We discussed the changes brought by the Ada Lovelace architecture in our GeForce RTX 4090 test and we refer you to this article to learn more about it. The GeForce RTX 4080 does not use the same graphics chip as the GeForce RTX 4090 (AD102), but a smaller version called the AD103. It has 45.9 billion transistors spread over an area of 378.6 mm² – compared to 76.3 billion transistors in 608.5 mm² of the top model.
Actually, this version has 7 GPC blocks on board. The classic rendering units (CUDA Cores) number 9728, which is halfway between the GeForce RTX 3080 (8704 CUDA) and the GeForce RTX 3080 Ti (10240 CUDA). But note that these units are not the same as those from the Ampère architecture. For others, there are 304 4th generation Tensor Cores to speed up processing dedicated to AI, 76 3rd generation RT Cores for processing related to raytracing, 304 units dedicated to texture and 112 rendering units. Also note the presence of 64 MB L2 cache, compared to 4 to 6 MB in the GeForce RTX 3000 series.
The number of calculation units in the end does not change much compared to the old generation of Nvidia graphics cards, but the switch to a fineness of engraving of 5 nm (named N4 by TSMC) allows the manufacturer to push the frequency cursor far enough. The GeForce RTX 4080 is thus provided for 2505 MHz in its Boost frequency, i.e. an increase of 795 MHz compared to the RTX 3080. A point where, together with the improvements in the architecture of the various processing units, it may be possible to achieve a significant. generation leap in performances.
Rounding out the GPU is 16GB of 2800MHz GDDR6X operating on a 256-bit bus. It achieves a bandwidth of 716.8 GB/s. The entire graphics card is listed for a maximum TGP of 320 watts, but Nvidia has changed its approach a bit. Thus, in GeForce RTX 2000 and 3000 series, TGP is always the power consumption found in most cases. In GeForce RTX 4000s, this is more than the power consumption limit. Nvidia, for example, shows that the average in-game consumption of the GeForce RTX 4080 is like 250 watts.
Cooling between 250 and 320 watts is usually not a big problem for graphics card manufacturers. Nvidia, however, saw things big for the cooling system, directly inherited from the GeForce RTX 4090. So we see a big card (30.5 cm long, 3 slots) and quite a bug -at (not less than 2, 2kg).
The cooling system is still based on a vapor chamber that is in contact with the GPU and the memory chips. Two 11.6 cm diameter fans circulate air on either side of the graphics card through aluminum fins. These fans have the good taste of using an oil bath rather than a ball bearing system. It offers better stability and more noise – we find this type of system with Noctua fans, for example.
If the graphics card operates on a PCIe 4 interface, it is powered via a PCIe 5 type 12HPWR cable. A new format that allows the power supply of the graphics card to be limited to one cable, but even then it seems that different cable manufacturers have not fully mastered it – some GeForce RTX 4090s that see their power socket melted. However, Nvidia provides a 12HPWR adapter to the 3 8-pin PCIe sockets so that the graphics card can be used with a power supply without a specific cable.
The reference card – called Founders Edition – also offers four video outputs: three DisplayPort 1.4a and one HDMI 2.1a. The latter is able to output an 8K HDR signal at 60Hz or 4K HDR at 120Hz and also supports variable refresh. Note that the video encoder and decoder currently handles H.264, HEVC, VC1, VP9 and AV1 formats.
As expected, the GPU operating frequency is very high. In all our tests, we measured a maximum of 2790 MHz and an average of 2758 MHz in 4K – always in combination with games with classic rendering, raytracing and raytracing with active DLSS. In WQHD and Full HD, the average is 2769 MHz.
This high frequency, combined with the evolutions brought about by the architecture of Ada Lovelace, makes it possible to observe a wonderful surge of performance. In games that run only in rasterization and in 4K definition, the GeForce RTX 4080 is 50% faster than the GeForce RTX 3080 10 GB. It leads the GeForce RTX 3090 in a very good way (a lead of 30% on average). The GeForce RTX 4090 is even faster with an average frame rate 26% higher. Obviously, the Radeon RX 6900 XT is well beaten (+ 45%), but that’s not the opposite of it; however, we will wait for the verdict once the Radeon RX 7900 XTX and RX 7900 XT have been tested.
In raytracing, the evolution is just as interesting, with an average of 67% higher than the RTX 3080, and 43% higher than the GeForce RTX 3090. The GeForce RTX 4090, on the other hand, is 41% faster than it exercise. . Two worlds! So the newcomer from Nvidia excels in the art of running raytracing games, and it’s more than that it helps DLSS to see the amount of the image increase; a point that is even more true with the arrival of more and more DLSS3 compatible games where the image rate is almost double compared to the execution of DLSS2. This is a big plus for Nvidia graphics cards.
In fact, the GeForce RTX 4080 is an accomplished graphics card in its design (exemplary finish, controlled heating and noise) and that excels in all other areas. Very powerful in games, it makes short work of the most technically demanding titles, and especially those that use and abuse raytracing. A good point that comes at the moment when, oh surprise, the games that use the increased number of rays are announced for 2023.
Nvidia also continues to design its graphics cards as a complete experience, adding hardware to software. This resulted in the arrival of DLSS3, which is now exclusive to the GeForce RTX 4000 series.upscaling technology that literally increases the frame rate tenfold in games, without a noticeable effect on the quality of rendering. Developers tend to be enthusiastically following Nvidia in integrating its technologies, there is no doubt that most games will use this solution in the coming months.
Going by the nomenclature of the map, we can estimate that the generation jump in games is very important. The GeForce RTX 4080 is on average 55% faster in frame rate than the GeForce RTX 3080; but at the cost of a sharp price increase. The newcomer is sold for €1,469 in its Founders Edition version, when the previous model was launched at €719 – then adjusted to €759. A price that is doubled for performances that are not – the player will never make a profit, which undoubtedly explains in part why DLSS3 is exclusive to the RTX 4000 – Nvidia arguing that it has a material limitation; a moot point.
However, we can also compare the performance / price ratio with the GeForce RTX 3090, which was launched at €1,549 in its time and is now adjusted to €1,199. There, the GeForce RTX 4080 is actually more interesting, with an average speed increased by 33% for a price increase of 22% over the current price.
In short, the positioning of the price can be seen in different ways, the fact remains that these graphics cards remain out of reach of many budgets, the majority of players spend a little more than € 700 to equip themselves; and the average basket is much lower. It will be very interesting to see the performance offered by Radeon RX 7900 XT and Radeon RX 7900 XTX, its two real competitors from AMD.
Efficient cooling system.
HDMI 2.1a video output.
Nvidia software ecosystem.
Shutdown ventilation if the card is idle.
Requires a well ventilated enclosure.
Cooling system that damages CPU air heatsinks.
Power connector in the middle of the board.