RTX 4090 graphics card with NVIDIA Ada Lovelace architecture has begun to be sold in the market, and its unprecedentedly powerful performance has been unanimously sought after by enthusiast players. The RTX 4080, which was released at the same time as the RTX 4090 on September 20, was officially launched at the agreed time. As the current sub-flagship of the RTX 40 series, the RTX 4080 also has a series of new features and black technologies of the Ada Lovelace architecture, such as the third-generation RT Cores and the fourth-generation Tensor Cores. Of course, it also includes DLSS 3, which can greatly improve the visual effects of the game and the frame rate of , and the price is relatively dessert. Therefore, the RTX 4080 has also attracted widespread attention from gamers and high-performance users, and on November 15, the RTX 4080 also officially ushered in the performance lifting. In this test, we used the GeForce RTX 4080 Founders Edition (hereinafter referred to as RTX 4080 FE) from NVIDIA. I want to know how powerful this original faith-level artifact is? Then let’s have a look with us.
Ada Lovelace architecture has been fully upgraded. RTX 4080 can be regarded as a dessert in the flagship
Before we conducted a comprehensive analysis of the NVIDIA Ada Lovelace architecture. Here we will briefly summarize its technical highlights.
●The newly designed streaming multiprocessor in the NVIDIA Ada Lovelace architecture has up to 83 TFLOPS shading capability, with a throughput of 2 times higher than the previous generation.
●The third generation RT Cores has an effective ray tracing computing power of 191 TFLOPS, which is 2.8 times that of the previous generation of products.
●The fourth generation Tensor Cores has a new FP8 engine, with tensor processing performance up to 1.32 PFlops, 5 times higher than the previous generation.
●Supports shader execution reordering (SER), and improves execution efficiency by rescheduling the shader load instantly. SER can bring up to 3x performance improvements to ray tracing, and overall gaming performance can be up to 25%.
●Add Ada optical flow accelerator brings a 2x performance improvement, allowing DLSS 3 to predict motion changes between frames in a scene, allowing neural network to improve frame rate while maintaining image quality.
●Architecture has been greatly improved, and TSMC custom 4N process is adopted. This process is specially optimized for GPU, achieving up to 2 times the energy efficiency leap.
●12GB video memory and above models provide dual NVIDIA encoder (NVENC), which can shorten the video export time by up to half and provides hardware encoding support for AV1.
Shader Executing Reordering (SER) Shader The execution of reordering
GPU is the most efficient when completing a large number of parallel calculations. As the ray tracing effect becomes more and more complex in the game, the shader may face a large number of divergent and disordered inefficient work tasks, which affects the rendering efficiency. With the emergence of SER technology, these messy work tasks can be classified and dynamically reorganized into more efficient tasks, thereby improving the execution efficiency of shaders. According to NVIDIA's official data, SER technology will bring about 2 times the performance improvement of shader.
The third generation RT Core
From RTX 20 to RTX 30 to RTX 40, RT Core evolved to the third generation. Simply speaking, from the perspective of effective ray tracing computing power, the third-generation RT Core of the NVIDIA Ada Lovelace architecture GPU has reached 191 TFLOPS, which is 2.8 times that of the previous generation.
In addition, RT Core in NVIDIA Turing and Ampere architecture GPUs include acceleration units for accelerating the traversal of boundary volume hierarchy (BVH) data structures and performs intersecting test calculations of ray triangles and ray bounding boxes. In Ampere's RT Cores, BVH traversal is accelerated by Box Intersection Engine, and the ray triangle intersection test is accelerated by the triangle intersection engine, which has proven to be the most efficient engine to handle ray tracing workloads to date. In the third-generation RT Cores of Ada GPU, in addition to the above two functions, they also introduce "micromap transparency engine" (OMM) and "micromesh displacement engine" (DMM).
The function of the micromap transparency engine is to optimize ray tracing rendering and greatly reduce the workload of the shader, which is a great improvement in handling the ray tracing characteristics of complex objects. The specific principle is to bake the ray tracing characteristics into opaque masks, so that irregular shapes and translucent objects can be rendered faster and more accurately.
microgrid replacement engine can simplify complex graphics with a very large number of faces and create a simpler model, thereby achieving the purpose of rendering complex geometric figures with basic triangles, greatly reducing the amount of computing and the use of video memory. According to official data, the micro-grid displacement engine can increase the BVH speed of ray tracing by up to 15 times, and the video memory usage can be reduced to 1/20 of the original. In short, the more complex the model, the more outstanding the optimization effect of the micro-grid displacement engine.
