The best graphics card sits at the heart of any gaming PC, but quantifying what is ‘best’ can be tricky. There’s no single solution that’s right for everyone: Some want the fastest graphics card, others the best value, and many are looking for the best card at a given price. Balancing performance, price, features, and efficiency is important because no other component impacts your gaming experience as much as the graphics card.
We test and review all the major GPUs, and we’ve ranked every graphics card in our GPU benchmarks hierarchy based on performance. We’ve also done extensive testing of graphics card power consumption, using proper hardware, and we’ve looked at the broader AMD vs Nvidia GPUs breakdown. For those not running a top-end Intel CPU, we’ve also tested the top ten current GPUs on AMD’s Ryzen 9 3900X and compared performance with the same GPUs running on Core i9-9900K. Here we cut things down to a succinct list of the best graphics cards you can currently buy.
We’ve provided 10 options for the best graphics cards, recognizing that there’s plenty of potential overlap. Also, Nvidia just dropped the GeForce RTX 3080, which blows away the competition at a lower price than the outgoing RTX 2080 Ti. You can also see how the RTX 3080 scales with CPU performance. Hint: You’ll want something made in the past few years, preferably with at least 6-cores and 12-threads. There’s also the upcoming GeForce RTX 3090 and GeForce RTX 3070, with release dates of September 24 and October 15, respectively—all using the new Ampere architecture. AMD’s Big Navi is coming on October 28 (or at least being revealed), and Intel’s Xe HPG are also coming, though HPG won’t be here until 2021.
If you’re looking to buy a new high performance GPU, no one should pay more today for hardware that’s soon to be outdated. We might see a few price drops on existing product lines, but Nvidia at least likes to discontinue rather than discount, and prices on many PC components are on the rise, likely related to COVID-19 shortages. With the arrival of the RTX 3080 and Ampere, basically everyone looking at the $400 price point or above should either buy the RTX 3080, or wait for the 3070 and Big Navi to arrive and then make a decision. Maybe we’ll even get RTX 3060 or RTX 3060 Ti before the end of 2020 rolls around.
Our current recommendations reflect the changing GPU market, factoring all of the above. The GPUs are ordered mostly by performance, but price, features, and efficiency are still factors so in a few cases a slightly slower card may be ranked higher. We expect to a massive shakeup at the top of the performance rankings in the next few weeks, naturally.
When buying a graphics card, consider the following:
• Resolution: The more pixels you’re pushing, the more performance you need. You don’t need a top-of-the-line GPU to game at 1080p.
• PSU: Make sure that your power supply has enough juice and the right 6- and/or 8-pin connector(s). For example, Nvidia recommends a 650-watt PSU for the RTX 2070 Super, and you’ll need two 8-pin and/or 6-pin PEG connectors.
• Video Memory: A 4GB card is the minimum right now, 6GB models are better, and 8GB or more is strongly recommended.
• FreeSync or G-Sync? Either variable refresh rate technology will synchronize your GPU’s frame rate with your screen’s refresh rate. If your monitor supports G-Sync tech (for recommendations, see our Best Gaming Monitors list), you’ll need a GeForce GPU. AMD’s FreeSync tech works with Radeon cards, and Nvidia has certified some FreeSync displays as being G-Sync Compatible.
|GPU||Performance Rank||Value Rank (fps/$)|
|RTX 3080||1 (142 fps)||7 ($699?)|
|RTX 2080 Ti||2 (118 fps)||10 ($1,140)|
|RTX 2080 Super||3 (102 fps)||9 ($700)|
|RTX 2070 Super||4 (91 fps)||8($500)|
|RX 5700 XT||5 (87 fps)||4 ($360)|
|RTX 2060 Super||6 (77 fps)||6 ($370)|
|RX 5600 XT||7 (71 fps)||2 ($260)|
|RTX 2060||8 (68 fps)||5 ($295)|
|GTX 1660 Super||9 (58 fps)||3 ($230)|
|GTX 1650 Super||10 (43 fps)||1 ($160)|
Best Graphics Cards for Gaming 2020
The king is dead, long live the king! Nvidia’s GeForce RTX 3080 has arrived, sporting the new and improved Ampere architecture. It’s over 30% faster than the previous gen 2080 Ti, for $500 less. If you’re serious about maxing out all the graphics settings and you want to play at 4K or 1440p, this is the card to get — it’s mostly overkill for 1080p gaming, though enabling all ray tracing effects in games that support the feature makes 1080p still reasonable.
