Nvidia Geforce RTX 2080 vs RTX 2070 Super
Getting a high-end graphics card as a gamer is tough. You want to get the most performance, but is it really worth paying the premium for the best cards? That is a question ever user has to answer for themselves.
Oftentimes the second or thrid ranked GPUs offer a lot better price-performance ratios, meaning you get more performance for the dollar or pay less for more performance than you would with the best performing graphics card.
In this test we take a closer look at the Nvidia Geforce RTX 2080 & the Nvidia Geforce RTX 2070 Super.
The price difference can be around 100$-200$. But is it really worth it to pay that premium for the RTX 2080?
How much better is the RTX 2080 really?
Test Results: Nvidia Geforce RTX 2080 vs RTX 2070 Super
Ranking First: Nvidia Geforce RTX 2080
- Great for 1440P /1080P gaming
- DLSS is an exciting feature with a ton of potential
- Ray tracing and deep learning
- New and improved design
- 650W power supply requirement
Best performance on the market
With the Geforce RTX 2080 Super in the test, Nvidia releases the third and for the time being last super model of the Turing series after the RTX 2070 Super and RTX 2060 Super.
The 2080 Super positions itself against the old RTX 2080, which is currently sold out due to the release of the RTX 2070 Super. In the comparison 2080 Super vs. RTX 2080 Ti the clearly more expensive 2080 Ti remains the fastest graphics card for gamers.
Simultaneously with the publication of this review, the release and sale of the RTX 2080 Super also took place on July 23, 2019. Interested parties could choose between the 2080 Super Founders Edition reference card and various 2080 Super custom designs from many manufacturers.
Price lower than RTX 2080 at launch
The official price of the RTX 2080 Super is 739 Dollar according to Nvidia. This makes it cheaper than the RTX 2080 Founders Edition, which cost 849 Dollar at the start in late summer 2018.
At the moment the old RTX 2080 is available from 640 Dollar, but it has a tough position against the almost equally fast RTX 2070 Super from 530 Dollar.
For a short time the RTX 2080 Super is also available in a game bundle with the raytracing titles Control and Wolfenstein: Youngblood.
Control was released in about a month’s time after the Nvidia release on August 27th, 2019, Youngblood was already available on July 26th, 2019, but the latest serial part of the Wolfenstein universe does not yet support raytracing – this will be added later by patch.
RTX 2080 Super vs. RTX 2080
Compared to the normal Geforce RTX 2080, the successor with super addition also relies on the TU104 graphics chip, but offers more shader and texture units as well as higher clock rates.
We have listed the most important differences of the RTX 2080 Super below:
- higher chip clock rates: 1.650 MHz (Boost 1.815 MHz) instead of 1.515 MHz (Boost 1.710 MHz)
- higher VRAM clock: effective 15.5 GHz instead of 14 GHz
- more shader and texture units: 3,072 and 192 instead of 2,944 and 184
- increased power consumption (TDP): 250 instead of 215 watts (Custom Designs) or 225 watts (Founders Edition)
The Geforce RTX 2080 Super has a higher raw performance than the normal RTX 2080, as our table below illustrates, which, together with the attracted chip and video memory clock rates, makes it a few percentage points faster in games.
As with all Super models, the RTX 2080 Super increases the number of RT and Tensor cores relevant for raytracing and DLSS edge smoothing compared to the old 2080 variant.
However, you should not hope for a significant difference in performance in the respective application scenario due to the very small increase.
|GPU||Clocking Rate||Shader||Memory||RT-/ Tensor cores||TMUs/ROPs||Memory Bandwidth|
|RTX 2060||1.680 MHz||1.920||6,0 GB||30/240||120/48||336 GB/s|
|RTX 2060 Super||1.650 MHz||2.176||8,0 GB||34/272||136/64||448 GB/s|
|RTX 2070||1.620 MHz||2.304||8,0 GB||36/288||144/64||448 GB/s|
|RTX 2070 Super||1.770 MHz||2.560||8,0 GB||40/320||160/64||448 GB/s|
|RTX 2080||1.710 MHz||2.944||8,0 GB||46/368||184/64||448 GB/s|
|RTX 2080 Super||1.815 MHz||3.072||8,0 GB||48/384||192/64||496 GB/s|
|RTX 2080 Ti||1.545 MHz||4.352||11,0 GB||68/544||272/88||616 GB/s|
As expected, the Geforce RTX 2080 Super places itself in second place in our graphics card ranking behind the 2080 Ti and is on average a good eight percent faster than the old RTX 2080 Founders Edition, which Nvidia delivered overclocked ex works in contrast to the new Super variant – custom designs of the RTX 2080 Super should be correspondingly a bit faster.
