MSI GeForce RTX 5090 SUPRIM SOC 32 GB GPU Review – Enter The Realm of AI-Charged Gaming

NVIDIA GeForce RTX 50 Series - Blackwell Architecture & Features

It's been two years since NVIDIA introduced its Ada Lovelace GPUs, kicking things off with the RTX 4090 and finishing up the initial lineup with the SUPER family At CES, the company unveiled its new RTX 50 "Blackwell" family which features a brand new architecture and several changes such as new cores, AI accelerators, new memory standards, and the latest video/display capabilities.

Today, NVIDIA is releasing the last and fastest card within its "RTX 50" portfolio, the GeForce RTX 5090. The GeForce RTX 5090 is a top-of-the-line graphics card, designed for enthusiast gamers and prosumers, and features a gargantuan price point of $1999 US. This makes the RTX 5090 25 percent more expensive than the RTX 4090 which had an MSRP of $1599 so let's find out if the new features are enough to justify the massive price bump. Today, we will be trying out MSI's GeForce RTX 5090 SUPRIM SOC, a custom variant that is designed with a ton of cooling potential and retails for a premium.

NVIDIA GeForce GPU Segment/Tier Prices

Graphics Segment	2025	2023-2024	2022-2023	2021-2022	2020-2021	2019-2020	2018-2019	2017-2018
Titan Tier	GeForce RTX 5090	GeForce RTX 4090	GeForce RTX 4090	GeForce RTX 3090 Ti GeForce RTX 3090	GeForce RTX 3090	Titan RTX (Turing)	Titan V (Volta)	Titan Xp (Pascal)
Price	$1999 US	$1599 US	$1599 US	$1999 US $1499 US	$1499 US	$2499 US	$2999 US	$1199 US
Ultra Enthusiast Tier	GeForce RTX 5080	GeForce RTX 4080 SUPER	GeForce RTX 4080	GeForce RTX 3080 Ti	GeForce RTX 3080 Ti	GeForce RTX 2080 Ti	GeForce RTX 2080 Ti	GeForce GTX 1080 Ti
Price	$999 US	$999 US	$1199 US	$1199 US	$1199 US	$999 US	$999 US	$699 US
Enthusiast Tier	GeForce RTX 5070 Ti	GeForce RTX 4070 Ti SUPER	GeForce RTX 4070 Ti	GeForce RTX 3080 12 GB	GeForce RTX 3080 10 GB	GeForce RTX 2080 SUPER	GeForce RTX 2080	GeForce GTX 1080
Price	$749 US	$799 US	$799 US	$799 US	$699 US	$699 US	$699 US	$549 US
High-End Tier	GeForce RTX 5070	GeForce RTX 4070 SUPER GeForce RTX 4070	GeForce RTX 4070 GeForce RTX 4060 Ti 16 GB	GeForce RTX 3070 Ti GeForce RTX 3070	GeForce RTX 3070 Ti GeForce RTX 3070	GeForce RTX 2070 SUPER	GeForce RTX 2070	GeForce GTX 1070
Price	$549 US	$599 $549	$599 US $499 US	$599 $499	$599 $499	$499 US	$499 US	$379 US
Mainstream Tier	GeForce RTX 5060 Ti 16 GB GeForce RTX 5060 Ti 8 GB	GeForce RTX 4060 Ti GeForce RTX 4060	GeForce RTX 4060 Ti GeForce RTX 4060	GeForce RTX 3060 Ti GeForce RTX 3060 12 GB	GeForce RTX 3060 Ti GeForce RTX 3060 12 GB	GeForce RTX 2060 SUPER GeForce RTX 2060 GeForce GTX 1660 Ti GeForce GTX 1660 SUPER GeForce GTX 1660	GeForce GTX 1060	GeForce GTX 1060
Price	$429 US $379 US	$449 $299	$399 US $299 US	$399 US $329 US	$399 US $329 US	$399 US $349 US $279 US $229 US $219 US	$249 US	$249 US
Entry Tier	GeForce RTX 5060	RTX 3050 8 GB RTX 3050 6 GB	RTX 3050	RTX 3050	GTX 1650 SUPER GTX 1650	GTX 1650 SUPER GTX 1650	GTX 1050 Ti GTX 1050	GTX 1050 Ti GTX 1050
Price	$299	$229 $179	$249 US	$249 US	$159 US $149 US	$159 US $149 US	$139 US $109 US	$139 US $109 US

NVIDIA GeForce RTX 50 Gaming Graphics Cards

With Blackwell, NVIDIA is going full-on into the AI segment with loads of optimizations & AI-specific accelerators.

The Blackwell GPU does many traditional things which we would expect from a GPU, but simultaneously breaks the barrier when it comes to untraditional GPU operations. To sum up some features:

• New Streaming Multiprocessor (SM)
• New 5th Gen Tensor Cores
• New 4th Gen RT (Ray Tracing) Cores
• AI Management Processor
• Max-Q Mode for Desktops & Laptops
• New GDDR7 High-Performance Memory Subsystem
• New DP2.1b Display Engine & Next-Gen NVENC/NVDEC

The technologies mentioned above are some of the main building blocks of the Blackwell GPU, but there's more within the graphics core itself which we will talk about in detail so let's get started.

The NVIDIA GeForce RTX 5090 is as high-end as Blackwell gets for gamers. This graphics card is a beast and a powerhouse solution for content creators and gamers alike. NVIDIA has gone to lengths to design this engineering feat, so let's see what you get for a hefty price of $2000 US, the biggest-ever price bump for a GeForce GPU.

Updated with the new Blackwell GPU architecture, the NVIDIA GeForce RTX 50 GPUs utilize next-gen hardware technologies such as 5th Gen Tensor Cores, new streaming multiprocessor units, and 4th Gen ray tracing cores. The new architecture also supports FP4 algorithms for even faster AI processing capabilities that can be leveraged by technologies such as DLSS 4.

NVIDIA GeForce RTX 5090 Graphics Card Specifications

Starting with the details, we first have the NVIDIA GeForce RTX 5090, which will be the top GPU of the lineup. The graphics card is expected to feature the PG144/145-SKU30 PCB design and will incorporate the GB202-300-A1 GPU core. The GPU will have 170 SMs enabled out of the total of 192 SMs and will feature 21,760 cores instead of the total of 24,576 cores. This marks an 11.4% reduction, slightly more than the -11.1% reduction of the RTX 4090 versus its full AD102 die.