Fourth-generation Tensor Core
On the Ada GPU, Tensor Cores has evolved to the fourth generation, and this time the Tensor Cores upgrade is particularly important. It is equipped with a brand new FP8 engine and tensor processing performance up to 1.32 PetaFlops (a five times more improved compared to the previous generation). The most revolutionary advancement of the fourth-generation Tensor Cores is of course the newly added Optical Flow Accelerator, which is the core of NVIDIA DLSS 3 black technology.
DLSS 3 technology
DLSS technology is a special black technology for improving picture quality and frame rate in RTX series graphics cards. By the RTX 40 generation, it evolved to DLSS 3. The span proves that its improvement is revolutionary. NVIDIA officially said that it represents the "new era of neural network rendering."
is different from the previous DLSS. DLSS 3 further introduces optical multi-frame generation technology based on the original super-resolution function. Simply put, it is possible to generate new frames through AI calculations, rather than just generating pixels like before. The work of this frame generation, and the work of this frame generation is completed by the new optical flow accelerator in the Ada architecture, of course, this means that the "optical frame generation" function in DLSS 3 is exclusive to RTX 40.
Optical flow accelerator can calculate the optical flow field between two frames based on DLSS 2, capture the direction and speed of changing from the first frame to the second frame in the game screen, including pixel information such as particles, light irradiation, reflection, etc., and accurately reconstruct the picture by calculating the motion vector and the optical flow field.
In general, DLSS 2 does the calculation of the remaining pixels on the screen through AI, and DLSS 3 can also calculate the next frame of the complete picture. In this way, DLSS 3 is equivalent to completely reconstructing the pixels of the actual display screen 7/8, and it can also be said to reconstruct the pixels equivalent to 7 times the actual rendered pixels. In addition, since the "frame generation" of DLSS 3 is post-processed in the GPU, even if there is a bottleneck in CPU performance, the game frame rate can be improved.
According to official data, DLSS 3 can bring up to 4 times the game performance improvement. DLSS 3 includes three parts: "optical multi-frame generation", "super resolution" and "NVIDIA Reflex". As long as developers integrate DLSS 3, they can support DLSS 2 by default. At present, DLSS 3 has received extensive support from game developers and game engines. In the detailed tests below, we will further introduce the situation of DLSS 3 games.
NVIDIA Studio Optimization
In terms of productivity, the CUDA ecosystem created by NVIDIA GPU is an unrivalled existence in the industry, and RTX graphics cards are also an efficient tool that designers prioritize. Therefore, the optimization of the RTX 40 series in NVIDIA Studio combined with the NVIDIA Studio driver will also provide content creators with better choices. Overall, RTX can currently accelerate more than 100 mainstream creative applications, and the SDK is provided to make these applications more efficient and provides exclusive features such as Optix, DLSS and Maxine. NVIDIA Studio also offers a full suite of creative applications including NVIDIA Omniverse, Broadcast, Canvas, and RTX Remix.Compared with AI live
AV1 and AI live
AV1 and other encoding methods can provide clearer video images at the same code rate, and AV1 also supports some more advanced encoding features, such as HDR and low latency. The RTX 40 encoder (the RTX 40 series with more than 12GB of video memory, including RTX 4080, all provide NVENC A V1 dual encoder) provides support for AV1 hardware encoding, while streaming software such as OBS and Discord will provide support for NVENC AV1 soon, allowing players to enjoy a more exquisite and smooth live video picture. In terms of decoding, NVENC of RTX 40 also inherits the fifth generation hardware decoder on RTX 30, supporting MPEG-2, VC-1, H.264 (AVCHD), H.265 (HEVC), VP8/9 and AV1's hardware decoding . In the AI live broadcast part, NVIDIA's updates in live broadcast application technology include: facial expression prediction, eye tracking (making you look like you are watching a camera), and quality improvements of virtual green screens, and developers can easily integrate these functions of the NVIDIA Broadcast SDK in their APP.
