Is the Radeon RX 9070 XT really faster than the GeForce RTX 5090? It is a strange question, I know, but it is one quite a few of you have been asking me over the past week, and I think that's because of a few other videos that have popped up on YouTube. So, I am going to address it. Now, you might be thinking, "That is a super dumb question. Why would anyone ask that? And why am I addressing it?" Well, hear me out. Sometimes the 9070 XT can be faster than an RTX 5090, but also not really faster than an RTX 5090. It does depend. It is a bit complicated, and I'll explain everything right after this.
Today's sponsor spot is brought to you by MSI and their GeForce RTX 50 series. Built with Nvidia's Blackwell architecture and enhanced by MSI's premium board design, power delivery, and thermal solutions, RTX 50 series cards deliver major gains in games, AI workloads, streaming, and productivity. Enjoy many of the latest games such as Control Resonant with the full suite of RTX features such as DLSS 4.5, ray reconstruction, path tracing, and dynamic multi-frame generation. Ray reconstruction improves image quality in supported ray traced and path traced games, offering better lighting accuracy, motion clarity, and stability, while dynamic frame generation helps to better utilize high refresh rate monitors. MSI offers a number of higher quality models from the Shadow range to the Supreme. So for more information on MSI's entire range, please click the link in the video description.
Okay, so 5 years ago now, I investigated an issue that saw Radeon GPUs performing up to and sometimes in excess of 30% faster in games when compared to GeForce GPUs under CPU limited conditions. This issue did primarily affect lower-end CPUs that are more likely to result in CPU bound performance when gaming. As it turns out, Radeon GPUs don't lean as heavily on your CPU. And as a result, when CPU bound, a Radeon GPU will provide more performance headroom. In part two of the Nvidia driver overhead investigation, I was able to show real-world examples where the Radeon RX 5600 XT was faster than a GeForce RTX 3080, which was pretty alarming given that on average the 3080 is around 130% faster than the 5600 XT. Now, in order to clearly demonstrate this, I was required to use an incredibly slow processor in the Ryzen 5400, but the point wasn't necessarily to test a real world or realistic hardware configuration. Rather, we were looking to test a theory.
So, what is driver overhead or CPU overhead? In simple terms, it's the amount of processing power your CPU must use to translate game engine instructions into commands that your GPU can then execute. Now, although all graphics drivers require some CPU time, Nvidia's architectural approach demands significantly more than AMD's. And if you want more information on the subject, I suggest going and watching the original videos. I'll link them in the description. But it is important to make it clear that Nvidia's driver overhead issue isn't always going to be responsible for the slim margin seen between the Radeon RX 9070 XT and GeForce RTX 5090. That can simply just be a result of a plain old CPU bottleneck limiting performance of various GPUs to roughly the same level.
Now, recently it has been suggested that while the RTX 5090 looks glorious in reviews and benchmarks, in reality it's not nearly as impressive, often coming in slower or at best not much faster than the cheaper 9070 XT, especially when shooting for higher frame rates. I guess the theory here is when aiming for 200 plus frame rates, a lot more is asked of your CPU. And with Nvidia's driver overhead issue, this can cause issues for GPUs such as the RTX 5090. And for those of you unaware, before your graphics card can render a single pixel on your screen, your CPU has to prepare the scene. The CPU calculates the game logic, it determines player inputs, physics, AI movement, and collision detection just as some examples. So once it calculates all of that where everything should be, it then sends a command called a draw call to the GPU.
Because this pipeline happens sequentially, the CPU must repeat its workload for every single frame rendered. Therefore, at 60 fps, your CPU has to calculate game logic and issue draw calls 60 times every second. At 240 frames per second, it must be done 240 times per second. And that means the CPU would now be working four times harder. So if your CPU can throughput say 240 FPS in a given game, but your GPU is capable of 300 FPS, you'll still end up limited to 240 FPS. As in this example, the CPU is the weakest link. And that means if the 9070 XT can achieve the high frame rates that you're after while reaching the limits of your CPU using what should be a much faster GPU in the RTX 5090, it won't result in any additional performance because you've hit the limits of your CPU. And in fact, as we've seen in the past, in some instances, performance can actually go backwards with a GeForce GPU relative to what you'd see with a Radeon GPU due to the additional CPU overhead.