On the memory side, the NVIDIA GeForce RTX 5090 will be a beast with 32 GB of GDDR7 VRAM capacity running across a 512-bit interface. The VRAM will operate at 30 Gbps speeds for up to 1792 GB/s of bandwidth plus a bigger L3 cache and new memory compression techniques will also boost the overall bandwidth available to the GPU.

The card will come with a 575W TBP rating, though it should be mentioned that the TBP value shouldn't be compared with actual real-world or gaming power consumption as that ends much lower. The graphics card will come with the latest technologies such as PCIe Gen 5.0 support and also get DisplayPort 2.1b UHBR20 (8K 165Hz) support to utilize those newer QD-OLED panels that are being rolled out.

NVIDIA GeForce RTX 5090 Graphics Card Performance

In terms of performance, NVIDIA is showing over 2x faster performance for the GeForce RTX 5090 versus the RTX 4090 and that's using DLSS 4 in games such as Cyberpunk 2077, Alan Wake 2, and Black Myth Wukong, while standard titles such as Far Cry 6 will be featuring around a 30-40% uplift while not leveraging DLSS.

NVIDIA GeForce RTX 5090 Graphics Card Available, Founders Edition & Custom Designs

In terms of design, the NVIDIA GeForce RTX 5090 Founders Edition goes for a dual-slot and dual-fan cooling solution which makes use of a double flow-through design with two intake fans on the front and a very compact nature versus the RTX 4090. The FE model will feature 304mm in length and 137mm in height while leveraging liquid TIM and a 3-layered PCB, a first of its kind for a GeForce GPU. The card will also be available in custom flavors from board partners.

The NVIDIA GeForce RTX 5090 and RTX 5080 will be the fastest GPUs in the lineup and will launch on the 30th of January.

NVIDIA GeForce RTX 50 GPU Specs (Official):

Graphics Card Name	NVIDIA GeForce RTX 5090	NVIDIA GeForce RTX 5080	NVIDIA GeForce RTX 5070 Ti	NVIDIA GeForce RTX 5070	NVIDIA GeForce RTX 5060 Ti	NVIDIA GeForce RTX 5060	NVIDIA GeForce RTX 5050
GPU Name	Blackwell GB202-300	Blackwell GB203-400	Blackwell GB203-300-A1	Blackwell GB205-300-A1	Blackwell GB206-300	Blackwell GB206-250	Blackwell GB207-300
GPU SMs	170 (192 Full)	84 (84 Full)	70 (84 Full)	50 (50 Full)	36 (36 Full)	30	20 (20 Full)
GPU Cores	21760	10752	8960	6144	4608	3840	2560
Clock Speeds	2.41 GHz	2.62 GHz	2.45 GHz	2.51 GHz	2.57 GHz	2.49 GHz	2.57 GHz
Memory Capacity	32 GB GDDR7	16 GB GDDR7	16 GB GDDR7	12 GB GDDR7	16 GB / 8 GB GDDR7	8 GB GDDR7	8 GB GDDR6
Memory Bus	512-bit	256-bit	256-bit	192-bit	128-bit	128-bit	128-bit
Memory Speed	28 Gbps	30 Gbps	28 Gbps	28 Gbps	28 Gbps	28 Gbps	20 Gbps
Bandwidth	1792 GB/s	960 GB/s	896 GB/s	672 GB/s	448 GB/s	448 GB/s	320 GB/s
Power Interface	1 12V-2x6 (16-Pin)	1 12V-2x6 (16-Pin)	1 12V-2x6 (16-Pin)	1 12VHPWR (16-Pin)	1 12VHPWR (16-Pin)	8-Pin	8-Pin
Launch	30th January, 2025	30th January, 2025	20th February, 2025	5th March, 2025	16th April, 2025	19th May, 2025	July 2025
TBP	575W	360W	300W	250W	180W	145W	130W
Price	$1999 US	$999 US	$749 US	$549 US	$429/$379	$299	$249

The NVIDIA Blackwell or RTX Blackwell architecture is designed for gamers and content creators. This architecture will be offered on the RTX 50 graphics cards first, launching later this month.

NVIDIA RTX Blackwell Architecture Deep-Dive

What we have known so far about the NVIDIA RTX Blackwell Gaming GPUs is that they are based on TSMC's 4nm process node, feature up to 92 Billion transistors with 4000 AI TOPS, 380 RT TFLOPs, a 125 TFLOPS of FP32 compute, the fastest GDDR7 memory interface with up to 1.8 TB/s bandwidth and come with a brand new Founders Edition design.

With Blackwell, NVIDIA had a few design goals in mind to accelerate the graphical capabilities for the next generation of gaming. The architecture was to be designed and optimized around new neural capabilities and workloads. It aims to reduce the overall memory footprint, it also focuses vastly on energy efficiency, and all the while adding new quality of service capabilities.

So Blackwell had to introduce a lot of changes. The main ones include the addition of 5th Gen Tensor Cores, offering high-speed FP4, compute and up to 4000 AI TOPS of performance, 4th Gen RT (Ray Tracing) cores with up to 360 RT TFLOPs and designed for Mega Geometry, a next-gen AI Management Processor which enables Simultaneous AI models and graphics workloads to be executed, a brand new Blackwell SM with 125 TFLOPS of peak FP32 compute and the addition of GDDR7 memory which offers the world's fastest memory speeds of up to 30 Gbps (on the RTX 5080).

Other notable RTX Blackwell GPU architecture additions include DisplayPort 2.1 (UHBR20), PCIe Gen5 support, and 4K NVDEC/NVENC with 4:2:2 colors.

Diving into the Blackwell SM, we first compare it with the Ada SM, which was mostly optimized for traditional shaders, & most of its Tensor cores were used either for DLSS or content creation apps. Ada also partitioned the FP32 cores into two blocks, one that could purely execute FP32 and one that could manage both FP32 & INT32 formats.    With Blackwell, NVIDIA has doubled its INT32 GPU throughput which can help accelerate workloads such as Work Graphs and Shader Execution, and 5th Gen Tensor Cores also offer the aforementioned doubled throughput.

Other microarchitectural changes allow multiple workloads to be executed efficiently. Blackwell also improves SER (Shader Execution Reordering) by 2x by reordering the neural models and the standard shading models and putting the same work together in an organized fashion. These models are then passed through tensor cores (if ML models) or shared cores (if shading models) for final execution.

