AMD Radeon RX 6700 XT Review: Big Navi Goes on a Diet
Tom’s Hardware Verdict
The Radeon RX 6700 XT delivers good performance, and the official price looks reasonable. It’s doubtful AMD will meet the demand at launch, but long-term this should prove a competitive offering in the GPU market.
Cons
- —
Not particularly efficient
- —
Price is a bit high
- —
Weaker RT performance
- —
You know it’s going to sell out, right?
The AMD Radeon RX 6700 XT has landed — or at least, it will land tomorrow at retail, and sell out 0.67 seconds later. (For a bot, that’s practically an eternity!) This also marks the debut of AMD’s smaller Navi 22 GPU, which we’ll call Little Big Navi or Medium Navi or something. Officially priced at $479, AMD pits the RX 6700 XT against both the RTX 3060 Ti and the RTX 3070, targeting the sweet spot for price and performance among the best graphics cards and landing in the upper ranks of our GPU benchmarks hierarchy. Let’s find out how it stacks up in our comprehensive review.
Architecturally, Navi 22 doesn’t add or remove features from Big Navi and RDNA2. It comes with full DirectX Raytracing (DXR) support and implements the full DirectX 12 Ultimate features list, including Variable Rate Shading (VRS), mesh shaders, and sampler feedback. What AMD has done echoes what we’ve seen in past generations of GPUs: It trimmed the fat, shrinking the die size by reducing the number of shader cores, memory controllers, and Infinity Cache. Here’s a quick overview of the specs for AMD’s latest GPUs, plus the previous generation Navi 10-based RX 5700 XT for comparison.
(Image credit: Tom’s Hardware)
| Graphics Card | RX 6700 XT | RX 6900 XT | RX 6800 XT | RX 6800 | RX 5700 XT | |
|---|---|---|---|---|---|---|
| Architecture | Navi 22 | Navi 21 | Navi 21 | Navi 21 | Navi 10 | |
| Process Technology | TSMC N7 | TSMC N7 | TSMC N7 | TSMC N7 | TSMC N7 | |
| Transistors (Billion) | 17. 2) |
336 | 519 | 519 | 519 | 251 |
| CUs | 40 | 80 | 72 | 60 | 40 | |
| GPU Cores | 2560 | 5120 | 4608 | 3840 | 2560 | |
| Ray Accelerators | 40 | 80 | 72 | 60 | N/A | |
| Infinity Cache (MB) | 128 | 128 | 128 | 96 | N/A | |
| Game Clock (MHz) | 2424 | 2250 | 2250 | 2105 | 1755 | |
| VRAM Speed (Gbps) | 16 | 16 | 16 | 16 | 14 | |
| VRAM (GB) | 12 | 16 | 16 | 16 | 8 | |
| VRAM Bus Width | 192 | 256 | 256 | 256 | 256 | |
| ROPs | 64 | 128 | 128 | 96 | 64 | |
| TMUs | 160 | 320 | 288 | 240 | 160 | |
| TFLOPS FP32 (Boost) | 12.4 | 23 | 20. 7 |
16.2 | 9 | |
| Bandwidth (GBps) | 384 | 512 | 512 | 512 | 448 | |
| TBP (watts) | 230 | 300 | 300 | 250 | 225 | |
| Launch Date | Mar-21 | Dec-20 | Nov-20 | Nov-20 | Jul-19 | |
| Launch Price | $479 | $999 | $649 | $579 | $399 |
AMD made some serious cuts with Navi 22 compared to Navi 21, with half the potential CUs and shader cores. That’s the biggest change, but there are others. The Infinity Cache checks in at 96MB now, 25% smaller than on Big Navi. Similarly, there are now six 32-bit memory interfaces instead of 8 interfaces, which reduces bandwidth by 25%. AMD compensates for the hefty reduction in core counts by delivering the highest official GPU clocks so far, with a Game Clock rated at 2424MHz — and like other RDNA2 chips, it can and often will exceed that in gaming workloads, with a maximum boost clock of 2581MHz (it’s actually a bit higher than that according to AMD’s drivers, but that’s what AMD lists on the official spec sheet).
