Positioning & Architecture
This is our seventh GeForce RTX 5090 review, featuring the ASUS RTX 5090 TUF. We've previously reviewed the ASUS RTX 5090 Astral flagship, and hoped that ASUS would be willing to provide the TUF for review, too, but their focus is on the Astral. Thanks to one of our readers, who provided the card on loan so we can still bring you the TUF review. That's also why the TUF in this review is not the "TUF OC," but the normal one—vendors usually send their OC models, while a lot of gamers purchase the non-OC variants, because they are cheaper and the factory OC is tiny anyway.
The Blackwell architecture introduces several architectural improvements under the hood, like giving all shaders the ability to run FP32 or INT32 instructions, on Ada only half the cores had that ability. The Tensor Cores are now accessible from the shaders through a new Microsoft DirectX API, and they now support FP4 and INT4 instructions which run at lower precision, but much faster with less memory usage. There's numerous additional architecture improvements, we talked about all of them on the first pages of this review.
Compared to last generation's flagship, the RTX 4090, today's RTX 5090 increases the number of GPU cores to 21,760, up from 16,384. Other unit counts are increased, too. One of the highlights is the switch to the brand-new GDDR7 memory, which further increases bandwidth per pin, and NVIDIA bumped the memory bus from 384-bit to 512-bit, which results in a staggering 1.8 TB/s memory bandwidth. From a fabrication perspective nothing has changed though—Blackwell is built on the same 5 nanometer "NVIDIA 4N" TSMC node as last generation's Ada. NVIDIA claims this is a "4 nanometer process," but during Ada it was confirmed that NVIDIA 4N is actually not TSMC N4 (note the order of N and 4), but 5 nanometer. At the end of the day the actual number doesn't matter much, what's important is that NVIDIA is using the same process node.
Performance
We upgraded our test system in preparation for this wave of GPU launches, which is now built on AMD technology with the outstanding Ryzen 7 9800X3D. We've updated to Windows 11 24H2, complete with the newest patches and updates, and have added a selection of new games. At 4K resolution, with pure rasterization, without ray tracing or DLSS, we measured a 41% performance uplift over the RTX 4090. While this is certainly impressive, it is considerably less than what we got from RTX 3090 Ti to RTX 4090 (+51%). NVIDIA still achieves their "twice the performance every second generation" rule: the RTX 5090 is twice as fast as the RTX 3090 Ti. There really isn't much on the market that RTX 5090 can be compared to, it's 81% faster than AMD's flagship the RX 7900 XTX. AMD recently released new GPUs, the RTX 5090 is roughly twice as fast as the Radeon RX 9070 XT, and AMD has confirmed that they are not going for high-end with RDNA 4. Compared to the second-fastest Ada card, the RTX 4080 Super, the performance increase is 82%—wow! The RTX 5090 is also 56% faster than the RTX 5080, which is a huge jump.
Interestingly, despite being a non-OC model, the ASUS TUF does gain quite a bit of performance over the NVIDIA RTX 5090 Founders Edition, which is a bit unexpected. I looked into this and the primary reason is that the ASUS TUF is running MUCH lower temperatures than the Founders Edition (66°C vs 77°C). NVIDIA's Boost algorithm will automatically lower the boost clocks as temperatures go up—if they don't go up as much, the GPU will be clocked higher. You can see this on the temperature chart, note how the GPU clock starts out very high and then drop over time, as the card heats up. It also seems that the card is binned slightly better, in the end it runs around 100 MHz higher clocks on average, despite both cards having a boost clock rating of 2407 MHz. Just to confirm, the ASUS TUF in this review is a retail card, not a press sample, so it is definitely not cherry-picked for maximum out-of-the-box performance.
There really is no question, RTX 5090 is the card you want for 4K gaming at maximum settings with all RT eye candy enabled. I guess you could run the card at 1440p at insanely high FPS, but considering that DLSS 4 will give you those FPS even at 4K, the only reason why you would want to do that is if you really want the lowest latency with the highest FPS.