RTX 4080 specification parsing
Complete AD103 GPU core diagram
So what are the specifications of RTX 4080 as the sub-flagship of the RTX 40 series? The RTX 4080 uses AD103 chip. The AD103 has made some reductions compared to the AD102. The GPC has 7 groups. In addition, 1 GPC has also reduced 2 groups of TPCs and only 4 groups. On this basis, the RTX 4080 further reduced two sets of TPCs, with a total of 38 sets of TPCs.
RTX 4080 GPU core map
Therefore, the RTX 4080 GPU has 9728 CUDA cores, which can provide 49 Shader-TFLOPS traditional raster rendering capabilities; 304 fourth-generation Tensor Cores provide 780 Tensor-TFLOPS computing performance for AI processing and DLSS; 76 third-generation RT Cores have 113 RT-TFLOPS computing performance, which provides strong power for the next generation of ray tracing; and is equipped with 256 bit/16 GB Micron GDDR6X video memory. According to official data, the RTX 4080 can perform up to 2 times that of the RTX 3080 Ti, while the power consumption is reduced by nearly 10%. Of course, the RTX 4080 also has several technologies in the Ada Lovelace architecture, including shader execution reordering (SER), third-generation RT Cores, and DLSS 3.
In addition, careful players may notice that the number of RT Cores/Tensor Cores/TUs of RTX 4080 is even smaller than that of RTX 3080 Ti, and the memory bit width is lower. So how does it achieve higher performance? As mentioned earlier, the SM unit of the RTX 4080 has been further optimized and supports SER technology. At the same time, RT Cores and Tensor Cores are one generation higher, so the efficiency is naturally incomparable. In addition, the GPU's second-level cache is 10.7 times that of the RTX 3080 Ti, and the core frequency is more than 50%. Based on these factors, the final performance of RTX 4080 is actually far exceeding that of the RTX 3080 Ti. In terms of TGP, the RTX 4080 is set to 320 W, which is 30 W lower than the RTX 3080 Ti, which shows that its advanced process does bring extremely high energy efficiency ratio.
Next, let’s take a look at the real object of the faith artifact RTX 4080 FE.
"90 Flagship" same faith-class heat dissipation solution, RTX 4080 FE physical appreciation
NVIDIA's Founders Edition has always been a belief artifact in N Fan's mind. It can be said to be a solitary feature in industrial design. In addition to its strong performance, it also has extremely high collection value, and is also a reference benchmark for many AIC manufacturers in terms of electrical performance. The appearance of
RTX 4080 FE is exactly the same as that of RTX 4090 FE. The combination of silver and black complements, and the X-shaped shape in the center of the graphics card shows the power of technology and faith. There is an axial fan on the front of the radiator, and the other axial fan is located on the back of the graphics card to be responsible for the wind, thereby creating a more efficient cooling air duct in the chassis to improve the overall cooling effect.
RTX 4080 FE radiator is also equipped with the new heat sink fin and airflow design. Compared with the previous flagship FE radiator, the maximum airflow can be generated by the RTX 4080 FE by 20% higher, and at the same noise level, the airflow can be 15% higher, thus providing higher heat dissipation efficiency.
is the same as RTX 4090 FE. The thickness of the RTX 4080 FE has reached 3.5 slots, the weight is close to 2.2KG. The volume of the radiator is quite exaggerated, and the flagship is full of domineering sense. Of course, the auxiliary power interface is also 12Pin+4pin, and the packaging box also comes with a 12Pin+4Pin to 4×8Pin adapter cable. On the side of the RTX 4080 FE, it is equipped with a GEFEORCE RTX faith light, which will light up white light after power on, making the sense of faith full of vitality. A magnetic cover plate is designed at the rear of the graphics card. After opening it, the rack installation hole will be exposed, making it convenient to install in server cabinet .
interface part, RTX 4080 FE provides 3 DP 1.4 interfaces and 1 HDMI 2.1 interface, which can support 8K/60Hz HDR10 output.
RTX 4080 FE adopts a special-shaped PCB design, the overall layout is somewhat different from the RTX 4090 FE in details. The total power supply phase is 18. The powerful power supply design can provide a more stable high-load output curve, allowing the RTX 4080 GPU to stably perform all performance.