GDDR7 also brings a much-needed upgrade over GDDR6/X memory, offering twice the bandwidth and data rate of G6 memory with higher frequency and lower wattage. GDDR7 also supports PAM4 signaling and the PCB materials used on RTX 50 GPUs are top-of-the-line from an engineering point of view. This is the first full-fledged architecture for desktop PCs to utilize GDDR7 and PCIe 5.0 materials in full conjunction. The new memory interface also offers twice the efficiency of GDDR6 in terms of PJ/bit. This will be very useful in mobility "Max-Q" designs where efficiency matters the most.

Moving over to the Ray Tracing enhancements, the 4th Gen RT Cores introduce various new capabilities, such as a Triangle Cluster Intersection Engine which replaces the previous Triangle Intersection Engine which is optimized for Mega Geometry and can handle clusters of Mega Geometry and standard geometry much more efficiently.

The Mega Geometry engine also has a new Triangle Cluster Compression format which can be decompressed using Blackwell's on-chip engine. Lastly, the new Linerar Swept Spheres block accelerates RTX Hair and Fur rendering. To sum up, the new RT cores bring an 8x ray triangle intersection rate while reducing the memory footprint to 0.75x.

The FP4 format introduced on Blackwell's 5th Gen Tensor Cores will offer up to 32x throughput versus the Pascal generation and 2x versus the Ada generation of GPUs. These new cores will be taking full advantage of Neural Shading and Rendering techniques featured in next-gen AAA titles.

This also leads us to the next topic, which is about Blackwell's scheduling and how it processes various workloads. In Blackwell, NVIDIA is introducing a new programmable Coprocessor known as Amp which sits at the front of the GPU, and interacts differently with the different cores on the GPU while understanding what's running on them, what's being done on them, and scheduling precisely the specific workload for the right core.

NVIDIA also talked about Blackwell's new Power Gating modes. In Blackwell, the entire clock tree can be disabled even while the GPU is active. So, if the memory system or portions of the memory system are idle, power savings can be achieved this way.

Another way to save power is to disable logic and SRAM when entire engines are idle. Blackwell also introduces a secondary rail that separates the core and the memory system which runs them at different voltages and, for different workloads, captures more performance within a power budget. It also allows a 15x reduction in the time it takes from the rail gate to the core. The new rail gate system is particularly useful in laptops as it reduces leakage by a major margin.

A new aspect of Blackwell is also its accelerated frequency switching capability, which improves clock responsiveness by 1000x. For example, a workload such as physics which doesn't utilize the full width of the GPU can switch to a higher frequency, while a tensor core workload which can use the full width of the GPU can move to a lower frequency. But when the CPU hasn't fed the GPU with any work, Blackwell can drop frequency fast and this is done because Blackwell can switch back to a faster frequency faster.

In terms of clock frequency uplift, Blackwell achieves a 300 MHz higher frequency in active state versus Ada GPUs.

Lastly, we have Blackwell's Display & Video capabilities. New support on Blackwell includes support for DisplayPort 2.1b (UHBR20) with High-Speed hardware flip metering which improves the pace of frames using DLSS 4. There's also the 9th Gen Encoder and 6th Gen Decoder, offering AV1 UHQ & 2x H.264 Decode capabilities, while MV-HEVC and 4.2.2 Encode/Decode are also included in the RTX Blackwell Video engine block.

NVIDIA DLSS 4 - Continuing The Deep Learning Innovations

Since the advent of DLSS back in 2018, the technology has been improving continuously. The DLSS model is being trained on a supercomputer housed within the NVIDIA HQ which is running 24/7 & uses the latest and greatest of their GPUs for the past 6 years. The last major iteration of DLSS was DLSS 3.5, which introduced ray reconstruction. This new feature was part of the failure process in which the model detects various issues such as blurriness, ghosting, and flickering.

NVIDIA's in-house team of engineers then try to figure out what went wrong in the model and why the image wasn't created as intended. New approaches are defined to augment the model set which is retrained and tested across 100s of games to achieve the desired image quality and these result in upgraded versions of DLSS, with the latest one now being DLSS 4, which improves upon all aspects of the Super Sampling technology.

With DLSS 4, NVIDIA is transitioning to a completely new neural architecture model from 2020s DLSS 2. The main change is the new transformer engine which can be trained across multiple data sets while being computationally efficient, offering 2x the parameter size and 4x the compute horsepower.

DLSS 4 also adds the new MFG mode or Multi-Frame Generation which, instead of running two models per frame, runs five models per frame with super-resolution and ray reconstruction. This leads to 15 out of 16 pixels or frames being generated by AI, all the while improving the image quality.

NVIDIA also dives a bit into why they only thought of doing multi-frame frame-gen with Blackwell and there were two reasons, one was that DLSS's image quality was not that great to begin with and needed more training time, while the second was the time it takes to generate these new frames could result in frame pacing and artifacting issues.

So, as the DLSS model was trained, the image quality became much better, which is fairly noticeable in recent DLSS 3 and DLSS 3.5 titles, while for frame pacing, NVIDIA's flip metering system is the solution that displays frame and this has been upgraded, and it can now reduce the frame variability by 5-10x, leading to similar or better latency versus last-gen DLSS solutions even when MFG is enabled.

Best of all, while MFG will be limited to RTX 50 and Frame Generation will be limited to the RTX 40 and RTX 50 series, the image quality enhancements and Reflex 2 highlights will apply to all RTX GPUs as we reported here, so all RTX GPU owners are in for a small treat even if they don't own the latest and greatest hardware.

Unlike previous generations of DLSS, DLSS 4 will come with Day-0 support for a total of 75 games and apps, the largest library of DLSS-enabled titles on the first day and more are on the way. To be clear, NVIDIA confirmed that developers will be able to leverage DLSS 4 quite easily if they have already integrated DLSS 3 or DLSS 3.5 within the engines.

NVIDIA will also integrate the option to enable DLSS 4 for non-DLSS-4 titles such as those with DLSS 3 support within its NVIDIA App. You can select between the older CNN mode, which will be a little bit faster but with lower image quality, and choose between various frame generation modes such as 2x, 3x, and up to 4x. Users can also select between resolution override modes such as DLAA if they prefer higher image fidelity or go with DLSS Ultra-Perf, which enables faster performance on RTX GPUs.

