Nvidia RTX 4080 16GB (MSI Suprim X) Review

Nvidia

Today, Nvidia took the covers off the RTX 4080 16GB graphics card. Based on the AD103 chip, you get everything the Ada Lovelace architecture brings except for the flagship-class price. Starting at $1,199, it’s slightly more reasonable than the $1,499-plus RTX 4090 we reviewed previously. You get 4th gen Tensor cores and Optical Glow, 3rd gen Ray Tracing cores, DLSS 3, 8th gen AV1 encoders, and everything the new Ada Lovelace architecture offers.

MSI was kind enough to send their RTX 4080 Suprim X, which comes with the monster 3-plus slot Tri-FrozR 3S cooling solution, more robust power delivery, and slightly increased clocks out of the box. We’ll take a more detailed look at the features and let you know where the RTX 4080 stands performance-wise and see where it stands against the RTX 4090 and other comparable graphics cards.RTX 4080

RTX 4080

ADA Lovelace Architecture and Technologies

For those who read our RTX 4090 Suprim Liquid X review, the Ada Lovelace architecture shouldn’t be anything new. Below we’ve kept a lot of the information in there but updated it with the RTX 4080 clock speeds and specifications in case you’re new or just want a refresher.

Nvidia Ada Lovelace Core
Nvidia Ada Lovelace Core

Nvidia’s new Ada Lovelace architecture is fabricated on TSMC’s 4N manufacturing process. The smaller process allowed Nvidia to dramatically increase the transistor count, which turns into a lot more cores (70% more CUDA cores than GA102). The new GPU consists of a whopping 76.3 billion transistors, which they state makes it one of the most complex chips ever made.

With the new core, the Ada architecture also operates at higher clock frequencies using “… high-speed transistors in critical paths that could otherwise restrict the rest of the chip”. The RTX 4080 and AD103 chip inside comes in with clocks of 2.205 GHz on the core and 1,400 MHz on the 16GB GDDR6X with a 256-bit bus. Power consumption is listed as 320W with partner cards like our MSI, likely increasing that limit.

The AD103 GPU found on the RTX 4080 comes equipped with 7 Graphics Process Clusters (GPCs), 40 Texture Processing Clusters (TPCs), 76 Streaming Multiprocessors (SMs), along with a 256-bit memory interface (8x 32-bit memory controllers). There are also 152 FP64 cores (2 per SM) not shown in the diagrams. Nvidia states the TFLOP rate for the FP64 cores is 1/64th the TFLOP rate of FP32 operations, and the small FP64 core count is included to ensure compatibility with FP64 code which includes Tensor Core code.

RTX 4080 Block Diagram
RTX 4080 Block Diagram

The third generation RT core in these Ada GPUs adds a dedicated unit called the Opacity Micromap Engine and the Displaced Micro-Mesh Engine (DMME). The former evaluates Opacity micromaps that are used to accelerate alpha traversal. The DMME allows the Ada RT core to ray trace complex objects and the environment with significantly less BVH build time and storage costs which, when simplified, makes for a much faster and more efficient pathing. Together, Nvidia claims the Ada 3rd-gen RT core is the most powerful they ever made.

Ada Streaming Multiprocessor
Ada Streaming Multiprocessor

Another significant improvement over the previous generation is the upgrade from DLSS 2.0 to DLSS 3.0. The new sauce inside is a new frame generation technique that combines optical flow estimation with DLSS to improve the gaming experience. In other words, inserting accurately synthesized frames between existing frames improves the frame rate and provides a smoother gaming experience. The most significant performance uptick, you’ll see, is with using DLSS.

The Tensor cores have also been updated. Compared to Ampere, Ada provides more than double the FP16, BF16, TF32, INT8, and INT4 Tensor TFLOPS and runs the Hopper FP8 Transformer Engine, delivering over 1.3 PetaFLOPS of tensor processing on the 4090.

For those who like to stream content, Nvidia’s NVENC encoder and existing optimizations for OBS (Open Broadcaster Software) eliminated the need for a dedicated PC for video capture so you can play and stream at the PC without sacrificing quality or performance. Ada takes that a step further, incorporating support for AV1 video encoding the 9th gen NVENC hardware encoder. Previous generations only decoded but didn’t encode. The AV1 encoder is said to be 40% more efficient than the H.264 encoder found in the 3000 series GPUs. With this update, users can increase from 1080p to 1440p while running at the same bitrate and quality. The 4080 and 4090 use two NVENC encoders for the work.