The RX 6700 XT’s clock speed represents a bit of an interesting compromise. To push clocks into the 2.5GHz range, AMD had to increase the power limits. Considering the smaller die and reduced VRAM, normally we’d expect a pretty significant drop in power requirements compared to the RX 6800, but AMD lists a TBP (Typical Board Power, meaning it includes all power and not just the GPU) of 230W. That’s not bad, and it’s certainly within reach of most modern gaming PCs as it only requires a decent 500W PSU (AMD recommends 650W or higher). However, it’s slightly more than Nvidia’s competing RTX 3060 Ti and 3070 GPUs.
Navi 22 actually looks very similar to Nvidia’s GA106 used in the recent RTX 3060 12GB. It has the same amount of memory and a similar die size (slightly larger for AMD due to the Infinity Cache). Had AMD kept clock speeds closer to 2.25GHz, it probably could have shaved off a decent amount of power — certainly less than 200W is possible. But higher clocks come with performance benefits, and AMD apparently felt sacrificing some efficiency to boost performance was the better course.
MORE: Best Graphics Cards
MORE: GPU Benchmarks and Hierarchy
MORE: All Graphics Content
- 1
Current page:
AMD Radeon RX 6700 XT Specs
Next Page Meet the RX 6700 XT Cards
Jarred Walton is a senior editor at Tom’s Hardware focusing on everything GPU. He has been working as a tech journalist since 2004, writing for AnandTech, Maximum PC, and PC Gamer. From the first S3 Virge ‘3D decelerators’ to today’s GPUs, Jarred keeps up with all the latest graphics trends and is the one to ask about game performance.
AMD Radeon RX 6900 XT Review: Powerful and Pricey
Tom’s Hardware Verdict
The Radeon RX 6900 XT is now AMD’s fastest GPU. However, it’s only incrementally faster than the RX 6800 XT, making the $1,000 asking price a tough sell.
TODAY’S BEST DEALS
The AMD Radeon RX 6900 XT is officially AMD’s new halo product, surpassing the Radeon RX 6800 XT and RX 6800 by offering the fully-enabled Navi 21 GPU with 80 Compute Units.
The latest generation GPUs sit atop our GPU Benchmarks hierarchy and rate as some of the best graphics cards currently available (if we use the term «available» loosely). The 6900 XT is also the most expensive AMD graphics card we’ve seen in quite some time — the last time AMD sold a GPU priced at $1,000 or more was the Radeon Pro Duo (2016), a very limited quantity dual-GPU card. With RX 6800 XT taking on RTX 3080, the goal is clear: AMD wants to offer competitive performance to the GeForce RTX 3090 while undercutting Nvidia’s price by $500.
All of that is sort of a moot point, of course. Even eight months after the official launch, the RX 6900 XT remains virtually impossible to find in stock, just like all of the other RX 6000 and RTX 30 series graphics cards. Our GPU pricing index looks at eBay prices for the various cards, and while the RX 6900 XT technically only costs 55% more than the MSRP, it’s still a lot more than we’d recommend paying. We’re refreshing this review with the latest details but keeping the original performance results.
You can see the full view of how performance stacks up in more recent reviews like the RTX 3080 Ti, but until prices reach something approaching normalcy, buying a latest generation graphics card remains a serious pain in the nether regions.
Undercutting the 3090 price is all well and good, but there are a couple of problems when comparing the GeForce RTX 3080 and 3090 to the Radeon RX 6800 XT and 6900 XT. First, RTX 3090 more than doubled the VRAM of the 3080. It also has a wider memory bus and 20 percent more shader cores. Finally, it increased the power target by 10 percent and often delivered superior performance in professional workloads. In contrast, the RX 6900 XT has 10 percent more shader cores than the 6800 XT. And that’s it.
Let’s get straight to the point. On paper, the RX 6900 XT doesn’t look like a good deal. We’re not saying the RTX 3090 is a better deal for gaming purposes, but at least it offers some tangible benefits over the RTX 3080. Again, all of this is somewhat academic right now, as supplies of the latest generation graphics cards have proven woefully inadequate.