The Geforce RTX 2080 Ti clearly remains on the performance throne, but with the RTX 2080 Super it gets the strongest competitor so far.
In WQHD resolution the 2080 Ti has a lead of about ten percent, in UHD the overclocked 2080 Ti Founders Edition can then stand out more clearly by 17 percent from the 2080 Super.
In the comparison RTX 2070 super vs. RTX 2080 super the 2080 super can claim a comfortable projection of approximately 15 per cent for itself – however at a surcharge of nearly 40 per cent.
The bottom line is that the Geforce RTX 2080 Super is a very fast graphics card, which can smoothly display current games in UHD resolution with maximum details and very often break through 60 fps.
Single benchmark Full HD, WQHD and UHD
The Geforce RTX 2080 Super Founders Edition clocks with an average of around 1,900 MHz in our gaming benchmarks.
Therewith, it slightly outperforms its already overclocked predecessor as a reference model: We have determined an average clock rate of about 1,860 MHz for the RTX 2080 FE.
There are no surprises in the cooling design of the Geforce RTX 2080 Super Founders Edition.
The two reference cards (Super and Non-Super) have identical dimensions, are two slots high and use axial cooling with two 90 mm fans.
However, the fans of the RTX 2080 Super, like the two other Super models RTX 2070 Super and RTX 2060 Super, rotate at increased speeds compared to their predecessors.
In idle mode, the fan speed is 1,500 revolutions per minute, under load 2,020 revolutions per minute. Nvidia might want to counteract the somewhat higher TDP (Thermal Design Power) with that.
Thus, the RTX 2080 Super FE is a bit louder in comparison to the normal RTX 2080 FE – both in Windows mode and in games.
While it still works relatively unobtrusively in idle mode with 39.6 decibels, the fans turn up quite audibly in load scenarios with 42.6 decibels.
The chip is also sufficiently well cooled at all times. We measured a temperature of 75 degrees in games with our open test system, and 32 degrees in idle.
The power consumption increases only slightly, despite the TDP of 250 watts. We measured 318 watts for the overclocked predecessor. The new RTX 2080 Super Founders Edition needs 323 watts (entire test system without TFT).
The Geforce RTX 2080 Super in the test completes Nvidia’s super trio.
Once again, the focus is on more performance at a lower price, because apart from the slightly increased tech specs, only the highest effective GDDR6 clock rate to date (15.5 GHz) stands out in the 2080 Super.
Verdict: Best performance on the market
As expected, the RTX 2080 Founders Edition performed somewhat faster than the RTX 2080, but still maintains a sufficiently high gap to the Geforce RTX 2080 Ti, which only really comes into its own in UHD resolution and remains by far the fastest (and most expensive) gaming graphics card.
Compared to the RTX 2070 super the RTX 2080 super can maintain a projection of approximately 15 per cent for itself, is at the same time with a noncommittal price recommendation of 739 Dollar in addition, nearly 40 per cent more expensive – a clear surcharge per fps.
Altogether Nvidia increases the performance of the old 2080 with the new RTX 2080 Super and lowers the price at the same time, however the added value for buyers seems smaller with the 2080 Super than with 2070 Super and 2060 Super –
AMD has nothing to oppose the 2080 Super with Radeon RX 5700 and RX 5700 XT however (differently than with the more favorable super variants).
Ranking Second: Nvidia Geforce RTX 2070 Super
- Excellent gaming performance on par with $700 RTX 2080
- A GeForce RTX 2080-sized cooler
- high operating clock rates, slower-spinning fans, and lower operating temperatures
- Superb Founders Edition design
- Bigger and more power hungry than RTX 2070
Best price-performance ratio
Nvidia refreshed the current RTX 2000 generation with the Geforce models RTX 2060 Super and RTX 2070 Super as well as with the RTX 2080 Super coming at the end of July 2019 – the Turing refresh brought more performance for the same price in the Geforce RTX 2070 Super in the test.
The Geforce RTX 2070 Super offers a performance on GTX 1080 Ti level and competes with the RTX 2080.
The RTX 2080 is thus replaced by the 2070 Super, which will be available as RTX 2070 Super as Founders Edition from July 9th, 2019, custom designs from other manufacturers will follow shortly.
