When looking to buy a mid-range graphics card for a gaming PC, you’re probably choosing between the GeForce RTX 5060 Ti and the Radeon RX 9060 XT, and per our recommendation, 16GB versions for both GPU models. In today’s review, we will be comparing these GPUs with multiple upscaling modes enabled to determine which model performs best in realistic gaming configurations. We know many of you will be using DLSS 4 on a GeForce GPU and FSR 4 on a Radeon GPU, and this review includes performance data for both.
This comparison allows us to explore a range of important insights. You’ll be able to see both the native and upscaled FPS performance of these cards across various games, giving you a clearer idea of the experience each one offers.
We will examine whether DLSS 4 or FSR 4 provides a greater performance uplift, and whether that shift affects the overall performance difference between the two GPUs.
You will also see the impact of using Balanced or Performance upscaling modes instead of Quality, which is increasingly relevant as modern upscaling technologies have become highly effective even at lower settings.
The goal of this benchmarking feature is to provide more data and a different perspective than our initial day-one review, which focused primarily on native performance at various resolutions.
Testing multiple upscaling modes on top of all the other evaluations is extremely time-consuming, which is why the original review focused on more standardized, apples-to-apples data. Now that these cards have been available for some time, it’s the perfect opportunity to dive deeper into additional upscaling analysis.
We also have a growing collection of upscaling-related content on TechSpot, which is important background material since DLSS 4 and FSR 4 do not produce identical visual outputs. As a result, performance testing between the two is not a direct comparison.
Our in-depth features explore how these upscalers work, the visual differences between DLSS 4 and FSR 4, and how the various upscaling settings compare. These are valuable companion resources that complement this performance-focused review.
Test Setup and System Specs
Today, we are testing 15 titles at 1440p that support both DLSS 4 and FSR 4. For each game, we are evaluating the three main upscaling modes: Quality, Balanced, and Performance. This, in addition to native TAA. The TAA output is visually identical on both cards, serving as a consistent reference point.
DLSS 4 is used on the RTX 5060 Ti, while FSR 4 is used on the RX 9060 XT. On the GeForce card, most titles have been upgraded to DLSS 4 using the driver override feature, which ensures the latest upscaling model is in use. A few titles have native DLSS 4 support; in those cases, the transformer model is selected in-game and the latest model is enabled in the driver.
On the Radeon card, all games are configured to use FSR 4 via the driver software toggle.
We are fortunate to have the Asus Prime models of both the 5060 Ti 16GB and the 9060 XT 16GB for this testing. This adds a valuable real-world element to the comparison, as these are intended to be equivalent models from the same brand.
The Prime is Asus’ mid-range offering – better than the Dual series, but below the TUF Gaming line. Both cards are factory overclocked to similar levels and feature similar coolers and designs. While not identical, they are comparable in quality and pricing.
The Prime 5060 Ti 16GB is currently priced at $530, a $100 premium over its $430 MSRP. The Prime 9060 XT 16GB is priced at $440, a $90 premium over the $350 MSRP. These are significant markups, but they reflect the current state of the GPU market, where neither model is typically available at its MSRP.
Our test system includes a Ryzen 7 9800X3D CPU, 32GB of DDR5-6000 CL30 memory, and the latest game updates, Windows patches, and drivers. We are using driver version 577.00 for Nvidia and 25.6.3 for AMD.
Let’s get into the results.
Benchmarks
Horizon Zero Dawn
Starting with Horizon Zero Dawn Remastered, there is very little difference between the two models at native 1440p. These margins remain relatively unchanged when upscaling is applied. The RTX 5060 Ti is 3% faster using native TAA, 1% faster with Quality upscaling, 2% faster with Balanced, and again 2% faster with Performance. All of these differences fall within what we consider a tie. On both GPUs, Quality upscaling provides a 19% performance uplift at 1440p using the highest in-game preset.
Ratchet & Clank: Rift Apart
Ratchet & Clank: Rift Apart is quite playable on these mid-range GPUs, even while using the highest preset with ray tracing set to High. In this configuration, the GeForce GPU is 14% faster across all upscaling settings. Despite the 5060 Ti’s advantage in ray tracing, the 9060 XT still delivers playable performance. With FSR 4 set to Quality, it achieves an average of 75 FPS, which is solid for this type of game.