2)
There’s really not much to say about the specs that we haven’t already covered.
The RX 6900 XT has the full 80 CUs and 5120 shader cores that the Navi 21 GPU offers, and the remaining specs, including memory speed and TDP, are the same as the RX 6800 XT. Which isn’t to say that there aren’t differences in the silicon.
Binning and sorting chips from silicon wafers is nothing new, but the best chips often come from closer to the center of the wafer, where there are fewer defects. Since the 6900 XT requires a fully functioning Navi 21 chip, it almost guarantees other characteristics like power and voltage requirements will be better as well. How much does that matter in practice? If you’re hoping to overclock and reach maximum performance, the silicon lottery can get you an extra few percent. Considering the 6900 XT already costs 50 percent more than the 6800 XT, however, that’s a big jump in price for a likely minor improvement in overall performance.
AMD Radeon RX 6900 XT: Unboxed and Unwrapped
Image 1 of 12
(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)(Image credit: Tom’s Hardware)
The Radeon RX 6900 XT looks identical to the RX 6800 XT, except for the product label.
The packaging isn’t quite the same, however. Inside the main box, the RX 6900 XT comes wrapped in a mouse pad. It’s a nice little extra if you need a mouse pad or AMD swag. It’s a relatively large pad as well — not big enough to cover your entire desk, but much larger than a basic pad.
The RX 6900 XT is a dense graphics card, weighing slightly more than the RTX 3080 while sporting similar dimensions. It’s a bit thicker, occupying 2.7 slots, but it’s also about 2cm shorter. You could potentially fit it into a smaller build, but we recommend exercising some caution as ejecting 300W (or more) or heat into a small case will hurt performance. Actually, if you’re doing a smaller case, just save yourself some money and buy the 6800 XT — we’d bet once thermals level out, the two cards will perform nearly the same.
Like the RX 6800 cards, the 6900 XT includes a single HDMI 2.1, two DisplayPort 1.4, and one USB-C video output. If that doesn’t meet your needs, keep an eye out for third party cards.
We’ve heard there will be RX 6900 XT cards from AMD’s AIB partners (contrary to some earlier rumors), which makes sense as anyone making an RX 6800 XT could use the same design for RX 6900 XT. We’re not sure when those cards will be available for purchase, however.
There are two 8-pin PEG connectors for power, rated for 150W each. Combined with the 75W from the PCIe slot, power should be more than sufficient for normal operation. We did see slight power spikes when starting up certain workloads (FurMark, specifically), but these shouldn’t pose a serious problem to any quality power supply that can deliver 850W. We do recommend using a single rail PSU if possible.
Radeon RX 6900 XT Overclocking
(Image credit: Tom’s Hardware)
MSI Afterburner still doesn’t support RX 6000 series GPUs at present, leaving us with AMD’s built-in overclocking tools. We could just use Rage Mode, but all that seems to do is raise the power limit, so we set about doing some manual tuning.
Memory seemed easy enough to max out, and cranking the power slider to maximum as well doesn’t generally pose any problems. That leaves the GPU clock, voltage, and fan speeds, the latter of which we tuned to favor cooler temperatures over quieter operation. We set the RAM to 2150 MHz, dropped the GPU voltage to 1125 mV (down from 1175 mV), and set about testing.
We started by looping the Unigine Heaven 4.0 benchmark in a window at 1440p extreme, then bumped up the GPU clock to 2600 MHz to check for stability. We found that the power limit becomes a factor long before the maximum boost clock comes into play. With a 115 percent power limit in place, that gives the GPU up to 345W of power, and that’s pretty close to where it landed. Clock speeds then fluctuate according to the demands of whatever game or workload is running, with average clocks in the 2400-2550 MHz range.
While Heaven proved stable in our initial tests, we did encounter some issues in a few other games, which eventually led to the slightly revised overclocks you can see above.
We dropped the RAM 10 MHz to 2140, bumped the voltage back to 1150 mV, and that seemed to fix the problems. Additional time spent tuning might eke out a few more MHz, but for most users, simply enabling Rage Mode would do most of the dirty work — or set the RAM speed to 2100 and max out the power slider for a slightly higher result.
