Three weeks ago Nvidia launched their RTX 2060 SUPER and RTX 2070 SUPER and I had the chance to check out both on those in their Founders Edition versions. At the same time, they also announced the RTX 2080 SUPER as well but performance details weren’t released. So while we have been checking out the Founders Edition cards, the new AMD cards, and later the new aftermarket SUPER cards I’ve been waiting to check out the RTX 2080 SUPER. Well, today's the day. I’ve had a little while to put the RTX 2080 SUPER Founders Edition through our test suite and now we can check out what is different with the card and how it performs compared to the old RTX 2080 as well as the 2080 Ti, Radeon VII, and Vega 64. The other SUPER cards had significant performance jumps, will the 2080 SUPER do the same? Let's check it out.
Product Name: Nvidia RTX 2080 SUPER Founders Edition
Review Sample Provided by: Nvidia
Written by: Wes Compton
Pictures by: Wes Compton
Amazon Affiliate Link: HERE
So before diving in I did want to run through the specifications to see what has changed for the RTX 2080 SUPER. For comparison, I included the RTX 2070 SUPER, RTX 2080, and the RTX 2080 Ti. The RTX 2080 and RTX 2080 Ti are the two I’m most interested in comparing the 2080 SUPER with to see what has changed and how it compares with the gaming flagship card. So the first thing I wanted to check was which GPU the 2080 SUPER went with. Nvidia stuck with the TU104 that the RTX 2080 had so right away we know this isn’t jumping up into the range of the RTX 2080 Ti. That’s okay though, there is a big gap between the RTX 2080 and the RTX 2080 Ti. So what Nvidia did change was the SM count by adding two. This bumped the CUDA core count up from 2944 to 3072 and increased things like the RT Core Count and Tensor core count as well. The other big change was in clock speeds. Base/Core clock went from 1515 MHz on the 2080 to 1650 MHz on the 2080 SUPER. Then for boost clock, they went from 1710 MHz to 1815 MHz. That isn’t a huge jump, in fact, the clock speed of the RTX 2080 SUPER is still lower than you can find aftermarket RTX 2080’s clocked at. The combination of the two changes does translate to the INT32 and FP32 performance increasing from 10 TFLOPS each to 11.2 TFLOPS each. Nvidia didn’t push it in their comparison specifications but I also noticed that the memory clock speeds increased on the 2080 SUPER. The original RTX 2080 ran at 14 Gbps and the 2080 SUPER’s memory is running at 15.5 Gbps. This accounts for the increased memory bandwidth while the memory capacity is still 8GB and it still has the 256-bit memory interface.
Now just to confirm everything I did fun GPUz before getting into my testing. I’ve included that below. I just wanted to confirm that our card has the same clock speeds as in the specifications, which it did. You can also see the BIOS revision I tested on, the driver which was the pre-launch media driver for this launch, and that our card had Samsung memory should that change later (which it seems to always).
Packaging
No changes at all with the packaging if you have seen the Founders Edition box for the original RTX launch. It has a grey and black lined design in the background in a box with a slide off top. The Nvidia wrap around is on the left side where it is on the right for the aftermarket boxes. The wrap around has the GeForce RTX branding and then at the bottom the bright green background with the model name including the SUPER logo. Around on the back, the box doesn’t have anything going on at all other than the geforce.com website and a few things in fine print at the bottom.
The top of the box slides up and off and has foam inside of it to lock the card into place. The bottom portion has foam cut to the shape of the card as well. There is also a slot which comes with a tray in it for the documentation and accessories. That includes a support guide and a quick start guide. There is also a DisplayPort to DVI adapter.