Another feature that goes hand in hand with the new DLSS 4 technology is Reflex 2, which is designed to decrease latency and improve responsiveness in games, especially eSports titles. Reflex 2 introduces a new technology called Frame Warp which improves system responsiveness by 75%.

How NVIDIA achieves this is by sampling the mouse position right before the frame is rendered, and the camera is updated based on the user input and then warping the scene to a new position right before the frame is displayed. Reflex 2 will be natively coming to The Finals and Valorant. All RTX GPUs will support Reflex 2.

NVIDIA RTX AI For Gamers: Moving From Programmable Shaders To Neural Shaders

As we stated at the start of this deep dive, the RTX Blackwell architecture is designed with AI in mind and one of the major shifts that NVIDIA is making is to access next-gen Neural Shading technologies.

The company has already announced how it's cooperating with Microsoft to leverage DirectX's Neural Rendering capabilities that will fully unleash the tensor core prowess of RTX 50 GPUs.

With Blackwell's Neural Shaders, NVIDIA will harness various means of doing graphics, such as Neural Textures, Neural Materials, Nerual Volumes, Neural Radiance Fields, Neural Radiance Cache, and Neural Compression techniques to help with efficiency and optimization while adding in more performance by leveraging tensor cores.

The aim is to replace some or entire parts of the traditional graphics pipeline which has mostly relied on programmable shaders for a while now.

With traditional materials in a real-time scene, it's usually a few dozen lines of code to several thousand for film rendering, so to faithfully represent the same data, neural materials take the same code and take assembly collective of layers that are associated with a material and lay it out in a neural space (neural recompressed texture 7:1) & then a very small neural network with a few layers that are executed for each pixel on the screen that is using that material. In a demo shown by NVIDIA, Standard Materials take up 47 megs of memory while Neural materials take up just 16 megs of memory, so that's a factor of 3x.

Next is RTX Neural Radiance Cache for path trace, indirect lighting, and faster performance. NRC is training in real-time using the GPU to create a model on the fly. It collects the transport of light through the scene and caches it geo-spatially, so that you can have virtually unlimited bounces of light throughout the scene of light being transported around and be able to have that in the cache so that you can have one lookup in the cache and represent an unlimited number of bounces on the scene.

One area where Neural Shaders shine is in materials such as skin and, for that, NVIDIA is adding a new feature called RTX Skin. For this purpose, NVIDIA has collaborated with Unreal Technology & Disney, the subsurface scattering algorithm into the real-time realm for path tracing. The demo to showcase RTX Skin was Half-Life 2 Remix, where the bricks are missing indirect lighting by default but enabling NRC brings a lot more detail to the scene. The same can be seen in a Zombie, which looks much more vibrant and shadows are displayed more realistically.

Another feature is RTX Neural Faces, which combine Generative AI Faces with photorealistic characteristics that are trained at different angles, and different lighting conditions and display different emotions too. These can be used to enhance NPCs within games.

Even Ray Tracing hair and fur is discussed with the traditional methods being very expensive, since a character with individual strands may have 6 million strands, which requires a lot of horsepower. In Blackwell's RTX architecture, this new approach requires fewer polygons and fewer ray-traced spheres, leading to 3x less data to compute and requires less VRAM while offering higher FPS.

Blackwell also introduces a new groundbreaking technology called RTX Mega Geometry which allows you to have an uncompromised solution where you can have a full-fat Nanite mesh, fully path-traced, with no rasterization involved. This is done with a new API that quickly, effectively, and efficiently compresses these clusters over time.

In the Zorah Demo, which uses RTX Mega Geometry with half a billion triangles per scene, a lot more details are present when the tech is enabled, and it looks very fluid running on an RTX 50 GPU. RTX Mega Geometry will run on all RTX GPUs, but Blackwell has specific technologies that will accelerate the tech further.

The MSI GeForce RTX 5090 SUPRIM SOC graphics card comes inside a large cardboard box. The front of the package has a large "GeForce RTX" brand logo along with the "MSI" logo in the top left corner and the "SUPRIM SOC" series branding in the lower-left corner. A large picture of the graphics card itself is depicted on the front which gives a nice preview of the SUPRIM design.

The packaging has put a large emphasis on the RTX side of things as the first feature enlisted by AIBs will be NVIDIA Blackwell architecture, Ray Tracing & DLSS support. NVIDIA has bet the future of their gaming GPUs on Ray Tracing support as these are the first cards to offer support for the new feature.

The back of the box is very typical, highlighting the main features and specifications of the cards. The key highlights listed by MSI include the new HYPER Frozr Thermal design with Stormforce fans, Core Pipes, Advanced Vapor Chamber, Wave Curved 4.0 & the Air Antegrade Fin 2.0 heatsink.

There's also a focus towards GeForce.com on each AIB card through which users can download the latest drivers and GeForce Experience application which are a must for gamers to access all feature sets of the new cards.

Once again, the sides of the box greet us with the large GeForce RTX branding. There's also the mention of 32 GB GDDR7 (RTX 5090) memory available on the card. Opening the box, you are greeted with a nice SUPRIM logo.

Outside of the box, the graphics card and the accessory package are held firmly by foam packaging. The graphics card comes with a few accessories and manuals which might not be of much use for hardcore enthusiasts but can be useful for the mainstream gaming audience. The only two useful accessories are the GPU mounting anti-sag bar and the 16-pin to 4x 8-pin power adapter.

The card is nicely wrapped within an anti-static cover which is useful to prevent any unwanted static discharges on various surfaces that might harm the graphics card. The most interesting accessory that I found in the package was a graphics card support mount which is a great add-on that keeps your card sturdy inside your case.

After the package is taken care of, I can finally start talking about the card itself. This thing is a beast and I can't wait to test it out to find what kind of performance improvement I get over current-gen cards.

The card is nicely wrapped within an anti-static cover which is useful to prevent any unwanted static discharges on various surfaces that might harm the graphics card. The most interesting accessory that I found in the package was a graphics card support bracket. This bracket connects the graphics card to the casing, offering better durability and preventing any sort of bending that may occur due to the heavy weight of the SUPRIM series graphics cards.

After the package is taken care of, I can finally start talking about the card itself. This thing is a beast and impressively, it is slightly bigger than the previous SUPRIM variant which was already gargantuan.