All of that said, the only way the 9070 XT could possibly match or even beat the RTX 5090 would be when CPU limited. With a slow CPU, that's likely going to happen at sub 100fps. But the claim is if you're after say 200 FPS or better, the 9070 XT is often going to end up better than the RTX 5090, either with better performance or I suppose some are arguing much better value, which is fair enough given current pricing. But we're going to tackle the performance angle for this one. So, is it true? Can the RX 9070 XT really beat the RTX 5090? Well, to find out, I'll be comparing them at 1440p and 4K with and without upscaling using a range of quality settings. Please note, for all of the testing, I'm using the Ryzen 7 9800X3D because it really only makes sense to test a graphics card like the RTX 5090 with the fastest possible gaming CPU.
So, let's get into it. We'll start with Space Marine 2 because from previous testing, I know this is a game that has problems with Nvidia's overhead issue. Looking at the native data first, using the ultra preset, we see that at 4K, the RTX 5090 is a massive 96% faster than the 9070 XT, going from just 67 FPS right up to an impressive 131 FPS. However, at 1440p, that margin is heavily reduced to just 21%. And while in part this margin can be blamed on Nvidia's CPU overhead issue, it is largely just a result of a plain old CPU bottleneck. So there's nothing particularly unusual about this data. Space Marine 2 is an extremely CPU heavy game, which makes pushing frame rates over 100 FPS in our Horde battle test very difficult. For those playing with performance upscaling enabled at 4K using either DLSS for the 5090 or FSR for the 9070 XT, you'll find the GeForce GPU to be a mere 15% faster. However, if you're playing at 1440p in the more CPU demanding sections of the game, the Radeon GPU will actually end up slightly faster, 7% in this example, and this is a result of Nvidia's additional CPU overhead. So, if you've bought an RTX 5090 to play Space Marine 2 at anything other than native 4K using the maximum quality settings, you've wasted your money. But I don't suspect this will be the case for many gamers.
But out of interest, I ran these tests again, this time using the low preset. This doesn't necessarily place more load on the CPU. In fact, if you often set a game to its lowest possible settings, you aren't just turning down how good the game looks, you're also turning down the complexity of the game. Testing a CPU with low settings essentially tests how fast it can run a visually barren empty world. It's more of a throughput test. It doesn't reveal how well the CPU will handle the intense logic and physics calculations of a complex AAA title. For example, settings such as draw distance and level of detail place serious load on the CPU. The CPU has to manage the geometry and issue thousands of additional draw calls to the GPU for all of those distant objects. Still, as I said, this is more of a throughput test, and it does further highlight Nvidia's overhead problem, as the 9070 XT is now able to deliver higher frame rates at native 1440p, along with all of the upscaled testing, providing up to 11% greater performance at 1440p with balanced upscaling. Even so, at native 4K, the RTX 5090 was still 44% faster, less than half the margins seen with the ultra settings. But of course, this is because we're CPU limited to around 140 FPS with the RTX 5090.
Next up, we have Star Wars Zero Company, a turn-based strategy game that's also quite heavy on the CPU. Testing with the ultra preset, we see that at 4K, the RTX 5090 is 63% faster than the 9070 XT. So, a pretty big margin there, I'd say. Though, the margin is cut in half at native 1440p to just 31%. However, when using upscaling, we see the margins reduced further at 4K with performance upscaling. The 5090 is now just 22% faster. And perhaps more notable were the 1% lows, where the GeForce GPU was just 10% faster. Then, for those playing at 1440p with balanced upscaling, the RTX 5090 was just 10% faster on average or 3% when looking at those 1% lows. Still, this data isn't that unusual, and for the most part, you can just chalk this up to a good old-fashioned CPU bottleneck because really, that's what it is. You're not pushing big frame rates in Zero Company, and frankly, you don't need to.