Details aside, the table below lists the specifications for the new ADA GPUs, including our review sample, MSI RTX 4080 Suprim X.

Specifications and Features

Nvidia RTX 4000 Series Specifications
ModelMSI RTX 4080 SUPRIM XRTX 4090RTX 4080
(16GB)
RTX 4070
(12GB)
ArchitectureAda Lovelace
(AD102, AD103, AD104)
Manufacturing TSMC 4N (5 nm)
CUDA Cores9,72816,3849,7287,680
RT Cores
(Gen 3)
801448060
Tensor Cores
(Gen 4)
320576320240
Texture Units
(3rd Gen)
320576320240
ROPs11219211280
L2 Cache64 MB96 MB64 MB48 MB
Base Clock2,205 MHz2,230 MHz2,210 MHz2,310 MHz
Boost Clock2,625 MHz2,520 MHz2,510 MHz2,610 MHz
Memory16 GB
GDDR6X
24 GB
GDDR6X
16 GB
GDDR6X
12 GB
GDDR6X
Memory Speed718 GBps1008 GBps736 GBps504 GBps
Memory Bus256-bit384-bit256-bit192-bit
Supplementary
Power
PCIe 5 16-pin
3x 8-pin
PCIe 5 16-pin
3x 8-pin
PCIe 5 12-pin
2x 8-pin
PCIE 5 12-pin 2x PCIe 8-pin
Standard Display
Connectors 
HDMI (2.1)
3x DisplayPort (1.4a)
Max Resolution8K (7680 x 4320)
TDP320+ W450 W320 W285 W
Release Date11/15/202210/12/2210/12/2210/12/22
MSRP$1,379$1599$1,199$899

Before we get into the MSI card, here are a few stock images of the RTX 4080 FE.

The MSI RTX 4080 Suprim

The RTX 4080 Suprim we have in hand includes more robust power delivery, the massive Twin Frozr 3S, and a 320W+ TBP out of the box. Like the RTX 4090, the new 4080 card sources its power from the latest PCIe 5.0 12(+4) 12VHPWR connector that offers up to 450 W requiring three (instead of four and 600W) 8-pin PCIe connectors. MSI recommends a. MSI recommends an 850 W power supply for this card with a minimum of 750 W.

The RTX 4080 Suprim X includes a dual BIOS for silence and a more aggressive profile for gaming. Both BIOS offer the same performance but with different temperature thresholds and fan speeds/noise. The 2,625 MHz boost clock is the same as the 4090 Suprim Liquid X and one of the fastest listed on the market. But as many already know, performance among its peers will be very close regardless. The Suprim X variant is priced higher than the Founders Edition by $170 ($1379).

The RTX 4080 Suprim X lands on store shelves and improved over the Founder’s Edition (FE) card with impressive power delivery and that massive Twin Frozr3S cooling solution. But suppose the flagship SuprimX is still too expensive. In that case, MSI has other options like the RTX 4080 Gamin X Trio at $1309.99, which uses a custom cooler and is likely working with the reference card hardware specification.

As far as the appearance, the Suprim X sports a metallic shroud with gold trim covering the business parts on the front, while a matching grey backplate protects the rear and helps keep the card cool. Also on the front are three large 120mm Torq 5.0 fans that sit inside octagon-shaped cutouts removing the heat from the memory and VRMs generated by this 320 W card. The fans also utilize MSI’s Zero Frozr technology that keeps the fan off until cooling is needed. The fans never spun up at idle/desktop and watching videos and the card was completely silent. Even when fully loaded, the cooling solution was barely audible over the system it was installed in. If the noise is still too much, flip the dual bios switch to the Silent mode instead of gaming, and the fans ramp up more slowly.

If RGB lighting is your thing, the Suprim X sports illuminated chevrons next to the fans, MSI and Suprim branding on top, and a lit-up Suprim design feature on the back. While the features are certainly visible, it’s not bright enough to steal the thunder from anything inside the chassis. Still, it compliments an RGB-laden motherboard or other integrated lighting (controlled through the Mystic Light application).

