Nvidia kicked off 2025 by announcing its latest series of graphics cards led by the Nvidia GeForce RTX 5090. Here’s how the flagship RTX 5090 compares to the second-ranked RTX 5080.
The RTX 5090 and the RTX 5080 are both powered by Nvidia’s latest Blackwell architecture. With the 5080 costing close to half the price of the 5090, you might find yourself wondering whether it’s worth paying $1000/£960 more for the top-end 5090.
Keep reading to learn how the two 50 Series GPUs compare when it comes to price, performance, memory, display support, power requirements and more.
| RTX 5090 | RTX 5080 | |
| Nvidia CUDA cores | 21760 | 10752 |
| Shader cores | Blackwell | Blackwell |
| Tensor cores (AI) | 5th Generation, 3352 AI TOPS | 5th Generation, 1801 AI TOPS |
| Ray tracing cores | 4th Generation, 318 TFLOPS | 4th Generation, 171 TFLOPS |
| Boost clock (GHz) | 2.41 | 2.62 |
| Base clock (GHz) | 2.01 | 2.30 |
| Memory | 32GB GDDR7, 512-bit | 16GB GDDR7, 256-bit |
| Maximum digital resolution | 4K at 480Hz, 8K at 165Hz with DSC | 4K at 480Hz, 8K at 165Hz with DSC |
| Nvidia architecture | Blackwell | Blackwell |
| DLSS | DLSS 4 | DLSS 4 |
| Maximum GPU temperature | 90°C | 88°C |
| Total graphics power | 575W | 360W |
| Required system power | 1000W | 850W |
Price
The Nvidia GeForce RTX 50 Series was announced on January 6 2025, with the GPUs arriving in stores three weeks later on January 30. As of February 17, both the 5090 and 5080 remain out of stock on Nvidia’s UK website.
Prices for the Nvidia GeForce RTX 5090 start at $1999/£1939, while the Nvidia GeForce RTX 5080 is the more affordable of the two, with prices starting at $999/£979. This makes the RTX 5090 $1000/£960 cheaper than the flagship 5090 – or around half the price.
if( window.ipcTags.section === ‘mobile’ ) {
var widget_1_sid = null;
window.addEventListener(‘message’, (event) => {
if (
event?.data &&
typeof event.data === ‘string’ &&
event.data.indexOf(‘lre:playerReady://’) === 0)
{
let sessionData = JSON.parse(event.data.slice(‘lre:playerReady://’.length));
console.log(‘lre:playerReady’, sessionData);
let widget = window.anyclip.getWidget(null, sessionData.sessionId);
if (widget) {
console.log(‘LRE WIDGET READY’, widget);
widget.pauseMon();
} else {
console.log(‘LRE WIDGET NOT FOUND’);
}
}
});
}
.keystone-video-block .flexible-container .ac-lre-player-ph-ext {
position: relative;
padding-bottom: calc((2 / 3.1) * 100%);
padding-top: 30px;
height: 0;
overflow: hidden;
}
.keystone-video-block .flexible-container .ac-lre-player-ph-ext.ac-lre-carousel-vertical {
padding-bottom: calc((2 / 3.9) * 100%);
}
.keystone-video-block .flexible-container .acw1 {
position: absolute;
top: 0;
left: 0;
width: 100%;
height: 100%;
}
@media only screen and (min-width:1301px){
.keystone-video-block .flexible-container .ac-lre-player-ph-ext {
padding-bottom: 440px;
}
.keystone-video-block .flexible-container .ac-lre-player-ph-ext.ac-lre-carousel-vertical {
padding-bottom: 350px;
}
}
The Nvidia GeForce RTX 5090 is more powerful
So, why is the RTX 5090 more expensive? The most obvious reason is performance power. The RTX 5090 packs more than twice the number of Nvidia CUDA cores, stronger AI performance and more powerful ray tracing cores.
More specifically, the RTX 5090 is equipped with 21760 CUDA cores compared to the RTX 5080’s 10752 cores. CUDA cores are small processing units that execute computing tasks for Nvidia’s GPUs – the more CUDA cores there are, the faster the GPU’s performance.
Likewise, the Tensor cores in the RTX 5090 are capable of completing 3352 AI TOPS, or Tera Operations per Second. Whereas, the RTX 5080 tops out at 1801 TOPS. This means you can expect more efficient AI and DLSS performance from the higher-end 5090.
Both graphics cards are equipped with Nvidia’s 4th-generation ray tracing cores. However, the RTX 5090 delivers 318 teraflops (TFLOPS) of performance, whereas the RTX 5080 lands a bit shorter at 171 TFLOPS.
When it comes to clock speeds, the RTX 5080 does boast higher base and boost clock speeds. However, you can still expect faster overall performance from the RTX 5090.
The Nvidia GeForce RTX 5090 packs more memory
When it comes to memory, the Nvidia GeForce RTX 5090 is equipped with a standard configuration of 32GB GDDR7 RAM with a 512-bit memory interface width.
The RTX 5080, on the other hand, comes with 16GB of GDDR7 RAM with a 256-bit memory interface width. This means you get double the memory with the RTX 5090, enabling the GPU to offer faster performance and run more applications simultaneously.
The Nvidia GeForce RTX 5080 offers the same display support
Moving on to display support, the GeForce RTX 5090 and RTX 5080 offer matching specs and capabilities, meaning there’s no benefit to choosing one over the other.
Both graphics cards are capable of displaying 4K content at 480Hz, or 8K content at 165Hz with Display Stream Compression (DSC). Both GPUs can also support up to four monitors at one time.
When it comes to connectivity, both options offer three DisplayPort inputs as well as one HDMI input.
The Nvidia GeForce RTX 5080 requires less power
Finally, the Nvidia GeForce RTX 5080 requires less system power to run. The RTX 5080 needs 850W of power, whereas the RTX 5090 requires 1000W.
Likewise, the RTX 5090 can hit a higher maximum temperature of 90°C and produces 575W of graphics power, compared to the 88°C temperature of the RTX 5080 and 360W of total graphics power.
Early verdict
The RTX 5090 is the most powerful GPU in the Nvidia GeForce 50 Series, beating the RTX 5080 in many areas.
We have yet to review either GPU ourselves, meaning we can’t judge the performance based on experience just yet. However, the RTX 5090 definitely comes out on top on paper.
You might like…
The post Nvidia RTX 5090 vs 5080: Which GPU should you buy? appeared first on Trusted Reviews.