Next, the performance testing part that everyone is most concerned about. Let’s take a look at the true strength of RTX 4080 FE!
practical test: RTX 4080 leap upgrade, performance doubled
test platform
graphics card: NVIDIA GeForce RTX 4080 Founders Edition
processor: Core i9 13900K
memory: Kingston DDR5 6000 16GB×2
motherboard: Asus ROG MAXIMUS Z790 HERO
hard disk: WD_BLACK SN850 2TB
Power supply: Asus ROG Thor II 1200W
Operating system: Windows 11 Professional Edition 22H2
In the test, we unlocked the processor power consumption wall and turned on the Resize BAR option in the BIOS. Memory selection XMP DDR5 6000 mode, and the frequency division is set to automatic. In addition, the RTX 3080 Ti is also selected to compare it with the RTX 4090.
benchmark performance test
First of all, the
DLSS part, and in DLSS 2 mode, it is also leading the RTX 3080 Ti about 45%. In the DLSS 3 project, the RTX 3080 Ti does not support optical multi-frame generation, so it is only equivalent to running 4K tests of DLSS 2. The RTX 4080 with built-in optical flow accelerator and fully supports DLSS 3 is naturally more efficient, leading the RTX 3080 Ti about 128%! It's a leap-by-step upgrade. At the same time, we can also see that the benchmark performance of RTX 4080 is about 70% to 77% of that of RTX 4090, which is relatively consistent with its chip scale.
rasterized game performance test
rasterized game part, "Tomb Raider: Shadow" has 75 fps more than 3080Ti at 2K resolution, with an increase of 44%, and 37 fps more than 39% higher at 4K resolution. Compared to the RTX 4090, the performance under 4K is about 67% of it.
In "COD 19", the RTX 4080 has a frame rate increase of nearly 40% compared to the 3080Ti in 2K or 4K, and in 4K, it is about 73% of the RTX 4090, which is better than "Tomb Raider" Shadow.
"Red Dead Redemption 2" is a fair game. The RTX 4080 has improved by more than 30% in both 2K and 4K, and the improvement in 4K is slightly higher than 38%. Its frame rate at 4K is about 76% of the RTX 4090, and the gap is not too big.
Let’s look at the Ubisoft series games. First, it was "Assassin’s Creed: Valhalla". RTX 4080 also has an advantage of over 30% in 2K and 4K compared to the RTX 3080 Ti, and here it is only about 15% lower than the RTX 4090.
" Far Cry 6" is not very effective in optimizing flagship graphics cards. There is only a 5% gap between RTX 4080 and RTX 4090 under 2K, which may cause bottlenecks in the processor. And under 4K, the gap widened to 21%. The RTX 4080 also has a frame rate of about 30% higher than the RTX 3080 Ti at 4K.
RTX 3080 Ti's 4K "Watchdogs: Legion" can barely reach a little more than 60 fps, while RTX 4080 can still approach 90 fps, an increase of about 38%. The RTX 4080 has an RTX 4090 frame rate of about 75% at 4K. In the 3A masterpiece "Doom: Eternal" with excellent optimization, the RTX 4080 can exceed 240 fps at 4K. Although the frame rate of the RTX 3080 Ti is not low, the increase of the RTX 4080 has reached nearly 50%, and the increase is very considerable.
There is also the game where there is such a big performance gap in
. The frame rate advantage of RTX 4080 over the RTX 3080 Ti under 2K and 4K is more than 45%.
RTX 4080 performs relatively plainly in "Control", which has an advantage of about 28% frame rate in 4K compared to the RTX 3080 Ti and 29% behind the RTX 4090.
racing game "Forza Horizon 5", the RTX 4080 under 4K has 35% more frame rate than the RTX 3080 Ti and 18% lower than the RTX 4090. The frame rate at 2K is quite similar to that of the RTX 4090, which is obviously a bottleneck in the processor.
Let’s look at several e-sports online games. If you play " PlayerUnknown's Battlegrounds ", the RTX 4080 can also reach the highest image quality standard of 4K/144Hz. It is 33% higher than the RTX 3080 Ti at 4K and 31% lower than the RTX 4090, just in the middle.
