Seven months after Nvidia’s new RTX 40-series graphics card debuted with blazing speed, but it costs $1,600 GeForce RTX 4090, the company’s new Ada Lovelace architecture is finally making its way to the masses.Today Nvidia announced not one, not two, but three A member of the GeForce RTX 4060 series, the first model — the $399 GeForce RTX 4060 Ti — will be available next week, May 24.
Nvidia also revealed that a $499 version of the RTX 4060 Ti with 16GB of memory (and no other tweaks) will launch in July alongside the non-Ti GeForce RTX 4060 for $299.
With the launch of the RTX 4060 Ti, Nvidia appears to be responding to the latest pair of graphics card controversies.First, the RTX 4060 Ti stuck to the same price as its predecessor, halting the downward price trend painful high RTX 40 Series Pricing; Second, the introduction of the upgraded 16GB model addresses concerns about 8GB of VRAM is simply not enough for PC gaming No longer, despite the high price, probably not worth it considering some of the technical details of this GPU.
Let’s dig deeper.
Further reading: The Best Graphics Cards for PC Gaming
Meet Nvidia’s GeForce RTX 4060 Ti
Nvidia
Both versions of the GeForce RTX 4060 Ti are identical aside from raw memory capacity, a welcome change after Nvidia’s attempt atand abort) RTX 4080 12GB launch, which originally meant that there would be two completely different GPUs with the same RTX 4080 name. (The card was eventually relabeled as RTX 4070 Ti.)
Nvidia says the RTX 4060 Ti is tuned for great 1080p gaming performance, as surveyed Steam hardware is dominated by xx60-class graphics cards, and most of those people use 1080p monitors. Based on the company’s new Ada Lovelace GPU architecture, the RTX 4060 Ti at last Bringing Ada’s Incredibility Performance Supercharged DLSS 3 TechnologyNVENC and Cutting edge AV1 encoding, and superior power efficiency at a more mainstream price point. The early performance results Nvidia shared with the press centered around games supporting ray tracing and DLSS, fueling GeForce’s recent focus on AI. That’s nice and pretty, but unfortunately muddles the potential boost in raw GPU horsepower over its predecessors.
Nvidia
The company also shared some specs for the RTX 4060 Ti, further emphasizing improvements in ray tracing, DLSS performance, and power efficiency, but the chart only lists raw teraflop performance for various GPU hardware cores, not spelling out explicit shaders, RT and Tensor core count. early leak RTX 4060 Ti is recommended to have less CUDA core ratio RTX 3060 Ti (4,352 vs. 4,864), which, if true, may be why Nvidia took this approach. (The Ada Lovelace cores are more powerful than the RTX 30-series Ampere cores, though.)
One thing is for sure: the memory bus width is shrinking, from 256 bits in the RTX 3060 Ti to 128 bits in the 4060 Ti. Think of it like a traffic jam; if the memory requests are cars, the bus width is the size of the road. A hundred cars can go much faster on a four-lane highway than a single-lane dirt road at a rate similar to the card’s overall memory bandwidth. On paper, the GeForce RTX 4060 Ti offers 288GB/s of total memory bandwidth, which is a huge gap compared to the 3060 Ti’s 448GB/s.
But it’s not that simple.
An illustrative example of how traditional GPU memory works, without the larger on-chip L2 cache.
Nvidia
Comparison image showing how the larger L2 cache in Nvidia’s RTX 40-series (“Ada Lovelace”) GPUs can help reduce calls to traditional memory.
Nvidia
Nvidia from AMD’s Killer Infinity Cache Technology For the RTX 40-series products, the large on-chip caches are placed on the GPU itself, and the RTX 4060-series is no exception. The physical proximity not only means that this cache is faster, but deploying it also reduces the need to send many requests to the memory chips that line the GPU, as shown in the slide above. While AMD’s Infinity Cache uses L3 cache, Nvidia has opted for a 32MB L2 cache (compared to just 2MB for the 3060 Ti). The company says the larger L2 cache is especially helpful for ray tracing and DLSS performance, and leads to the “effective” memory bandwidth shown in the tech specs slide.