Switching to the low preset reveals some slightly more interesting data. At native 4K, the RTX 5090 remained much faster, delivering 48% greater average frame rate performance. However, at 1440p, it's now just 23% faster with comparable 1% lows. Upscaled 4K reduces those margins further. Now, the 5090 is just 11% faster on average, but 3% slower for the 1% lows. Then, at 1440p with balanced upscaling, we hit the CPU limits. Both GPUs average just shy of 140 FPS, though the 1% lows were 12% stronger for the Radeon GPU.
Of course, not every game is going to be as CPU limited as Space Marine 2 or Star Wars Zero Company. In fact, most aren't, and Doom: The Dark Ages is a really good example of this. And yeah, look, sure, we do have path tracing enabled in this particular set of results, but this rendering technique does require quite a bit of your CPU. And despite that, the RTX 5090 is able to destroy the humble 9070 XT. At native 4K, the 5090 might have delivered a rather unimpressive 35 fps average, but that did make it 250% faster than the 9070 XT. The native 1440p data is a little more realistic, but even so, here the RTX 5090 was still 89% faster. Then, with performance upscaling enabled, the GeForce GPU thrives, smashing the 9070 XT by a 178% margin. And this is of course at 4K or 4K with upscaling. Then at 1440p using balanced upscaling, the GeForce GPU was an impressive 160% faster, rendering 122 FPS opposed to just 47 FPS.
Given how high frame rates were with the Unreal preset in this game, which is the third highest preset in the game, though we're not enabling path tracing with this one, I decided not to go any lower because 100 fps at native 4K from the RTX 5090 seems pretty good to me and that made it almost 80% faster than the 9070 XT. Now, decreasing the resolution to native 1440p did also decrease the margin. But even so, here the RTX 5090 remained 60% faster with 162 FPS versus 101 FPS. Then at 4K with performance upscaling enabled, the RTX 5090 was 54% faster than the 9070 XT and then just 28% faster at 1440p using balanced upscaling. So, the margins did come down as you'd expect, but the 5090 was also averaging over 200 FPS.
Next up, we have Forza Horizon 6. And here, the RTX 5090 using the Extreme Plus RT preset at 4K is 90% faster than the 9070 XT, rendering 74 FPS opposed to just 39 FPS. Then at native 1440p, the GeForce GPU remained much faster, though the margin did also come down quite a bit to 62%. Now, at 4K with performance upscaling enabled, the RTX 5090 was 90% faster. So, a massive increase there. But as we see, if you're going to game at 1440p with balanced upscaling enabled, the 5090 really isn't much faster than the 9070 XT, rendering just 21% more frames on average with the same 1% low performance.
Now, all of that said, if you were buying an RTX 5090 to play Forza Horizon 6 using the medium preset, which I can't imagine anyone has done, but if you have, things get a bit messy in relation to the 9070 XT. The GeForce GPU is still 41% faster at native 4K, so that's something. But at native 1440p, that margin was reduced to just 11%. And in fact, the 9070 XT was 12% faster when comparing 1% lows, and this is a result of the Nvidia overhead issue. Then at 4K with performance upscaling enabled, the 9070 XT and RTX 5090 delivered the same 1% low performance despite the average frame rate of the GeForce GPU being 22% higher. Then at 1440p using balanced upscaling, the 1% lows of the 9070 XT were 16% higher, though the average frame rate was much the same.
Next up, we have Fortnite. Tested using the DirectX 12 epic preset, which leverages Unreal Engine 5's Lumen and Nanite technology. Looking at the native data first at 4K, the RTX 5090 is a massive 93% faster than the 9070 XT, delivering 87 FPS compared to just 45 fps, alongside a 60% lead in 1% lows. Dropping down to native 1440p sees that advantage reduced to 54% with the RTX 5090 averaging 117 FPS versus 76 FPS. Then once performance upscaling is enabled at 4K, the GeForce lead narrows further to 32% on average. And we see that the 1% lows are virtually neck and neck with the 5090 just 4% faster here. At 1440p with balanced upscaling enabled, we run directly into CPU limits and Nvidia's driver overhead issue. While the RTX 5090 still manages a 25% higher average frame rate, the 9070 XT actually pulls ahead for the 1% lows by 6%, delivering 100 FPS against 94 FPS from the GeForce GPU.