Torq Fan 5.0
Torq Fan 5.0

TORX FAN 5.0

Design improvements to TORX FAN 5.0 result in +23% airflow compared to an axial fan and +10% airflow compared to TORX FAN 4.0.

Vapor Chamber
Vapor Chamber

Vapor Chamber

A Vapor Chamber is excellent for rapidly stabilizing heat on a planar surface. The GPU and VRAM are covered with a Vapor Chamber which transfers heat to Core Pipes.

Core Pipes
Core Pipes

Core Pipes

Core Pipes are precision-crafted to make maximum use of the available space. A squared section of heat pipes fully touches the vapor chamber to spread heat along the full length of the heatsink.

Backplate
Backplate

GOT YOUR BACK

Fine details elevate the backplate, such as its brushed metal finish and the beveled edges polished with a light-gold color. Thermal pads beneath the sturdy plate provide additional cooling to the graphics card.

Ball Bearing
Ball Bearing

Ball bearings

Two sets of durable ball bearings spin the TORX FANs for years of intense and lengthy gaming sessions.

Zero Frozr
Zero Frozr

ZERO FROZR

The fans completely stop when temperatures are relatively low, reducing the noise significantly when active cooling is not needed. The fans will automatically start spinning again when the heat is on during gaming.

Retail Packaging and Accessories

The retail packaging for the MSI Suprim X comes in a predominately white box that includes a picture of the card along with the typical Nvidia Geforce RTX colors and branding. The back of the box shares some features about the card inside. After opening the box, you’re greeted by a black pad that covers the card inside and the included accessories. Lifting that exposes the card sitting snug inside the form-fitting foam. Additionally, you’ll find the included mouse pad, instructions, a GPU strut to support the heavy card, and the all-important 3x 8-pin PCIe to 12+4-pin PCIe 5.0 adapter to power the card.

Meet the MSI RTX 4080 Suprim X

Card Front
Card Front

Among the aftermarket cards, the Suprim X is right up there, with the rest sporting a premium appearance and using substantial cooling solutions. The chevrons to the fan’s right have RGBs below to provide a little bling on the front. The top of the card also sports RGB lighting to illuminate more branding. Ultimately, it’s a beautiful card and would be a nice showpiece for your build.

A Closer Look at the MSI RTX 4090 Suprim Liquid X

Zooming in on the I/O, we see three DisplayPort ports (v1.4a) and one HDMI (v2.1) port. The former supports a maximum resolution of 8K (7680×4320), while the HDMI port handles up to 4K @ 120Hz. The Suprim X’s Tri Frozr 3S cooler sends most of the air inside the case while some go out through the IO plate. You’ll want good airflow for this card to keep case thermals down and the card running as it should.

Power is sent through the new 12+3-pin PCIe 5.0 connector, which takes up much less space than the traditional 6+2-pin PCIe connectors we’re familiar with. Not only does the new connector take up less real estate on the card, but the PCIE-SIG-designed 12VHPWR plug can also deliver up to 600 W (this one 450W), which is a lot more than a single 6+2-pin is capable of (150 W). MSI includes the proper adapter to connect with the card, which, in this case, requires THREE independent PCIe 6+2-pin connections – do not use piggybacked connectors!

Also pictured in this image is the dual-BIOS switch. As it’s labeled, the Silent BIOS has a slower ramp-up and Zero Frozr capability, while the Gaming BIOS ramps up quicker and peaks at faster speeds offering more cooling (also uses Zero Frozr). While the clock speeds and power limits don’t change, fan behavior does.

Rear IO
Rear IO –

Power and BIOS Switch
Power and BIOS Switch

We also have some pictures of the core and memory and the PCB. You can see the Micron Memory, the AD103-360-A1 chip, and the upgraded power delivery (from FE), including ON Semiconductor NCP302150 50A MOSFETs for the 18-phase power delivery.

