Shulou(Shulou.com)11/24 Report--

As a veteran AIC maker, Gengsheng also launched their high-end graphics card at the launch of RTX 4080: Gengsheng GeForce RTX 4080 16GB dazzling SOC. This RTX 4080 features a unique obsidian glow and heat dissipation design, as well as the ultra-high factory frequency of the SOC version.

CTOnews.com also received the GeForce RTX 4080 16GB glare SOC in advance. In this test, we still filled all the configurations as follows:

Appearance ID design ploughing GeForce RTX 4080 16GB dazzle is a standard size three-slot three-fan graphics card, the general ATX chassis can fit.

The heat dissipation module of the graphics card is composed of black and white elements, and the rugged texture on the surface is inspired by obsidian. If you look at it this way, the card seems so simple that it doesn't look glamorous at all.

But after lighting, all the lighting effects will break through the opaque obsidian surface, bringing a dazzling feast for players.

The positive light effect is even more eye-catching, like a tattoo full of power, but also like the circuit board of cyber technology.

The above light effects can also be regulated by software through the brand-new control center.

On the side of the graphics card body, it is 3*DP 1.4a + 1*HDMI 2.1and supports up to 8K output.

The power supply interface is consistent with the public version, using the latest 16Pin power supply interface, which is more than enough to bring 320W TGP.

The package also includes 3 8Pin to 1 16Pin conversion heads, an ARGB graphics card bracket, and the corresponding light synchronization line.

Simple technical analysis our hands of Tengsheng GeForce RTX 4080 16GB glare is based on the current positioning of the second highest AD103 core, there are seven groups of GPC, of which four are full of blood. There are 9728 CUDA cores, 304 Tensor cores and 112 ROP units inside, which is similar to that of the previous generation RTX 3090.

There are three versions of Gengsheng RTX 4080 glare, the default version, the OC version and the SOC version, with increasing frequency in turn. The default core frequency is 2505MHz, the OC version frequency is 2565MHz, the SOC version frequency is 2625MHz, what we have is the SOC version with the highest frequency, this improvement is not small. Video memory is achieved in one step to the 256-bit bit width of 16GB GDDR6X. The large-scale improvement of video memory makes 4K games no longer worry about exploding memory, while also taking into account the current popular AI productivity work.

Tengsheng GeForce RTX 4080 16GB dazzling architecture is also the latest NVIDIA Ada architecture, which is based on TSMC 4N NVIDIA custom process, thus achieving a leap in performance and power consumption of up to 2x. The throughput of streaming multiprocessors is twice as high as that of the previous generation. The effective ray tracing computing power of the third generation RT Cores is 2.8 times higher than that of the previous generation. The fourth generation of Tensor Cores adds a new FP8 engine with up to 1.32 petaflops of Tensor processing performance, which is five times higher than that of the previous generation. SER brings up to three times the performance improvement for RayTracing, and up to 25% improvement for overall game performance.

As for the concrete function, the new architecture brings a lot of new functions: due to the introduction of Ada optical flow accelerator, DLSS 3 can predict the motion changes between frames in the scene, realize AI frame insertion, and improve the frame rate while maintaining the image quality. At the same time, it supports the application of dual AV1 encoder, which can not only shorten the video export time, but also have better picture quality.

Theoretical performance test next we carry on the theoretical performance test. First of all, the single-baked doughnut, after 15 minutes, the core temperature is stable at 65.1 ℃ and the power consumption is about 320W, which accords with the calibrated power consumption of 320W. The most important thing is that the fan is far from fully loaded, only running more than 40%, so it is quiet to use.

Next, there is a 3DMark pressure test, which can detect whether the performance of the graphics card has declined under continuous running minutes. Generally, more than 97% can be regarded as a qualified graphics card. The measured dazzling score of RTX 4080 is 99.6%, and the performance release is extremely stable.

In the 3DMark TimeSpy DX12 test, Tengsheng GeForce RTX 4080 16GB dazzling card score reached 26631 points, compared with 26552 points for RTX 4080, 17306 points for GeForce RTX 3080 10G version, 18981 points for RTX 3090, and 21862 points for RTX 3090Ti. In other words, the GeForce RTX 4080 glare surpasses all the graphics cards of the previous generation RTX 30 series, and is about 50% higher than the previous generation of RTX 3080.

In the 3DMark Fire Strike DX11 test, Tengsheng GeForce RTX 4080 16GB dazzling card score reached 55079 points, compared with 53379 points for RTX 4080, 40932 points for GeForce RTX 3080 10G version, 46045 points for GeForce RTX 3090, and 52630 points for GeForce RTX 3090Ti. This score is more obvious than the improvement of the public edition.

In the 3DMark Portal Royal light pursuit test, the dazzling score of GeForce RTX 4080 16GB is 17322, which is greater than the performance of the grating. It is also about 50% higher than GeForce RTX 3080.