If you skipped the first round of RTX GPUs, the RTX 30-series might finally get you you on board the ray tracing train. With potentially double the ray tracing performance of Turing, and games like Cyberpunk 2077 on they way with even more ray tracing effects, the RTX 3080 is your best chance south of $1,500 to play games in all their ray traced glory.
Ampere also brings improved tensor cores for DLSS, a technology we’re bound to see more of in future games now that it doesn’t require per-game training by a supercomputer. The biggest problem with RTX 3080 right now is going to be finding one in stock, at less than scalper prices. By the time the supply increases, we may also have RTX 3070 and RX 6900 XT cards as an alternative.
Yesterday’s champion is set to fall to fourth place overall once the RTX 3090 and RTX 3070 arrive. For now, it’s still the second fastest gaming GPU (not counting Titan RTX), and you can even find cards for sale on eBay for significantly lower prices than last month. Except, most of those prices are still at least as high as the RTX 3080, which is 30% faster, sometimes more.
Powered by Nvidia’s Turing architecture, the RTX 2080 Ti ushered in the era of consumer GPUs with ray tracing hardware. It also brought along tensor cores for DLSS and other tasks, plus concurrent floating-point and integer execution. And it was priced as high as the former Titan Xp, which was a major turn off.
If you already have an RTX 2080 Ti, there’s no need to worry just yet. Nvidia claims the RTX 3070 will be «faster than the 2080 Ti,» but then it also said RTX 3080 was «up to twice as fast as the RTX 2080.» Our testing shows the 3080 is closer to a 70% improvement, and very likely the 2080 Ti will still have plenty of games where it’s faster than a 3070. But seriously, don’t buy a 2080 Ti today — not unless you can get it for 3070 pricing (meaning $500).
It’s not a good time to buy a brand new RTX 20-series GPU, at all. Our list of the best GPUs is in the midst of some serious upheaval, what with the RTX 3080 launch and the upcoming 3090 and 3070 launches. The problem is, 30-series cards are probably going to sell out, or sell at highly inflated prices, for the next couple of months.
In the meantime, maybe you can find a good deal on the RTX 2080 Super. Don’t pay more than $500 for such a card, and probably not even that, but the 2080 Super is still very capable. It’s just not RTX 3080 levels of capable. 1440p and 4K gaming are totally possible, just not necessarily at maximum quality (especially 4K). The good news is that the difference between ultra and high quality in many games is difficult to see, while the jump in performance can be significant.
There’s still the question of what will happen with ray tracing adoption in the future, of course. The first round of DirectX Raytracing (DXR) games has often seen performance drop by 30-40% when the feature is enabled. DLSS can often make up for that drop, but the implementations of both DXR and DLSS vary by game. If you take a game like Control, which features ray traced reflections, contact shadows and diffuse lighting — the most complete implementation of ray tracing in a game to date (not counting Quake II RTX and Minecraft RTX, which are special cases of porting an old/simple game to full path tracing) — performance dropped by half, and DLSS 2.0 only mostly recovers the lost fps.
When games in the future start using more ray tracing effects, even the 2080 Super may not keep up. That doesn’t mean you should ignore ray tracing hardware, however. Just keep your expectations in check.
Like all of the other GPUs other than the RTX 3080, right now is a tough time to recommend any high-end card. AMD’s RX 5700 XT is reasonably priced for previous gen GPUs, but in the next month or two we could very likely see a big jump performance for the same $400 asking price.