Against the RX 5700 XT
The time of publication is not chosen at random. On July 7th 2019 AMD returned with the Radeon RX 5700 XT and RX 5700 navigation graphics cards in the upper mid-range graphics card segment and aims at the RTX 2070 (RX 5700 XT) and RTX 2060 (RX 5700) in terms of performance.
The RTX 2070 Super appeared at the same time as the RTX 2060 Super, to which we dedicated a separate review, but you can also find the 2060 Super’s benchmark results here.
RTX 2070 Super vs. RTX 2070
Compared to the conventional Geforce RTX 2070, the successor relies on the TU104 graphics chip known from the RTX 2080 and offers higher clock rates.
The most important differences of the RTX 2070 Super in comparison to the RTX 2070:
- different graphic chip: TU104 instead of TU106
- higher clock rates: 1,605 MHz (1,770 MHz) instead of 1,410 MHz (1,620 MHz)
- more shader and texture units: 2,560 and 160 instead of 2,304 and 144
- Increased power consumption (TDP): 215 instead of 175 watts (Customs) or 185 watts (Reference)
RTX 2070 Super vs. RTX 2080
Due to the higher raw power of the Geforce RTX 2070 Super, the new model is not only faster than the previous RTX 2070, but the Geforce RTX 2080 also becomes dangerous.
So dangerous that Nvidia discontinued the 2080 and brought the RTX 2080 Super to market on July 23, 2019.
The official price of the RTX 2070 Super is 529 Dollar. Thus Nvidia offers a bit more performance for the same price (RRP) about eight months after the RTX 2070 was released in mid-October 2018.
The RTX 2070 Super initially wasn’t sold as a Founders Edition, which this time will be neither more expensive nor faster than custom designs from various manufacturers, which were also expected to be available in July 2019.
On average, the Geforce RTX 2070 Super performs on the level of the Geforce GTX 1080 Ti and is thus only slightly slower than the Geforce RTX 2080, which Nvidia is phasing out accordingly.
According to our measurement results, the Super variant of the RTX 2070 is about 12 percent faster than the normal variant, but the custom design from KFA2 used for the test is not one of the fastest, but one of the cheapest models of the RTX 2070 to date.
The Geforce RTX 2070 Super also has more RT and Tensor cores for raytracing and Deep Learning Super Sampling (DLSS) calculations than its predecessor, but does not quite match the number of units of the RTX 2080.
The performance in corresponding games should be slightly higher, but you shouldn’t expect too big differences.
Single benchmarks Full HD, WQHD and UHD
In our benchmark course, the Geforce RTX 2070 Super Founders Edition clocks with an average of 1,860 MHz. Together with the slightly higher raw power, the RTX 2070 Super is thus faster than the KFA2 RTX 2070 EX selected by us.
In most cases the RTX 2070 Super can also hold its own against the Geforce GTX 1080 Ti Founders Edition, which is however somewhat held back by the radial cooling system.
However, the super newcomer doesn’t quite reach the performance of the overclocked RTX 2080 FE.
Volume, temperature and power consumption
The Founders Edition of the RTX 2070 Super has the same axial cooling design with two 90 mm fans as its reference predecessor.
The higher performance is accompanied by a 30 (Founders Edition) or 40 Watt (Custom Designs) increase in power consumption to a total of 215 Watt.
In order to keep the super variant relatively cool, Nvidia has tightened the fan speeds. The fans now rotate faster both in idle and under load.
In Windows mode we still measure an inconspicuous 39.6 decibels at 1,500 rpm, under load the RTX 2070 Super turns up audibly, but not disturbingly loud at almost 1,900 rpm and 42.1 decibels – silent lovers, however, would be better off with a quieter custom design or configure the reference card according to their wishes.
The temperatures turn out low with 33 degrees in desktop mode and still harmless under gaming load with 74 degrees, even if a bit hotter than previous models with the new reference cooling.
We don’t find any surprises in the power consumption.
With a performance on par with the GTX 1080 Ti FE, the RTX 2070 Super FE has an equally high power consumption.
With the Geforce RTX 2070 Super in the test, Nvidia sends a successor into the race almost nine months after the release of the Geforce RTX 2070, which offers more performance for the same price (RRP).
Verdict: Best price-performance ratio
A welcome development, because the RTX 2000 series has attracted a lot of criticism, especially in the price-performance ratio, and the Super models improve on this.