Blizzard 's Overwatch 2 can also achieve the highest image quality of 4K/144Hz standard. Compared with the RTX 3080 Ti, the frame rate increase of 34% in 4K, and 29% lower than the RTX 4090.
"Final Fantasy 14"'s performance is in line with the previous gap. The RTX 4080 has increased by about 30% compared to the RTX 3080 Ti, and 26% behind the 4K compared to the RTX 4090.
Judging from the rasterization performance test results of these mainstream games, the increase of RTX 4080 compared with RTX 3080 Ti under 2K is between 24% and 50%, with an average of 36%. The increase in 4K is between 28% and 46%, and the average is 36%. Compared to the RTX 4090, the average gap at 2K is 15%, and about 25% at 4K.
ray tracing game performance test
Of course, ray tracing and DLSS are the real essence of RTX graphics cards. Of course, we must use ray tracing games under extreme image quality to test the powerful performance of RTX 4080.
Let’s take a look at “Tomb Raider: Shadow” first. Before, the RTX 4080 did not focus on ray chasing at 4K, and the increase of RTX 4080 was 39% higher than that of RTX 3080 Ti. After turning on ray chasing, the improvement was further increased to 43%. Although the core of ray chasing is not as much as that of RTX 3080 Ti, the ray chasing unit of RTX 4080 is one generation higher after all, and the frequency is much higher, so it is reasonable to significantly improve performance.
"Metro: Departure Enhanced Version" 4K, the RTX 3080 Ti does not have DLSS, and even if DLSS quality mode is turned on, it cannot reach 60 fps. The advantage of RTX 4080 in 4K reaches 49%, and you can run to 80 fps by turning on DLSS.
"Watchdogs: Legion" has relatively high hardware requirements, and the same is true. The RTX 3080 Ti can barely reach 4K/60 fps when DLSS is turned on. The RTX 4080 has a nearly 40% advantage in 4K, and it can run to 79 fps when DLSS is turned on. There is no surprise in
"Control", but here the advantage of RTX 4080 over the RTX 3080 Ti is slightly smaller, with an increase of 26% under 4K and an increase of 31% after turning on DLSS.
"Far Cry 6" once again demonstrates its insufficient optimization of flagship graphics cards. The frame rates of RTX 4080 and RTX 4090 at 2K resolution in the game are actually the same. However, the performance under 4K is relatively normal, with the RTX 4080 being 27% higher than the RTX 3080 Ti.In the RTX version of the hardware killer game "Minecraft" "Minecraft", DLSS is bound to ray tracing by default. DLSS is turned on by default. If it is turned off manually, the RTX 3080 Ti is only 36 fps, but the RTX 4080 can still approach 60 fps, an increase of 50%. If DLSS is turned on, the advantage can reach 55%. In the
racing game "F1 22", the improvement of RTX 4080 compared with the RTX 3080 Ti is also above 30%, while it is about 25% behind RTX 4090.
In addition, we also tested "Memory of Light: Infinity", "Doom: Eternity", " Resident Evil 8". Among these three games, the improvement of RTX 4080 in 4K compared to the RTX 3080 Ti after turning on ray tracing is also above 40%.
Overall, based on the performance of these games, after turning on ray tracing, the improvement of RTX 4080 compared with the RTX 3080 Ti is about 40%, reaching about 72% of the performance of RTX 4090. After turning on DLSS, the increase of RTX 4080 compared with the RTX 3080 Ti is 41%, and the gap between the RTX 4090 is about 26%.
DLSS 3 Test
We know that DLSS 3 is the highlight of the RTX 40 series, and it consists of three major parts: DLSS 2 super resolution + frame generation + REFLEX. From the release of RTX 4090 to the current DLSS 3 support, games are increasing almost every month, and this time we selected several of them for testing (using the NVIDIA FramView tool to monitor frame rate and PCL latency).
"Cyberpunk 2077" With the 4K super ray tracing preset, after RTX 4080 turns on the DLSS 3 performance mode, the frame rate surges from 30 fps to 114 fps. If compared with the RTX 3080 Ti turn on the DLSS 2 performance mode, it is nearly doubled.
In "Life and Death Reincarnation", RTX 4080 can increase by nearly 100 fps after turning on DLSS 3. The RTX 3080 Ti is only 92 fps if it turns on DLSS 2. The advantages of DLSS 3 of RTX 4080 are still relatively obvious.