What’s more, AMD’s Navi 10 architecture doesn’t support hardware or software ray tracing, though in traditional rasterization techniques AMD’s RDNA architecture is very competitive. AMD’s GPUs also tend to do better in games that use either DirectX 12 or the Vulkan API, while DX11 games favor Nvidia. Overall, across our test suite, the 5700 XT beats the RTX 2060 Super by about 7% in performance (as well as the earlier RTX 2070 by 4%), and typically costs less. There have been some concerns with AMD’s drivers since Navi launched, but the latest updates appear to have addressed the biggest problems.
Compared to AMD’s own previous generation products, the RX 5700 XT also puts up an impressive performance. It basically matches the Radeon VII at 1080p and 1440p, all while using 75W less power. That’s partly the benefit of the 7nm FinFET manufacturing process, something Nvidia will utilize in Ampere, but the foundational RDNA architecture definitely improved resource utilization over the previous GCN architecture cards. Our biggest concern is that AMD’s RDNA 2 will add ray tracing later this year, and apparently 50% higher performance per watt. That’s definitely worth waiting for at this point.
We’ve also nixed the vanilla RX 5700 from our list, as current prices are nearly the same as the faster 5700 XT.
You know the drill by now: Wait for Ampere or Navi 2x GPUs. But if you can’t wait, or if you can find a good deal on an RTX 2070 Super, it’s still a good card. The law of diminishing returns is in full effect as you move up to the top of the GPU hierarchy, which means you’ll pay more for proportionately less performance every additional step. The best way to avoid that is to drop back a few notches, which is where you’ll find the RTX 2070 Super. 4K gaming is a stretch, but 1440p ultra is still viable.
Just as the RTX 2080 Super is a sensible step down from the 2080 Ti, the 2070 Super is only about 10% slower than the 2080 Super but costs almost 30% less. It’s also only a few percent slower than the earlier vanilla RTX 2080, and you still get 8GB of GDDR6, ray tracing and DLSS, and a card that beats the previous generation GTX 1080 Ti — along with every current AMD GPU. This is all at the same nominal price as the slightly older and slightly slower RTX 2070, and there’s no Founders Edition tax this round.
But the GeForce RTX 3070 with the new Ampere architecture is waiting in the wings, so it doesn’t make sense to spend a lot of money right before new graphics cards come out. The 2070 Super isn’t bad, but it’s yesterday’s high-end card staring in the shop windows at the shiny new Ampere GPUs.
Out of all the Nvidia RTX cards, the RTX 2060 Super is the hardest to recommend. It’s not that it’s a bad card — it’s plenty fast and has the same features as other RTX models — but performance and pricing end up being eclipsed by AMD’s 5700 XT. Or at least you could before the supply of previous gen graphics cards started to dry up. In a direct face off between the RX 5700 XT and the RTX 2060 Super, we gave AMD the edge. Still, the 2060 Super handles 1080p and 1440p gaming without difficulty, but anyone paying $400 or more for a 2060 Super will likely end up with buyer’s remorse in the next month or so.
We know, we’re sounding like a broken record, but the RTX 3070 and possibly even RTX 3060 Ti are coming. AMD also have Big Navi in the works. These newer and faster GPUs should all arrive by November is our bet, so keep using what you’ve got if possible and wait for the new hotness to arrive. Or keep an eye out for a good clearance. We’d suggest aiming for $300 or less on the 2060 Super at this point. Yes, things have changed that much, and arguably more.
If you want an 8GB card that can do ray tracing, for the lowest price possible, the RTX 2060 Super fills that niche. The extra memory does actually have an impact on ray tracing performance as well, so there are reasons to spring for the upgrade over the vanilla RTX 2060. Just don’t be surprised when an RTX 3060 shows up that performs better and costs less later this year or in early 2021.
The Radeon RX 5600 XT launch is the perfect showcase for how competitive the mid-range graphics card market has become. Initially intended to compete with the GTX 1660 Ti at the $280 price point, it would have easily won that matchup — it’s about 15% faster. Then Nvidia preemptively dropped the price on the RTX 2060 Founders Edition to $299, and worked with EVGA to launch a special RTX 2060 KO for $300. That’s when things got interesting.