MSI’s Tri Frozr heat sinks are some of the biggest heatsink cooling solutions that I have ever tested. I first tested the Gaming X Trio when MSI released the 1080 Ti variant back in 2017 and that was a very aggressive design in its own right. Since then, I have tested the RTX 2080 Ti, RTX 3090, RTX 3090 Ti and the last RTX 4090 SUPRIM in their Tri-Frozr iterations.

With the RTX 50 series cards, MSI has further refined the Tri Frozr design. The card measures even larger at 359 x 150 x 76 mm and weighs in at 2840 grams (427 grams more than the 4090 SUPRIM X). The card features a 3.8 slot height which is expected of today's high-end cards.

You would have to keep in mind the height when going for a triple or quad-slot card solution as your case or motherboard PCIe slot combination may not allow such a setup. The cooling shroud extends to the back of the PCB and it requires a casing with good interior space for proper installation.

The back of the card features a solid metal backplate that looks stunning. The backplate offers a lot more functionality than just looks which I will get back to in a bit.

In terms of design, we are looking at an updated version of the Tri Frozr heatsink known as HYPER FROZR which is now in its ninth variation while for the SUPRIM series, this is the 3rd iteration.

The first variation started with the GTX 780 Ti Lightning, the second was the 980 Ti Lightning, then came the 1080 Ti Gaming X Trio, the 1080 Ti Lightning, then the RTX 20 & RTX SUPER Gaming X Trio graphics cards while the seventh generation was introduced on the RTX 30 series. The 8th Gen was introduced with the RTX 40 series.

The new heatsink looks like a beefed-up version of the SUPRIM X heatsink that we saw on the 4090 with the main changes being the shroud and heatsink design that feature a neater shroud design on the front, absorbing the black and silver color platelets while featuring the RGB emitting V-shaped acrylic cutouts at the front.

The sides also come with a new RGB accent bar which lights up when the card is powered on.

Coming to the fans, the card features the latest fan designs based on the Stormforce design. All three fans feature a ring-based design to allow for higher airflow to be channeled within the main heatsink. All fans deploy a double ball bearing design and can last a long time while operating silently. Each fan has three blades that form three sets and each fan has three sets of them that make up a total of 7 fan blades. Each blade features claw textures to enhance air pressure.

MSI also features its Zero Frozr technology on the HYPER Frozr cooling solution. This feature won’t spin the fans on the card unless they reach a certain threshold.

In the case of the HYPER FROZR heatsink, that limit is set to 60C. If the card is operating under 60C, the fans won’t spin which means no extra noise would be generated.

I am back at talking about the full-coverage, full metal-based backplate that the card uses. The whole plate is made of solid metal with rounded edges that add to the durability of this card. The brushed silver finish on the backplate gives a unique aesthetic.

The graphics card also comes with a compact PCB design which means that the shroud, heatsink, and backplate are all extended beyond the PCB. The third fan blows air through the heatsink and blows it out from the cutouts that are situated at the very end of the backplate.

There are cutouts in screw placements to easily reach the points on the graphics card. We can also see the mew SUPRIM logo which drops the Dragon design and goes for a Diamond shape on the back which looks stunning.

MSI is also using heat pads beneath the backplate which offer more cooling to the electrical circuitry on the PCB. The most interesting thing to spot on the back aside from the backplate is the large retention metal bracket which adds more mounting pressure to effectively disperse heat from the GPU to the heatsink.

With the outside of the card done, I will now start taking a glance at what's beneath the hood of these monster graphics cards. The first thing to catch my eye is the humungous fin stack that's part of the beefy heatsink that the cards utilize.

The large fin stack runs from the front and to the back of the PCB and is so thick that you can barely see through it. It also comes with the wave-curved 4.0 fin stack design which I want to shed some light on as it is a turn away from traditional fin design and one that offers better cooling on high-end graphics cards such as the RTX 5090. The card also uses antegrade fins 2.0 on the back that direct and optimize air pass-through on the back by channeling air through a V-shaped fin-stack cutout, allowing more warm air to pass out of the card like a nozzle.

The heatsink has been designed to be denser by using a wave-curved and filled-fin design. It allows more air to pass through the fins smoothly, without causing any turbulence that would result in unwanted noise.

Talking about the heatsink, the massive block is comprised of 11 copper squared-shaped heat pipes with a more concentrated design to transfer heat from the copper base to the heatsink more effectively. The base itself is a solid nickel-plated base plate, transferring heat to the heat pipes in a very effective manner. To top it all off,

MSI adds extra protection to its impressive PCB by including a rugged anti-bending plate. This also acts as a memory and MOSFET cooling plate while the PWM heatsink with micro fins keeps the VRM cool under stressful conditions.

I/O on the graphics card sticks with the reference scheme which includes three Display Port 2.1b & a single HDMI 2.1 port.

There's also a dual-BIOS switch on the card which comes pre-configured with Silent & Gaming modes. The BIOS doesn't affect the clock profiles but rather affects the maximum power limit, enabling higher fan speeds for better cooling and more stable clocks.

The MSI GeForce RTX 5090 SUPRIM SOC is a very power-hungry graphics card as showcased by its custom design. Being so, the card utilizes a single 16-pin connector which can deliver up to 600 Watts of power to the graphics card. The card is rated at 575W but ends up around 600W with its full power limit.

MSI GeForce RTX 5090 SUPRIM SOC RGB Lighting Gallery:

MSI SUPRIM SOC series cards utilize their Mystic Light RGB technology to offer you a visually pleasing lighting experience on your graphics cards.

There are a total of 5 different RGB effects that you can choose from and the cards have 3 RGB accent points on the front, one on the back, and one lightbar surrounding the side of the card which looks good. You can fully customize the RGB lights to your preference using the MSI Mystic Light application from MSI's web page. Following is what the graphics card looks like when lit up.

We used the following test system for comparison between the different graphics cards. The latest drivers that were available at the time of testing were used by AMD, Intel & NVIDIA on an updated version of Windows 11. All tested games were patched to the latest version for better performance optimization for NVIDIA, Intel, and AMD GPUs.