Card layout and Photos
So Nvidia didn’t make any significant changes to the Founders Edition design for the RTX 2080 SUPER and unlike the RTX 2070 SUPER, there isn’t a larger cooler design to go up to than this. This is the same dual fan cooler design that the RTX 2080 and RTX 2080 Ti both use and that I have loved. It has a thick cast metal shroud design which has been a key feature in Nvidia’s Founders Edition and Reference cards for a while now. Only when the RTX cards launched they switched to this dual axial fan design. It has kind of a cassette tap look which the 80’s kid in me loves. The shroud has silver panels on each end of the card that run into the two fans and then in the center everything is chromed. The older RTX 2080 was similar but the center was black, the chrome look is new for all three of the SUPER cards and it also includes the model name in the middle like before but with the letters filled in with black (which were chrome before) and then bright green for the SUPER logo. I don’t normally like chrome that much but as I have said in the other two SUPER launches I actually do dig this look, even though I do miss the black. Both fans have a machined lip around them because they stick out just slightly and I don’t think you can say any other card looks like the Founders Edition design.
The fans have 13 blades on them with a heavy angle on them, between the angle and density you can’t see between the blades at all. In the center, they have a glued on cap which is machined. That cap is one of the carryovers from the older blower designs but it still looks good with this design.
Up on the top edge, the chrome finish from the front doesn’t carry on. The center section has the same black finish that the older RTX cards had and it has the GeForce RTX logo as well. The logo looks chrome but it is actually clear plastic with lighting behind it. Nvidia only allows this to light up in their signature green color and I’ve said it more times than I can count, but this really should be RGB to allow people to match the card to their build. I complain a lot about backlit branding but I don’t mind the GeForce RTX branding, I just wouldn’t want to see the Nvidia logo here so no complaints here on that this time around. Also up on the edge, the RTX 2080 SUPER still has the NVLINK connection like the RTX 2080 and RTX 2080 Ti had and the RTX 2070 SUPER added. Being the Founders Edition card, Nvidia also makes sure to match the metal casings design with a slide on cover that is metal. In fact, if you didn’t know to look you might even miss it when it's installed. Also up on top are the power connections and with a TDP of 250 Watts, I was surprised that they stuck with the 8-pin and 6-pin combination that the RTX 2080 has when the 2080 Ti had to have two 8-pin’s.
With the Axial fan design, the Founders Edition cooler is designed for the air to blow down and then through the heatsink and it has a vertical fin configuration. This is complemented with open exposure to some of the fin stack at the top and bottom edges. There is a wide V-shaped accent which blocks some of them but overall you can expect the warmed air to be heading out the top and bottom of the card. To be sure Nvidia also wrapped the metal fan shroud around the end and connected it with the backplate. The end of the card does have the two mounting holes for server and OEM use if you want to keep the card locked in place and prevent any PCIe slot damage or GPU sag.
I think the backplate is my favorite part of the Founders Edition cards, especially after taking a look at a few aftermarket designs just recently. Nvidia uses the same thick cast aluminum here which gives the card some weight but most importantly makes this a solid design, there isn’t going to be any sag to the PCB later with this design which should help with long term durability. The backplate is also the second most visible part of the card. Now the all silver design isn’t going to be for everyone and it is even more in your face on the backplate without the chrome and fans to split things up. But the backplate does have a line design molded into it that runs the length of the card, moving up and down to make room for the logo in the center. The RTX 2080 SUPER logo is filled in with chrome except for the SUPER logo which again has that bright green. The back also has two stickers with the serial number and barcode as well as a sticker with all of the required logos and certifications documented all together.
For the rear I? on the PCI bracket the RTX 2080 SUPER is is sporting the same configuration as the RTX 2080 and RTX 2080 Ti. You get one HDMI port and then three DisplayPorts. Nvidia also included the DisplayPort to DVI adapter for those still using DVI. Then in the bottom right corner, there is a small Type-C connection which can work as a USB port but is there for VirtualLink which is an alternative to USB-C. It is designed to feed power, display, and data over the one connection to your VR headset rather than having to hook all three of those up to a box to adapt to your headset. My favorite thing back here though is the all-black bracket. It’s a small touch, but it looks amazing when installed in a case with a black motherboard I/O shield as well and the Nvidia logo works well here.