AMD responded with a last-minute BIOS update that increased clock speeds on the RX 5600 XT, and allowed partners the option to run the memory at 14 Gbps instead of the ‘reference’ 12 Gbps. Needless to say, a 17% increase in memory bandwidth and GPU clocks helps quite a bit.
Now that the dust has settled, all current RX 5600 XT models should support 14 Gbps, either out of the box or via a BIOS update. Prices have also dropped, and the RX 5600 XT is currently the best overall value. If you have to find something to tide you over for a few months, a $250 RX 5600 XT is a sound choice.
Of course it lacks features like ray tracing and DLSS, and while we don’t know what Navi 2x will do to compete with DLSS, we do know it has ray tracing hardware. Hopefully in the next month or so we’ll hear about an RX 6600 or RX 6700 card that offers better performance, more features, and doesn’t cost too much more.
Nvidia’s RTX 2060 was a decent card when it launched in January 2019, and dropping the price on the Founders Edition one year later keeps it in the running. In a direct head-to-head between ‘reference’ RTX 2060 and RX 5600 XT models, Nvidia eked out a win, but lower prices have now pushed our recommendation back to AMD. A lot of it will depend on current pricing, the features you get, as well as potential factory overclocks. Non-reference RX 5600 XT cards beat the stock 2060 FE, but non-reference RTX 2060 cards like the EVGA RTX 2060 KO Ultra swap things back.
For many people, $300 is about as much as they’re willing to spend on a graphics card, making this an important part. Sure, the RTX 2080 Ti is almost twice as fast at 4K ultra, but the price is obscene. Even the new RTX 3080 and 3070 are still too much for a lot of gamers. $500 is enough for a complete Xbox Series X, while $300 will get you an Xbox Series S that’s far less capable. In fact, the RTX 2060 should beat the Xbox Series X in overall performance (6 TFLOPS vs. 4 TFLOPS).
The 2060 is still great for 1080p gaming, averaging 90 fps in our test suite at ultra settings, though Metro Exodus and Red Dead Redemption 2 both fall short of 60 fps at ultra settings. It’s also good to put things in perspective. The RTX 2060 at $300 is nearly 70% faster than the GTX 1060 6GB, and basically twice as fast as the GTX 970. If you’re in the market for a new graphics card today and you don’t want to wait around to see what Ampere and Navi 2x have to offer in the fall, this is a great 1080p gaming solution — and it will even handle ray tracing okay, if you enable DLSS.
Dipping down closer to $200, we’re given the choice between the GeForce GTX 1660 Super for $230, the vanilla GeForce GTX 1660 for around $200, or the RX 5500 XT 8GB starting at around $200. They’re all viable candidates, but we’ve done the testing (see below) and the GTX 1660 Super is 15% faster than the regular 1660, and nearly 20% faster than the RX 5500 XT 8GB. We’ve looked at the GTX 1660 vs. RX 5500 XT and declared the Nvidia card the winner, but we also think the GTX 1660 Super is better than the GTX 1660 for just $20-$30 more.
Despite Nvidia’s Turing GPUs still using TSMC 12nm FinFET, actual power use is basically identical to AMD’s Navi 14 chips made using TSMC 7nm FinFET. The fact that Nvidia is faster and draws the same power while using the older manufacturing node says a lot. For $230, the GTX 1660 Super basically gets you the same level of performance as the older GTX 1070 in a more efficient design, and it’s 1% slower than the GTX 1660 Ti (which is basically no longer worth considering). It also comes with the enhanced Turing NVENC that makes it a great choice for streaming video.
The main drawback to the GTX 1660 Super is that the RX 5600 XT and RTX 2060 aren’t that much more expensive, especially considering the performance increase. Both are around 20% faster, for just $30-$70 more money. When looking at the cost of the rest of a PC, for gaming purposes we prefer putting additional funds into the graphics card.