WCCFTECH GPU Test Bench (2025):

CPU	Intel Core i9-13900K @ Default
Motherboard	MSI MEG Z790 ACE
Video Cards	MSI GeForce RTX 5080 Vanguard SOC MSI GeForce RTX 5080 SUPRIM SOC GALAX GeForce RTX 5080 1-Click OC White MSI GeForce RTX 5090 SUPRIM SOC NVIDIA GeForce RTX 4090 FE Palit GeForce RTX 4080 SUPER GamingPro OC NVIDIA GeForce RTX 4080 FE Colorful GeForce RTX 4070 Ti SUPER BattleAx NVIDIA GeForce RTX 4070 SUPER FE ASRock Radeon RX 7800 XT Phantom Gaming ASUS GeForce RTX 4070 Ti TUF Gaming NVIDIA GeForce RTX 4070 FE NVIDIA GeForce RTX 4060 Ti FE MSI GeForce RTX 4060 Gaming X ASUS RX 7900 XTX TUF Gaming OC ASRock Radeon RX 7900 XT Phantom Gaming
Memory	G.SKILL Trident Z5 RGB Series 32GB (2 X 16GB) CL38 7200 Mbps
Storage	Teamgroup T-Force A440 Pro 2 TB Gen 4
Power Supply	MSI MEG Ai1300P 1300W PSU
OS	Windows 11 64-bit (24H2)
Drivers	AMD Radeon Adrenalin Edition 24.12.1 NVIDIA GeForce 572.16 WHQL Intel 6256 WHQL

• All games were tested at 3840x2160 (4K) resolution.
• Image Quality and graphics configurations are provided with each game description.
• The "reference" cards are the stock configs except where mentioned otherwise.

Speed Way

Developed with input from AMD, Intel, NVIDIA, and other leading technology companies, Speed Way is an ideal benchmark for comparing the DirectX 12 Ultimate performance of the latest graphics cards. 3DMark Speed Way’s engine is assembled to demonstrate what the latest DirectX API brings to ray-traced gaming, using DirectX Raytracing tier 1.1 for real-time global illumination and raytraced reflections, coupled with new performance optimizations like Mesh Shaders.

Firestrike

Firestrike is running the DX11 API and is still a good measure of GPU scaling performance. In this test, we ran the Extreme and Ultra versions of Firestrike which runs at 1440p and 4K, and recorded the Graphics Score only since the Physics and combined are not pertinent to this review.

Time Spy

Time Spy is running the DX12 API and we used it in the same manner as Firestrike Extreme where we only recorded the Graphics Score as the Physics score is recording the CPU performance and isn't important to the testing we are doing here.

Port Royal

Port Royal is another great tool in the 3DMark suite, but this one is 100% targeting Ray Tracing performance. It loads up ray-traced shadows, reflections, and global illumination to tax the performance of the graphics cards that either have hardware-based or software-based ray-tracing support.

Doom Eternal

DOOM Eternal brings hell to earth with the Vulkan-powered IDTech 7.  We test this game using the Ultra Nightmare Preset and follow our in-game benchmarking to stay consistent.

Red Dead Redemption 2

Developed by Rockstar San Diego, Red Dead Redemption 2 is one of the most visually stunning open-world games I've played to date. It is backed up by a rich story set about the protagonist, Arthur Morgan. The game is based on the RAGE engine which features an insane amount of graphics fidelity but also requires much power to run maxed out. For this test, we set the graphics settings to Ultra with AA turned disabled.

Wolfenstein: Youngblood

Wolfenstein is back in The New Colossus and features the most fast-paced, gory, and brutal FPS action ever! The game again puts us back in the Nazi-controlled world as BJ Blazkowicz. Set during an alternate future where Nazis won the World War, the game shows that it can be fun and can be brutal to the player and to the enemy too. Powering the new title is, once again, id Tech 6 which is much acclaimed after the success that DOOM has become. In a way, ID has regained its glorious FPS roots and is slaying with every new title.

Alan Wake 2

Alan Wake 2 sets you up in a horror thriller between two dimensions and lets you play better two different protagonists, Alan himself and Saga, who once again have to find a way to fix the darkness that erupted in Bright Falls.

Cyberpunk 2077 is an action role-playing video game developed by CD Projekt Red and published by CD Projekt. The story occurs in Night City, an open world in the Cyberpunk universe. Players assume the first-person perspective of a customizable mercenary known as V, who can acquire skills in hacking and machinery with options for melee and ranged combat. The game uses CD Projekt Red's in-house Red Engine, one of the most visually breathtaking and graphics-intensive engines designed to date.

Atomic Heart

Atomic Heart is set in an alternate universe where the Soviet Union achieved incredible technological breakthroughs thanks to Dr. Sechenov, who invented a liquid programmable module called Polymer that links robots in a so-called Kollektiv network.

Battlefield V

Battlefield V brings back the action of the World War 2 shooter genre. Using the latest Frostbite tech, the game does a good job of looking gorgeous in all ways possible. From the open-world environments to the intense and gun-blazing action, this multiplayer and single-player FPS title is one of the best-looking Battlefields to date.

Baldur's Gate III

2023's GOTY is well-deserved its title. The creation from Larian Studios is a turn-based RPG with gorgeous interiors and exteriors shown through a bird's eye top-to-bottom view. You can sink in countless hours in the game and if you're a fan of the D&D playstyle, then this epic is just for you.

Cyberpunk 2077

Cyberpunk 2077 is an action role-playing video game developed by CD Projekt Red and published by CD Projekt. The story occurs in Night City, an open world in the Cyberpunk universe. Players assume the first-person perspective of a customizable mercenary known as V, who can acquire skills in hacking and machinery with options for melee and ranged combat. The game uses CD Projekt Red's in-house Red Engine, which is one of the most visually breathtaking and graphics-intensive engines designed to date.

Dead Space (Remake)

Remaking Dead Space was a bold choice but I would say that the team at EA Motive nailed every bit and piece of this horror classic. The remake makes the USG Ishimura twice as scarily beautiful. The gore, endless corridors of terror, and the void of space all look incredible while the game remains true to its core to the original Dead Space formula. Modern cards can run the game well but can also be demanding if you crank the settings to the max with ray tracing enabled.

Death Stranding

Sam Porter Bridges has delivered one of PS4's most anticipated games to the PC community and opened a new world of possibilities. This was the first game to feature the Decima Engine on PC and unarguably did it the best. Death Stranding may not feature ray tracing effects, but it does showcase that DLSS can be used effectively even when RT isn't around. We tested this one just like in our launch coverage with DLSS enabled.

