NVIDIA's GeForce RTX 4060 Family was announced last month. It consists of the RTX 4060 Ti 8 GB ($400, our review of the Founders Edition), RTX 4060 Ti 16 GB ($500, launches next month) and the $300 RTX 4060 non-Ti 8 GB, which we're reviewing today. This launch follows the same embargo mechanics as other recent releases from NVIDIA. On Day 1 (today), only reviews of cards priced at $300 MSRP are allowed, and tomorrow, reviews for cards priced above that can be published. We have a total of 10 reviews for RTX 4060, six today: ASUS Dual OC, Galax EX, Gigabyte WindForce OC, MSI Ventus 2X, Palit Dual and Zotac Spider-Man.

With the GeForce RTX 4060, NVIDIA is going after gamers using monitors with the most popular 1080p Full HD resolution. While the RTX 4060 Ti was targeted at the same crowd, it offers higher performance at a higher price. The RTX 4060 in today's reviews is crafted to reach the psychologically important $299 price point. To achieve that, NVIDIA is using their AD107 graphics processor, the smallest Ada GPU available. RTX 4060 Ti uses the bigger AD106, both are fabricated at TSMC Taiwan using a 5 nanometer process. You get 3072 GPU cores, which is the maximum available on AD107, paired with 48 ROPs, 96 TMUs and 24 RT cores. Memory capacity is set to 8 GB, with a 128-bit memory bus, there will not be a 16 GB version of RTX 4060 non-Ti.

The Galax RTX 4060 EX is a factory overclocked custom-design that still sells for NVIDIA's MSRP of $300. With +105 MHz, this is one of the largest factory overclocks tested today. The 4.2% OC turns into a 3% performance improvement over the baseline RTX 4060—the best result today. Still, it's not something you'll be able to notice subjectively in your games, but that's just how factory overclocks work these days.

Averaged over the 25 games in our freshly updated H2 2023 test suite, at 1080p, we find the RTX 4060 just 6% ahead of the recently-launched AMD Radeon RX 7600. This means that RTX 4060 isn't able to beat last generation's RTX 3060 Ti, which remains 8% faster. The gen-over-gen performance gain is only 20%, at least more than what we saw with RTX 4060 Ti vs RTX 3060 Ti. Compared to the RTX 4060 Ti, the performance difference is 20%. AMD's Radeon RX 6700 XT 12 GB is 13% faster than the 4060, the Radeon RX 6600 XT is 10% slower. NVIDIA's aging RTX 2080 offers roughly the same performance as the RTX 4060, and the gap to the four year old Radeon RX 5700 XT is 20%. Intel's Arc A770 is within 5% of the RTX 4060, the A750 is 12% behind—not much. If Intel can bring their pricing down, they could steal some sales from NVIDIA and AMD in this segment. With these performance levels, RTX 4060 is a solid choice for gaming at Full HD—you'll be getting 60+ FPS in nearly all titles at maximum settings. Gaming at 1440p is in reach at decent FPS rates, too, but you'll have to reduce settings in some games, or enable upscaling with DLSS/FSR.

While I think that ray tracing isn't the most important technology to have in this segment, it's still some extra eye-candy that a lot of games come with these days. However, enabling ray tracing significantly impacts performance, which can be troublesome if you're struggling to maintain a frame rate above 60 FPS, even with RT off. On the other hand, in games where you have extra FPS to spare, activating ray tracing can further enhance the visual experience, beyond classic "ultra" settings. NVIDIA has been the leader in ray tracing for years and RTX 4060 isn't any different. While AMD has to execute ray tracing in their shader cores, NVIDIA has dedicated hardware units, which can take over that task. Compared to AMD's Radeon RX 7600, the RTX 4060 offers 22% better RT performance and is even able to beat the Radeon RX 6700 XT with its 12 GB framebuffer.

Probably the most important selling point for the GeForce RTX 4060 is support for DLSS 3 Frame Generation. The algorithm takes two frames, measures how things have moved in those two frames and calculates an intermediate frame in which these things moved only half the distance. While this approach is definitely not problem-free, especially when pixel-peeping at stills or slowed down video, in real-time it's nearly impossible to notice any difference. As you run at higher FPS and resolution it becomes even more difficult, because the deltas between each frame are getting smaller and smaller. Being able to double your FPS is a huge capability, because it means you can enable ray tracing for free, or game at higher resolutions. Of course you are limited to games with DLSS 3 support, of which there are currently around 40, mostly AAA titles, but not every title will support it. AMD doesn't have anything similar, they announced that FSR 3 exists last year and since then we haven't seen a single demo, with no updates at Computex either.

GeForce RTX 4060 comes with a 8 GB VRAM buffer—4 GB less than last generation's RTX 3060. This is a total non-issue though. Even in the worst-case (The Last of Us, a known memory hog), at 1080p, the RTX 4060 is still 8% faster than the RTX 3060. While it would be nice of course to have more VRAM on the RTX 4060, the 128-bit bus design limits the memory choices to 8 GB or 16 GB. With this test suite we do have all the new games and I find it very hard to spot significant FPS issues with the RTX 4060. No doubt, you can find edge cases where 8 GB will not be enough, but for thousands of games it will be a complete non-issue, and I think it's not unreasonable for buyers in this price-sensitive segment to set textures to High instead of Ultra, for two or three titles. If you still want more memory, then NVIDIA has you covered. The RTX 4060 Ti 16 GB launches in July, for $500, and gives people a chance to put their money where their mouth is. I'm definitely looking forward to testing the 16 GB version, but I doubt there will be enough of a difference to justify the cost.