Switching over to Fortnite's competitive performance mode produces some very interesting results. Upscaling wasn't tested here as it's not supported in this mode, but with frame rates soaring well into the hundreds, it's not really needed anyway. At 4K, the RTX 5090 is still GPU bound enough to command a 45% lead for the average frame rate, rendering 694 FPS versus 478 FPS. However, despite trailing significantly in averages, the 9070 XT delivered 9% higher 1% lows, and this is a clear indicator of Nvidia's CPU overhead kicking in at these extreme frame rates. Then at native 1440p, we slam right into a system-wide CPU bottleneck around the 700fps mark. Here, the overhead penalty along with the CPU limits leave the RTX 5090 pretty helpless against the 9070 XT at least. The 9070 XT matches and even slightly edges it out by 3% on average while offering 5% stronger 1% lows. So if you happen to be purchasing an RTX 5090 strictly for low-detail competitive gameplay, again, you're wasting your money. But this has been a very well-known fact for a long time now. So I don't expect this to be news to really anyone or hopefully not anyway.
Next up, we have Spider-Man 2. And with ray tracing cranked up using the very high quality preset along with ultimate RT, the RTX 5090 dominates at native 4K, delivering 76 FPS versus just 29 FPS for the 9070 XT. And that meant that the GeForce GPU was a massive 162% faster with nearly triple the 1% low performance. Dropping the resolution or enabling upscaling quickly sees us run into a CPU bottleneck. At native 1440p, the 5090's lead is cut to just 58%. Though, it bounces back to 96% at 4K using performance upscaling. However, at 1440p with balanced upscaling, we hit a hard CPU ceiling near 110 FPS on the GeForce GPU, shrinking its average lead to just 15%. Dropping to the medium preset strips away the ray tracing workload, and this completely changes the dynamic. At 4K, the RTX 5090 remains a solid 74% faster than the 9070 XT, averaging 183 FPS versus 105 FPS. Past this point, the RTX 5090 hits a brick wall, flatlining at 248 FPS across native 1440p, as well as both of the upscaled runs due to severe CPU limitations. At 1440p, using balanced upscaling, the 9070 XT closes the gap to within 9% on average. Though unlike many of the other titles we've tested, the GeForce GPU does maintain stronger frame pacing, boasting a 39% lead for 1% lows. So, the data is more indicative of a typical CPU bottleneck.
Next up, we have Pragmata. Tested using the quality preset with ray tracing enabled, the RTX 5090 is able to flex its muscles here at native 4K, coming in 78% faster than the 9070 XT with 144 FPS versus 81 FPS. And that's coupled with a 68% lead for the 1% lows. Dropping down to native 1440p sees that advantage compressed to 49%. Though the 5090 is still pushing an impressive 221 FPS. Performance upscaling at 4K produces a similar 56% margin in favor of the GeForce GPU. However, at 1440p with balanced upscaling, CPU limits begin to creep in, shrinking the 5090's lead down to 33% as it tops out at 252 FPS. Now, switching over to the performance preset pushes frame rates into the stratosphere. At native 4K, the RTX 5090 delivered a massive 85% performance advantage, averaging 200 FPS against 108 FPS for the 9070 XT. At 1440p, that lead comes down to 51%. And unlike some of the other titles that we've already looked at, the Nvidia GPU doesn't appear to be hindered by driver overhead. With balanced upscaling at 1440p, the RTX 5090 stretches its legs to average an absurd 374 FPS. And that's a 50% lead over the Radeon GPU. And that's also backed up by a strong 53% advantage for the 1% lows.