Forza Horizon 5

Forza Horizon 5 carries on the open-world racing tradition of the Horizon series.  The latest DX12-powered entry is beautifully crafted, amazingly well executed, and a great showcase of DX12 games.  We use the benchmark run while having all the settings set to non-dynamic with an uncapped framerate to gather these results.

Halo Infinite (DX12 Highest)

Next up, we have the latest entry in the Halo franchise, Halo: Infinite, which uses the brand new Slipspace engine (although there are rumors it will be ditched in the future for Unreal Engine) based on the DX12 API. The game rocks some incredible environments for Master Chief to visit on the Halo ring.

Hitman III (DX12 Highest Settings)

Hitman III is the highly acclaimed sequel to the 2016 Hitman & 2018 Hitman II, a redesign and reimaging of the game from the ground up. With a focus on stealth gameplay through various missions, the game again lets you play as Agent 47. The game runs on the IO Interactive Glacier 2 engine which has been updated to deliver amazing visuals and environments on each level while using DirectX 12 API.

Metro Exodus

Metro Exodus continues Artyom's journey through Russia's nuclear wasteland and its surroundings. This time, you are set over the Metro, going through various regions and different environments. The game is one of the premier titles to feature NVIDIA’s RTX technology and does well in showcasing the ray-tracing effects in all corners.

Resident Evil Village

Resident Evil Village is the latest in the horror franchise that was wonderfully rekindled with RE7 and onto the RE2 Remake. But now the RE Engine is back and better than ever with Ray Traced Reflections and Lighting that makes the world just come to life, unironically. The game was tested in the center of the village itself with all graphical settings maxed out and with raytracing enabled.

Resident Evil IV Remake

The remake of the beloved and highly acclaimed Resident Evi IV is here, boasting the latest RE engine which adds stunning visuals and even better ray tracing effects, the game looks just as incredible as it plays.

Starfield

Bethesda's latest RPG epic is set in space and takes place across a vast universe, filled with lots of planets to explore. Based on the latest iteration of the Creation Engine, Starfield offers a great amount of visual fidelity, whether you are exploring an abandoned base or just roaming a planet on which you just laid your foot.

Crysis Remastered (DXVK RT)

Crysis is back with a vengeance to reclaim its title of the graphics crown. The remastered version of the game uses DX11 API but has Vulkan extensions on top which enable Vulkan Ray tracing. That's also something that the original game didn't offer. DXVK, along with improved textures and visual effects, leads to higher performance demand making us question once again "Can It Run Crysis?"

Doom Eternal

DOOM Eternal brings hell to earth with the Vulkan-powered IDTech 7.  We test this game using the Ultra Nightmare Preset and follow our in-game benchmarking to stay as consistent as possible.

Alan Wake 2

Alan Wake 2 sets you up in a horror thriller that takes place between two dimensions and lets you play better two different protagonists, Alan himself and Saga, who once again have to find a way to fix the darkness that erupted in Bright Falls.

Note - FSR Frame-Gen mod applied across all GPUs. DLSS 3.5 has been applied to RTX 50/40 GPUs.

Battlefield V

Battlefield V brings back the action of the World War 2 shooter genre. Using the latest Frostbite tech, the game does a good job of looking gorgeous in all ways possible. From the open-world environments to the intense and gun-blazing action, this multiplayer and single-player FPS title is one of the best-looking Battlefields to date.

Cyberpunk 2077

Cyberpunk 2077 is an action role-playing video game developed by CD Projekt Red and published by CD Projekt. The story occurs in Night City, an open world in the Cyberpunk universe. Players assume the first-person perspective of a customizable mercenary known as V, who can acquire skills in hacking and machinery with options for melee and ranged combat. The game uses CD Projekt Red's in-house Red Engine, which is one of the most visually breathtaking and graphics-intensive engines designed to date.

Dead Space (Remake)

Remaking Dead Space was a bold choice but I would say that the team at EA Motive nailed every bit and piece of this horror classic. The remake makes the USG Ishimura twice as scarily beautiful. The gore, the endless corridors of terror, the void of space, all of it looks incredible while the game remains true to its core to the original Dead Space formula. Modern cards can run the game well but it can also be demanding if you crank the settings to the max with ray tracing enabled.

Hogwarts Legacy

Hogwarts Legacy, as the name suggests, is set in the world of Hogwarts and retains its landscape true to the books and the movies. The game looks visually stunning although it can be a total hog when running at the highest settings with all visual candy enabled.

Shadow of The Tomb Raider

The sequel to Rise of the Tomb Raider, Shadow of The Tomb Raider is visually enhanced with an updated Foundation Engine that delivers realistic facial animations and the most gorgeous environments ever seen in a Tomb Raider Game. The game is a technical marvel and shows the power of its graphics engine in the latest title.

Metro Exodus

Metro Exodus continues Artyom's journey through Russia's nuclear wasteland and its surroundings. This time, you are set over the Metro, going through various regions and different environments. The game is one of the premier titles to feature NVIDIA’s RTX technology and does well in showcasing the ray-tracing effects in all corners.

Resident Evil Village

Resident Evil Village is the latest in the horror franchise that was wonderfully rekindled with RE7 and onto the RE2 Remake. But now the RE Engine is back and better than ever with Ray Traced Reflections and Lighting that makes the world just come to life, unironically. The game was tested in the center of the village itself with all graphical settings maxed out and with raytracing enabled.

Resident Evil IV Remake

The remake of the beloved and highly acclaimed Resident Evi IV is here, boasting the latest RE engine which adds stunning visuals and even better ray tracing effects, the game looks just as incredible as it plays.

Stray (That Cat Game)

Stray is a 2022 adventure game developed by BlueTwelve Studio and published by Annapurna Interactive. The story follows a stray cat who falls into a walled city populated by robots, machines, and mutant bacteria, and sets out to return to the surface with the help of a drone companion, B-12. The game uses Unreal Engine 4, but DX12 Ray tracing can be enabled by adding the "-dx12" extension to the game.

To test DLSS 4, we were sent press builds which allowed us to enable Multi Frame Generation tech within the following titles. For now, we are focusing on the quality preset at 4K since you are getting over 200 FPS easily, so unless you want more FPS (240+ for high-refresh rate 4K QD-OLED or standard gaming monitors), "Performance" and "Ultra Performance" modes don't make sense.