PCB
PCB

AD103-360-A1 GPU
AD103-360-A1 GPU

Micron Memory
Micron Memory

ON Semiconductor 50A MOSFETs
ON Semiconductor 50A MOSFETs

Below is a screenshot of GPU-Z showing the clocks we achieved at stock speeds. Out of the box, the MSI RTX 4080 Suprim X has a listed boost clock of 2,625 MHz, which it maintained much higher (upwards of 2,800 MHz) throughout our testing, even on air.

GPUz - Stock
GPUz – Stock

MSI RTX 4080 Suprim X on the test bench…

MSI RTX 4080 Suprim X on the test bench
MSI RTX 4080 Suprim X on the test bench

Test System and Benchmark Methods

Test System Components
MotherboardGigabyte Z690 Tachyon ($349.99)
CPUIntel i9-12900K (stock)
CPU CoolerCorsair iCUE H150i
MemoryKingston Fury Beast 2×16 GB 5200 MHz CL40 ($161.99)
SSDMushkin Helix 1TB NVMe (OS + Applications – $80)
Power SupplyEVGA 850 W P6 ($119.99)
Video CardMSI RTX 400 Suprim X ($1,349) / 526.72 driver

 

Our test system is based on the latest (at the time of publishing) mainstream Intel z690 platform and uses the i9-12900K 8P,8E/24t CPU. The DRAM is in a 2×16 GB configuration at 5200 MHz with CL40 timings, a middle-of-the-road option balancing performance, and cost. The CPU runs stock.

Since the last update, we have made some changes and updated titles. More details can be found in the GPU Testing Procedure article, which we have updated with our latest benchmarks. Below is a quick summary for easy reference.

  • UL 3DMark Time Spy – Default settings
  • UL 3DMark Fire Strike (Extreme) – Default settings
  • UL 3DMark Port Royal – Default Settings (Ray Tracing capable cards only)
  • Unigine Superposition – 1080p High, 1080p Extreme
  • Shadow of the Tomb Raider – DX12, “Highest” preset
  • The Division 2 – DX12, Ultra preset, VSync Off
  • Assassin’s Creed: Odyssey – Ultra High preset, VSync Off
  • Far Cry 6 – Ultra defaults, HD Textures enabled
  • F1 2021 – DX12, Very High defaults, Bahrain track, show FPS counter.
  • Metro: Exodus – DX12, Ultra defaults

Synthetic Benchmarks

Our first set of benchmarks hails from Underwriters Laboratories, which acquired Futuremark in 2014. Earlier in 2018, a rebrand occurred, and since that time, Futuremark is now UL. The benchmarks have not changed, just the name. We chose to stick with 3DMark Fire Strike (Extreme) and 3DMark Time Spy as these tests give users a good idea of performance on modern titles.

3DMark Fire Strike (Extreme) is a DX11-based test that runs at 1080p resolution. UL says the graphics are rendered with detail and complexity far beyond other DX11 benchmarks and games. 3DMark Time Spy is a DX12 benchmark designed for Windows 10 PCs. It supports new API features such as asynchronous compute, explicit multi-adapter, and multi-threading and runs at 2560 x 1440. 3DMark Port Royal is the first Ray Tracing benchmark designed for Windows PCs and graphics cards with Microsoft DirectX Raytracing capabilities.

As we can see from the charts, the RTX 4080 Suprim X slots in above the RTX 3090 Ti and AMD’s 6900 XT and below the fastest graphics card out, the RTX 4090. The RTX 4080 is 27% slower than the  RTX 4090 and around 19% faster than the RTX 3090 Ti across these synthetic benchmarks.

Gaming Benchmarks

We have updated our testing suite for gaming benchmarks to bring more modern titles into the mix. Gone are Battlefield V, F1 2018, Far Cry 5, AOTS:e, and World of Tanks, which have been replaced with Metro Exodus, Assassin’s Creed: Odyssey, F1 2021, and Far Cry 6. We kept The Division 2 and Shadow of the Tomb Raider. The games should provide a good view of the card’s overall performance with many DX12-based. One thing worth noting with these high-end cards, they aren’t made for 1080p gaming, so the gaps between cards tend to get minimized.

1920 x 1080 (1080p) Results

In the 1080p tests, it’s tough to see a significant separation between such fast cards as you tend to be CPU limited. That said, the RTX 4080 again finds itself between the RTX 4090 and 3090 Ti in most titles.