It’s worth noting that while AMD lists the official boost clock of the RX 6900 XT as 2250 MHz, in AMD’s own drivers, the boost clock shows up as 2519 MHz. You won’t normally see clocks quite that high, but you could with Rage Mode. In that sense, AMD has taken a page out of Nvidia’s book by advertising a lower boost clock than what you’ll get in many games. We appreciate this, and the combination of increased power limits and a minor bump to the maximum boost clock gave us a decent result.
Overclocked results are present in our benchmark charts, but not in the DirectX Raytracing (DXR) suite. The net improvement was around 5 percent, and 7 percent at 4K, which isn’t that much of a boost to performance.
There’s a decent chance third party cards may come with better factory overclocks than what we achieved.
MORE: Best Graphics Cards
MORE: GPU Benchmarks
MORE: All Graphics Content
TODAY’S BEST DEALS
- 1
Current page:
Features and Specifications
Next Page Radeon RX 6900 XT — Test Setup
Jarred Walton is a senior editor at Tom’s Hardware focusing on everything GPU. He has been working as a tech journalist since 2004, writing for AnandTech, Maximum PC, and PC Gamer. From the first S3 Virge ‘3D decelerators’ to today’s GPUs, Jarred keeps up with all the latest graphics trends and is the one to ask about game performance.
Why Are Graphics Cards So Expensive?
Graphics cards have gotten progressively more expensive over the years. Whether you choose AMD or NVIDIA, you’ll need to spend well over half a grand at the very least to get your hands on a high-end GPU.
However, this wasn’t always the case. So, what exactly happened to the GPU market? Why are consumers forced to shell out thousands of dollars on a flagship graphics card today?
Here, we look at the reasons why graphics cards are so expensive. Many of these factors are obvious. Others, however, might surprise you.
1. The Lack of Competition
The lack of competition in the mid-2010s and early 2020s, especially from AMD, has been one of the driving factors behind the ever-increasing GPU prices. When NVIDIA launched the Pascal-based GeForce 10-series graphics cards in 2016, the company followed an aggressive pricing strategy as it was already facing heat from AMD with its RX 300 series of GPUs.
But following this release, AMD just fell apart with its RX Vega launch in 2017, as it had no GPUs that were a match for NVIDIA’s flagship 1080 Ti. And due to this lack of competition, NVIDIA became confident enough to launch the newer RTX 2080 Ti for a price tag of $1,299 in 2018—for a significantly worse price to performance than its predecessor.
Again, AMD didn’t have any cards to compete in the high-end market and instead focused on mid-range GPUs.
Two years later, a seemingly more confident NVIDIA launched the Ampere-based RTX 30-series of graphics cards, with the RTX 3090 sitting right at the top for a whopping $1,500. And if that wasn’t enough, the company did one better by launching a more powerful RTX 3090 Ti for $2000 in 2022. Although AMD has started competing again in the high-end GPU market this time around with its Radeon RX 6000 graphics cards, NVIDIA still holds the performance crown.
2. The Global Chip Shortage
Why are GPUs so expensive? Economics, my dear Watson. Supply, demand, and the allocation of resources all determine how available or expensive goods or services are, especially with what the general public sees and has access to.
We are currently amid a global chip shortage that could continue indefinitely. We use semiconductor chips in everything, from computers to airplanes; manufacturers from all around the world have had to scale back or limit their production due to the shortage.
Chip shortage has impacted not just graphics cards but other sectors as well, including laptops, phones, and video game consoles.
3. The COVID-19 Pandemic and Inflation
We get that the COVID-19 global pandemic is old news, but you can’t deny the fact that it has drastically changed how people and businesses choose to work. More people have started to work from home, which has resulted in an increased demand for personal electronics. GPU prices are high right now because the market is hungrier for them than ever.
Whether it’s for work or for entertainment, lots of people are looking to buy electronic devices like computers, smartphones, or tablets. This has drastically increased the demand for graphics cards, which is why graphics cards are so expensive.
Another consequence of the pandemic is broken supply chains. While COVID-19 is no longer the prime reason for this, other factors such as the Russia-Ukraine war have disrupted global supply chains in 2022.
