AMD vs Nvidia 2021: which makes the best graphics cards?
Is Nvidia finally winning the Nvidia vs AMD war? While the battle is hotter than it’s ever been since the release of the Ampere graphics cards and the Big Navi Radeon RX 6000 cards, the king of GPUs is still a matter of perspective.
Nvidia might be winning on sheer power, even if AMD has rolled out some potent graphics cards that can compete with the best of them. Meanwhile, AMD is still winning in terms of price, even if Nvidia has made its price points more accessible, and even though it’s also putting out expensive, high-end GPUs like the new RTX 3080 Ti.
The Nvidia vs AMD battle has produced some of the best and most affordable GPUs we’ve seen in years. It has given us affordable options like the new AMD Radeon RX 6700 XT, Nvidia GeForce RTX 3060 Ti and the Nvidia RTX 3060. It’s also given us entries that deliver sheer power like the Nvidia GeForce RTX 3080, RTX 3080 Ti and AMD Radeon RX 6900 XT, which are still within many people’s reach in terms of price.
So, even though there might never be one true winner, it only means good things for consumers. With these two manufacturers constantly trying to one up each other, we’ll always have compelling GPUs to choose from. The only thing to do is decide which one is best for you.
We’ve put the two manufacturers – Nvidia vs AMD – side by side to see which one has an edge over the other in terms of pricing, performance, and features. Hopefully, when all is said and done, you’ll have made your decision as to which team to go with.
Traditionally, AMD has always been known as the more affordable brand of graphics cards, and that's true to this day... to a point. Right now, especially in the mid-range, AMD has graphics cards like the Radeon RX 5500 XT, which provide excellent performance at the $199 (about £150, AU$280) price point. If your budget is around this level, AMD's generous helping of VRAM here means that you're getting much better performance in higher-spec games than Nvidia's equivalent GTX 1650 can offer.
Once you start going up the price stack, things change, however. At the top of the pile, AMD still comes out the winner in terms of affordability. The Radeon RX 6900 XT is much cheaper at $999 (£770, about AU$1,400) next to the Nvidia GeForce RTX 3090’s $1,499 (£1,399, around AU$2,030) and even the slightly more affordable RTX 3080, which will set you back $1,199 (£1,049, AU$1,949).
Once to get to the lower high-end, however, things are no longer so black and white. Both the AMD Radeon RX 6700 XT and the AMD Radeon RX 6800 sit a little higher on the dollar scale than their direct Nvidia rivals without delivering that much of a performance advantage.
AMD’s Big Navi may not be the Nvidia killer it was first rumored to be, but a few of the cards in the line are certainly giving Nvidia some stiff competition. If you want to play the best PC games at 4K and get a solid 60+ fps frame rate, you're no longer stuck with Nvidia. Although with the Nvidia RTX 3080 Ti now officially out in the world, AMD is going to need to roll out a serious contender soon.
In 2021, you can get a graphics card that will power high-end AAA PC games at 1080p settings with something like the AMD Radeon RX 5600 XT or the Nvidia GeForce RTX 3060. If you want to play AAA games at 1440p with no compromises, both Team Red and Team Green have great options with the Radeon RX 6700 XT and the Nvidia GeForce RTX 3060 Ti.
The wider availability of graphics cards that can push pixels at these resolutions on a budget has made PC gaming much more accessible than ever before, and these upcoming generations have done the same for 4K gaming on PC, especially with the PS5 and Xbox Series X both costing much less than the price of a high-end gaming PC.
For 4K, both manufacturers have terrific offerings, with Nvidia rolling out the Nvidia GeForce RTX 3080, RTX 3080 Ti, and even the Nvidia GeForce RTX 3090, and AMD releasing the Radeon RX 6900 XT, which has been developed to go toe-to-toe with Nvidia's RTX 3090.
Exclusivity and features
When it comes to features beyond just rendering games, Nvidia and AMD take much different approaches.
Typically, AMD's approach is much more consumer-friendly, as it releases features and technologies that can be used even on Nvidia graphics cards – though they will usually work best on AMD's own silicon.
Nvidia, on the other hand, likes to keep things close to its chest, launching features like DLSS that only work on its own platform. Team Green has been doing this for years, going back to PhysX. In fact, with that latter technology, you were even able to have a dedicated PhysX graphics card in your system to handle the computationally heavy workload.
Recently, however, Nvidia has launched a ton of features that are helpful outside of gaming, both as part of its ongoing Nvidia Studio driver program for creative and professional workloads, and just to aid people in their post-pandemic lives.
Most notably, with Ampere, you get Nvidia Broadcast, which is incredibly useful technology for pretty much everyone. With this program, you can replace backgrounds in any video conferencing app using AI. What's better is that you can also use it to filter out all background noise from your microphone while in a call, so you don't have to worry about disrupting that 10am meeting by drinking coffee and hurriedly eating breakfast.
Conversely, AMD is still very much centered on gaming with its mainstream graphics cards, and all features in its FidelityFX software suite introduced with RDNA are centered on delivering a better gaming experience. This includes things like contrast adaptive sharpening (CAS) that makes playing on a higher resolution display easier, and better ambient occlusion.
So, which is better? Neither
There’s so much to love about both Nvidia and AMD graphics. In the end, both of these companies rely on competition with each other to thrive. Suffice to say, the Nvidia vs AMD debate requires that you understand there’s a reason Radeon and GeForce GPUs are so similar in performance right now.
Each company is doing its best to keep up with the mindshare of the other, and that’s good for us. They’re basically fighting for our money, learning from each other’s mistakes and legislating marked improvements along the way.
It’s up to you who wins the fiery contest of Nvidia vs AMD, although we will say this: Nvidia is unmatched in the 4K market right now. If it helps any, the RTX 2080 Ti is probably your best bet if you want your PC to keep up with your Ultra HD display – as long as you can afford it. On the other hand, if you’re on a budget and looking into mid-range cards, Nvidia and AMD graphics cards will probably be about the same.
Jackie Thomas (Twitter) is TechRadar's US computing editor. She is fat, queer and extremely online. Computers are the devil, but she just happens to be a satanist. If you need to know anything about computing components, PC gaming or the best laptop on the market, don't be afraid to drop her a line on Twitter or through email.
AMD vs Nvidia: Who Makes the Best GPUs?
AMD vs Nvidia. If you're building a gaming PC, you'll inevitably be faced with choosing between the two GPU heavyweights. Both companies make GPUs that power the best graphics cards, fighting for supremacy in our GPU benchmarks hierarchy. AMD vs Nvidia isn't the only decision you’ll need to make when building a PC, of course. You'll also need to choose between AMD vs Intel CPUs. Our focus here will be on graphics, however, and we'll be looking at performance, features, drivers and software, power and efficiency, pricing and more.
The AMD vs Nvidia flame wars have been cooking since the late 90s, with Nvidia currently leading in the GPU arena by most metrics. Its graphics cards account for the majority of GPUs on the Steam Hardware Survey, for example, and in terms of pure finances, Nvidia is worth roughly three times as much as AMD (with a large chunk of AMD's resources devoted to CPUs).
But we're not interested in the distant past or finances. We want to find a winner in the current battle of AMD vs Nvidia GPUs. That primarily means looking at AMD Big Navi and Nvidia Ampere graphics cards, and perhaps adding Intel Xe-HPG to the mix whenever that arrives (probably this fall, based on the Xe-HPG Scavenger Hunt and other PR stunts).
It's important to keep the big picture in view throughout this analysis. We're not just focusing on the fastest GPU, or the most power-efficient GPU, or the best bang-for-the buck GPU. We'll consider all of the factors in each category, from budget to mid-range to high-end and extreme GPUs, along with the tech behind the GPUs. We will declare a winner today, but of course this isn't the end of the war. It's more like owning the heavyweight GPU title: A victory today doesn't mean your opponent won't come back leaner and meaner next year.
With that preamble out of the way, let's pull out the boxing gloves and go the rounds with AMD vs Nvidia.
AMD vs Nvidia: Gaming Performance
For decades, faster GPUs have enabled game developers to create increasingly detailed and complex worlds. While you can find everything from budget GPUs to high-end offerings from both AMD and Nvidia, when it comes to outright performance, Nvidia has a slight overall lead thanks to the chunky GeForce RTX 3090. Beyond the pole position, however, it's a closer match. If you look at our GPU benchmarks hierarchy, you'll see that the top five positions consist of two GPUs that use Nvidia's GA102 architecture (the 3090 and 3080) and three GPUs using AMD's Navi 21 architecture, with AMD bookending the Nvidia cards.
There's a catch, of course: We're only looking at games running APIs and settings that work on all GPUs, which means we haven't included ray tracing or DLSS in the results. We also haven't included any FidelityFX results, and CPU bottlenecks certainly play a role at lower resolutions. Here's the full rundown of performance, using the nine games and six settings/resolutions combinations from our GPU Benchmarks.