CTOnews.com has made a comparison table of GPU physical scores, and the GeForce RTX 4080 16GB glare has a certain improvement compared to the public card in all projects. Of course, the most important thing is that non-public graphics cards do not lock power consumption, in actual use, we can also carry out manual overclocking to further expand the advantages.

Although the theoretical performance of NVIDIA DLSS 3 parsing has surpassed that of all RTX 30 series graphics cards, the really big boost is DLSS 3 technology. DLSS 3 is a new optical flow accelerator driven by the new fourth generation Tensor Core and GeForce RTX 40 series GPU, performance multiplier technology based on AI technology, the latest version of NVIDIA deep learning supersampling technology, and groundbreaking optical frame generation technology, will usher in a new era of NVIDIA RTX neural network rendering games and applications.

The optical flow accelerator based on NVIDIA Ada Lovelace architecture can analyze two consecutive game images and calculate the motion vector data of objects and elements from frame to frame without the need for traditional game engine modeling. The frame generation model acquires both motion vector and optical flow data in the game engine, which greatly reduces the visual anomalies of AI when rendering elements such as particles, reflections, shadows and lights.

To put it simply, in the past, DLSS 2 improved the image quality and the number of frames by rendering a low-resolution image, and then AI zoomed in to high resolution. While the new DLSS 3 is compatible with DLSS 2, it also adds a frame generation function, which can insert a new frame calculated by AI between two real frames to double the number of frames. Combined with the super-resolution function of DLSS 2, AI can reconstruct up to 7/8 display pixels, and the game performance can be up to four times better than without DLSS!

So let's also take a look at the performance of the new card through actual measurement. The latest version of 3DMark has been adapted to DLSS 3's BenchMark, and we can directly see the difference by comparing the test results of GeForce RTX4080 16GB glare with DLSS 2's control variables.

According to the actual measurement of CTOnews.com, the average frame number of GeForce RTX4080 16GB glare in DLSS 2 mode is 98.65 frames, which is 2.5 times higher than that without DLSS 2.

With the latest DLSS 3, the average frame number of GeForce RTX4080 16GB glare has further increased to 126.92 frames, which is roughly 3.3 times higher than that without DLSS 2.

Of course, the biggest problem with the new technology is whether it can adapt to new games quickly. Obviously, NVIDIA has done a very good job in this respect. Not only is the first DLSS 3 game lineup very luxurious, but as long as it is adapted to DLSS 3, it will be downwards compatible with DLSS 2, which will greatly reduce the difficulty of adaptation for developers and maximize the benefits of players.

At present, DLSS 3 adapts to a number of new games. When GeForce RTX4080 releases collectively on November 15, there will be 10 DLSS 3 games escort. They are:

Legend of the Plague: the Requiem (APlagueTale:Requiem)

Memory of Light: infinity (BrightMemory:Infinite)

Destroy all Mankind 2: reprobed (DestroyAllHumansdestroy 2)

Shadow Torch City (F.I.S.T.:ForgedinShadowTorch)

"F1 ®22"

"against the current" (Justice)

The reincarnation of Life and death (Loopmancer)

Marvel Spider-Man: remaking (Marvel'sSpider-ManRemastered)

Microsoft simulated Flight (MicrosoftFlightSimulator)

"Super Human" (SUPERPEOPLE)

This time, we selected "Cyberpunk 2077", "Super Human", "memory of Light: infinite" and "reincarnation of Life and death" and other games that have been adapted to DLSS 3 to test the glare of Gengsheng GeForce RTX4080 16GB.

Let's first test "Cyberpunk: 2077", which is a great test of GPU performance, and carry out its own BenchMark test under the highest picture quality of 4K super light chase. From the results, it does put a lot of pressure on GPU performance. The number of frames when DLSS is fully off is only 75, while after opening DLSS 3 and selecting a performance file, the number of frames increases to 93 frames, an increase of 24%.

In a newly launched 3A sequel, the Legend of the Plague: Requiem, the effect of increasing the number of frames is even more obvious. By default, it still can not reach 60 frames, but with the blessing of DLSS technology, the number of frames can be nearly 3 times as many as when it is opened to the performance file. If DLSS 3 + super performance is turned on, the number of frames can even be increased by more than three times.

The DLSS 3 option is also available in Super Human, a new book by PUBG developer Blue Hole. Although there is still a slight problem with adaptation, the number of frames can be increased by 70% after opening the DLSS 3 performance file, and the number of super frames can be doubled in DLSS 3.

Next, let's test an online PC game "Magic Tower" that eats GPU performance very much. Once upon a time, "Magic Tower" was just a mobile game, but unexpectedly, the PC side actually pulled the picture quality to the maximum, and almost all the new features of the NVIDIA graphics card were used, which provided complete RTX+DLSS 2+DLSS 3 support. After opening RTX, the reflection of the ground is obviously more real, and the feeling of Cyber City is lifelike.

Of course, after actually turning on the light tracking effect, it really tests the performance of the graphics card. The highest picture quality of 4K is only more than 40 frames, so it is difficult to play smoothly, but when we open the performance file of DLSS 3, the number of frames is doubled directly, and when we come to 118frames, it suddenly becomes a competitive experience, and this is 4K light pursuit.