So here is a look at the RTX 2080 and the RTX 2080 SUPER next to each other and like I said there wasn’t much changed. The chrome in between the fans and the SUPER added to the logo are the only two things changed in the design and I’m glad. This is a great looking card. Short of RGB for the logo nothing needs to be touched for it to look good. Though come to think of it a design that used black where the silver is would look good!
When testing I also grabbed a picture of the backlit logo in action. The Nvidia green looks good no doubt. But it would look good in a few other colors as well!
Test Rig and Procedures
Test Rig
Motherboard: Asus Crosshair VIII HERO WiFi
Cooling: Noctua NH-U12S for cooling
Noctua NT-H1 Thermal Paste
Memory: G.Skill Trident Z Royal 3600MHz 16-16-16-36
Storage: Corsair MP600 2TB
Cooling - Noctua NH-U12S
Power Supply - Corsair AX1200w
Case - Primochill Wetbench
Synthetic Benchmarks
As always I like to start off my testing with a few synthetic benchmarks. 3DMark especially is one of my favorites because it is very optimized in both Nvidia and AMD drivers. It's nice to not have to worry about it being favored too much either way and the repeatability of the results makes it a nice chance to compare from card to card. In this case, I was curious how the RTX 2080 SUPER compares with the original RTX 2080 and the RTX 2080 Ti on the Nvidia side. Then with the Radeon VII, Vega 64, and new RX 5700 XT on the AMD side. To test that I started with 3DMark Fire Strike with all three of its detail settings. This is a DX11 test and the Nvidia cards haven’t been doing as well here lately but the 2080 SUPER FE did still perform well here. In the performance and extreme details, for example, it came in ahead of the Radeon VII when the original RTX 2080 didn’t. Getting up to the Ultra detail the Radeon VII was helped by its high VRAM but still, it wasn’t much of a gap.
Now in Time Spy I took a look at DX12 performance and the Turing based cards have really taken a liking to DX12 and it shows here. The top 7 cards in the normal test were RTX cards with the 2080 SUPER getting almost 1000 points on the RTX 2080 but it was still 2228 points behind the 2080 Ti. At the Extreme detail setting the Radeon VII did come up a little but the top is still the 2080 Ti, 2080 SUPER, then 2080 altogether.
I also ran the Port Royal test which takes a look at ray tracing performance. Sadly AMD hasn’t added support for DirectX Ray Tracing yet but of course, Nvidia has with their RTX cards leading ray tracing into the market. The 2080 SUPER improved on the 5933 of the original 2080 to 6802 for a really nice improvement. Of course, like on the other tests, the 2080 Ti is still at the top by a large margin.
I also tested in Superposition which is a Unigine Engine based test. I tested at medium and extreme detail levels at 1080p. The order of the results isn’t a surprise at all with the 2080 Ti, 2080 SUPER, then 2080 at the top.
VR Benchmarks
As for Virtual Reality, I love it but it is more demanding than traditional gaming. This is partially because of the resolutions needed to render for two eyes and because they render more than what is immediately visible. But also because of post effects to get the proper “fisheye” effect for it to look proper in your eyes with the HMD. You also have to have much higher expectations for frame rates in VR, skipping frames or lower FPS can actually cause motion sickness in VR. Because of that, I ran a few tests.
My first test was again in Superposition. This time I tested the VR Maximum and VR Future tests using the Vive resolution. The RTX 2080 SUPER FE chewed through the VR Maximum test with a 151.48 FPS score which was 6 FPS higher than the RTX 2080. The gap at the VR Future detail wasn’t as big at 4 FPS and that future test is where the 2080 Ti really took off. None of the AMD cards were anywhere near the top with the 570 XT the highest at VR Maximum with 122.73 FPS and the Radeon VII the highest AMD card at VR Future with 44.08 FPS. Both were running with the RTX 2060 SUPER.