Nvidia's RTX 3090 isn't remotely affordable, of course, and that's before we take into account the current market conditions. We'll discuss that more in a bit, but the GeForce RTX 3080 and Radeon RX 6800 XT represent a better view of performance. Those two trade blows in traditional rasterization games, with AMD perhaps taking a slight lead, while Nvidia easily jumps ahead — often by a large margin — as soon as ray tracing and/or DLSS get turned on. We give Nvidia a slight lead at the top of the performance ladder, but that's not the only category to consider.
Looking at the mainstream market ($400, give or take), things get a bit messy. AMD's Radeon RX 6700 XT takes on the GeForce RTX 3060 Ti, with slightly better performance — provided once again that we ignore RT and DLSS. In our review, we called it a net tie. However, AMD's card (theoretically) costs 20% more than Nvidia's offering, never mind the GeForce RTX 3060 12GB. We're still waiting to see other Navi 22 models arrive, however, and an RX 6700 (non-XT) could make for a better option. Older model GPUs are mostly outclassed by the new arrivals, so Nvidia nominally wins the mainstream market as well, depending on the availability of cards.
What about in the budget category? There haven't been any new budget GPUs since last year. Looking at existing cards, the RX 5500 XT 8GB vs GTX 1660 goes to Nvidia again, and in general we'd normally recommend avoiding the lower tier GTX 1650 or RX 5500 XT 4GB. But these are not normal times. Anyway, until RTX 3050 / 3050 Ti or RX 6500 XT cards arrive, Nvidia has a lead in the budget sector, even if many of the categories are close. The GTX 1660 is a touch faster and uses a bit less power, for roughly the same price. Well, not really the same price these days, but we'll hit that below.
What about cards that cost even less? Right now, that means older GPUs that have 4GB of VRAM, because GPU and component shortages combined with cryptocurrency mining have inflated prices on everything. Some older generation cards like AMD's Radeon RX 570 4GB or Nvidia's GTX 1060 can be found for under $200, but supply is spotty at best.
Winner: Nvidia Across a large suite of games, Nvidia wins in most categories. AMD does okay at the top sector with the RX 6700 XT, RX 6800, RX 6800 XT, and RX 6900 XT, but a tie at the top and losses everywhere else give Nvidia the overall win for now.
AMD vs Nvidia: Power Consumption and Efficiency
Prior to AMD's Navi, GPU power efficiency was decidedly in favor of Nvidia. But Navi changed all that, and Big Navi has further improved AMD's efficiency. Using chips built with TSMC's 7nm FinFET process and a new architecture that delivered 50% better performance per watt, Navi started to close the gap. Except, it was so far behind that even a 50% improvement didn't fully address the efficiency deficiency.
But Nvidia's Ampere architecture pushed higher clocks at the cost of efficiency, while AMD's Big Navi gets a healthy boost in efficiency from the Infinity Cache. The net result is that Ampere and Big Navi are pretty close to tied.
Using Powenetics hardware to capture the real graphics card power use of GPUs, we've tested all of the current and recent graphics cards from both companies. We've also tested third party cards from both sides, but we'll confine the charts to the reference designs as much as possible.
While power use favored Nvidia's older GPUs, sometimes by a wide margin, the differences on the latest generation hardware go the other way. RTX 3090 uses the most power, followed by the RTX 3080 and then the three Navi 21 cards: RX 6900 XT, 6800 XT, and 6800. Previous generation cards like the Titan RTX, 2080 Ti, and 2080 Super fall in between the 6800 and 6800 XT, but we're less interested in those.
Below that, differences are small enough that there's no clear winner. AMD's RX 6700 XT uses a bit more power than the RTX 3070 and a bit less than the RTX 3060 Ti, but they're all in a 10W range. The 3060 12GB uses 35W less than the others, but it's at a lower performance tier as well.
Among the previous gen budget GPUs, Nvidia's GTX 16-series cards have a small lead, particularly in FurMark — which isn't nearly as important as gaming power use. Still, Nvidia's GTX 1660 Ti and GTX 1660 Super use a bit less power but perform up to 20% faster than the RX 5500 XT 8GB. It's not a major difference, but it's still a win. Similarly, dropping down to Nvidia's GTX 1650 line, the GTX 1650 Super edges out the RX 5500 XT 4GB by a hair in performance while using 25W less power, while the GTX 1650 GDDR6 shaves off another 20W but also drops 17% in performance.
Winner: Tie Focusing just on the current generation AMD Big Navi and Nvidia Ampere GPUs, power and efficiency are relatively close, especially when looking at non-reference designs where power limits often end up around 350W. AMD takes a slight lead at the top, the middle's a tie, and the previous gen budget cards favor Nvidia. TSMC's N7 process also factors in, helping out AMD, while Nvidia's use of Samsung 8N likely reduces overall efficiency.
AMD vs Nvidia: Featured Technology
Most of the features supported by AMD and Nvidia seem similar, though the implementations do vary. Both support ray tracing now, which allows for some nice effects, but it's not required to get a good gaming experience. Nvidia's DLSS is a bigger factor, as FidelityFX Super Resolution (FSR) isn't ready yet. There are other aspects as well. G-Sync takes on FreeSync, Radeon Anti-Lag goes up against Nvidia's ultra-low latency mode plus Reflex, and there other areas where features effectively match up as well.
Supporting the same APIs and similar hardware features doesn't make things equivalent, however. Ampere and RDNA2 also support mesh shaders and variable rate shading (VRS), as well as some other features that are all part of the DirectX 12 Ultimate spec. But Nvidia's performance in ray tracing tends to be quite a bit higher than AMD, even without DLSS.
While FreeSync and G-Sync might seem equivalent on the surface, the best G-Sync displays are almost invariably higher quality and better latency than FreeSync displays. The same goes for AMD's anti-lag and Nvidia's ultra-low latency and Reflex: Similar in theory, but in practice Reflex implementations come out on top.
Video encoding and decoding are another important aspect, and here Nvidia definitely comes out ahead. The Turing and Ampere codec support higher-quality encodes and lower GPU utilization, both good things. There's no need for CPU-based video encoding with Nvidia's latest GPUs.
There is one area where AMD has a clear advantage, though it's also perhaps to AMD's detriment. TSMC's N7 7nm process that AMD uses for RDNA2 (and RDNA and Zen 3 and the PS5/XSX) chips clearly delivers better performance and power characteristics than Samsung's custom 8N 8nm (really just an improved version of Samsung's 10LPP process). The catch being that lots of other companies also want to partake of TSMC's goodness — AMD, Apple, Nvidia GA100, Qualcomm, and even Intel all use TSMC, along with various other smaller players. This becomes a problem when TSMC doesn't have enough capacity to meet the demands of all of those companies.
Winner: Nvidia While AMD and Nvidia have superficial parity on most features, Nvidia's implementations are generally superior — and cost more. G-Sync, Reflex, DLSS, and NVENC all end up being at least slightly better than AMD's alternatives.
AMD vs Nvidia: Drivers and Software
Trying to determine a clear winner in the drivers and software category is difficult. Quite a few people previously encountered black screen issues with AMD drivers on RX 5000 Navi series GPUs, while others didn't have any difficulties. Newer drivers have fixed these problems, but some user complaints continue. Nvidia drivers aren't foolproof either, and depending on the game and hardware, issues crop up for both companies. But is one company doing better with drivers?
AMD makes a lot of noise about its yearly driver overhaul. The Radeon Adrenalin 2020 drivers consolidated everything under one large umbrella, aiming to simplify things, though it can be confusing at first if you're used to the older drivers. AMD tends to skip WHQL (Windows Hardware Quality Labs) testing, which means fewer hoops to jump through and potentially more bugs slip through, but Microsoft's ensurance of a minimum level of functionality doesn't really mean much for gaming purposes. You can generally count on at least one new AMD driver per month, often more if there are major game launches.
Nvidia's driver schedule follows a similar cadence. You'll get new drivers for major game launches or new graphics card hardware. Nvidia's releases (outside of hotfixes) are all WHQL certified, and Nvidia also has a separate Studio Driver branch for content creators.
One of the big differences between AMD and Nvidia drivers is that Nvidia has two separate user interfaces. The Nvidia Control Panel handles things like resolutions and certain graphics settings, while GeForce Experience tackles game optimizations, driver updates, and extra features including ShadowPlay, Ansel, and more. Just say no to data mining plus drivers.
Winner: Tie We prefer AMD's unified driver approach, as it's one less interface to navigate, but there's just so much stuff in the current releases. Nvidia's Q&A is arguably better, though plenty of bugs and issues end up slipping through on both sides. Quantifying drivers ends up being an incredibly subjective affair, however, so we're calling this one a draw.
AMD vs Nvidia: Pricing and Availability
Who offers the better value in the battle of AMD vs Nvidia? Ha. Hahahaha... The wheels fall off this comparison when we get to pricing, as everything remains on backorder. That's been the case since the RTX 3080 launch last September, and while there are indicators — like the Steam Hardware Survey — that a fair number of people have managed to buy one of the latest generation GPUs, there's no stock available at any normal retail outlet.