Ray Tracing & Neural Rendering
NVIDIA is betting on ray tracing and Blackwell comes with several improvements here. Our results show a several percent smaller performance loss from enabling RT, which definitely helps. On top of that, the company is introducing several new optimization techniques that game developers can adopt. The most interesting one is Neural Rendering, which is exposed through a Microsoft DirectX API (Cooperative Vectors). This ensures that the feature is universally available for all GPU vendors to implement, so game developers should be highly motivated to pick it up. Performance in RT is good, at 4K we see a +32% improvement over the RTX 4090, +61% vs RTX 5080. AMD's fastest the RX 7900 XTX is way behind, the RTX 5090 is 2.5x (!) faster in this scenario. With the RDNA 4 architecture on the RX 9070 Series, AMD has given their RT cores some extra love, the RX 9070 XT is faster than the RX 7900 XTX in ray tracing, but still falls behind the RTX 5090, offering roughly half the performance in RT, at much lower pricing of course.
VRAM
RTX 5090 comes with a staggering 32 GB VRAM size, which is a first for any consumer card. Modern games can't even make use of the 24 GB on RTX 5080/4090, so this increase isn't something that benefits gamers today. I'm having serious doubts that we'll see 32 GB VRAM usage in any game in the next few years, but you never know, maybe next-gen consoles will bump the memory to 32 GB, which could motivate game studios to use that much memory. For creators and AI, this capability is useful though, as it lets them run larger problems, but that also means that these people, with deep pockets, will snatch away cards from gamers. When looked at from a different angle, 32 GB makes sense though. If you want to build a graphics card with a 512-bit memory bus, you will end up with 16 memory chips, which means your choices are either 16 GB or 32 GB. 16 GB wouldn't make sense for the RTX 5090, so 32 GB is the only logical choice, despite the increase in cost and design complexity.
DLSS 4 Upscaling & Frame Generation
NVIDIA made a big marketing push to tell everyone how awesome DLSS 4 is, and they are not wrong. First of all, DLSS 4 Multi-Frame-Generation. While DLSS 3 doubled the framerates by generating a single new frame, DLSS 4 can now triple or quadruple the frame count. In our testing this worked very well and delivered the expected FPS rates. Using FG, gaming latency does NOT scale linearly with FPS, but given a base FPS of like 40 or 50, DLSS x4 works great to achieve the smoothness of over 150 FPS, with similar latency than you started out with. Image quality is good, if you know what to look for you can see some halos around the player, but that's nothing you'd notice in actual gameplay.
Want lower latency? Then turn on DLSS 4 Upscaling, which lowers the render resolution and scales up the native frame. In the past there were a lot of debates whether DLSS upscaling image quality is good enough, some people even claimed "better than native"—I strongly disagree with that—I'm one of the people who are allergic to DLSS 3 upscaling, even at "quality." With Blackwell, NVIDIA is introducing a "Transformer" upscaling model for DLSS, which is a major improvement over the previous "CNN" model. I tested Transformer and I'm in love. The image quality is so good, "Quality" looks like native, sometimes better. There is no more flickering or low-res smeared out textures on the horizon. Thin wires are crystal clear, even at sub-4K resolution! You really have to see it for yourself to appreciate it, it's almost like magic. The best thing? DLSS Transformer is available not only on GeForce 50, but uses Tensor Cores on all GeForce RTX cards! While it comes with a roughly 10% performance hit compared to CNN, I would never go back to CNN. NVIDIA has constantly been improving the list of supported DLSS 4 games, it's at over 75 now, most through NVIDIA App overrides, and many more are individually tested, to ensure best results. NVIDIA is putting extra focus on ensuring that there will be no anti-cheat drama when using the overrides.
Physical Design, Heat & Noise
The ASUS TUF looks fantastic thanks to its high-quality all-metal construction that gives the card an excellent look and feel. Visually, the design theme almost matches the GeForce 40 TUF identity, which is perfectly fine of course. The quad-slot, triple-fan cooling solution achieves excellent temperatures—just 66°C, which is very low, especially when considering that this is a 575 W graphics card. Unfortunately noise levels are fairly high, reaching 41 dBA, which makes the card even a little bit louder than the NVIDIA Founders Edition, which is a compact dual-slot design. I think ASUS went a bit overboard with their fan settings, focusing on low temperatures way too much, which results in a noisy gaming experience. The good thing is that the card comes with a dual BIOS feature, which lets you enable a "quiet" BIOS. Now the card really runs quiet, reaching only 32 dBA, which is one of the best results in our test group. Temperatures are still fine—73°C, which is only a small increase.