2560 x 1440 and 4K UHD Results

Below are the higher resolution results starting with 2560 x 1440 and the gaining-in-popularity 3840 x 2160 (4K UHD) resolution.

2560x1440
2560×1440
3840x2160
3840×2160

At more GPU-board resolutions, the RTX 4080 finds its footing between the RTX 4090 and RTX 3090 Ti again. It’s much closer to the 4090 than the 3090Ti as the load shifts to the graphics card. At 4K UHD, the RTX 4090 pulls away from the 4080, but the latter can run this high resolution and averages over 120 FPS.

Ray Tracing and DLSS Testing

Below, we tested Shadow of the Tomb Raider with ray tracing at 1440p and 4K. Even with the frame-reducing ray tracing enabled (without DLSS), the RTX 4090 easily handled this title.

Folding @ Home Performance

For all of the dedicated folders for Team 32 and others in the cause, we gathered a small data set of the card folding at stock settings. During our nine hours of productivity using the card’s overclocked settings, it averaged 2,893 Mhz on the core and 266W. Peak power consumption for the card was around 320W. It ran incredibly quiet throughout. Below is a screenshot of GPUz and the folding stats from my stats page. I expect this to be somewhere around 17-18M give or take a million or so (WU, OS, etc) at stock speeds.

Folding @ Home
Folding @ Home

Overclocking the MSI RTX 4080 Suprim X

For overclocking, we used the latest version of MSI Afterburner without issue. The OC Scanner also did the job of finding a stable overclock. Ultimately, we settled on +104 Mhz for the core and +400 for the memory. While the scores didn’t go up much (a couple of percent) on these mid-resolution benchmarks, there was still some improvement to be had. This card also allows for a massive 25% power limit increase. With these settings, the card peaked at 2,910 MHz and ran around that speed consistently through most of our benchmarks. Even when overclocked, the card peaked at around 320W.

GPUz - 2,910 MHz core when overclocked
GPUz – 2,910 MHz core when overclocked

Temperatures and Power Use

We test power consumption by running through the game benchmarks of Shadow of the Tomb Raider and F1 2021 at stock speeds and while overclocked. We monitor temperatures throughout this testing, with the peak temperature listed in the data below. The benchmarks are extended (time) to allow the card to settle to simulate actual gaming conditions more accurately.

Temperatures on the liquid-cooled MSI RTX 4090 Suprim Liquid X reached 64°C in F1 2021 and 63°C SOTR at stock settings. After overclocking the card, temperatures increased slightly in each game, peaking at 66°C in F1 2021 and SOTR. Longer gaming sessions will increase the temperatures, but this gives you a general idea of the capabilities. For cooling up to 450W, the temperatures are well in order and much better than any air-cooled card. Keeping the temperatures in this range allows the card to use all the boost bins and isn’t limited by temperatures.

Temperatures
Temperatures

Power use for the RTX 4080 Suprim X peaked at 496 W (total system power) at stock and reached 510 W while overclocked (both in F1 again). At minimum, a high-quality 750 W power supply will be adequate for your needs, but MSI recommends an 850 W model.

Power Use (System)
Power Use (System)

Conclusion

Nvidia’s RTX 4080 is an incredibly fast card, beating out last generation RTX 3090 Ti handily, and uses significantly less power than the flagship RTX 4090. The RTX 4080 gives you every improvement the ADA Lovelace architecture offers. MSRP on the RTX 4080 FE is $1199, while the Suprim X we have in hand will run you $1349, or almost the cost of a 4090 FE. Although the price of these cards remains high, I’m not heartbroken about paying more for the upgraded power delivery and improved cooling it offers.

Performance-wise it’s a 4K/120Hz capable video card across most titles and will support 2560×1440 165Hz+ without flinching. Power use peaked at the listed TBP of 320W, so everything was in line. It’s also a lot easier to cool 320W compared to 450W, and your case internals will thank you.