Our GPU Pricing Index takes a look at eBay pricing as an alternative, since it's the one place where you actually can find cards. They're just sold at exorbitant prices, typically double to triple the official launch price. If you're not willing to pay eBay pricing — and you shouldn't! — you'll need to exercise your patience muscle and just keep trying. Newegg has periodic 'lotteries' (several of us have entered multiple times and still come up empty), and we're told AMD and Nvidia continue to ship cards "as fast as possible" to their partners. Unfortunately, it's not fast enough and other component shortages are also a factor.
At the top of the pricing spectrum, the RTX 3090 currently goes for around $3,000, which is double its already extreme $1,500 launch price. RTX 3080 meanwhile sells for roughly $2,200, over triple its official $700 launch price. Even if you're lucky and manage to score a 'deal,' you'd be looking at $1,000 or more for the 3080 and $2,000 or more for the 3090. The only winner there is the retail outlet or AIB that's raking in the money.
Stepping down to the RTX 3070 and RTX 3060 Ti, both sell for around $1,300–$1,400 on eBay. There's a very strong correlation between eBay prices and Ethereum mining profitability, across all GPUs, and that extends down even to cards that only have 4GB and technically can't mine Ethereum very well. (There are other cryptocurrencies that can still be mined on such cards, unfortunately for gamers.) Rounding out the latest GPUs, Nvidia's RTX 3060 12GB 'only' costs $850–$950, which is pretty awful for a card that basically matches the previous generation RTX 2070 in performance.
AMD's Big Navi GPUs aren't much better. The RX 6900 XT sells for $1,800–$2,000, the RX 6800 XT goes for $1,500–$1,700, and the RX 6800 costs $1,350–$1,500. Technically, those are all much better relative prices than Nvidia's Ampere GPUs, but no one should be spending such large sums of money for cards that ostensibly should sell in the $579–$999 range. Oh, and the new RX 6700 XT that has a $479 launch price costs $900–$1,100.
Mainstream and budget GPUs, or at least GPUs that should qualify as mainstream and budget offerings, aren't any better. GTX 1660 variants sell in the $500+ range instead of $200-$250, and GTX 1650 variants cost $400 or more. The same goes for AMD's previous-gen RX 5500 XT and RX 5600 XT.
Winner: NobodyIt's a terrible time to buy a graphics card, and that's been the case since about last August. In retrospect, paying $700 for an RTX 2080 Super last summer would have actually been a 'good' deal, as such cards now generally sell for $1,000. If you can find a card at close to MSRP, AMD and Nvidia match up pretty well, but we probably won't see cards regularly available at or below MSRP until 2022 — and that's assuming the mining profitability drops.
|Round||Nvidia GeForce||AMD Radeon|
|Drivers and Software||✗||✗|
|Price and Value|
AMD vs Nvidia: Bottom Line
With an overall score of four to two, Nvidia continues to hold the crown as the king of the GPU world. Cue the band and light the fireworks. That said, all of this ends up mostly being meaningless when we look at what you can actually buy right now at reasonable prices.
Your best bet if you want a new graphics card might be a pre-built gaming PC. Normally, that's not what we'd recommend, and lead times are still a couple of months at most places. But given prices at places like eBay, you could almost buy a complete PC for the cost of just the graphics card.
The biggest battle right now ends up being the supply of actual cards. There are indications Nvidia sells about ten times as many Ampere GPUs as AMD sells RDNA2 chips, specifically the Steam Hardware Survey and number of sold GPUs at eBay, but getting hard data on the actual number of cards from either vendor remains nebulous at best.
Overall, while AMD lands some solid blows, Nvidia holds the lead in technology, features, and performance (when we factor in DLSS and RT). AMD mostly counters with promises of cards that are even more difficult to find in stock than Nvidia's offerings. Perhaps the best way to buy an AMD GPU illustrates the problem: Buy a console. The supply of PlayStation 5 and Xbox Series X consoles isn't awesome, but Sony and Microsoft appear to be getting more of AMD's silicon than anyone else.
There are still more products planned for this year, like Nvidia's RTX 3050 Ti, RTX 3050, and the fabled RTX 3080 Ti and 3070 Ti. AMD has the RX 6700 likely coming soon, and Navi 23 (RX 6500 series) as well. Those will sell out just as quickly as everything else, but perhaps sometime in 2022 we'll finally see things settle down and a return to normalcy.
- Homes for sale reno
- Sig sauer p365 9mm
- Character modelling in 3ds max
- Farberware 13 piece self sharpening
- Shadow priest bis classic
The war between the two titans of the GPU industry has just been rekindled. The AMD vs NVIDIA debate might have been a one-sided affair for a while, but now AMD is catching up and is ready to make an attempt for the throne.
This GPU wardates back to the1990s. Although AMD has a much longer history in tech, NVIDIA has dominated the field in recent times and is far more successful financially, being valued at about twice as much.
It is worth keeping in mind that AMD devotes a large percentage of its resources to CPUs, giving them less of a focus on GPUs than NVIDIA.
However, history is rarely given much consideration in the world of technology; few people care that AMD has been around since the 1950s. Today, the GPU market has become an NVIDIA versus AMD situation. Scroll down to find the best choice for you.
AMD vs NVIDIA: Performance
If you are considering getting a new GPU, you’re probably curious about the potential performances of each card. Hitting 60 FPS seems like the bare minimum in today’s gaming world, and a good GPU is the key to achieving that performance.
Even so, building a new PC and having it produce the best possible in-game performance isn’t just about getting the best GPU. It’s also important to know that CPU and RAM will need to be on par with the GPU to avoid bottlenecking.
There are three general GPU classifications and each one represents a different part of the market. These are low-end or budget, mid-tier or mid-range, and high-end. As each of these categories is beneficial in different ways, it’s only fair to compare AMD and NVIDIA for each.
For this category, we will look at the RX 5500 XT and GTX 1660 as they are probably the best budget cards that AMD and NVIDIA have to offer in the $200 price range.
They are both good representations of their respective manufacturer’s previous flagship technologies (AMD’s RDNA and NVIDIA’s Turing) and actually stack up quite well.
The reason that we aren’t making this about RDNA 2 and Ampere is that there aren’t any budget cards near this price point.
While the RX 5500 XT offers a better base clock rate at 1685MHz compared to the GTX 1660’s 1530MHz, NVIDIA wisely used this to their advantage and offered a better boost rate at 1785MHz, which is higher than AMD’s game rate, at 1737MHz. Although you are unlikely to notice a difference, it’s interesting to see how this competition has extended to even the smallest details.
AMD further showcases its capabilities with an 8GB GDDR6 RAM, which is definitively better than NVIDIA’s 6GB GDDR5. It also features higher memory bandwidth and more L2 cache, but, as you might have already guessed, NVIDIA uses less power for the GTX 1660.
However, the hardware is nothing without accompanying software and, in this regard, NVIDIA reigns supreme. Despite the previously mentioned specifications favoring AMD, and although it consumes less power, NVIDIA ultimately performs better.
As gamers are far less interested in power consumption than performance, NVIDIA is the winner in this category.
For this category, we will look at AMD’s RX 6700 XT and NVIDIA’s RTX 3060Ti as they are two great options that represent their respective companies’ foray into mid-range GPUs.
This tier’s battle is far more interesting as cards are less evenly matched in the mid-range category than they used to be.
The consensus among the benchmarkers is clear: the RX 6700 XT is simply the better card. The RTX 3060 Ti has a comparable performance and tightens the gap even further at 1440p but, overall, the 6700 XT will simply give you more FPS.
However, one of the most important questions here is the price. The RTX 3060 Ti is considerably cheaper at $399, while the RX 6700 XT’s comes at a $479 price point. This leads to our next question: which card has the better value?
It is questionable whether or not these cards should even be compared but, at this point, they are the best mid-range cards from each manufacturer.
The fact of the matter is that the RTX 3060 Ti’s performance is appropriate for the price level at which it is being offered, which can also be said for the RX 6700 XT. Despite the price difference, RX 6700 XT is the better card, which is a win for AMD.
Here’s where things get a little trickier. This is because AMD didn’t really offer anything that could compete with NVIDIA until the release of the RX 6000 series, which is why NVIDIA has a huge advantage in this area.
However, we can still compare an AMD and an NVIDIA card but first, we need to talk about the elephant in the room.
With their latest generations, both companies released excellent enthusiast-class cards. Although AMD’s RX 6900 XTis only $200 more than NVIDIA’s high-end RTX 3080, and one could argue that these should be in the same bracket, we will compare the RX 6800 XT with the RTX 3080.
These prices are inverted when compared to the mid-range as NVIDIA’s RTX 3080 is the more expensive card. The $50 price difference probably doesn’t seem like much when you’re spending $650-$700, but what makes this particularly interesting is the performance.
Overall, the RX 6800 XT is the better card in terms of raw rasterization performance.
However, the RTX 3080 performs better in the ray-tracing department, which shouldn’t be overlooked. This technology is still relatively new, but it has still been around for long enough to expect better from AMD.