Game delay testing, of course, the essence of DLSS 3 technology is that AI generates new frames, so it is bound to bring the side effect of increased delay. For games that require high precision, ms-level delay differences will also affect the player's experience and feel. So after turning on the frame generation function of DLSS 3, NVIDIA Reflex will also be forced to open to minimize the rendering delay caused by the generated frame.

Let's take "Superhuman" as an example, the delay after turning on DLSS 3 is lower than opening nothing at all.

In Watchman: return, which is also the FPS competitive online game, the NVIDIA Reflex low latency option is also provided, which enables CPU to start submitting rendering work to GPU before the previous frame is completed, thus greatly reducing or even eliminating the rendering queue and faster response time. After testing, the delay is reduced by half when the enhanced Reflex option is turned on.

In addition to supporting DLSS 3's masterpiece, CTOnews.com has also tested several mainstream games for everyone. After all, at present, the number of "old" 3A games that support DLSS 2 is very large.

In Tomb Raider: shadow, it is measured that even if DLSS is not turned on at the highest resolution of 4K, it can run up to 140frames while running BenchMark, which is basically enough for 4K / 144Hz display.

The game also supports DLSS 2 technology, and after turning on DLSS 2, the number of frames can be increased by up to 80%. All the projects are also slightly higher than the public domain, and can already challenge the future 8K games.

Finally, we tested a light-chasing 3A masterpiece "Control: the Ultimate Edition". The 4K highest-quality and highest-light DLSS can hardly play smoothly, while Tengsheng GeForce RTX4080 16GB can almost double the number of frames after opening the initial DLSS, which is an excellent experience. After some testing, Gengsheng GeForce RTX4080 16GB dazzle to control 4K games without any problems, for gamers, it is definitely a handy game tool.

Creative production test GeForce RTX4080 16GB glare has improved greatly, as well as video memory. With these hardware and technical upgrades, RTX4080 graphics cards will certainly bring higher efficiency and greater potential in creative production and in-depth learning. By installing the NVIDIA Studio driver, you can support more than 75 creative software to call the AI computing power of the NVIDIA CUDA core.

In modeling and rendering software, GeForce RTX4080 can use CUDA to provide hardware acceleration for work. At present, almost all modeling software will optimize the GPU of NVIDIA, so it can achieve the coexistence of efficiency, stability and compatibility. For example, in the commonly used rendering tool V-Ray, you can use RTX accelerated ray tracing to achieve high-performance final frame rendering. With the noise reduction feature of AI, GPU can further speed up interactive rendering.

In V-Ray Benchmark, Tengsheng GeForce RTX4080 16GB glare scored 3094 points, which can handle most of the modeling rendering needs.

Thanks to ADA's new flagship architecture, the performance of light chase has been greatly improved, and the score of GeForce RTX4080 16GB glare in V-Ray GPU RTX has been increased to 4198 points, even the high rendering pressure of real-time light chase can be handled calmly.

In addition, dual NVIDIA encoders support AV1 coding. AV1 is the next generation video coding technology, which can provide faster video coding and higher quality streaming media transmission performance while occupying the same space. This encoding format is already very common in mainstream multimedia applications. For example, domestic editing software clipping professional version already supports AV1 video encoder acceleration, as well as OBS live streaming software, which also supports NVENC AV1 coding recording.

The trend of AV1 is not only the result of technological advances, but also the pain point of the industry's demand for productivity, which is to significantly compress the size of video files while ensuring that the image quality is almost unaffected.

We also carried out a simple test, through the editing professional version of the software to use H264 and AV1 these two different coding to export the same video material. In the end, the file size encoded by H264 is 159MB, while the file size encoded by AV1 is only 127MB, and the size has been reduced by 21%, but there is no difference in image quality.

Finally, there is the rendering test of PugetBench For Davinci. GPU acceleration effects based on NVIDIA CUDA technology can achieve faster real-time video editing and frame rendering, and the output speed using NVIDIA coding can be increased by up to 5 times. Gengsheng GeForce RTX4080 16GB scored 2547 on the test, and 8K video clips no longer have to be an agent.

To sum up, this year's RTX4080 graphics card is menacing. A few years ago, it was hard for many people to imagine what it would be like to play smoothly with a light chase on 4K resolution. Now 4K light chase games with the help of NVIDIA DLSS 3 technology can not only run up to 60 frames in full picture quality, many games can even reach the high frame number of electronic games, there is no doubt that players in this era are very happy.

When it comes to the dazzling GeForce RTX4080 16GB card, the dark style and the unique RGB rugged texture are a major feature of this card in ID design. If I use it myself, I will install it vertically, highlighting its RGB effect. For game enthusiasts, Gengsheng GeForce RTX4080 16GB glare can fully meet your game needs, and it is also a good choice for RTX40 graphics cards this year.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.