"Plague Legend: Requiem" The highest image quality of native 4K can run to nearly 60 fps, and after turning on DLSS 3, it skyrockets to 160 fps, close to 2.8 times the native. Compared with the performance of the RTX 3080 Ti on DLSS 2, it is 1.7 times.
"Music of Light: Infinity" After turning on DLSS 3, the RTX 4080 can run to a frame rate of about 2.5 times the native 4K, which is about 1.5 times that of DLSS 2 on RTX 3080 Ti.
"Microsoft Flight Simulation 2020" has performed better after the latest UPDATE11 update. The test uses the route from Chongqing to Chengdu. The RTX 4080 can run to 74 fps under native 4K. After turning on the DLSS 3 performance mode, it can be increased to 172 fps, which is about 2.3 times the native one. In terms of
delay, since the RTX 40 series comes with an optical flow accelerator, it achieves a higher frame rate through optical multi-frame generation of DLSS 3, and effectively reduces the delay through Reflex. Therefore, we can see that in all tested games, the PCL delay is far lower than the 4K native setting when DLSS 3 is turned on. Therefore, when playing games that support DLSS 3 with RTX 40 series graphics card, please remember to turn on DLSS 3 in the game. Not only will the screen details be richer and the game frame rate is higher, the control delay will also become lower, and the experience will be faster and smoother. In terms of image quality of
DLSS 3, we can see from the screenshot of "Cyberpunk 2077", even the performance mode of DLSS 3 is almost no different from the original, and the quality mode is completely unmatched. Players can fully feel at ease to turn on DLSS 3, while enjoying exquisite graphics and smooth frame rates.
In addition, after the release of the RTX 40 series, many DLSS 3 games have also arrived one after another.When GeForce RTX 4080 was released on November 15, 10 DLSS 3 games have been released, including "Plague Legend: Requiem", "Memory of Light: Infinity", "Destroy All Humanity 2: Redetection", "Shadow Torch City", "F1 2022", "Against the Cold", "Life and Death", "Marvel Spider-Man: Remake", " Microsoft Flying Simulation " and "Super Humanity". In addition, "WRC Generations", "Need for Speed: Unruly" and "Warhammer 40K: Dark Tide" will be released soon after the launch of the RTX 4080, and players can play these DLSS 3 games before Christmas.
productivity test
SPEC contains a large number of industrial software testing projects. In the latest 2020V3.1 version, the improvement of RTX 4080 compared with RTX 3080 Ti is still relatively obvious. Among them, Energy's energy test set has the largest improvement, which is 2.1 times that of the RTX 3080 Ti. In fact, the performance of the RTX 4080 is even better than the RTX 3090 Ti.
In terms of regular 3D rendering software, in Blender Benchmark 3.3, the improvement of RTX 4080 compared to RTX 3080 Ti is also obvious. The performance of monster and classroom is about 1.6 times that of RTX 3080 Ti, and junkshop also has a 30% improvement. The performance of
RTX 4080 in V-Ray is also OK, and it can achieve a 50% rendering efficiency improvement compared to the RTX 3080 Ti. The performance improvement in OC renderers is as high as more than 35%. In terms of video editing of
, judging from PR2022, RTX 4080 is still stronger than RTX 3080Ti, which is still the case where PR does not support dual encoders.
As for Da Vinci, the efficiency improvement of MagicMask magic mask is also very obvious. The RTX 3080 Ti takes 31 seconds to complete the work RTX 4080 in just 20 seconds.
As for the new AV1 encoding, let’s take a look at the output efficiency of Da Vinci. The difference between the output time of the RTX 4080 and RTX 4090 is very small, but the dual encoder has indeed achieved exponential efficiency improvement compared to the RTX 3080 Ti.
In addition, if the video memory capacity is not taken into account, the single card productivity performance of RTX 4080 is even better than the RTX 3090 Ti.
In short, RTX 4080 can be said to be the strongest consumer-grade productivity GPU under RTX 4090. The comprehensive performance in 3D rendering exceeds that of the RTX 3080 Ti, and the video encoding performance is exponentially improved. Considering the unrivalled position of the CUDA and NVIDIA STUDIO ecosystem in the industry, RTX 4080 is indeed worthy of the choice of designer users.