You can still give them some credit as this is their first crack at it, but the fact remains that it didn’t perform as well as NVIDIA. Still, future driver updates and better AMD-oriented game optimization should lead to better peformance.
Ray tracing is still relatively rare on the market, including the amazing new feature from NVIDIA: DLSS. This feature can considerably increase your performance while only minimally reducing visual quality.
In the end, it wouldn’t be fair to give NVIDIA a point here considering a huge percentage of games do not yet utilize ray tracing or DLSS. So, this point goes to AMD.
Total score: AMD 2 – NVIDIA 1
AMD vs NVIDIA: Features
While features may seem less critical than actual specs, they are an important part of what makes a good GPU. Both AMD and NVIDIA offer similar GPUs in terms of hardware and price, but the devil is in the details or, in this case, the features.
If there is something that needs to be talked about, it’s this. Although ray tracing isn’t a requirement for good GPU performance, it makes a clear difference by offering better and more realisticvisuals.
What is ray tracing anyway?
Technical explanations aside, ray tracing is a rendering technique that allows for lighting to be presented more accurately by accounting for variables such as object materials and how lighting is reflected off of them.
Ray tracing awards a point to NVIDIA as they were the first to implement it in their GPUs. With the arrival of Big Navi, AMD has a chance to prove themselves in this regard. So far, the RX 6000 series has performed well enough, but still not as well as the best from NVIDIA.
NVIDIA has pursued ray tracing technology since the 2000s, having introduced it to the world in 2018. This move highlighted their dominance in the GPU market and AMD has yet to fully recover from it.
As the next generation of gaming consoles, the Playstation 5 and Xbox Series X wouldn’t have allowed their names to be associated with something less than top-of-the-line. AMD has been very upfront about its intention to introduce ray tracing in its next GPUs.
While things do look promising for AMD with the introduction of ray tracing, the point here has to go to the market innovator NVIDIA.
Variable Rate Shading
VRS is a technology first brought to the market by NVIDIA that has found its best use in VR. It essentially calculates which frames in your field of view are going to be fully shaded or even rendered at all. This significantly reduces the burden on the GPU, redirecting that extra energy to more useful things.
AMD still hasn’t incorporated this tech into their GPUs, but it’s heavily rumored that it will be introduced in their RDNA 2 line, as AMD filed for the patent for VRS all the way back in early 2019.
There have also been talks of perfecting eye-tracking technology and using that to further improve upon VRS, which sounds like a sci-fi idea.
As we were able to see this cool piece of technology in action on NVIDIA’s side, it deserves this point too.
Deep Learning Super Sampling
Designed as another way to increase the efficiency of the GPU, DLSS is a path-breaking piece of technology. It can even be considered ahead of its time because of the process required to fully enjoy its benefits.
The biggest issue here is that game developers are required to enable DLSS support when making the game. In order for the player to see the improvement, it needs to be sent to NVIDIA, who then let an AI run through the game, analyze the images, and automatically upscale it to a higher resolution.
Initially, the fact that NVIDIA does the heavy lifting was one of the biggest drawbacks of DLSS. The process wasn’t as optimized as it needed to be and this resulted in DLSS looking like an interesting idea rather than an effective technology.
With the arrival of NVIDIA’s Ampere architecture, this process was streamlined and we certainly expect further improvements.
G-Sync vs FreeSync
These are NVIDIA’s and AMD’s adaptive synchronization technologies designed to eliminate screen tearing during gameplay. Screen tearing occurs when the GPU’s output is mismatched with the display’s refresh rate.
The communication between the GPU and the monitor basically works in this way: if the monitor refreshes at 60Hz, it requires 60 frames to be sent from the GPU (same for 120Hz, 144Hz, and so on).
The issue usually occurs when the GPU is unable to produce the required frames, causing screen tearing.
Adaptive sync technology allows for the GPU to effectively change the monitor refresh rate, depending on the number of frames being produced.
If the game dips to 40 FPS, the GPU will limit the monitor to refresh at 40Hz. However, this doesn’t make the game run any smoother, it just prevents screen tearing.
In the not so distant past, the solution for this was software-based, most notably with VSync, but this is being phased out in favor of newer technologies.
G-Sync is NVIDIA’s solution to screen tearing and has drawn some criticism. As the first to experiment with adaptive sync tech, NVIDIA used that to their advantage, although with some hardware requirements. Monitors must be G-Sync compatible and, although not specifically stated, this brought an extra cost of anywhere from $100 to $300 in price.
To run G-Sync, monitors require a proprietary NVIDIA G-Sync scaler module which means they will all have similar on-screen menus and options. This is AMD’s FreeSync’s biggest advantage: by using the Adaptive Sync standard built into the DisplayPort 1.2a specification, it allows manufacturers to choose any cheaper scaler.
Nevertheless, G-Sync has a better way to handle the GPU producing too many frames. It will lock the GPU’s frame rate upper limit to that of the monitor. FreeSync, when in-game VSync is turned off, will allow the GPU to produce extra frames. This can lead to screen tearing but also lowers the input lag.
The biggest issue which divides the community is the fact that not all NVIDIA cards work with FreeSync monitors, just as not all AMD cards will work with G-Sync monitors. This is being ironed out, but the fact remains that you will need to check if your monitor will work properly with your GPU.
While both sides have their pros and cons when it comes to adaptive sync technology, the fact that FreeSync is more readily available is what ultimately earns AMD the point here.
Total score: AMD 3 – NVIDIA 4
Related:FreeSync vs G-Sync – Which Is Best?
AMD vs NVIDIA: Drivers And Software
The fact of the matter is that good hardware requires good software. Drivers are programs that control how a certain device (such as a GPU) interfaces with the CPU. It allows the software to use the hardware it controls to the best of its ability without needing to control every aspect of how that particular part operates.
As previously mentioned, AMD’s RX 5000 series failed miserably when it launched, with driver issues creating black screens and crashes. Unfortunately, this issue persists despite newer drivers attempting to fix it. NVIDIA’s issues are equally problematic, as they are often smaller and therefore more difficult to identify.
AMD has significantly improved its driver capabilities with its yearly Radeon Adrenalin updates. The 2020 version allegedly offers an impressive 12% improvement over the 2019 version.
AMD also makes a conscious effort to simplify things and only uses one piece of software to update its drivers. They have followed a schedule of at least once-per-month with major releases.
The biggest setback for AMD is their products’ persistent issues, which take a long time to fix.
NVIDIA’s driver update schedule consists of two diverse applications to control their hardware.
Their NVIDIAControl Panel enables the configuration of aspects such as 3D settings and display resolution.
GeForce Experience handles game optimization, driver updates, and extra features.
The NVIDIA Control Panel has not seen a UX or UI change for more than a decade. The design is outdated and it can be incredibly slow at times.
GeForce Experience in general sounds like a great idea, but it isn’t what users hoped it would be. Users must log in to use the available features such as automatic driver updates, recording, FPS counter, etc. For many, GE is considered to be bloatware.
In comparison, Radeon Software is much quicker, far more intuitive, requires no account and provides other useful features such as Radeon Chill, Radeon Boost, manual and automatic overclocking, undervolting, manual fan curve, and more.
In the end, while AMD has its downsides, the efficient simplicity with which their software operates earns them a point in this category.
Total score: AMD 4 – NVIDIA 4
AMD vs NVIDIA: Power Consumption and Efficiency
When AMD introduced Navi and announced their gamble at TSMC’s 7nm FinFET process, they probably thought this amazing 50% per watt performance would bridge the efficiency gap.
Unfortunately, this was not the case: Navi didn’t even outperform older NVIDIA GPUs that were built on the TSMC’s last-gen 12nm node.
The future looks brighter for AMD as its RDNA 2 cards were able to produce another 50% upgrade over RDNA.
This isn’t completely black and white. In the extreme performance range, NVIDIA’s RTX 3090 uses a lot of power with only the RX 6900 XT coming close.
We could say that RX 6800 XT is not as power-draining as the RTX 3090, but that would completely disregard the fact that the latter is simply a better GPU.
This does switch in the medium range where the RTX 3070 is the better performer, but only with a slight margin. That margin is the same in terms of overall performance so we can’t really chalk this up as an NVIDIA win.
For now, NVIDIA is simply a better performer in the budget category and there isn’t much room for debate here. Both companies are preparing their respective budget cards for their latest generation, so this section might need an update soon.
NVIDIA narrowly edges out AMD in the performance-per-watt metric in their latest generations, but also for the previous few. Although AMD is catching up, it should be concerned with NVIDIA’s efficiency at using previous-generation lithography.
Total score: AMD 4 – NVIDIA 5
AMD vs NVIDIA: Dollar Value
While top-level performance is what most gamers are looking for in their GPUs, the price also needs to be considered. As previously discussed, there are three basic categories for both price and performance.
NVIDIA has a clear advantage in the extreme price range performance-wise, but that gap is gradually closing with AMD’s RX 6900 XT coming close to the RTX 3090 while being a whole $500 cheaper.
If we drop a level below, where the RX 6800 XT is $50 cheaper than the RTX 3080, it wouldn’t be unreasonable to claim it’s a better option.