Galax's thermal solution is compact enough that it will fit into all cases—an important capability for upgraders. Temperatures are excellent, reaching only 64°C, and noise levels are "OK" with 34.8 dBA. When we first published our review, we measured 41 dBA. After discussing with Galax, they released an updated VBIOS with improved fan settings. This VBIOS will be used on all production cards that make it into sales, we've uploaded the new BIOS here. Our apples-to-apples cooler comparison test at normalized noise and heat output confirms that the underlying cooler is excellent. It's the second best MSRP design, right behind the ASUS Dual, and is able to beat all other coolers tested today by quite some margin. Given these results, I feel like Galax could have allowed another 1°C or 2°C more for even less noise. Just like all other recent graphics card releases, the RTX Galax 4060 EX will stop its fans in idle, desktop work, internet browsing and light gaming.

Improving energy efficiency has been a major focus for all GPU makers for years. RTX 4060 is a highly energy-efficient design that consumes around 130 W during gaming—very impressive. This result is a little bit better than the ~150 W on Radeon RX 7600, and considerably lower than the 180 W we've measured on RTX 3060. Still, it's only marginally better than the 150 W result of the RTX 4060 Ti. Especially when FPS is taken into account, it seems that AD107 is a less energy-efficient design than AD106, but the differences aren't that big. Energy efficiency is the lowest of all Ada cards, but still better than all older cards ever released. AMD's Radeon RX 7600 achieves amazing idle power of just 2 W, the 10+ W result on RTX 4060 is five times higher. Multi-monitor desktop has always been problematic for AMD, but even here both competing cards are similar in their power draw.

While their power draw is pretty high, with this rebench, Intel Arc becomes a first-class citizen in our test suite, and I have to say I'm very pleased with the experience, no serious issues to report and their driver team is doing a good job improving the experience. For example, the Arc A770 is roughly comparable to RTX 4060 performance, but consumes almost twice the power, with 235 W. Assuming 15 hours gaming per week at a kWh rate of EUR 0.4 or USD 0.2 this results in a €62 / $31 savings in two years going with the NVIDIA option. Not a ton, but still something to consider. PSU requirements of the RTX 4060 are minimal, any PSU that has an 8-pin power cable will be able to run the RTX 4060 just fine. For cost-savings, the RTX 4060 doesn't have dedicated board power monitoring hardware on the PCB, so the power draw is estimated, which results in unexpected readings in GPU-Z, and the power throttling mechanism is very aggressive in Furmark, which runs at only 110 W, while actual games reach up to around 130 W.

Galax is able to match the $300 MSRP set by NVIDIA, and even gives you a factory overclock at no price increase, I like. Still, the RTX 4060's price point seems rather high, considering that performance is similar to the RX 7600, which launched at $270 and is currently sold for $250. No doubt, the RTX 4060 is a decent card, but when looking at alternatives, combined with its price point, it's very "meh." While AMD lacked a clear unique selling point with the RX 7600, NVIDIA has DLSS 3. As awesome as it is, I feel that paying $50, or +20%. for DLSS 3 a bit much, especially considering the relatively short game support list. Where it works, it is a game-changer, especially in single-player. While this pricing strategy against RX 7600 is nothing new, it's actually textbook NVIDIA, there's just so many good alternatives available on the market at this time. For example, Radeon RX 6700 XT costs $310, yet offers with a more future-proof 12 GB framebuffer, with slightly better performance in rasterization, but worse RT capability. The Radeon RX 6750 XT 12 GB for $360 offers even better performance, but is probably out of reach for the $299 crowd. GeForce RTX 3070 can be found for $320 these days, with considerably higher performance in both rasterization and ray tracing. If you prefer GeForce over Radeon, a used RTX 2080 costs $240 at the moment, and offers virtually the same FPS as the RTX 4060—it just lacks DLSS 3—not an unreasonable tradeoff for $60 saved. On the other hand, $60 is the price of a single AAA game, a cost that you should be able to absorb as PC gamer. Still, at $300 the RTX 4060 will be a tough sell, at $270 and below things will be better and at sub-$250 it's probably the best option available. Back in the day GTX 1060 was a great and affordable choice, which really had an impact and got people into mainstream PC gaming. Ideally the market should return to that premise, but this seems unlikely with NVIDIA's monopoly and AMD's willingness to sell for NVIDIA's price minus 10%.

A rather curious, but noteworthy alternative is AMD's Radeon RX 5700 XT. This four-year old card still has enough horse power for a decent 1080p gaming experience (but without RT), and is available used for well under $200. While it comes with more noise and higher power draw, it could be a solid short-term choice, especially if you don't game a lot, which means power cost doesn't matter. I've mentioned Intel's Arc graphics cards before, they have matured a lot in the last year and are an interesting alternative for tech-savvy people. Intel's value proposition is not unsimilar to what RTX 4060 and RX 7600 offer, which means Intel needs to bring their price down to make a splash—they'll get there, I'm sure.

Given these high hardware prices I wouldn't be surprised if a lot of PC gamers start considering the various game consoles. Microsoft's Xbox Series S sells for $300, the Sony PS5 Digital for $400 - the price of just the graphics card. Both will give you a first-class gaming experience without shader compilation stutters and other PC port troubles. Gaming at 4K works well, and ray tracing is included in most titles. No doubt, the visual quality will be lower than on a high-end gaming PC, but it'll be good enough for a lot of people who just want to enjoy their games and feel like they are getting priced out of the PC gaming market.