This means that people living in different parts of the world may not have immediate access to the graphics cards they need, forcing them to look into importing from foreign markets. These outside markets, attracting new business for one reason or another, then become overburdened; inflation, corruption, and other indicators of imbalance may then ensue.
Long story short: this is why graphic cards are so expensive. There are fewer GPUs floating around than people with the means, the need, and the cash to buy them, and it’s partially COVID’s fault.
4. US Tariffs on Imports
During Trump’s tenure as US President, new tariffs were being imposed on imports from overseas weekly, including a tariff that added a 25 percent tax on top of the asking price for graphics cards imported from China. This extra cost is passed along to US consumers through higher prices.
The Biden Administration has not announced any plans to dial back Trump’s tariffs on China.
The team intends to reevaluate its current trade agreements before moving forward with any new strategies.
5. Cryptocurrency Miners
Miners need a good graphics card and software to mine cryptocurrencies such as Ethereum efficiently. They often have high requirements for what kind of graphics cards they need for their set-ups; the faster the card, the faster they can mine, so higher-end cards are more desirable.
Although miners have stayed away from stocking up graphics cards in 2022, thanks to the bear crypto market, the tables could turn at any moment. A bullish market often results in miners buying as many graphics cards as possible, further disrupting the demand and supply.
Crypto miners are often knowledgeable, well-connected, and adequately funded enough to buy out graphics cards and acquire the best of the best before anybody else has a chance to cut in. Thankfully, NVIDIA has managed to alleviate this problem with its LHR graphics cards, which cuts the mining hash rate in half, effectively making them undesirable to miners.
6. Scalpers Selling Hardware for Profit
Crypto miners aren’t the only ones crowding out the graphics card market; scalpers looking for monetary gain have also disrupted the graphics card market, although they employ different tactics to meet their ends.
Scalpers resell goods for a quick (and usually large) profit. While scalping isn’t limited to graphics cards, they are hot commodities at the moment due to the surging demand. As a result, they’re perfect for reselling at high prices to desperate consumers.
Sometimes, scalpers may obtain their stock through questionable means. In many parts of the world, however, scalping in any form is still a legal practice. There are no federal laws prohibiting scalping in the US, although a few states have implemented laws that discourage the practice.
During the 2021 holiday period, US politicians unveiled a new bill—the Stopping Grinch Bots Act—to prevent scalpers from using bots to secure goods online. At the time of writing, this bill was still only a proposal and was yet to become law.
7. Better Specs and Performance
Unlike the graphics cards of the early 2010s, modern GPUs use advanced materials and top-of-the-line designs. Generally speaking, they’re higher-quality and more difficult to manufacture than the ones from the previous generations. Simply put, you can consider higher manufacturing costs one of the main reasons why modern GPU prices are so high.
Producers can only manufacture what they can afford. Due to increased manufacturing costs, producers must get creative about maximizing their profits without compromising a product’s quality or quantity.
The current crop of graphics cards from NVIDIA and AMD is pushing the boundaries of computer graphics in more ways than just frame rates. With cutting-edge technologies like ray tracing and FSR (DLSS), these modern GPUs are designed to deliver the best gaming experience possible at higher resolutions like QHD and 4K.
In many ways, you are technically getting what you pay for, but it’s simply another factor to consider when scoping out your options the next time you’re ready to upgrade your graphics card.
All the Reasons Why GPU Prices Are So High
We’ve listed all the reasons why graphics cards are so expensive, but unfortunately, GPU prices are not within our direct control. As long as the demand remains higher than the supply, graphics cards will continue to be expensive.
With inflation factored in, you should no longer expect high-end graphics cards with sub-$500 price tags as you did five or six years ago. Times have changed, and technology continues to move forward to deliver new and better experiences to the consumer—for a price, of course!
0007
9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000
9000 9000
GTexels/s
9000 1018GB/S
for one cycle. This is an important factor in memory performance, and therefore the overall performance of the graphics card. 9000
9000 shadows and reflections in games.
