16 PCIe 5.0
Video memory 16GB 32Gbps GDDR7
memory interface 256 bit
memory bandwidth 960GB/sec
power consumption 360W
power connector 1 x 16 pin

The Nvidia GeForce RTX 5080 will be built on the new architecture, using one of the company’s new Blackwell RTX 50 GPU designs. While the RTX 5090 was rumored to use a large new chip codenamed GB202, the latest RTX 5080 leaks claim the new graphics card will use a smaller chip codenamed GB203.

It’s likely that the chip will also be used in other GPUs in Nvidia’s stack, such as the RTX 5070 Ti, where parts of the GPU are deactivated, but the RTX 5080 is rumored to use a variant codenamed GB203-400-A1.

Nvidia says the RTX 5080 will have 10,752 CUDA cores, which is less than half the number of the RTX 5090, but to be fair, the price is only half of the RTX 5090. This number is similar to the 10,240 CUDA cores currently in the RTX 4080 Super, but much lower than the RTX 5090’s rumored number. The flagship GPU is rumored to have 21,760 CUDA cores, more than double the 10,750 cores rumored for the RTX 5080.

Likewise, the RTX 5080 launched with 16GB of GDDR7 memory, and while that’s faster than the GDDR6X VRAM used in the RTX 4080, the amount remains the same. However, recent leaks claim that a 24GB RTX 5080 may launch later, giving the card more headroom to run high-resolution games with tons of detail.

What’s more, the 24GB card is rumored to feature the same core GPU specs as the 16GB card, a feat now technically possible thanks to the just-released Samsung 24Gb GDDR7 chip. The chip has a 3GB capacity in the same package size as a 2GB chip, meaning it can directly replace a 16GB card.

But unfortunately, the width of the video memory interface is the same as the RTX 4080, 256 bits, which is disappointing at this price. Given that the RTX 5090 has a 512-bit interface, the RTX 5080 should have at least a 320-bit bus and more VRAM.

On the plus side, using GDDR7 VRAM should provide more bandwidth. In fact, Nvidia says the memory bandwidth is 960GB/s, which is not far behind the RTX 4090, showing the benefits of upgrading to GDDR7.

Finally, the RTX 5080 consumes 360W and requires a 16-pin power connector and at least an 850W PSU.

Nvidia GeForce RTX 5080 Benchmark Estimates

Our Nvidia GeForce RTX 5080 benchmarks estimate the card to be slightly faster than the RTX 4080 Super at rasterization, but significantly faster at ray tracing.

There aren’t a lot of leaks about the RTX 5080’s performance yet, but Nvidia claims to have made great strides in ray tracing in partnership with Blackwell, and one leak claims that the RTX 5080 will have faster ray tracing than the RTX 4090. According to XpeaGPU news, the performance of the GB203 GPU rumored to be used in the RTX 5080 is “close to AD102 in terms of raster and faster in terms of RT.”

Over the past few generations, Nvidia’s new GPU architecture has made big improvements in ray tracing performance over previous generations, but the improvements in non-ray tracing game performance (rasterization) have been smaller. This appears to be the case with the RTX 5080 as well.

Nvidia has also introduced new neural rendering technology in this GPU series, as well as DLSS 4, which supports multi-frame generation as long as the game supports it to further improve performance.

Despite its purportedly lower specs, this leak suggests the RTX 5080 will still be faster than the RTX 4090, thanks to improvements to the Blackwell architecture. However, we won’t know for sure until we have the cards in hand to test for ourselves.

In the meantime, while we wait for the RTX 5080 to launch, you can read our complete guide to Nvidia DLSS, where we explore Nvidia’s upgrade and the power of the frame-generation technology.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button
×

Adblock Detected

*We Appreciate Your Visit!*

To enjoy all the features and content on our website, please consider disabling your ad blocker. Our site relies on ads to provide you with quality content and a seamless experience. Thank you for your understanding!