Our apples-to-apples cooler comparison test reveals that the ASUS cooler sits roughly in the middle of our test group, definitely stronger than the FE thermal solution, but not nearly as strong as what we've seen on the MSI Suprim, for example. Compared to the Founders Edition, at the same heat load and noise level, the GPU runs 12.6°C cooler.
PCI-Express 5.0
NVIDIA's GeForce Blackwell graphics cards are the first high-end consumer models to support PCI-Express 5.0. This increases the available PCIe bandwidth to the GPU, yielding a small performance benefit. Of course PCIe Gen 5 is backwards compatible with older versions, so you'll be able to run the RTX 5090 even in an older computer.
Just like we've done over the years, we took a detailed look at
PCI-Express scaling in a separate article. Testing includes x8 Gen 5, for instances when an SSD is eating some lanes. The popular x16 4.0 was tested, which is common on many older CPUs and entry-level motherboards. Finally, some additional combinations were run, down to PCIe x16 1.1. The results confirm that unless you are on an ancient machine, PCIe bandwidth won't be a problem at all.
Power Consumption
The ASUS TUF runs at the same 575 W power limit as the Founders Edition, you can increase that to 600 W. Despite the high absolute power number, the efficiency (with FPS results taken into account) is good—slightly better than the RTX 5090 FE, actually. On the Founders Edition we measured very high non-gaming power draw of 30+ W, the TUF runs a tiny bit better than that, but still high. Hopefully NVIDIA can address this.
Overclocking
Overclocking the ASUS RTX 5090 TUF worked quite well, and we gained +7% in real-life performance, which is more than we see usually. Unfortunately NVIDIA is limiting the maximum overclocking for the GDDR7 memory chips to +375 MHz—usually NVIDIA doesn't have any OC limits. I guess that's just another bug that will be fixed eventually—we really want to push these memory chips to the max!
Pricing & Alternatives
The official ASUS MSRP for the RTX 5090 Gaming OC is $2450, but of course, right now the card is sold out and scalpers are asking for thousands of dollars—just like for any other RTX 5090. It seems new supply is still a few weeks away, so it doesn't look like the situation will be improving any time soon. At the MSRP, the ASUS card is 23% more expensive than the $2000 NVIDIA Founders Edition, which is a pretty steep increase for a better cooler with amazing build quality, no factory OC, a tiny RGB lighting element and a dual BIOS. Compared to some other premium RTX 5090 cards the cooler is definitely not the most powerful, which makes me wonder why pick this over other cards with better cooling—unless of course you have no choice because everything is sold out. With the quiet BIOS, the card is definitely one of the quieter options on the market, so it could be interesting for gamers looking for a quiet RTX 5090 model—and I can see how this crowd would be willing to pay the extra cost.
Unfortunately the GPU market at this time is a huge mess—everything decent is out of stock and scalped to insane pricing. And apparently people are still paying these prices or the price would go down. I'm also wondering why NVIDIA had to launch so many cards in such a short time, with very little inventory, which makes people extremely unhappy. The same can be said for MSRP price points that are unrealistic, and board partners are adding hundreds of dollars on top, or "launch" SKUs at MSRP, in tiny volume, that will never be replenished. Another big problem are the various tariffs that are coming into effect right now, which will push prices even higher. Not the best time to be shopping for a GPU.
Generally, the RTX 5090 is very expensive, but it is the best and everyone wants to have one. AMD, the next-closest competitor, lacks the raw performance, and they don't have anything to counter software features like DLSS multi-frame-generation and the new Transformer upscaler model. The new AMD RX 9070 Series targets a much lower performance segment, at much more affordable pricing, too, of course, and they are all sold out, too. NVIDIA's dominance with game studios will ensure that DLSS 4 support is coming to virtually all new releases—no doubt. Previous generation RTX 40 cards aren't obsolete at all, though. Thanks to DLSS Transformer you're getting a free image quality upgrade on all RTX cards. Also, isn't gameplay the most important thing? Rather than graphics? Are games fun at 1080p, with medium settings? Absolutely, as long as the game is good.