My second round of tests was in VRMark using the Cyan Room and Blue Room benchmarks. I didn’t include the Orange Room test because it isn’t demanding enough for any of the newest video cards. The Cyan Room test focuses on high detail current day VR performance and Blue Room is a future-looking test which is why there is such a big gap in FPS between the two on all cards. The RTX 2080 Ti is the only card close to the target FPS needed in the Blue room. The 2080 SUPER did gain 6 FPS in that test over the original RTX 2080, in the Cyan Room test it gained 9 FPS. Once again none of the AMD cards were coming close at all, with the Radeon VII being the closest at 30 FPS less than the 2080 SUPER.
In-Game benchmarks
Now we finally get into the in game performance and that is the main reason people pick up a new video card. To test things out I ran through our new benchmark suite that tests 8 games at three different resolutions (1080p, 1440p, and 4k). Most of the games tested have been run at the highest detail setting and a mid-range detail setting to get a look at how turning things up hurts performance and to give an idea of if turning detail down from max will be beneficial for frame rates. In total, each video card is tested 42 times and that makes for a huge mess of results when you put them all together. To help with that I like to start off with these overall playability graphs that take all of the results and give an easier to read the result. I have one for each of the three resolutions and each is broken up into four FPS ranges. Under 30 FPS is considered unplayable, over 30 is playable but not ideal, over 60 is the sweet spot, and then over 120 FPS is for high refresh rate monitors.
So how did the RTX 2080 SUPER Founders Edition perform? Well, I would consider the RTX 2070 SUPER to be a solid 1440p card and at 1080p a high refresh rate card. The bump in performance over the RTX 2080 puts the RTX 2080 SUPER up over that with this being a great 1440p high refresh rate card. As well as 1080p. At 1080p 8 of the results were up over 120 FPS and the other 6 were between 60 and 119 FPS. 1440p was similar with 4 in the 120+ FPS range and 9 between 60-119. Now keep in mind I’m not testing the less demanding esports titles, these are new high-end games running at sometimes their highest detail settings and the RTX 2080 SUPER just ate them all up at both resolutions. 4k did show a drop off, but not as significant as other cards and everything at 4k is still playable with a majority being smooth. The RTX 2080 Ti is still THE 4k card, but the RTX 2080 SUPER is edging close to that as well. You might have to turn a game or two down to medium settings but you aren’t going to have touch with gaming at 4k as well. But this is the card for high refresh rate fans. This will push those high refresh rate 1440p monitors.
Digging into the detailed graphs I wanted to see if the trend of the top cards being the 2080 Ti then the 2080 SUPER continued or if any of the AMD cards have disrupted that at all but none of them have. The Radeon VII did come close in some of the tests but at the lower resolutions, the RTX 2080 still pulled ahead. This range of card right now is basically uncontested, though AMD is slowly getting closer. Beyond that, there weren’t really any surprises in the results at all for once!
Compute Benchmarks
Now some people don’t need a video card for gaming, they need the processing power for rendering or 2D/3D production or in some cases people who game also do work on the side. So it is also important to check out the compute performance on all of the video cards that come in. That includes doing a few different tests. My first test was a simple GPU Compute benchmark using Passmark’s Performance Test 9 and the RTX 2080 SUPER FE did improve over the RTX 2080 but it wasn’t enough to edge past the RX 5700 XT in this test. The 100 point increase still left it behind the XT by 264 points. The 2080 Ti is still the top dog here with its huge 1096 point lead over the RX 5700 XT.
I love blender as a test because it is an extremely popular 3D modeling and production program making this a very real world benchmark. This is time based so the lower the score the better. The 2080 Ti is of course up top. I wasn’t all that impressed that the 2080 SUPER only took off a little over 2 seconds with the additional cores and GPU and memory clock speeds. Sadly the RX 5700 cards aren’t here due to a driver issue that would cause the benchmark program to crash before even getting to the test.