9000 million 9000 76300 million
9000 76300 million 9,000 400 million 9000.000 MENTER METREMEN new process of creating a chip.
Newer versions support higher throughput and provide better performance. PC video cards. Types of graphics cards.
Video adapter.
Video card.
A video card (also video adapter, graphics adapter, graphics card, graphics card, graphics accelerator) is a device that converts a graphic image stored as the contents of a computer’s memory (or the adapter itself) into a form suitable for further display on a monitor screen. The first monitors built on cathode ray tubes worked on the television principle of scanning the screen with an electron beam, and a video signal generated by a video card was required for display.
However, this basic function, while remaining necessary and in demand, has gone into the shadows, ceasing to determine the level of imaging capabilities — the quality of the video signal (image clarity) has very little to do with the price and technical level of a modern video card.
First of all, now a graphics adapter is understood as a device with a graphics processor (graphics accelerator), which is engaged in the formation of the graphic image itself.
Modern video cards are not limited to simple image output, they have an integrated graphics processor that can perform additional processing, removing this task from the computer’s central processor. For example, all modern Nvidia and AMD (ATi) graphics cards render the OpenGL and DirectX graphics pipeline in hardware. Recently, there has also been a trend to use the computing power of the GPU for non-graphical tasks.
Typically, a video card is made in the form of a printed circuit board (expansion board) and is inserted into an expansion slot, either universal or specialized (PCI Express, AGP). Video cards built-in (integrated) into the motherboard are also widespread — both as a separate chip, and as part of the north bridge of the chipset or CPU (but in this case, the device, strictly speaking, cannot be called a video card).
Discrete video cards.
Discrete video cards are the most high-performance class of graphics adapters.
As a rule, it is connected to the high-speed PCI Express data bus. Previously, there were video cards connected to AGP buses (a specialized data exchange bus for connecting only video cards), PCI, VESA and ISA. At the moment, modern video cards are connected only through the PCI Express bus, and all other types of connections are outdated. In computers with an architecture other than the IBM-compatible, there were other types of video card connections.
The discrete card can optionally be removed from the device (for example, on laptops, the discrete card is often soldered to the motherboard). It is called discrete due to the fact that it is made in the form of a separate chip (or chipset) and is not part of other computer components (unlike graphic solutions built into motherboard system logic chips or directly into the central processor). Most discrete graphics cards have their own random access memory (VRAM), which can often have faster access speeds or a faster access bus than regular computer RAM.
Although, earlier there were video cards that completely or partially used the main RAM for storing and processing graphic information, at present, almost all modern video cards use their own video memory. Also, sometimes (but quite rarely) there are video cards whose RAM is not installed in the form of separate memory chips, but is part of the graphics chip (in the form of separate crystals, or on the same chip with the graphics processor).
Implemented as a separate chipset rather than as part of other chips, discrete graphics cards can be quite complex and much more powerful than integrated graphics. In addition, having their own video memory, discrete video cards do not need to share RAM with other computer components (primarily with the central processor). Own operational allows you not to waste the main RAM to store information that is not needed by the central processor and other computer components. On the other hand, the video processor does not have to wait in line to access the computer’s RAM, which can be accessed by both the central processor and other components at the moment.
All this has a positive effect on the performance of discrete graphics cards compared to integrated graphics. Technologies such as Nvidia’s SLI and AMD’s CrossFire allow multiple graphics cards to be used in parallel for a single task.
Integrated graphics. Built-in GPU.
Integrated graphics cards do not have their own memory and use the computer’s RAM, which affects performance for the worse. Although Intel Iris Graphics since the Haswell processor generation have had 128 megabytes of L4 cache at their disposal, they can take the rest of the memory from the computer’s RAM. Modern integrated graphics solutions are used in portable devices due to low power consumption. Their performance is already at a high enough level and allows you to play simple three-dimensional games.
Modern integrated graphics processors are located on the same chip as the CPU (for example, Intel HD Graphics or Intel Iris Graphics), previous generations (for example, Intel GMA) were located as a separate chip.
Hybrid solutions.
Hybrid solutions are used where both energy efficiency and high graphics performance are required, allowing you to use the integrated graphics adapter in everyday tasks and use the discrete graphics adapter only where it is needed.