Next up is Marvel Rivals. And starting with the ultra preset, we find this hero shooter is reasonably demanding at 4K. The RTX 5090 is 85% faster than the 9070 XT, delivering 87 FPS against just 47 FPS. That's alongside 79% greater 1% lows. Dropping down to native 1440p sees that lead compressed to 52% with the GeForce GPU averaging 129 FPS versus 85 FPS. And then with performance upscaling enabled at 4K, the 5090 remained 46% faster. While at 1440p with balanced upscaling, that did narrow the margin to 35% as the CPU limits did begin to become a factor. Though the 5090 still pushed a solid 166 FPS.
Switching to the low preset dramatically increases frame throughput across the board. At native 4K, the RTX 5090 maintained an 86% advantage, averaging 145 FPS compared to 78 FPS. Now, lowering the render resolution to native 1440p yields a 62% lead for the GeForce GPU at 252 FPS. And finally, at 1440p, using balanced upscaling, the RTX 5090 cleared 320 FPS, making it 37% faster than the 9070 XT, while retaining a healthy 29% advantage for the 1% lows.
Up next is Assassin's Creed Black Flag Resynced. And starting with the high preset and extended RT, we see solid scaling for the GeForce GPU at 4K. Here, the RTX 5090 was 59% faster than the 9070 XT, delivering 81 FPS compared to 51 FPS with a similar 63% advantage for the 1% lows. Dropping down to native 1440p reduced that lead to 48%. Here, the 5090 hit 120 FPS versus 81 FPS for the Radeon GPU. And upscaling really does tell a pretty similar story. At 4K with performance upscaling, the 5090 remained 50% faster. But at 1440p using balanced upscaling, the CPU limitations did start to become a problem as the margin shrank to 35%. And the 5090 also only gains an extra 12fps over its native 1440p performance. Switching over to the ultra low preset further shifts the bottleneck over to the CPU. At 4K, the RTX 5090 maintained a respectable 46% lead with 130 FPS versus 89 FPS. However, at 1440p, that gap narrowed to 30%. And at 1440p with balanced upscaling, it fell to just 27%. More telling though were the 1% lows which dropped to just a 9% advantage for the GeForce GPU, demonstrating how heavier frame throughput exposes system bottlenecks and limits the RTX 5090's ability to really stretch its legs like what we see in native 4K.
Moving on to Battlefield 6 and we'll start by looking at the overkill preset data which reveals a game that's heavily GPU bound. At 4K, the RTX 5090 was an impressive 88% faster than the 9070 XT, averaging 122 FPS compared to 65 FPS with a similar 83% advantage for the 1% lows. Dropping to 1440p sees that massive margin held up at 88% rendering 188 FPS versus 100 FPS. And even with upscaling enabled, the CPU limits don't aggressively strangle performance here. At 4K with performance upscaling, the 5090 remained 83% faster. And while 1440p balanced upscaling trims the lead down to 74%, the GeForce GPU still comfortably pushed well over 200 FPS.
Now, dropping down with the medium preset does begin to relieve the GPU load and as a result introduce the typical CPU constraints. At 4K, the RTX 5090 was good for a 57% performance advantage, rendering 166 FPS compared to 106 FPS for the Radeon GPU. However, lower resolutions and with the help of upscaling, that will shrink those margins significantly. For example, at native 1440p, the 5090 is 40% faster. And then once we enable balanced upscaling, the advantage falls to just 19%, averaging 242 FPS versus 204 FPS. And the 1% lows, they also compressed to a 21% margin.
Finally, we have Cyberpunk 2077 Phantom Liberty. And testing with the path traced ray tracing overdrive preset is an absolute bloodbath for the Radeon GPU. At 4K, the RTX 5090 is an astonishing 200% faster than the 9070 XT, turning an unplayable 12fps slideshow into an almost passable 36 FPS average with triple the 1% low performance. Now, dropping to 1440p sees the RTX 5090 deliver, making it 180% faster. Even with upscaling engaged, AMD's 9070 XT simply cannot cope with this workload. At 4K, with performance upscaling, the GeForce GPU remained 163% faster. And at 1440p with balanced upscaling, it was still 126% faster, easily clearing 120 FPS, while the 9070 XT failed to hit 60 FPS. However, switching over to the standard rasterized ultra preset did provide far healthier numbers across the board. Though the GeForce GPU still commands a very hefty lead. At 4K, it was 85% faster, pushing 115 FPS versus just 62 FPS. Then at native 1440p, that margin came down to 64%, while 4K performance upscaling saw a 53% advantage going the way of the GeForce GPU. And finally, at 1440p with balanced upscaling enabled, CPU limits start to come into play, trimming the 5090's lead down to 41% with 244 FPS versus 173 FPS.