Next, we have Basemark with a look at OpenGL and DirectX 12 compute performance. As always the 2080 Ti is at the top, way ahead of everything else. The 2080 SUPER, on the other hand, gained 623 points over the RTX 2080 in the DX12 test and 530 in the OpenGL test. A nice jump, but still only about a ¼ of the way to the RTX 2080 Ti.
The last compute test was Geekbench and this was run with the newly updated 4.4 version along with retests on all of the other cards. 4.4 fixed a bug that affected Intel and AMD GPUs. Of course, that dropped the AMD cards down the chart. In this test the RTX 2080 SUPER increased even more over the RTX 2080, getting between 1/3 and ½ of the way into the gap between the 2080 Ti and the 2080.
Cooling, Noise, and Power
Once you decide which GPU you want to go with, you have to decide between all of the different models available. Clock speeds play a part as do pricing and aesthetics. But there are a couple of other areas that you should consider that are card specific. Power is one but it is in some ways tied with the overclock or at least how the overclock is done. There is also cooling performance and noise performance which both have to do with the cooler design.
My first test was to take a look at power usage and for that, I hooked the test bench up to our Kill-A-Watt and recorded total power draw at the plug. I then put the system under load using 3DMark Fire Strike, specifically the combined test which tests the GPU and physics at once which gets a similar load to in game. The RTX 2080 SUPER Founders Edition along with the 3900X pulled 399 watts which for comparison the RTX 2080 did 394. This was more than the RX 5700 XT but below the crazy numbers that Vega 64 did. Radeon VI and the RTX 2080 Ti were right together as well and above the 2080 SUPER. The second test I wanted to focus on the GPU so I used AIDA64’s stress test to put the GPU under load but didn’t load any other part of the system. There is still some idle power draw but this number is obviously lower with the RTX 2080 SUPER Founders Edition pulling 287 watts here vs 279 for the RTX 2080 FE. I’m surprised that the 2080 SUPER still stayed as close as it did with the RTX 2080 given the increase in TGP from 215 watts to 250 watts. This put the 2080 SUPER FE up under the RX 5700 XT by a big margin as well.
Next, I wanted to check out the noise performance of the Founders Edition cooler. For this, I ran three tests using the decibel meter setup 18 inches away from the test system. I tested at 50% and 100% fan speeds to get a look at the overall potential for noise. Then I tested with the card under load for 30+ minutes, then measuring the sound level of wherever the fans landed to keep it at that temperature using the stock fan profile out of the box. I also documented the 100% fan speed RPMs because that sometimes will give a good look as to why a card is quieter or louder, or if it is noisy but doesn’t have a very high RPM it tells us that the design is noisier. So at 100% fan speed, the Founders Edition cooler falls right in the middle of the chart with the water cooled and aftermarket designs being quieter and the blower cards being louder. With no cooler change, the 2080 SUPER and regular 2080 were relatively close in noise as well as all of the other Nvidia dual-fan cards. 50% fan speeds weren’t too bad as well. It was the under load noise level that I was most curious about though and the 2080 SUPER was a lot higher than the original 2080 in that test. It would fall in the middle of the pack overall, but I was surprised to see such a big change from card to card, you have to wonder if the fan profile has been changed between the two or if it is just more heat from the slightly more cores and higher memory and core clocks.