The Last of Us Part I
In The Last of Us Part I, both graphics cards perform similarly using the Ultra preset. At native 1440p, the GeForce card leads by 3%, which increases slightly to 4% with Quality upscaling. Interestingly, the margin widens at lower upscaling modes, where DLSS 4 on the 5060 Ti is 6 to 7% faster. FSR 4 doesn’t scale as well in this title, although Balanced mode still offers a 37% uplift over native TAA, compared to 42% for DLSS 4.
Marvel’s Spider-Man 2
In Spider-Man 2, tested with the Very High preset and Very High ray tracing, the GeForce GPU starts with a modest 5% lead at native 1440p. This advantage increases with upscaling: DLSS 4 boosts performance by 49% on the 5060 Ti, while FSR 4 improves performance by only 33% on the 9060 XT. As a result, the 5060 Ti extends its lead to 16% with Quality upscaling and 20% with Balanced. Both cards reach a 60 FPS average with upscaling, though the GeForce card is clearly faster.
The Last of Us Part II
The Last of Us Part II shows similar average performance between the GPUs at native 1440p, although the 9060 XT provides higher 1% lows. When upscaling is enabled, the 5060 Ti scales slightly better, showing a 5 to 7% performance lead on average. However, in 1% lows, the two models are evenly matched in the tested area.
Ghost of Tsushima
Ghost of Tsushima is one of the more favorable titles for the Radeon GPU. The 9060 XT is a few percent faster in all tested configurations, although the differences are minor. Overall, both cards deliver smooth and highly playable performance with no unusual scaling behavior.
Stalker 2: Heart of Chornobyl
Stalker 2 is a straightforward case. The RTX 5060 Ti begins with a 7% performance lead, which stays consistent within a point or two when upscaling is applied. The Epic preset at 1440p can be demanding, and both cards struggle in the most intense scenes. The 5060 Ti hits a 60 FPS average with Quality upscaling, while the 9060 XT requires Balanced mode. Having 16GB of VRAM is critical to reach this performance level, as noted in our day-one review.
Marvel Rivals
Marvel Rivals performs better on the RTX 5060 Ti with the Ultra preset, offering 12% higher performance at native 1440p and 11% more with Quality upscaling. However, at lower upscaling modes, the Radeon card closes the gap. Balanced mode shows a 7% lead for the GeForce GPU, and just 6% with Performance. For actual gameplay, reducing the preset from Ultra is recommended, as neither card provides exceptional performance with upscaling enabled.
Kingdom Come: Deliverance II
Kingdom Come: Deliverance II behaves similarly to Spider-Man 2, despite using a different game engine. At native 1440p, the margin is only 3% in favor of the 5060 Ti. But with Quality upscaling, the GeForce card becomes 36% faster, while the Radeon only improves by 28%. This results in the 5060 Ti being over 10% faster with upscaling and able to surpass 100 FPS using Balanced mode.
Star Wars Outlaws
Star Wars Outlaws presents the opposite trend. At native 1440p, the 5060 Ti leads by 7%, but with Quality upscaling, the 9060 XT pulls ahead. The Radeon sees a 45% gain from FSR 4, compared to just 22% for the GeForce GPU. This swings the performance lead to 10% in favor of the 9060 XT. However, the advantage shrinks when using Balanced or Performance modes.
Hunt: Showdown
Hunt: Showdown performs poorly on Radeon GPUs, a trend we observed previously with the RX 9070 XT. At native 1440p, the 5060 Ti is 45% faster. Even with Quality upscaling, the gap remains at 29%. FSR 4 Performance only manages to match DLSS 4 Quality, and the 9060 XT struggles to exceed 60 FPS. AMD will need to provide further optimization for this title, despite existing FSR 4 support.
God of War: Ragnarok
God of War: Ragnarok runs well on both GPUs. Our test targets one of the most demanding sections of the game, which does reduce 1% lows. Average FPS is similar between the two cards across native and upscaled modes, but the 9060 XT delivers noticeably better 1% lows in this specific area.
Stellar Blade
Stellar Blade is another title where the 9060 XT underperforms. At native 1440p, the GeForce card is 29% faster, and this margin increases slightly with upscaling. In 1% lows, the differences are even more stark, especially in the intense beach scene filled with fire effects. With Quality upscaling, the 5060 Ti achieves 55% higher 1% lows. This is another game where AMD could benefit from additional tuning.