Okay, so across the 11 games tested, focusing on the higher quality settings, the RTX 5090's dominance is undeniable. At 4K, the GeForce GPU was a staggering 105% faster on average when compared to the 9070 XT, delivering 78 FPS compared to just 38 FPS, alongside a massive 114% advantage for the 1% lows. Dropping the resolution to native 1440p saw the 9070 XT average 72 FPS, but the RTX 5090 still maintained a commanding 61% performance advantage. And when we look at upscaling, the RTX 5090 remained 68% faster at 4K using performance upscaling. And then at 1440p with balanced upscaling, the margin did compress slightly to 40%. So greatly reduced when compared to what we saw at 4K, but of course this is due to things such as CPU limits becoming more of a factor. So what I'm saying is these results are very much to be expected.
Now, lowering the visual quality across our 11 game suite naturally shifts things towards again more of a CPU bottleneck, but the RTX 5090 still remained comfortably ahead at native 4K where it was 64% faster than the 9070 XT, averaging 144 FPS compared to 88 FPS. It also maintained a strong 63% lead for the 1% lows, pushing 114 FPS versus 70 FPS. However, once we move to lower resolutions, system and CPU limits become much more obvious. At native 1440p, the GeForce GPU's lead drops to 36%, averaging 204 FPS versus 150 FPS for the Radeon GPU. Then, once we enable upscaling, the margins compress even further. At 4K with performance upscaling, the RTX 5090 was 36% faster, rendering 193 FPS versus 142 FPS. Then at 1440p balanced upscaling, we really do hit a wall. The RTX 5090 managed 224 FPS while the 9070 XT pushed 189 FPS and that shrank the average performance delta to just 19%.
So, is the 9070 XT as fast as the RTX 5090? Obviously, no, not even close. When testing just the GPU without any limitations, the 5090 was an average 105% faster. Now, technically speaking, there are examples where the 9070 XT can end up faster than the 5090, even with a flagship CPU, such as the 9800X3D. And this is due to Nvidia's overhead issue. There are a few caveats, though. Using the highest or lowest quality settings in any of the games that I tested, and I did test quite a few, more than the 11 that were shown in this video, I didn't find a single example where the 9070 XT was close to matching the 5090 at native 4K, let alone beating it. I also didn't find a single example where the 9070 XT matched or beat the 5090 at 1440p using higher ultra settings. And this was also true at 4K using the performance upscaling mode. So, if you're fortunate enough to be playing games using an RTX 5090 at 1440p or 4K using high quality settings, there's simply no chance the 9070 XT is coming close to delivering the same or better levels of performance.
Where the 9070 XT can get very close to match or perhaps even beat the RTX 5090 using a flagship CPU is at 1440p using balanced upscaling or while running low to medium quality type settings. But even then, this really will only apply to a select few games. For example, the 5090 was still comfortably ahead of the 9070 XT at 1440p using balanced upscaling and the higher visual quality settings in games such as Assassin's Creed Black Flag Resynced, Battlefield 6, Cyberpunk 2077, Pragmata, Marvel Rivals, Doom: The Dark Ages, and Spider-Man 2. So, seven of the 11 games tested. Thing is though, in a lot of the games, the RTX 5090 wasn't much faster when using balanced upscaling at 1440p when compared to native 1440p performance. So, if anything, you would just enable DLAA to boost the image quality rather than using upscaling to try and increase frame rates. Also, something worth considering when comparing the results with DLSS upscaling enabled is that the DLSS algorithm has its own overhead. And this is the time it takes for the AI algorithm that reconstructs a lower resolution image into a higher resolution image. In short, although DLSS is designed to boost your frame rate, running the upscaling process isn't entirely free. And I should note that FSR upscaling has the same kind of performance overhead.