The last test was to take a look at the cooling performance of the Founders Edition card. To do this I used AIDA64’s stress test to load the GPU and to document temperature tests. I tested by running that load for a half hour or longer if needed, waiting for the temperature to level off and no change. I tested twice, once with the stock fan profile and again with the fans turned up to 100%. The idea is to see out of box cooler performance and then see how much headroom is left. The delta between the two will show that. So the RTX 2080 SUPER Founders Edition reached 68 degrees with the stock fan profile were the original RTX 2080 reached 69. Given the higher clock speeds, I think that shows my theory from before that the fan profile is slightly changed. At 100% fan speed the 2080 SUPER dropped down to 51 degrees where the original 2080 dropped down to 47 degrees confirming that the 2080 SUPER does run warmer with the fan speed being the same with the same cooler design. The 17-degree delta between the stock and 100% results on the RTX 2080 SUPER Founders Edition also shows that this cooler still has more room in it for an overclock as well.
While doing thermal testing with the stock fan profile I also pulled out our thermal camera to take a look at the RTX 2080 SUPER Founders Edition to see how it was handling the additional heat of the 2080 SUPER. I was impressed to see that the areas behind the fans were running cool, in fact, the hot spots would be closer to the cool spots on a few of the cards I’ve tested recently. The thick metal shroud hadn’t soaked up too much heat as well on the fan side. The top down view showed that the heatsink area at the top where a majority of the heated air is coming out is doing its job. The entire middle of the card is warmer with ti much cooler down at the end. Then around on the backplate, there was a noticeable hot area down where the VRMs are. The backplate was also running a lot warmer than the front of the card but still not bad other than that one VRM area.
Overall and Final Verdict
With all of the testing done and having taken a look at the specifications and the Founders Edition design, I can finally sort through all of my thoughts. Now right off the bat, I’m going to say that I was really hoping to see the RTX 2080 SUPER did what the other SUPER cards did, but it didn’t. What I mean by that is the other SUPER cards caught up with the card slotted above it and it was clear even just from looking at the specifications of the RTX 2080 SUPER that Nvidia wasn’t doing that with this card. This isn’t the RTX 2080 Ti killer, it is just a refresh of the RTX 2080 to help compensate for the RTX 2070 SUPER jumping up closer to the RTX 2080. Now a bigger jump would have been nice but what we ended up with is still not a bad increase. In some tests, it helped bridge the gap between the RTX 2080 and the RTX 2080 Ti by about ¼ to 1/3. Frankly, there was such a big gap behind the 2080 Ti that it makes the 2080 SUPER look worse than it is. When I got into in game performance I was really impressed with the performance, this card is going to play anything you throw at it at any resolution. The 1080p and 1440p performance is setup for the high refresh rates and at 4k everything is going to be playable with a majority smooth at over 60 FPS.
As for the Founders Edition design itself, Nvidia didn’t really change much here at all. The addition of chrome in the middle section just like the other SUPER cards and the bright green SUPER logo are the only two visual updates. I would have liked to see the top logo lighting finally get the RGB upgrade, but it didn’t. Beyond that I still love this cooler design, it is one of the best looking card designs on the market and the thick cast shroud and backplate really set it apart when it comes to overall feel and construction. This is the only card design that I know I won’t have to worry about GPU sag later in life. The silver design is still going to be very polarizing and adding chrome to that is going to only increase that. I like it, but I think a black design might still be better. I also can’t forget that black PCI bracket that I love so much.
As for pricing, this refresh is actually slotted at the same price as the original RTX 2080 Founders Edition. $699 still isn’t cheap, but more performance for the same amount will never be turned down. Nvidia has also been spicing up their RTX cards with the addition of a game bundle as well. You get Wolfenstein: Youngblood and Control which are both upcoming top tier games that support ray tracing. This helps make sure everyone has at least some ray tracing supported games while we wait for more games to come out and I love that you get free games that aren’t even out yet, not an old game that most people who wanted it have already played. That $80 value helps take a little edge off the $699 price point. Obviously, unless you are running a high refresh rate monitor or gaming at 4k this is more card than you might need. But for those who do need it, this is a much better value than stepping up to the RTX 2080 Ti! It will be interesting to see how the 2080 SUPER aftermarket cards perform, but I wouldn’t discount the Founders Edition either. This is a solid card.
Live Pricing: HERE