We’ve long recommended AMD’s RX 570 4GB as our budget pick, and while it’s still available for $120-$140, we’re ready to move on to something better. Nvidia’s GTX 1650 Super is about 30% faster than the RX 570, but that’s only a small part of the upgrades. It also uses substantially less power, plus it includes Nvidia’s latest Turing NVENC hardware to help with video encoding and decoding. You don’t actually need a ton of CPU power to livestream your gameplay, as the 1650 Super is more than capable of doing the dirty work all on its own.
Even if price is your driving concern, saving $40 to end up with an older and less efficient architecture doesn’t really make sense. Polaris first launched in 2016, and while the 1650 Super won’t necessarily pay for itself in power savings (it would take a couple of years to break even on power cost at eight hours per day of gaming), more performance, better efficiency, and better video support make this an easy recommendation.
If you want to stick with AMD hardware, the RX 5500 XT 4GB offers nearly identical performance, though it still uses up to 25% more power, and we’d suggest the 8GB model that costs $20-$40 more … which then leads back to the 1660 Super above. You can also look at the vanilla GTX 1650 (or the GTX 1650 GDDR6) to save $10-$20, but the TU117 GPU in those cards uses the less capable Volta/Pascal NVENC hardware.
If you just need any old GPU, there are less costly options like the GT 1030 and RX 550, but those have even less performance. They’re not really good for anything beyond very light gaming, though the GTX 1050 and 1050 Ti are the fastest GPUs right now where you don’t need a 6-pin PCIe power connector.
How We Test Graphics Cards
Determining pure graphics card performance is best done by eliminating all other bottlenecks — as much as possible, at least. Our current graphics card testbed consists of a Core i9-9900K CPU, MSI MEG Z390 Ace motherboard, 32GB Corsair DDR4-3200 CL16 memory, and an XPG SX8200 Pro 2TB SSD. We test across the three most common gaming resolutions, 1080p, 1440p, and 4K, at medium and ultra settings.
Where possible, we use ‘reference’ cards for all of these tests, like Nvidia’s Founders Edition models. Most mid-range and lower GPUs do not have reference models, however, and in some cases we only have factory overclocked cards for testing. One specific card deserves a call-out: the RX 580 8GB card we tested is a heavily overclocked Sapphire Nitro+ Limited Edition, which is why it sometimes even manages to beat the RX 590. It’s the only 580 currently in our GPU labs, however, so we rolled with it. (A ‘normal’ RX 580 would be about 10% slower.)
Our current test suite of games consists of nine titles. The data in the following charts is from testing conducted in June 2020. Note that a few games (Red Dead Redemption 2 and The Division 2) may not run on all GPUs, and we have interpolated data for the overall score to keep the charts consistent. You need at least a 4GB card for 1080p ultra in RDR2, and 6GB or more for 4K ultra. (Omitting the results for RDR2 skews the average performance chart slightly, which is why we interpolated results.)
The following charts contain the most common 40 GPUs of the past six years, give or take. We have included everything from the Titan RTX down to the ‘ancient’ GTX 970 and R9 390, plus all of latest AMD Navi and Nvidia Turing GPUs. The charts are color coded with AMD in red/grey and Nvidia in blue/black to make it easier to see what’s going on.
Besides performance, we also test graphics card power consumption. We recently retested all current GPUs using Powenetics equipment and software, and found that Nvidia generally maintains an efficiency lead. Here are the main power charts from our testing:
All GPUs Ranked
Our full GPU hierarchy ranks all current in previous generation GPUs by performance, using aggregate data from the gaming test suite. Below is the abbreviated hierarchy with all the cards you can still buy (plus a few extras) ranked in order of performance, from best to worst. The score represents aggregate performance, scaled relative to the Titan RTX.