Power and temperature test
There is actually nothing to worry about in terms of temperature. After all, the RTX 4080 FE uses a radiator as big as the RTX 4090 FE. Since this radiator is fine to handle the RTX 4090, it is even more fine to deal with the RTX 4080 with TGP 320W.
Speaking of power consumption, I have to say a few more words here. From the release of RTX 4090 to the launch of RTX 4080, many players may see that the nominal TGP power consumption of this generation of flagships and sub-flagships is not low. For TGP power consumption, it is actually the upper limit of the GPU running in Boost state. In terms of easier to understand, it is actually the upper limit of the power consumption wall. As the upper limit of power consumption, it does not refer to the power consumption that the GPU must achieve in daily use and games.
Due to the new TSMC 4N NVIDIA customization process, the RTX 40 series GPU has a better energy consumption ratio, and it consumes less power when achieving the same or even higher performance. This brings up a phenomenon, that is, before the RTX 40 series GPU reaches the upper limit of TGP power consumption, the frequency or voltage may have reached the upper limit, so it does not require higher power consumption to meet the performance release needs. On GPUs of the previous generation Ampere architecture, they usually need to run to the limit of power consumption to achieve the highest performance release. This creates the illusion that the RTX 40 series GPUs are always unsatisfied with power consumption. In fact, this is because the energy consumption ratio is improved and no higher power consumption is required.
We used the NVIDIA FramView tool to monitor the power consumption performance of some RTX 4080 FE at 4K resolution in some games. It can be seen that the average power consumption in these games does not reach TGP power consumption of 320W. However, the RTX 3080 Ti, which is compared in comparison, is almost always running against the upper limit of 350W.
If we simply calculate it, take "Microsoft Flight Simulation 2020" (with the highest image quality in native 4K) as an example. In this game, the frame speed of RTX 4080 is about 1.32 times that of RTX 3080 Ti, but the power consumption is only 73% of the latter. This can be calculated that the energy consumption ratio of RTX 4080 in this game is more than 1.7 times that of RTX 3080 Ti, which can be said to have performed very well.
As a test tradition, we also ran the 3DMark Speedway cycle test machine and used the NVIDIA FramView tool to monitor power consumption. It can be seen that in this test, the power consumption of the RTX 4080 FE is 308W, the maximum GPU temperature is only 60℃, the node temperature is only 70℃, and the maximum GPU frequency reaches 2820MHz. The energy efficiency ratio improvement brought by the new TSMC 4N NVIDIA customization process is indeed quite considerable.
Summary: Dessert in the flagship, Ada Lovelace "80" new combat power definition 4K ray tracing ultimate experience
NVIDIA Ada Lovelace GPU architecture has brought revolutionary progress to the RTX 40 series. In addition to the super-large chip scale and high frequency gain obtained from the TSMC 4N NVIDIA customization process, the third-generation RT Cores and the fourth-generation Tensor Cores are also powerful, especially the newly added optical flow accelerator, which allows the optical multi-frame generation of DLSS 3 technology to be realized, thereby greatly increasing the game frame rate and reducing latency. As the sub-flagship of the RTX 40 series,
RTX 4080 can also run smoothly with the support of DLSS with the support of DLSS. The frame rate increase in games that support DLSS 3 is even more amazing, and it can be regarded as a dessert flagship for enthusiast players.
In addition, from the previous productivity test, we can also see that the RTX 4080 with 16GB of large video memory also has a huge performance increase compared to the RTX 3080 Ti in terms of 3D rendering output and video editing, especially the video encoding performance has even doubled, which can also be regarded as a very practical and efficient tool for productivity applications. While
has greatly improved its performance, the energy consumption ratio of RTX 4080 is also quite good. The actual power consumption in most games is below 300W. While providing high performance, it can also better alleviate players' power consumption anxiety.
Therefore, for players who pursue enthusiastic gaming experience and designer users who pursue high efficiency, the RTX 4080 can definitely be regarded as the current "dessert flagship" choice. As a faith-level gaming artifact from NVIDIA, GeForce RTX 4080 Founders Edition not only provides the ultimate experience of 4K ray tracing games and excellent productivity performance, but also has extremely high collection value, which is worth buying at the first time!
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-