Before the advent of hybrid graphics, manufacturers built in a discrete adapter in addition to the built-in one, switching between them required a reboot, which was not very user-friendly. Hybrid adapters only use the integrated graphics adapter for display output, but some calculations can be transferred to a discrete graphics card instead of being performed by themselves. For the user, switching between video adapters becomes invisible. Examples of such solutions are Nvidia’s Optimus technology and AMD’s DualGraphics. APU (short for Accelerated Processing Unit, accelerated processing unit) is a hybrid central processing unit that combines a central processing unit with a graphics processing unit in a single chip.
As a result of combining the CPU and GPU, there is an overall reduction in power consumption and system cost. Hybrid CPUs make it possible to make compact systems suitable for most users who do not require powerful graphics tasks.
External graphics card (eGPU).
The term eGPU refers to a discrete graphics card located outside the computer. It can be used, for example, to increase performance in 3D applications on laptops. Typically, PCI Express is the only suitable bus for this purpose. An ExpressCard, mPCIe (PCIe x1, up to 5 or 2.5 Gb / s, respectively) or a Thunderbolt 1, 2, or 3 port (PCIe x4, up to 10, 20, or 40 Gb / s, respectively) can be used as a port. In order to streamline the nomenclature, AMD carried out work to standardize external video adapters.
Which video card to choose — GeForce or Radeon
To begin with, we want to note that our article is aimed at people who are not specialists in the field of graphics cards.
We know how difficult it can be to choose a graphics card when you consider the variety of models, the amount of video memory, and the gaming technologies offered by manufacturers. We will present you with an overview of video cards, explaining the basic aspects of the functioning of this important component of a computer.
Video card — GeForce or Radeon
As it turns out, the choice of video card manufacturer determines, first of all, the fact that there is such a model that fully satisfies the user’s needs:
- Corresponding performance (satisfactory test results)
- Stability and «work culture» (noise, heating)
- Reliability (other users do not report repeated errors)
Price also matters, of course, but if you’re aiming for high-shelf, 4K, or virtual environment gaming, then be prepared to part with a hefty sum.
The choice of a particular video card is also influenced by the image synchronization technology on the monitor.
If your monitor supports AMD FreeSync technology, selecting a Radeon card will allow you to use this feature. Buying an NVIDIA design will, in turn, allow you to take advantage of G-Sync technology.
Incidentally, it is worth mentioning the issue of compatibility with other computer components. There is no contraindication to combine Intel and AMD components in the same kit. This will not affect performance. So you don’t have to focus on one manufacturer’s solutions.
Graphics cards for gaming and professional systems
These two families of graphics cards are separated by a clear line that is worth exploring to avoid disappointment.
Office video cards
By the above term we mean video cards for low-demanding games, amateur processing in Adobe Photoshop and high-definition movie playback.
Unlike integrated graphics cards, dedicated cards have their own video memory, otherwise known as VRAM. Its quantity depends on the technical characteristics of a particular model.
Meanwhile, more than 1 GB is rarely installed in the integrated video memory, which may be too small.
For office and home use, the NVIDIA GeForce GT and AMD Radeon R7 graphics cards will suffice.
Graphics cards for gaming
The name in this segment VGA wanders the most misleading, because, it is worth noting that gaming graphics cards are suitable not only for games. Due to the large processing power, it is excellent at rendering graphics, even in 4K resolution, VR projection and semi-professional designing.
In games, the role of VGA does not end with providing certain graphics settings or maintaining a certain level of FPS. With a gaming graphics card, you can stream your gameplay online without wasting gaming power, as well as play on a high-resolution monitor.
If higher GPU load power is important to you, choose a graphics card that lists overclocking capability in the description.
For these purposes we recommend video cards NVIDIA GeForce GTX series 10xx , AMD Radeon RX series 5xx and Radeon RX Vega
Professional video cards
Professional video cards differ from other models not so much in terms of device operation.
Interestingly, despite the huge computing power, they are not ideal for gaming. On the other hand, they are brilliant at dealing with specific and complex calculations in engineering applications, geological surveys, architectural design, etc.