|Nvidia RTX 3080||117.4%||GA102||1440/1710 MHz||10GB GDDR6X||320W||RTX 3080|
|Nvidia Titan RTX||100.0%||TU102||1350/1770 MHz||24GB GDDR6||280W||Titan RTX|
|Nvidia GeForce RTX 2080 Ti||94.4%||TU102||1350/1635 MHz||11GB GDDR6||260W||Nvidia GeForce RTX 2080 Ti|
|Nvidia GeForce RTX 2080 Super||83.1%||TU104||1650/1815 MHz||8GB GDDR6||250W||GeForce RTX 2080 Super|
|Nvidia GeForce RTX 2080||78.1%||TU104||1515/1800 MHz||8GB GDDR6||225W||GeForce RTX 2080|
|Nvidia GeForce RTX 2070 Super||74.0%||TU104||1605/1770 MHz||8GB GDDR6||215W||GeForce RTX 2070 Super|
|AMD Radeon VII||73.5%||Vega 20||1400/1750 MHz||16GB HBM2||300W||Radeon VII|
|AMD Radeon RX 5700 XT||70.3%||Navi 10||1605/1905 MHz||8GB GDDR6||225W||AMD Radeon RX 5700 XT|
|Nvidia GeForce RTX 2070||66.2%||TU106||1410/1710 MHz||8GB GDDR6||185W||RTX 2070|
|AMD Radeon RX 5700||63.7%||Navi 10||1465/1725 MHz||8GB GDDR6||185W||AMD Radeon RX 5700|
|Nvidia GeForce RTX 2060 Super||62.9%||TU106||1470/1650 MHz||8GB GDDR6||175W||GeForce RTX 2060 Super|
|AMD Radeon RX Vega 64||60.5%||Vega 10||1274/1546 MHz||8GB HBM2||295W||Gigabyte Radeon RX Vega 64|
|AMD Radeon RX 5600 XT||57.7%||Navi 10||?/1615 MHz||6GB GDDR6||150W||Radeon RX 5600 XT|
|Nvidia GeForce RTX 2060||55.9%||TU106||1365/1680 MHz||6GB GDDR6||160W||Nvidia GeForce RTX 2060 FE|
|AMD Radeon RX Vega 56||53.3%||Vega 10||1156/1471 MHz||8GB HBM2||210W||Radeon RX Vega 56|
|Nvidia GeForce GTX 1660 Ti||47.2%||TU116||1365/1680 MHz||6GB GDDR6||120W||GeForce GTX 1660 Ti 6GB|
|Nvidia GeForce GTX 1660 Super||46.9%||TU116||1530/1785 MHz||6GB GDDR6||125W||GeForce GTX 1660 Super|
|Nvidia GeForce GTX 1660||41.0%||TU116||1530/1785 MHz||6GB GDDR5||120W||Geforce GTX 1660|
|AMD Radeon RX 590||40.4%||Polaris 30||1469/1545 MHz||8GB GDDR5||225W||Radeon RX 590|
|AMD Radeon RX 5500 XT 8GB||39.4%||Navi 14||?/1717 MHz||8GB GDDR6||130W||AMD Radeon RX 5500 XT 8GB|
|AMD Radeon RX 580 8GB||38.6%||Polaris 20||1257/1340 MHz||8GB GDDR5||185W||AMD Radeon RX 580|
|Nvidia GeForce GTX 1650 Super||35.5%||TU116||1530/1725 MHz||4GB GDDR6||100W||Nvidia GeForce GTX 1650 Super|
|AMD Radeon RX 5500 XT 4GB||35.4%||Navi 14||?/1717 MHz||4GB GDDR6||130W||AMD Radeon RX 5500 XT 4GB|
|AMD Radeon RX 570 4GB||31.5%||Polaris 20||1168/1244 MHz||4GB GDDR5||150W||Radeon RX 570|
|Nvidia GTX 1650 GDDR6||29.8%||TU117||1410/1590 MHz||4GB GDDR6||75W||GeForce GTX 1650 GDDR6|
|Nvidia GeForce GTX 1650||26.1%||TU117||1485/1665 MHz||4GB GDDR5||75W||GeForce GTX 1650 Gaming OC 4G|
Want to comment on our best graphics picks for gaming? Let us know what you think in the Tom’s Hardware Forums.