However, due to the RTX 3080’s better ray-tracing performance, it’s hard to outright say that either is better. Due to the price difference, let’s say that AMD has a slight edge in this price bracket.
Dropping further down to the mid-range market, we have a much clearer picture. Although the RX 6800 is more expensive than the RTX 3070, it is also a better card and its cost is justified. However, as the question here is the dollar value, we feel that both cards perform appropriately for their prices so we will call this one a tie.
Although the comparisons seem mostly evenly matched, the performance per dollar favors AMD, which is why they get a point this time around. Keep in mind that it was a close match.
And The Winner is…
With a total score of 5 to 5, there is no better card manufacturer. Of course, the AMD vs NVIDIA debate is subjective and you shouldn’t blindly purchase either company’s card. The general advice before making any sort of investment in a PC part is to know your requirements and budget.
You should carefully research all available cards before making your final decision. In both company’s latest generations, the answer to which you should buy essentially boils down to ray tracing. If you don’t consider that of the utmost importance then AMD is a better option. However, if you want the highest possible image quality, you should go with NVIDIA.
Best GPU Temperature For Gaming
Alex is a Computer Science student and a former game designer. That has enabled him to develop skills in critical thinking and fair analysis. As a CS student, Aleksandar has very in-depth technical knowledge about computers, and he also likes to stay current with new technologies.
The 2010s haven’t been a good time for AMD, that’s for sure.
Besides struggling with their FX series of CPUs that were years behind what Intel was selling, AMD also had trouble competing with Nvidia in the GPU market. “Team Green” dominated the high-end and AMD could only keep up in the low-end and mid-range.
AMD only made a comeback in 2017, starting with the release of their long-awaited Ryzen CPUs that are now a very popular choice for gaming builds.
In 2019, AMD launched its 7nm RDNA-based GPUs that are now to be succeeded by new RDNA2 models.
In this guide, we’ll see just how well AMD manages to compete with Nvidia in 2021 and, ultimately, which company offers better gaming GPUs at the moment.
Pricing & Performance
Naturally, the main question when discussing GPUs is performance. How does it score in benchmarks, and what kind of performance can it manage in different resolutions?
Well, it’s impossible to generalize on this subject since performance obviously varies wildly depending on the model you get and the money you’re willing to spend on a GPU.
AMD’s budget GPUs used to almost consistently outperform whatever Nvidia could offer at the same price point. At the moment, however, when comparing the latest budget offerings from both companies, Nvidia pulls slightly ahead. Ironically, though, some of AMD’s old Polaris-based RX 500 cards are still the best picks if you’re pinching pennies, as they offer good value for your money.
As for Navi, the RX 5500 XT was a bit disappointing. The 4 GB ($169) and 8 GB ($199) variants of the card were on roughly even terms with Nvidia’s GTX 1650 Super ($159) and could barely come close to the GTX 1660 Super ($229), which ultimately made the RX 5500 XT a very unappealing card for those looking to get the best value for their money in this price range.
However, since the latest-gen RTX 3000 and RX 6000 budget models aren’t out yet, now is not an ideal time to build a budget gaming PC, especially if you want it to be a bit more future-proof.
Meanwhile, in the mid-range, the competition was a bit tighter, and AMD’s beefier RDNA-based GPUs could more than hold their own against both the GTX and the RTX cards that Nvidia is offering at the moment.
The RX 5600 XT ($279) performed significantly better than the similarly priced GTX 1660 Ti ($279), all the while keeping up with the original RTX 2060 ($349). However, the upgraded RTX 2060 Super ($399) did have a slight upper hand, plus it also had ray tracing. In turn, the RX 5700 XT ($399) remained a great overall value pick in this range to compete with the noticeably pricier RTX 2070 Super ($499).
However, the new RTX 3060 Ti($399) blows them all out of the water, as it offers performance comparable to last-gen high-end models such as the RTX 2080 Super. Until AMD comes forward with an RX 6700 XT or something else that can compete at the aforementioned price point, the RTX 3060 Ti will remain the best mid-range card currently on the market.
When it comes to high-end, it used to be the same old story – Nvidia pretty much had a monopoly here, as their RTX 2070 Super, RTX 2080 Super and RTX 2080 Ti dominated the market. While the RX 5700 XT could compete with the RTX 2070 Super and offered very good value for money, AMD had nothing that could challenge Nvidia’s flagship cards.
However, that changed.
AMD’s new RX 6000 series have proven to be more than capable of taking on Nvidia in the high-end. The RX 6800 ($579) outperforms the RTX 3070 ($499) by a noticeable margin in most games, thus potentially justifying the slightly higher price.
Meanwhile, the RX 6800 XT ($649) generally goes toe-to-toe with the pricier RTX 3080 ($699), although Nvidia does have a slight lead here. The AMD GPU does ostensibly offer better value for your money, but if you’re aiming at a high-end GPU in this price range, chances are that a $50 difference won’t matter much.
Finally, we have the best of the best: the RX 6900 XT ($999) and the RX 3090 ($1499), both of which are very niche products and unlikely to appeal to a broader audience.
The RX 6900 XT pulls slightly ahead of the RTX 3080 in terms of performance, but at a significant price premium. The overly expensive RTX 3090, with its whopping 24 GB of GDDR6X memory, will be more appealing to professional users rather than gamers.
Now, keep in mind that we used MSRP pricing and general performance for reference when commenting on the aforementioned GPUs. Performance will inevitably vary from game to game and the pricing will also depend on the model as well as on the market conditions.
If you’re shopping for a new graphics card right now, you might want to check out our selection of the best graphics cards of 2021. Note that we do our best to keep our buying guides up to date, so if you notice some outdated info, that means the article is likely slated for an update in the near future.
Real-Time Ray Tracing – Is It Worth It?
The most heavily marketed new feature of the 2018 Turing GPUs were their real-time ray tracing capabilities. So, what is real-time ray tracing and is it worth it in 2021?
As the name implies, with real-time ray tracing, in-game lighting can be rendered much more realistically as the GPU traces the paths of virtual rays of light and thus simulates the way that they interact with objects in the environment more accurately. Naturally, the benefits of ray tracing are most noticeable when there are a lot of reflective surfaces around.
For the past two years, ray tracing was an Nvidia-exclusive feature found only in their RTX GPUs, although AMD has now introduced ray tracing support with their latest RX 6000 models.
Now, there’s no denying that ray tracing is an important step forward when it comes to the ongoing decades-old quest for photorealistic graphics. But is it really that big of a deal at the moment?
Well, there are several factors that put the brakes on the ray tracing hype train, and those are the following:
It is demanding on the hardware. When turned on, ray tracing can deliver a big performance hit, sometimes outright cutting the FPS in half. This performance hit is most noticeable with weaker GPUs, and it also varies from game to game.
The benefits aren’t always obvious. Sure, tech demos and segments designed to show off ray tracing will look amazing with this feature turned on, but when there are no reflective surfaces around, ray tracing will offer little to no discernible improvement in terms of visuals.
Not all games support it. As of 2021, more games support ray tracing than before, but it is still far from being a universal feature, even when it comes to big AAA titles.
It’s worth noting that Nvidia GPUs do still have the advantage when it comes to ray tracing performance, especially in games that support DLSS. So, if you really care about ray tracing, Nvidia would probably be a better pick, although the new RX 6000 cards do fare admirably in terms of overall in-game performance and value.
Ray tracing is definitely becoming increasingly popular, what with both the PlayStation 5 and the Xbox Series X being compatible with it. But in any case, as we mentioned above, it’s still far from being a mainstream feature.
VRR – AMD FreeSync vs Nvidia G-Sync
While V-Sync can be good enough for 60Hz monitors, it’s simply not viable for monitors with high refresh rates such as 120 Hz, 144 Hz or 240 Hz, among others.
This is because V-Sync prevents screen tearing by imposing a cap on the number of frames that the GPU dishes out, thus ensuring that the framerate and the monitor’s refresh rate never fall out of sync. However, it comes with its share of drawbacks, such as stuttering and input lag.
Needless to say, if you intend to get a monitor with a high refresh rate, it will inevitably come with one of two technologies: AMD FreeSyncorNvidia G-Sync.
At their core, these technologies function in a similar way – they both use hardware to ensure that the refresh rate of the monitor is always the same as the framerate. This prevents the two from ever falling out of sync regardless of how wildly the framerate might fluctuate, thus removing screen tearing without any stuttering or input lag.
However, there’s a downside to everything.
Nvidia G-Sync monitors tend to be more expensive for a number of reasons. First, Nvidia runs tight quality control, and all G-Sync monitors need to meet their standards before being approved. Moreover, OEMs have to pay licensing fees to use G-Sync, and they have to buy the scaler modules directly from Nvidia, as they are the only ones who make them.
A scaler module is a piece of hardware built into the monitor that makes VRR possible. While Nvidia essentially has a monopoly on G-Sync scalers, AMD allows OEMs to use third-party scaler modules and they don’t require them to pay licensing fees to use the technology.