They are also valued by people involved in advanced 3D graphics, professional photo editing, and 4K video production.
Professional graphics cards: NVIDIA Quadro
How much video RAM do you need?
The choice of video memory size is affected by:
- The resolution in which you want to play or work is the minimum value for Full HD resolution — 4 GB of video memory. More is required for WQHD — 6 GB of VRAM, and to take advantage of 4K you need 8, and preferably 11 GB of video memory.
- Level of detail and additional effects such as v-sync. If you want to play at high settings, go for a card with at least 4GB VRAM, but for ultra settings, invest in a 6GB model.
- Monitor screen format. While it may not seem like a big deal, a 16:9 monitor will need 4 GB of VRAM, while a 21:9 display will require 6 GB of VRAM.
- Game segment. Esports games, i.e. CS:GO, LoL or Dota, are not particularly demanding and 4 GB of VRAM is enough for them. And AAA segments such as The Witcher or Battlefield 1 have a big appetite for VRAM, so you should buy a graphics card with at least 6 GB of memory, although it is no doubt more reliable to take a model with 8 GB of VRAM.
In addition to quantity, the type of video memory also matters. The most popular GDDR5 and GDDR5X modules with high bandwidth. In games where large amounts of data are transferred, this makes a huge difference.
HBM2 (High Bandwidth Memory) VRAM is used in Radeon RX Vega graphics cards. True, it works at lower clock frequencies, but thanks to a wider bus, it provides more bandwidth, and, therefore, convenience in games.
Video card size, appearance and technology
It is very important that the size of the video card matches the computer case.
Especially in the case of cases such as microATX and iTX, where free space is very limited and there is no chance of installing a full-sized component. Especially for such cases, video card manufacturers have created miniature versions of VGA.
The design of a video card seems to be of little importance, however, it plays an important role in the process of forming a set that will be placed in a case with transparent walls.
If you want components to be visually uniform, look at the colors of the LED lighting. Moreover, graphics cards can be connected to other devices that have additional lighting. Thanks to this, you can choose one backlight color or one theme for: components (not only for VGA, but also for the motherboard or LEDs in the case) and accessories (keyboard, mouse, monitor).
Under the term «work culture» we mean the volume level of the video card. A quiet but very efficient cooling system is a semi-passive system. At low load (at the beginning of work and in the initial stages of the game), the fans stand still, and heat is removed through the radiator.
Interestingly, video cards with massive heatsinks can work for a long time without activating active cooling.
Only when the temperature inside the system exceeds the set value, the fans will start to drive cold air towards the radiator fins.
The graphics card control panel usually allows you to change the fan speed, depending on the user’s preferences. The volume matters, first of all, during work, because in games the noise is drowned out by the sounds from the headphones or from the speakers.
Support for VR technology
Forming a set that will pull VR games, make the main emphasis on the video card. It is its power that determines the capabilities of the system, and therefore the comfort of entertainment in virtual reality.
In order for the created computer to work at maximum resolution, without defects, tearing or delays, we recommend taking GeForce GTX 1070, GTX 1070 Ti and GTX 1080 video cards. Among AMD chips, the Radeon RX 580, RX Vega 56 and RX Vega 64 models are best suited
Number and size of fans
One of the characteristics of video cards is the number of fans.
Standard and miniature models often have only one fan. In more advanced versions, you will find more — from two to three, and the number is more important than the diameter. Three 50mm fans are more efficient than two 100mm fans. The number of fan blades leads, in turn, affects the volume of the cooling system — the more, the quieter.
Cooling efficiency is also affected by the contact of the video memory and GPU with the copper plate. The metal back helps in dissipating heat, which also strengthens the entire structure, preventing deformation.
Video card ports
Pay attention to the number and types of ports when you plan to connect more than one monitor, VR glasses or use image synchronization technologies:
- VR displays.
- DisplayPort/mini DisplayPort is a connector compatible with NVIDIA G-Sync and AMD FreeSync technology.
- DVI and VGA ports are used to connect older generation monitors.
The
Also make sure your computer’s power supply has enough power.