Now, I suspect some people will point to the fact that across the 11 games tested, the RTX 5090 was 105% faster at 4K, but then just 40% faster at 1440p using balanced upscaling. And even though that is a jump from 104 FPS to 146 FPS, I do suspect some kind of mental gymnastics being performed here where this is used to claim that the RTX 5090 isn't as powerful as some people think it is when tested under real world conditions or something along those lines. It's kind of hard to predict how these things go, but I think something like that might materialize here. But that kind of argument is, in my opinion, obviously nonsense. And as I've already said, none of this should be news to anyone. In our original RTX 5090 review, it was 63% faster than the 7900 XTX at 4K, but then just 38% faster at 1440p, and that was across 17 games. So, the fact that the GPU becomes more limited by the CPU at lower resolutions shouldn't be news to anyone. And this also happens when comparing the 7900 XTX to the 7700 XT, for example. The XTX is much faster at 4K and the margin again compresses once you lower the resolution as the data does tend to become more CPU limited. And I feel these CPU limited results are being used to conflate the Nvidia overhead issue, making it appear more serious than it actually is.
And yes, Nvidia does have an overhead issue. I've dedicated a series of videos to proving this, but it's also not really something RTX 5090 owners realistically need to worry about or anyone running a high-end CPU needs to worry about. I also revealed years ago that this does really only become a problem for those of you running older CPUs, hoping to get a bit more out of them as you upgrade to a newer, faster GPU. In those examples, you will find more instances where Radeon GPUs can afford you more performance headroom. I've also seen claims recently that the Nvidia overhead issue is a problem for high-end gaming. So, those of you with powerful CPUs but are trying to target high frame rates. This claim also isn't true, at least not entirely true, and again really conflates a few issues. The truth is for those of you playing competitive shooters using competitive quality settings, so typically low quality settings, the GPU doesn't really matter. Using the performance mode, you can push crazy high frame rates in Fortnite with a 9060 XT or 5060 Ti. And therefore, you absolutely do not need an RTX 5090 or even a 9070 XT for that matter. Rather, what you need is a screaming fast CPU.
So, the fact the 5090 isn't any faster than a 9070 XT at 1440p in a game like Fortnite using the performance settings, well, that's just to be expected. In our example, we ended up being CPU limited at around 700 FPS. And again, yes, the 9070 XT was technically faster, but the margin was so small it's not worth talking about. And again, we really were primarily testing the CPU in that configuration. Moreover, you're not, for example, going to find a game where the 9070 XT can spit out, let's say, 500 FPS, while the 5090 is limited to less than 400 FPS, at least with a 9800X3D installed or any other high-end CPU. You could possibly achieve that kind of result with something like a Ryzen 5 5500, which I suppose is an interesting for science type test, but it's not exactly useful for anyone who owns a high-end GPU or is considering buying one. It's very much useful for those who have lower slower CPUs, which is what we looked at in our original series 5 years ago now. And I've done some GPU CPU scaling content since then, and it's really only revealed problems for this Nvidia overhead with those lower, more affordable CPUs.
Later this week, I will take a closer look at the Nvidia overhead issue with the 9070 XT and 5070 Ti, testing a wider range of games, settings, and CPUs. I'm certainly not trying to sweep this issue under the rug, but I also don't want the problem to be misunderstood. It's not something that is going to be an issue for those of you running high-end PCs. It's just that simple. And that is going to do it for this video. If you liked it, you know what to do. Subscribe for more content. As I said, more of this sort of content is coming up, and I'm sure there'll be a lot of questions around this one, so I'll do my best to answer them. So yeah, I'll be reading all of your comments in the comments section below. Yeah, like, subscribe, do that stuff. There's also the join button or Patreon if you want to become a Hardware Unboxed member. Get access to some pretty cool content there. We got Q&A stuff, behind the scenes content. I say Q&As maybe. There's stuff there. Check it out if you're interested. If not, that's perfectly fine. And I would like to thank you for watching this video. I'm your host, Steve. See you again next time.