As a result, FreeSync is much more popular and readily available in cheaper monitors, but its implementation isn’t always flawless, and some monitors will only work in a specific range.
On the bright side, there are many G-Sync Compatible monitors out there now i.e. monitors that don’t use Nvidia’s scaler modules and didn’t go through their testing process but are still compatible with G-Sync. However, they lack some of the features that you’d get with a certified G-Sync display such as ultra low motion blur, overclocking or variable overdrive.
Compatibility also used to be a big issue in the past as FreeSync only worked with AMD GPUs and G-Sync only worked with Nvidia GPUs. But the situation isn’t as black and white anymore. For instance, Nvidia GPUs now work with FreeSync monitors. AMD GPUs aren’t yet compatible with G-Sync, but this will also change soon.
At the end of the day, both of these technologies will get the job done, but FreeSync is obviously a budget choice, while G-Sync is the premium one. The proprietary nature of G-Sync still makes it quite expensive, but Nvidia is slowly shifting to a more liberal approach. So really, who knows what might happen further down the road.
So, all things considered, which company currently offers better GPUs, Nvidia or AMD?
The answer is neither.
Well, simply because it’s impossible to make generalizations without comparing specific GPU models. Both companies offer different solutions at different price points that can suit the requirements and budget constraints of a wide range of customers. On top of that, the situation can change drastically all the time.
Competition is definitely a lot better than it has been over the past few years, that much is certain. AMD used to be the definitive choice for budget and mid-range builds while Nvidia had a monopoly in the high-end. Now things are changing as Nvidia is offering more competition in the lower price ranges while AMD is finally taking Nvidia on in the high-end.
Overall, the two are on fairly even terms now. Ultimately, the more AMD manages to close the gap and provide adequate competition across the entire price spectrum, the better it will be for every consumer’s wallets.
HDMI vs DisplayPort vs DVI vs VGA – Quick, Simple and Easy Explanation
Samuel is GamingScan's editor-in-chief. He describes himself as a dedicated gamer and programmer. He enjoys helping others discover the joys of gaming. Samuel closely follows the latest trends in the gaming industry in order to keep the visitors in the flow.More About Samuel Stewart
Nvidia to gpu amd compared
GPU Benchmarks and Hierarchy 2021: Graphics Cards Ranked
Our GPU benchmarks hierarchy ranks all the current and previous generation graphics cards by performance, including all of the best graphics cards. Whether it's playing games or doing high-end creative work like 4K video editing, your graphics card typically plays the biggest role in determining performance, and even the best CPUs for Gaming take a secondary role.
The following table sorts everything solely by our performance-based GPU gaming benchmarks. We have a separate article that lists the best graphics cards, which looks at all factors, including price, graphics card power consumption, and overall efficiency. For this GPU benchmarks hierarchy, the most recent addition is AMD's Radeon RX 6600 XT that launched on August 11, along with the GeForce RTX 3080 Ti and GeForce RTX 3070 Ti that launched in June. The next card slated for launch looks to be the Radeon RX 6600, currently rumored to launch on October 13.
Deals on new graphics cards are still non-existent. Everything is selling out, at higher than "suggested" pricing, so there's no incentive for any company to cut prices. In fact, most of the AIB (add-in board) graphics card manufacturers have jacked up prices by 30% or more relative to the AMD and Nvidia 'official' starting prices. With continuing chip shortages across the globe affecting all industries, we don't expect pricing to improve on graphics cards until 2022. The profitability of Ethereum mining continues to be a major influence on GPU prices, despite the volatility of Ethereum and Bitcoin prices.
If you're looking to buy a graphics card right now, unfortunately the options are limited. You can always give the Newegg Shuffle a try, but fair warning: I've done that on most days and have only been selected for the opportunity to buy an overpriced bundle three times, and if you limit your selections to the better deals, your chances will be substantially worse. Otherwise, eBay has cards, at even more extreme prices — our GPU price index has the details on what you should expect to pay there, which usually means 50% to 100% over MSRP. With the chip and substrate shortages expected to last well into 2022, by the time retail outlets have inventory just sitting on the shelves, we might be looking at Nvidia's 5nm Lovelace GPUs and AMD's RDNA3 architecture.
Which graphics card do you need? To help you decide, we've created this GPU benchmarks hierarchy consisting of dozens of GPUs from the past four generations of hardware. Everything is ranked from fastest to slowest, using the results from our test suite consisting of nine games for our GPU benchmarks, running at 'medium' and 'ultra' settings with resolutions of 1080p, 1440p, and 4K. For comparison purposes, the fastest card, based on the combination of all nine GPU benchmarks, three resolutions, and two settings, gets normalized to 100 percent, and all others are graded relative to it.
Not surprisingly, the fastest cards use either Nvidia's Ampere architecture or AMD's Big Navi. We're not testing with ray tracing or DLSS here, as most of the previous generation cards don't support those features, but even in traditional rasterization rendering the new GPUs come out on top. You can check out full launch reviews of Nvidia's GeForce RTX 3090, GeForce RTX 3080 Ti, GeForce RTX 3080, GeForce RTX 3070 Ti, GeForce RTX 3070, GeForce RTX 3060 Ti, and GeForce RTX 3060 12GB; and there's also AMD's Radeon RX 6900 XT, Radeon RX 6800 XT, Radeon RX 6800, Radeon RX 6700 XT, and now Radeon RX 6600 XT. At present, only one of the ten highest performance GPUs doesn't use either Ampere or RDNA2 — and that's the Titan RTX, which hardly counts.
Of course it's not just about playing games. Many applications use the GPU for other work, and we've covered some professional GPU benchmarks in our RTX 3090 review. But a good graphics card for gaming will typically do equally well in complex GPU computational workloads. Buy one of the top cards and you'll can run games at high resolutions and frame rates with the effects turned all the way up, and you'll be able to do content creation work equally well. Drop down to the middle and lower portions of the list and you'll need to start dialing down the settings to get acceptable performance in regular game play and GPU benchmarks.
It's not just about high-end GPUs, of course. We also recently tested Intel's Xe Graphics DG1, which basically competes with integrated graphics solutions. The results aren't pretty, and we didn't even try running any of those at settings beyond 1080p medium. Still, you can see where those GPUs land at the very bottom of the GPU benchmarks list.
Again, all of the games and settings we're using for testing have to conform to the lowest common denominator. That means ray tracing and proprietary tech like Nvidia's DLSS aren't enabled, even where they're supported. You can see how the GPUs stack up in DXR performance in our AMD vs. Nvidia ray tracing article, and we've also included RT and DLSS results in recent reviews like the RX 6600 XT, RTX 3070 Ti, and RTX 3080 Ti launch reviews. None of those scores are factored into the GPU benchmarks rankings, but the short summary is that Nvidia is usually quite a bit faster at RT, and DLSS provides a significant boost to performance for a minimal loss in image quality.
If your main goal is gaming, you can't forget about the CPU. Getting the best possible gaming GPU won't help you much if your CPU is underpowered and/or out of date. So be sure to check out the Best CPUs for gaming page, as well as our CPU Benchmarks Hierarchy to make sure you have the right CPU for the level of gaming you're looking to achieve.
GPU Benchmarks: Which Cards Ranked Highest?
The Nvidia GeForce RTX 3090 takes top honors for raw performance, with a composite score of 152.7 fps across all 54 tests. That's the 100% mark, though it's worth noting that it also scored 98.7 fps at 4K ultra. It's nominally a $1,500 graphics card, which is out of reach of most gamers, but current shortages have rocketed pricing up to the $2,500 range. So much for "less than Titan" affordability.
Not too far behind the 3090 are the GeForce RTX 3080 Ti, Radeon RX 6900 XT, Radeon RX 6800 XT, and GeForce RTX 3080, theoretically priced at $1,200, $1,000, $650 and $700, respectively (good luck finding any of those in stock for anything close to official launch prices). The RTX 3080 Ti lands in an odd spot, with only slightly lower pricing and performance than the 3090. It's basically at the old Titan price of $1,200, but at that point why not just spend the extra $300 for the 3090? Similarly, the 6900 XT is a minor bump in performance for a relatively large bump in price compared to the 6800 XT, and we'd generally recommend sticking with the latter.
The 6800 XT is also technically faster (barely, by a basically meaningless amount) than the RTX 3080 by our ranking formula, though as mentioned above, ray tracing and DLSS very much change the picture. Add those in and the 3080 easily beats even the 6900 XT. This is why we continue to rank the RTX 3080 as the best overall graphics card, though that's contingent on actually finding one for a price at least somewhat close to the $700 MSRP (anything under $1,000 would be worth a shot these days).
The new GPUs make all of AMD's and Nvidia's previous generation GPUs look a bit weak. The Radeon RX 6800, GeForce RTX 3070 Ti, and GeForce RTX 3070 match or beat the outgoing RTX 2080 Ti with a theoretical starting price of just $580, $600, or $500, respectively. The RTX 3060 Ti meanwhile leads the old 2080 Super in performance and potentially costs 42% less. Only the RTX 3060 12GB seems a bit lackluster, with performance basically at the level of the old RTX 2070 (non-Super) that launched 2.5 years ago. If the $329 official launch price were anywhere to be found, it would be a great deal, but it's not.