No graphics card review is complete without evaluating its temperatures and thermal load. All of the graphics cards that we tested were running their default 'Performance' BIOS and the results are below:

I compiled the power consumption results by testing each card under idle and full stress when the card was running games. Each graphics card manufacturer sets a default TDP for the card which can vary from vendor to vendor depending on the extra clocks or board features they plugin on their custom cards. The default TDP for the RTX 5090 is rated at 575W.

NVIDIA's next generation is finally here and oh boy, it has a lot of things we need to talk about. So let's start with the performance.

NVIDIA Is The Only Next-Gen GPU Choice For Enthusiasts

NVIDIA isn't giving up on the enthusiast gaming segment like its competitors and while the gains aren't as big as we saw in the previous generation, they still claim a solid position in the gaming market with the RTX 5090 flagship.

Talking about rasterization and ray tracing-specific performance, you're looking at a 30-40% bump in performance which is very decent. Sure, most of us were expecting similar gains as we saw from the RTX 4090 (versus the RTX 3090) but NVIDIA is achieving these gains on the same node so those bigger jumps might have to wait till the next node shrink and architectural update. This 30-40% up-lift in performance does lead to higher gaming potential, especially at 4K, and just like the RTX 4090, the RTX 5090 will require the most powerful CPUs available on the market today to extract their full potential (go for AMD's Ryzen X3D CPUs based on Zen 4 or Zen 5).

DLSS4 Is The Game Changer, Better Image Quality Than DLSS3 With Transformers Gunning For 4K 240Hz Gaming

DLSS4 is huge! With the updated upscaling technology, NVIDIA is taking image quality to the next level. The model they have been training is in the most advanced state yet with the image quality improving day by day. DLSS4 is a visual upgrade over DLSS3 and all RTX gamers can enjoy its benefits. With the faster transformer model on the Blackwell GPUs, you will see limited losses in terms of performance when switching from the older CNN model to the newer transformer model.

Now DLSS has been plagued with the "Fake Frames" controversy in regards to the frame-gen tech but it does offer better performance and smoothness. NVIDIA is quadrupling the frame-gen with a 4x mode which boosts FPS dramatically. This is enough for those who want to utilize the potential of their 4K 240Hz monitors.

Yes, there will be certain artifacts when using DLSS4 with frame generation enabled but like I said, the DLSS model is improving and this is DLSS at its finest. We can expect DLSS4 to receive countless amounts of updates in the coming months and the time I spent trying it out, there were just minor artifacts that you can only notice when focusing on a certain section of the frame. It is almost visually indistinguishable from the native render.

NVIDIA's Blackwell Is Built For The Future of Gaming, But We Might Have to Wait

The NVIDIA Blackwell architecture is designed with AI in mind and while DLSS 4 and its 4x Multi-Frame Generation technology is one aspect, there's also support for FP4 which would help in certain AI workloads.

Things we liked about the MSI GeForce RTX 5090 SUPRIM SOC:

• The Ultimate & Fastest Gaming GPU
• Still Very Efficient For a 575W GPU
• Built For 4K 240Hz gaming monitors
• Designed With Next-Gen Neural Rendering
• Good Raster & RT Improvement
• Next-Gen Blackwell Architecture With Several AI Additions
• Massive 32 GB GDDR7 Memory
• DLSS4 Offers Better Image Quality
• DLSS4 MFG Offers 4x FPS Boost
• MSI SUPRIM SOC Design Looks Spectacular
• Cooling Performance is Superb
• Very low noise output with 0db operation
• Next-Gen Video Encode/Decode Engines
• DP2.1b (UHBR20) support
• PCIe Express 5.0 technology

The big gun is the neural shaders support along with several new Neural rendering approaches that the company has highlighted and discussed. While the company hasn't yet announced any major game partnerships that will make use of these technologies, we are anxiously waiting for NVIDIA's update in this regard as that would further unlock the architectural potential of NVIDIA's Blackwell architecture.

Custom Coolers Are Supreme, & MSI's SUPRIM SOC Knocks It Out of the Park

Cooling an almost 600W behemoth is not an easy job but MSI has been constantly at work improving its high-end cooling solutions such as the SUPRIM SOC. The card is bigger than ever but the extra mass and heatsink real estate do a great job of keeping this juggernaut at 65C at peak loads and closer to 60C during gaming. The card also runs a bit more silently than the previous RTX 4090 SUPRIM design at both Silent and Max RPM modes. The only thing you should keep in mind is to have 4 slots worth of room available inside your PC.

Things we would have liked about the MSI GeForce RTX 5090 SUPRIM SOC:

• Lower Price Point
• More overclocking potential without limitations (voltage locked)
• Huge Graphics Card (4-slot design)
• GDDR7 Memory runs hot
• A few games with Neural Shaders at launch

The 12V-2x6 connector and the supplied adapter cable are also great with MSI outlining the section with a yellow tint to make sure users have the connector plugged in all the way. This solution was already featured in MSI's PSU lineup and it's great to see it being expanded to the GPU lineup too. Overall, MSI knocked things out of the park with its impressive SUPRIM SOC design and it's right there at the top of our list as a strong custom variant of the RTX 5090 GPU.

Big Performance Comes At Big Cost

With 32 GB of memory, a supercharged GPU architecture built with AI in mind, and no competition in sight, NVIDIA has gone with another price hike for its enthusiast-tier 90 card. The RTX 5090 costs $1999 US with premium models easily going for around $2200 US+ and availability is also a major question. You'll be paying at least 25% more than the 4090, with 28% higher power consumption for 30-40% better performance. Enthusiasts will take every bit of extra performance that they can get but unlike the RTX 4090, the RTX 5090's value doesn't look that enticing. If we were getting a 50 percent performance bump, it would've attracted the same kind of appeal as the RTX 4090 but that isn't there so fully recommending the RTX 5090 to all gamers is not a wise choice.

Things

It will be up to the enthusiasts to decide whether the pricing justifies an upgrade over existing GPUs but from a technological perspective, the RTX 5090 is a big deal with lots of future potential. NVIDIA made Blackwell and the RTX 50 series with the future in mind, they knew their competition wasn't getting anywhere close to them so they went a very different route rather than focusing on traditional improvements such as better raster and ray tracing capabilities. If NVIDIA can pull it off and deliver on the Neural AI and Neural Shading promises that Blackwell is designed for, then we are in for a truly spectacular era of next-gen PC gaming.

Follow Wccftech on Google to get more of our news coverage in your feeds.