The thing is this: as we Previous generation Radeon RX 6600 XT, which has 32MB of infinite cache and a 128-bit bus, pairing a large on-chip cache with a neutral bus width helps the GPU excel in terms of performance and power at its target display resolution, but typically results in higher Lower resolution results than expected. Higher resolutions have higher memory requirements, which results in a lower successful on-chip cache hit rate. When this happens, memory requests go to traditional memory, and the RTX 4060 Ti’s 128-bit bus is much smaller, which can be problematic when this happens.
We’ll have to wait for reviews to see if this applies to the RTX 4060 Ti as well (it does to the RTX 4070 series, which use similar technology tuned for 1440p gaming). That would be miserable.this RTX 3060 Ti One of the best GeForce cards of the last generation, precisely because it delivers solid 1440p performance and great 1080p gaming. (For what it’s worth, GeForce product manager Justin Walker says the RTX 4060 Ti can play games at 1440p, though you may need to tweak some settings in some titles, and games that support DLSS will also see a performance boost.)
Performance results for 16GB RTX 4060 Ti with DLSS; note that Resident Evil Remake and A Plague Tale: Requiem are set to Ultra settings on the 16GB chart, but reduced to High on the 8GB RTX 4060 Ti performance chart above.
Nvidia
If the RTX 4060 Ti does end up being the sweet spot for uncompromising 1080p gaming, its different memory options might become less important. While some cutting-edge games can exceed 8GB at 1080p and cause stuttering if you don’t tweak the graphics settings, the vast majority of games won’t. We’ll have to see what testing reveals when we get the cards on May 24th (8GB) and sometime in July (16GB).
Both cards appear to be very power efficient, with the 8GB model rated at 160W and the 16GB model rated at 165W — a difference of just 5W between the two.
Meet the GeForce RTX 4060
But that’s not all! Nvidia also unveiled the $299 GeForce RTX 4060, revealing technical specifications and early performance results.
Nvidia
Everything we said about the 4060 Ti is tuned for 1080p gaming really Also works with vanilla 4060. The card actually uses a different core GPU than the 4060 Ti (codenamed AD107 and AD106, respectively), and based on the specs provided by Nvidia, it should be significantly weaker than the Ti model. It still offers DLSS 3, AV1 encoding, extreme power efficiency and all the other Ada Lovelace benefits. Nvidia’s early performance trailers show it beating its predecessor in games that support DLSS.
Nvidia
There are also some notes about the memory subsystem here. First, the on-chip L2 cache is reduced to 24MB compared to the Ti’s 32MB. Second, the RTX 4060 also comes with 8GB of RAM, compared to 12GB for its RTX 3060 predecessor. Embarrassing – although to be fair, the RTX 3060 cost $329 at launch, while the RTX 4060 cost $299 at more advanced manufacturing nodes.
Nvidia
Nvidia has also taken the step of revealing “1% lower” performance results for the RTX 4060 compared to its predecessor in an attempt to avoid memory issues. When a game runs out of GPU VRAM, it bumps everything to your much slower main system memory, which can cause severe stuttering. If this is the case in Nvidia’s selected games, it will be shown in the table above. That said, activating ray tracing and DLSS changes the technical requirements, and all games in this chart have DLSS and ray tracing enabled where possible. These differences may be more or less noticeable in-game without these features enabled.
Again, we’ll have to see what independent testing reveals when the RTX 4060 launches in July.
Budget GPU renaissance?
After nearly five years of pandemic and cryptocurrency-induced neglect, suddenly there’s a budget war against PC gamers.Intel’s big improvement $250 Arc A750 and $350 Arc A770 Compete in the same field as the $299 4060 (A770 offers 16GB for that price) Discounted Last-Gen AMD Optionswhile leaks suggest that AMD is also planning to release the new affordable Radeon RX 7600 soon.
The final benchmarks will tell the whole story, but friends, affordable PC gaming seems to be suddenly back on the menu. It’s damn time.