AMD's Navi 21 GPUs, aka Big Navi, finally break into the top three overall, even including Titan cards. That's something AMD hasn't managed since the Vega 64 launch (where it came in third). AMD is also mostly at feature parity with Nvidia now, with both companies supporting ray tracing. Except, Nvidia has Tensor cores that help with other tasks like DLSS, Nvidia Broadcast, RTX Voice, and potentially future applications, plus Nvidia's ray tracing performance is definitely still faster in the majority of DXR (DirectX Raytracing) games. The first Navi 22 card, the RX 6700 XT, lands in the ten spot if we discount the Titan RTX, while the first Navi 23 card, the RX 6600 XT, sits down in position 20, just ahead of its RX 5700 XT predecessor.
If you're in the market for a new sub-$500 graphics card right now, the best options are the RTX 3060 Ti and the RX 6700 XT. Neither one actually sells for under $500, sadly, but if you keep searching you can probably find a card for close to $500. The RTX 3060 Ti is only a bit slower than the RTX 3070, while the RX 6700 XT falls between the two Nvidia cards.
If you're willing to pay eBay prices, right now the RX 6700 XT costs around $700, as does the RTX 3060, while the RTX 3060 Ti costs closer to $1,000. Again, whatever GPU you're hoping to buy, it's still a terrible time to buy a graphics card, as all of the most desirable GPUs are either out of stock or seriously overpriced. In the Newegg Shuffle, for example, we routinely see graphics cards priced at 25% to 50% above the nominal MSRP, and your chances of getting selected are quite slim in our experience.
If you can find a reasonable deal on a latest generation GPU right now, great! But don't pay more for a previous gen GPU just because there aren't enough RX 6000-series or RTX 30-series GPUs to meet the current demand. Eventually, supply will catch up, and that will be the right time to buy. If you can't wait, our advice is to just try and find any old GPU that still works to hold you over. Based on current eBay prices, the best FPS per dollar card you can find — used — is the relatively ancient GTX 970. <Sigh>
That brings us to the bottom third of the list, the home of budget GPUs like the GTX 1650 Super, RX 5500 XT, and more. These cards give up a lot of performance in order to keep pricing down, and there are older generation GPUs that can perform just as well (or better) if you shop around. But component shortages have affected even these, with $300 and higher prices even on relatively weak cards like the GTX 1650 and RX 5500 XT 4GB — and we can't blame miners, as mining performance on 4GB cards is very poor these days.
Theoretically, the GTX 1660 Super, GTX 1650 Super, and RX 5600 XT are the best budget options, or at least they used to be before prices launched into the stratosphere. The higher you go on price, the worse things get, so take a careful look at historical pricing before you buy anything. GTX 1660 Super should cost $250, and GTX 1650 Super should cost around $175. The best prices we can see right now are around $450 and $375, respectively. Hard pass.
Unless you already have the hardware, or can get it for cheap, we don't recommend going below the GTX 1650 Super. Wait out the current shortages, and spend some time with indie games that can often run just fine on ... well, practically anything, even Intel's integrated graphics solutions! And you don't even need to buy a graphics card if you go that route. AMD's APUs are an even better option if you're on an extreme budget.
If you're looking at something like an RX 550 or GT 1030, you should consider AMD's integrated graphics on its Ryzen APUs as a viable alternative. If you have an older PC and are looking at adding a GPU, a motherboard and CPU upgrade might end up being a better option. Or not, as even a basic motherboard, CPU, and RAM can set you back $200 or more. Plus, the APUs are also sitting at inflated prices now. <Sigh again>
AMD did launch its latest Ryzen 5000G series processors, with the fastest integrated graphics we've seen so far. Prices aren't too bad, all things considered, but you get performance roughly on the level of the GT 1030 — a card that can actually be found in stock for around $125. Meaning, you could skip the full motherboard, RAM, and CPU upgrade and just get a low-end dedicated card to try and pass the time until things return to 'normal.'
Also worth noting is that the scoring assigned to each GPU uses all six test resolutions and settings, except on integrated graphics where we scale the result — because, come on, no one is going to try and run Borderlands 3 at 4K on an iGPU. (It will probably just crash.) If you want to check performance at just 1080p medium, or one of the other options, you can see the ranking order for the main GPUs in the charts below.
Test System for GPU Benchmarks
Our overall GPU benchmarks scores are based on the average frames per second (fps) of our testing of Borderlands 3, The Division 2, Far Cry 5, Final Fantasy XIV, Forza Horizon 4, Metro Exodus, Red Dead Redemption 2, Shadow of the Tomb Raider, and Strange Brigade. If you want to do your own GPU benchmarking, see our complete list of the best GPU tests, which includes a lot more games and synthetic tests as well.
That's nine games, six settings and over 40 cards from the current and previous generations. We have a solid mix of game genres and APIs, plus AMD and Nvidia promoted titles, making this the definitive GPU benchmarks and performance hierarchy for gaming purposes. Due to the mix of various generations of GPUs, note that we don't include ray tracing or DLSS testing in any of the figures. That does penalize Nvidia's RTX cards quite a bit, and the RX 6000 series as well, since previous generation GPUs can't even try to run ray tracing in most games.
GPU Benchmarks and Performance Hierarchy Charts
Here you can see the average performance charts for our testing at 1080p, 1440p, and 4K (medium and ultra on all three). If you want to see the full suite of individual game tests, check out the charts in our Best Graphics Cards article. We've focused on the 'executive summary' and have omitted individual game charts as well as a few GPUs that don't fully qualify. We've left off the integrated graphics solutions as well as many older GPUs. That gives us 28 GPUs in the charts, color coded for your viewing pleasure. You can find additional charts with the 'retired' GPUs below the main charts.
Again, our GPU benchmarks scoring uses the average of all 54 scores (nine games, three resolutions, two settings). Including all 54 scores means the fastest cards are somewhat penalized because they run into CPU limitations at 1080p and even 1440p — especially at medium settings — and the slower GPUs can also end up penalized because they were never intended to run games at 1440p or 4K — especially at ultra settings.
If you intend to play at 1440p or 4K, the charts below can help you focus in on just those results. For example, the RTX 3080 overall scored 20.8% higher than the RTX 2080 Ti, but if you only look at 4K ultra performance, it's 33.5% faster.
Here's the same information as above, this time with all the older GPUs (which we generally aren't retesting on a regular basis these days). This chart also includes all the Titan GPUs, which we don't generally count as full participants in our GPU rankings due to cost. Not that price seems to matter much these days...
AMD vs. Nvidia: the midrange graphics card battle is heating up
AMD has been trying to put a dent in Nvidia’s gaming GPU dominance for years. So far, its attempts — like the Radeon VII and its new 7nm process, which are designed to compete with Nvidia’s RTX 2080 — have fallen a little flat. Now, AMD is setting its sights on Nvidia’s midrange cards with its Radeon RX 5700 ($349) and RX 5700 XT ($399). They’re both built on AMD’s new RDNA architecture, and the big promise is that it can provide up to 1.25x performance per clock and 1.5x performance per watt compared to the Graphics Core Next (GCN) architecture that powered its previous Polaris and Vega GPUs.
Price, performance, and power usage are key at the $300 to $400 price points, and AMD is trying to find a delicate balance of all three with the RX 5700 series. Not to be outdone, Nvidia has also launched its upgraded midrange RTX cards: the RTX 2060 Super ($399) and RTX 2070 Super ($499). These cards offer modest speed bumps over the existing RTX cards they are replacing at the same price points. They were also enough to force AMD’s hand to cut the price of the Radeon RX 5700 and 5700 XT graphics cards just two days before their release last month.
That price cut was hugely important because if the Radeon RX 5700 XT had debuted at its original $449 price instead of $399, then Nvidia’s $499 RTX 2070 Super would have been easily worth the $50 upgrade. As it stands, AMD’s aggressive pricing makes the choices at the midrange a lot more difficult, especially as the RX 5700 and 5700 XT are now going head-to-head with Nvidia on performance.
I’ve been testing Nvidia and AMD’s latest midrange cards over the past couple of weeks, with the intention of finding a clear winner for 1440p gaming. While AMD has failed to match Nvidia in terms of performance at the high end, competition at the midrange couldn’t be closer. The graphics card wars are finally heating up.
AMD’s new cards are both based on the company’s 7nm process, RDNA architecture, and use 8GB of GDDR6 graphics memory. AMD might be using a new architecture here, but the company isn’t pushing any real new hardware advantages to using RDNA just yet. There’s no support for ray tracing or promises of AI-powered anti-aliasing. AMD is focusing on raw performance instead of eye candy.
AMD’s big architecture changes do mean that the company has higher bandwidth memory and improvements to its compute units. We’ve seen AMD use 7nm for the Radeon VII, but this card was based on the previous GCN architecture. AMD claims that RDNA cards will have 50 percent better performance than previous GCN cards, which could mean we’ll see some interesting developments in the high-end of the market soon.