The release of the super models was also a planned move to put pressure on AMD’s upcoming Navi-10 graphics cards Radeon RX 5700 XT and RX 5700 and to take the wind out of their sails in advance.
Competition is good for business and in the end it’s the buyers who benefit – because AMD lowered the announced prices for the Super models depending on the actual performance of the RX 5700 (XT) compared to the RTX 2060 and 2070 Super.
Verdict: Nvidia Geforce RTX 2080 vs RTX 2070 Super
All in all the question of RTX 2080 vs RTX 2070 Super is not that hard to answer. Yes the performance of the RTX 2080 is slightly ahead of the 2070 Super.
But why would you pay a premium of over 100$ for a performance improvement of less than 4%.
Even then the RTX 2070 Super is the newer card with great performance. For 99% of users the 2070 Super is the right choice.
For the other 1%: They know why they take the RTX 2080.
If you want tochek out the price of the RTX 2070 Super, click here:
Finding the perfect Graphics Card
Computer games are only really fun when the graphics are right. Apart from the fact that some games won’t start at all without the right graphics card, no one likes to have to recognize their opponents through the pixel mud.
So that you can play all big (and small) titles with top stats, we give you tips about the best graphics card. This must not only fit your wallet, but also your requirements.
If you prefer to play LoL, we can promise you that you will need less power than in the latest RPGs.
Why you need a graphics card
The graphics card is still one of the most important components in a gaming PC.
While the graphics units of the CPUs are sufficient for most office applications and even for occasional image and video editing, this is only the case with very few games.
Because these tiny chips simply don’t offer the performance you need for gaming.
OnBoard cards are also seldom sufficient – and even if you get a title to start with, you usually have to make heavy sacrifices in terms of FPS and the general game feeling.
So if you want to play an AAA title every now and then, if you love to be in a very real virtual world or if you just want a smooth gaming experience, then it’s time for a decent graphics card.
You can choose from a wide variety of models and from 50$ to over 1000$ every budget is covered. Now it’s time to find out what the best graphics card is for you.
Gaming graphics cards: 2 top providers
Of course there are countless manufacturers of graphics cards on the market.
However, two stand out and dominate the scene: NVIDIA with its GeForce models and AMD with its Radeon GPUs. Both manufacturers have some fundamental differences that need to be weighed up.
Usually the Radeon graphics cards bring more computing cores. This makes them suitable for gaming as well as for lengthy calculations, such as those required for economic simulations.
They are also ahead of the game when it comes to displaying 3D models.
However, NVIDIA has optimized the built-in computing cores for gaming processes.
This is also the reason why less power is usually needed. If you want to use your PC mainly for gaming, the GeForce cards often have an advantage.
This advantage becomes even greater when the games run on DirectX 11. In contrast, DirectX 12 based games are optimized to take full advantage of AMD’s higher core count.
Basically, the clock frequency is first of all the basic clock of a graphics card.
That means the frequency with which the GPU works minimally. If a higher value is stored here, you can assume that the card will perform better in the games.
NVIDIA usually equips its cards with higher clock frequencies. However, depending on the price range you are in, you can expect more (or less) performance.
By the way: Often a turbo or boost clock is specified in addition to the base clock.
The gaming graphics card can only reach this if it was previously well cooled.
Thus, the original performance can be exceeded even for a limited time. However, if you play for a very long time and the chip runs slowly hot, the performance is reset to the original clock.
Graphics memory is critical to keep your games running at peak performance over time.
Most modern graphics cards still use GDDR5 memory. Only a few, very expensive models use GDDR5X & GDDR6 technology, which offers some, but in everyday use rather insignificant advantages.
If we look at our two top suppliers, it looks like this: The cards from AMD often have a HBM2 (High-Bandwith-Memory 2) memory built in.
This convinces with faster loading times thanks to its high bandwidth. The best graphics cards from NVIDIA on the other hand rely on the current GDDR6 memory. This is also faster than the GDDR5 system.
To ensure that your gaming graphics card is really future-proof, you should choose 8GB memory if possible.
4GB is the minimum for most games, below that you should only use absolute economy versions. The newest (and most expensive) GPUs even come with 11 GB.
Conclusion: AMD or NVIDIA?
Whether you choose a GeForce or a Radeon card is basically up to you.
However, we would like to point out here that NVIDIA is often ahead of the pack when it comes to the technical details.
In particular, the manufacturer has significantly optimized and improved the power consumption of its current GPUs.
AMD, on the other hand, convinces above all with its – mostly but significantly – lower prices.