There’s a lot of competition in this space, with each board partner coming out with high-end versions (Gigabyte Aorus and Asus Strix). Performance among these combatants will be close and any RTX 4080 is a 1440p/165Hz card or even 4K/120 in a lot of titles so it’s FAST. It comes down to price, appearance, and noise levels. Without a doubt, we love the look of the new MSI cards. The silver gives off a premium vibe, and the neutral coloring allows it to fit in with any build.

Overclocker Approval Rating
Click to find out what this means.

– Joe Shields (Earthdog)

About Joe Shields 326 Articles
Joe started writing around 2010 for Overclockers.com covering the latest news and reviews that include video cards, motherboards, storage and processors. In 2018, he went ‘pro’ writing for Anandtech.com covering news and motherboards. Eventually, he landed at Tom’s Hardware where he wrote news, covered graphic card reviews, and currently writes motherboard reviews. If you can’t find him benchmarking and gathering data, Joe can be found working on his website (Overclockers.com), supporting his two kids in athletics, hanging out with his wife catching up on Game of Thrones, watching sports (Go Browns/Guardians/Cavs/Buckeyes!), or playing PUBG on PC.

Loading new replies...

Avatar of ratbuddy
ratbuddy

Member

8,974 messages 56 likes

Getting a 404, sure ya clicked publish? :)

Reply 1 Like

Avatar of Evilsizer
Evilsizer

Senior Forum Spammer

19,659 messages 217 likes

some of the charts don't even show the 4080 listed, any reason the 3070 is missing?

Reply 1 Like

Avatar of EarthDog
EarthDog

Gulper Nozzle Co-Owner

76,614 messages 3,325 likes

Yikes. That's my fault. Pushed to get it ready, no time for a second edit.... sick. Bleh. Charts updated. This whole launch was just a debacle for me, lol. Sorry about that, everyone.

As far as the 3070... I don't have that dataset yet. I'll have one and will add it to a future GPU review.

Reply Like

Avatar of ozzlo
ozzlo

Senior whose title will pier

2,478 messages 2 likes

you get everything the Ada Lovelace architecture brings except for the flagship-class price.

This statement pains me, Any single component that costs more than an entire decent mid-range PC is still "flagship-class price" IMO.

Did everyone so easily forget they have Jacked up the price of 80 series cards $500 in just 3 years?

Reply 1 Like

Avatar of EarthDog
EarthDog

Gulper Nozzle Co-Owner

76,614 messages 3,325 likes

Without a doubt, I agree with your sentiment. These cards are just too damn expensive. But sadly, flagship pricing, at least for NV and the performance it offers, is another 20% higher. Big oof for sure.

Reply Like

Avatar of Janus67
Janus67

Benching Team Leader

17,246 messages 585 likes

Seems like something that should cost about $300 less to even be a 'decent' value proposition. Otherwise they are just upselling the 4090

Reply 1 Like

Avatar of EarthDog
EarthDog

Gulper Nozzle Co-Owner

76,614 messages 3,325 likes

I think we'll see a reset of NV's midrange and lower cards (not these, lol) once AMD's cards release. Rumored pricing on their new gen is a lot lower, but, so is performance. Thing is, not sure we'll see it as consumers as the mid-range and lower cards from NV will come out after AMD releases a couple of cards.

...but such as product life cycle, yes? New cards come out, the market adjusts as needed.

Reply Like

Avatar of Janus67
Janus67

Benching Team Leader

17,246 messages 585 likes

True, and at this point we don't have crypto [at least] to be holding the prices up. I'm wondering how much artificial scarcity there will be for the new cards either by manufacturing or just scalpers picking them up to resell on StockX/ebay/etc. Ultimately, from my 3080 the only good option it looks like would be a 4090 as an actual upgrade, but again looking at about a 1:1 $/performance ratio, where I'd be spending twice as much as my 3080 cost me for about double the performance on the 4090.

Ultimately, even at 4K my 3080 does well, granted I don't run raytracing (truthfully I can't tell much of a difference in most cases, and most of the games I play are either faster-paced or don't offer it).

Reply 1 Like

click to expand...
Avatar of BugFreak
BugFreak
2,421 messages 663 likes

While I give Nvidia credit for pricing their cards at the levels people were paying scalpers anyways the pricing for the 4x series is totally off the rails. Good thing my 3080 has many years until it gets "too old" for what I play.

Reply 1 Like