That 8GB of VRAM is the same amount found on both the RTX 2060 Super and RTX 2070 Super. Nvidia has done very little in terms of hardware to its upgraded RTX Super cards, but there are now more raw CUDA cores and higher clock speeds. Both the RTX 2060 Super and RTX 2070 Super look physically the same as their predecessors, though, which means they’re very well-built and look great.
AMD and Nvidia midrange GPUs (2019) specs
|Specs||GeForce RTX 2060 Super||GeForce RTX 2070 Super||Radeon RX 5700||Radeon RX 5700 XT|
|SP compute||7.2 TLFOPS||9.1 TFLOPS||7.9 TFLOPS||9.7 TFLOPs|
|Memory||8GB GDDR6||8GB GDDR6||8GB GDDR6||8GB GDDR6|
|Architecture||Turing 12nm TSMC||Turing 12nm TSMC||RDNA 7nm TSMC||RDNA 7nm TSMC|
While Nvidia moved to a new dual-fan setup for its Founders Edition RTX cards last year, AMD has stuck to its blower-style on its reference cards. This feels like an obvious mistake because the RX 5700 and RX 5700 XT both run hotter and are way louder than Nvidia’s RTX competition as a result. Blowers have benefits for space-constrained cases, but for everyone else, a dual- or triple-fan system will be far more efficient. On the 5700 XT, there’s also a dent in the casing, which is designed to improve airflow, but it just ends up looking strange rather than providing an obvious improvement to how the card performs.
Both of AMD’s new cards also support PCIe 4.0, but you’ll need a new motherboard and AMD’s third-generation Ryzen CPUs to see the benefit from that right now. While PCIe 4.0 doubles the bandwidth of PCIe 3.0, this isn’t really required at current resolutions and refresh rates, and it will become far more important in the future when resolutions push beyond 4K. In a card designed for 1440p monitors, it’s nice to have for future-proofing, but it’s totally unnecessary. Fortunately, it is backwards compatible with PCIe 3.0, so you can still use the 5700 cards with existing motherboards and CPUs.
I’m genuinely impressed by how well both of these new AMD cards perform in 1440p gaming. In Far Cry 5, frame rates regularly exceeded 90 fps on max settings, and the 5700 XT outperformed Nvidia’s RTX 2060 Super and came close to the higher-priced 2070 Super in most of my tests. However, I did find that frame rates between the AMD and Nvidia cards can vary a lot depending on the game.
Destiny 2 performed a lot better on both the 2060 Super and 2070 Super than AMD’s 5700 series in terms of raw frame rates, but Division 2 seems to be better optimized for AMD’s cards. I ran both at maxed-out settings, and all of the cards handled well. Likewise, the Fire Strike Extreme benchmark performed far better on both the 5700 XT and 5700 than Nvidia’s cards. Time Spy ran better on Nvidia’s cards.
AMD vs. Nvidia midrange 2019 benchmarks
|Benchmark||GeForce RTX 2060S||GeForce RTX 2070S||Radeon RX 5700 XT||Radeon RX 5700|
|Fire Strike Extreme||9,782||10,784||10974||10016|
|Far Cry 5||87fps||96fps||94fps||87fps|
|Shadow of the Tomb Raider||50fps||59fps||50fps||45fps|
I tested all games with a 27-inch Asus ROG Swift PG279Q monitor (G-Sync disabled) at 1440p resolution and max settings. Shadow of the Tomb Raider, a recent DirectX 12 title, performed similarly on both the $399 5700 XT and the $399 RTX 2060 Super.
I was most surprised with the RTX 2070 Super performance, bringing it very close to what the standard RTX 2080 can offer. If you’re looking at spending $699 on the new RTX 2080 Super, then I would definitely look at the 2070 Super, which has most of the performance of an RTX 2080 for $200 less. I briefly tested the 2080 Super, too, and the performance gains over the regular RTX 2080 weren’t as significant as the 2060 Super and 2070 Super.
All four cards seem ideal for 1440p gaming, even at max settings in modern games. You can obviously adjust the settings down to improve frame rates for a 144MHz monitor setup or get usable frame rates with all the eye candy enabled.
As with all games and tests, I’d expect both AMD and Nvidia’s cards to perform better over time once titles are optimized and drivers are tweaked and improved. Even without that, the performance gap in the midrange has definitely shrunk, thanks to AMD’s new cards, though.
This surprisingly close performance at the same $399 price point makes the decision between AMD or Nvidia a lot more difficult. That choice will probably come down to whether you value less heat, less power consumption, and a lot less noise from your PC.
During my testing, both of AMD’s cards regularly exceeded 80C, while Nvidia’s cards hovered around 73C under load. AMD is using a different way to measure temperature than Nvidia, which does result in higher reported temps. But even with that considered, the AMD card made the overall temperature inside my PC higher than either Nvidia card. Temperature aside, the difference in the cooling between the 2060 Super and the 5700 XT is night and day. AMD’s cards are noticeably louder than Nvidia’s equivalents, and if you value a silent PC, then I can’t recommend AMD’s cards here. Nvidia’s cooling means the 2060 Super and 2070 Super stay at similar noise levels (in terms of what you notice) whether your PC is idle or pushing out a demanding AAA title at 1440p.
AMD’s RX 5700 XT and 5700 blower spins up almost immediately into a game, and it’s distracting unless you’re using headphones. AMD’s cards also draw a lot more power than Nvidia’s new Super line. The RTX 2060 Super has a TDP of 175 watts, whereas AMD’s similarly priced 5700 XT draws up to 225 watts. The RX 5700 comes in lower at 185 watts, while the $499 RTX 2070 Super has a TDP of 215 watts.
I’m hoping that third-party RX 5700 XT and 5700 cards will deal with the noise and heat situation a lot better than AMD’s reference blower, as they usually come with multiple fans. I’d recommend waiting on those if you’re impressed by the performance gains but not by the heat and noise produced or if you’re interested in overclocking these cards.
Another consideration will be the features that Nvidia offers with its cards compared to AMD’s latest. AMD has clearly optimized for raw performance here, but Nvidia offers similar performance plus extra features like real-time ray tracing on top. The promise of cinematic effects in games is still early days, but Nvidia is investing a lot of time and money into getting game developers to take ray tracing seriously. This has been bolstered, thanks to the fact that Sony’s next-generation PlayStation and Microsoft’s Project Scarlett Xbox console will both support ray tracing.
That should mean more game developers start focusing on ray tracing, as both next-gen consoles will support it. We haven’t seen enough games with ray tracing to really make it worth it just yet, but the promise is there, and without support for ray tracing, it does mean AMD’s Radeon 5700 and 5700 XT aren’t exactly future-proof. You also miss out on Nvidia’s Deep Learning Super-Sampling (DLSS) that leverages the company’s supercomputer farms to scan games before they are released (or even afterwards) and work out the most efficient way to render graphics.
AMD does offer Radeon Image Sharpening to upscale games without big frame rate penalties and Radeon Anti-Lag to improve competitive games, but neither comes close to the impact real-time ray tracing has on a gaming experience. I’ve found Battlefield V is one of the good examples of ray tracing effects, thanks to the reflections on objects and guns, and Shadow of the Tomb Raider uses ray tracing in a number of shadowy scenes.
The choice between AMD and Nvidia for $399 or below has suddenly become a lot more complicated. AMD’s price cut was brilliantly timed to react to Nvidia’s Super cards, and the performance promises match up. If you’re looking at spending $400 on a new graphics card, then you now have two great choices (or a $499 RTX 2070 Super option if you’re willing to spend a little more to get even better performance).
If, like me, you value a cool and quiet PC, then the obvious choice is still Nvidia right now. AMD has certainly closed the performance gap in the midrange, thanks to some impressive price points, but it still has a long way to go with its reference cards to compete on heat, power consumption, and noise.
AMD has laid some important and exciting groundwork with the move to 7nm and the RDNA architecture. We’re unlikely to see Nvidia move to 7nm this year, so AMD’s new architecture is the big new challenger for graphics cards in 2019. It’s a challenge that AMD has pulled off on the performance side in the midrange. Now, it has to look beyond the sub-$400 price points and get a lot more competitive at the higher end of the market to make things truly interesting.
Photography by Tom Warren / The Verge
Vox Media has affiliate partnerships. These do not influence editorial content, though Vox Media may earn commissions for products purchased via affiliate links. For more information, see our ethics policy.
- Paladin leveling guide classic
- Ford f150 wiper arm replacement
- Mexican fan palm seeds dogs
- 2009 impala power steering fluid
- Kenwood car stereo remote control
- Moto x developer edition verizon
- Com samsung android personalpage service
- Town of lancaster highway department
- Go out plex walmart
This blond, having wrapped my hair around his hand, sucks my mouth. From time to time he takes out his wet cock and slaps it on his lips. I try to catch him with my lips, lick, and kiss. The guys' hands crawl all over the body, crumple their breasts, twist their nipples, they hurt but pleasantly hit the ass. Mma.