F1 25
The final title is F1 25. This is another case where the Radeon card catches up when upscaling is used. At native 1440p with the Ultra High preset (including ray tracing), the 5060 Ti is 9% faster. However, with Quality upscaling, both GPUs deliver the same average frame rate. Still, 1% lows remain stronger on the GeForce card. The 5060 Ti goes from being 28% faster in 1% lows natively to 14% faster with Quality upscaling.
Performance Summary
Here are the average results across the 15 tested games, calculated using a geometric mean. What stands out is that, although we observed a range of outcomes with upscaling – sometimes the 5060 Ti gained more ground, other times the 9060 XT did – on average, the performance margins between the two GPUs remain consistent.
In this sample of 15 games, we found that the 5060 Ti 16GB was 9% faster on average at native 1440p. With Quality upscaling, it was 10% faster, then again 10% faster using Balanced upscaling, and 9% faster using Performance upscaling.
This is good news for those who chose either card based on our day-one benchmarks, which focused on native rendering. In real-world scenarios where DLSS 4 or FSR 4 is used, the relative performance gap between the two models does not change meaningfully. Of course, one could argue that image quality varies between DLSS 4 and FSR 4, and in some titles, Quality vs Balanced might feel closer subjectively, but when comparing the same upscaling modes, the margins are consistent.
This also means the typical performance uplift from enabling upscaling is similar across both GPUs. On average, Quality mode provided a 30% increase over native rendering, Balanced delivered about 45%, and Performance added roughly 55%. These results come from games that run at approximately 60 FPS natively.
On average, Quality mode upscaling provided a 30% increase over native rendering, Balanced delivered about 45%, and Performance added roughly 55%.
You can expect a larger uplift if the baseline FPS is lower and a smaller one if the baseline is higher. Dropping from Quality to Balanced yields about a 10% gain, which we consider the minimum acceptable setting for gaming at 1440p using either DLSS 4 or FSR 4. Performance mode provides an additional 8 to 9% boost but comes with more noticeable visual compromises.
You may have noticed these results are more favorable to the RTX 5060 Ti compared to our day-one review of the 9060 XT, where the 5060 Ti was just 1% faster on average at 1440p. This is almost entirely due to the game selection in this review, which focuses solely on titles that support both DLSS 4 and FSR 4.
The current sample includes games that perform significantly better on GeForce GPUs, such as Assassin’s Creed Shadows and Stellar Blade, which were not part of our initial coverage. Unfortunately, some of the stronger titles for Radeon, like Call of Duty: Black Ops 6, could not be included as they do not yet support DLSS 4.
Takeaways and Buying Recommendations
Ultimately, these results do not substantially alter our overall assessment of the GeForce RTX 5060 Ti and Radeon RX 9060 XT 16GB. With upscaling enabled, the performance margins remain in line with what we observed in native rendering tests, so there is no reason to adjust our previous recommendations.
Another encouraging takeaway is that upscaling enables highly playable performance in most of the games we examined. In 12 out of the 15 titles tested at 1440p ultra settings, upscaling using acceptable modes like Quality or Balanced delivered at least 60 FPS in the benchmarked areas.
Many of the games were running at 80 to 90 FPS using Quality upscaling, which provides a great experience for a mid-range or mainstream graphics card. And this is without using frame generation, which many gamers do not like, just standard upscaling.
At the moment, the best-value option in this price range remains the Radeon RX 9060 XT 16GB, which can currently be found for as low as $360 to $370. The GeForce RTX 5060 Ti 16GB is occasionally available at $450 but is more often priced at $480 or higher.
That makes the GeForce card roughly 25 to 35 percent more expensive, for at best a 10 percent performance advantage, as shown today when comparing Quality upscaling at 1440p. Across a broader set of games, we would expect the performance margin to drop into the single digits.
The more meaningful difference between these two products lies not in raw performance, but in the overall quality of the upscaling technology. DLSS 4 generally produces better visual results than FSR 4 at resolutions like 1440p, which are typical for mainstream GPUs.
While FSR 4 is certainly usable and has its strengths, DLSS 4 often provides more consistent visual stability. You can learn more about this in our DLSS 4 vs FSR 4 feature articles.
On top of that, Nvidia continues to offer broader upscaling game support. AMD has done a decent job expanding FSR 4 support, particularly in titles released in 2025, but the total number of supported games remains limited. It also excludes many popular older titles that people still play today. When comparing DLSS 4 to older versions of FSR (or DLSS 3) the difference is significant. DLSS still leads by a wide margin. AMD has made a promising start with FSR 4, but there is clearly more work ahead.