Ditching High-End Graphics Cards Could Give AMD an Edge in the GPU Wars

PCMag editors select and review products independently. If you buy through affiliate links, we may earn commissions, which help support our testing.

Sometimes ceding ground is the right strategy, and AMD plans to lose the flagship graphics-card battle with Nvidia to win the GPU war. The graphics-card market is relatively young, starting in the late 1980s, and it’s been a profoundly turbulent one all through the decades. Several companies have claimed market leadership in the graphics-card industry over that time, and all but one—Nvidia, of course—have faded away.

Nvidia has held the GPU crown for decades, with AMD doing all it can to claim that position. But Radeon has for the longest time been relegated to also-ran status. Now, it appears that AMD will relaunch an old strategy and drop its high-end graphics-card lines to try and surpass Nvidia's market share in certain areas. I think that's a winner move.

AMD Focuses on Casting a Wider Net

Sure, the idea of AMD dropping out of the high-end graphics-card market may give the impression that the company is performing poorly or struggling. That’s not the case here, though, or why AMD is taking this step.

Instead, the company has its eyes set on simply gaining as much market share as possible. Nvidia has been the market leader for decades, but the history of the GPU market has shown that this certainly can change. Before Nvidia, the market leader was a company called 3dfx; before that, it was S3 Graphics, and even further back, Hercules Computer Technology. Depending on your age, you may recall some or all of them. These companies have since gone out of business or been bought out. Nvidia is the exception, and that's not likely to change in the near future. But AMD would unquestionably like to grow to be the most widely used graphics-card manufacturer and be able to claim market leadership on volume.

In an interview with Tom’s Hardware’s Paul Alcorn, Jack Huynh, AMD’s senior vice president and GM of its Computing and Graphics Business Group, said the following:

“I’m looking at scale, and AMD is in a different place right now. We have this debate quite a bit at AMD, right? So the question I ask is, the PlayStation 5, do you think that’s hurting us? It’s $499. So, I ask, is it fun to go King of the Hill? Again, I’m looking for scale. Because when we get scale, then I bring developers with us.

So, my number one priority right now is to build scale, to get us to 40% to 50% of the market faster. Do I want to go after 10% of the total addressable market or 80%? I’m an 80% kind of guy because I don’t want AMD to be the company that only people who can afford Porsches and Ferraris can buy. We want to build gaming systems for millions of users.”

Often, with new generations of graphics cards, the fastest and most potent cards get the most attention. Dubbed "flagship" cards, they are spiffy and always fun to look at, but their overall market share is minuscule due to their excessive pricing. None of the current-gen flagships has even a 1% market share, going by Valve's Steam Hardware Survey, though the Nvidia GeForce RTX 4090 is surprisingly close to that mark. Still, that’s just a piffle compared with the Nvidia GeForce RTX 3060, which has 5.51% of the market by itself, or the Nvidia GeForce GTX 1060, which launched eight years ago and still holds 2.93% of the market at the moment.

Valve’s data convincingly shows that midrange graphics cards hold far more market share than flagships, which is why AMD is laser-focused on this market segment. It may forfeit the flagship market share, but in return AMD will be able to compete better in more mainstream market segments.

The Upsides: Plentiful, More Precise Processors With Peak Performance

I see two primary ways exiting the flagship market can help AMD better compete in other market segments. First, it is architecturally advantageous. Most graphics-card families start with a high-end graphics chip for the flagship part, which then is scaled down as the basis for lower-grade GPUs. This traditionally works well, but it's also true that designing a product for a specific purpose tends to work better than applying a high-end part scaled down for a lesser or different platform. That’s how Arm came to dominate the smartphone processor market, not AMD or Intel. The same dynamic is valid here.

The other significant benefit of not making a flagship is...well, not having to make them. The chip fabs and factories used to produce the various parts that go into a graphics card all have production limits, and the more resources you dedicate to flagships, the less you’ll have elsewhere. Forgoing the flagships will enable AMD to produce more midrange graphics processors, which, in turn, will likely help to reduce their price, due to increased economies of scale.

I also see some significant secondary advantages to doubling down on the budget market. As Huynh said in his interview, gaining more market share will also help attract more developers to keep AMD graphics cards at the fore when designing future PC games. Games developed with a particular graphics architecture in mind are typically better optimized for that hardware and, as a result, perform better on that hardware. Developers often try to optimize for as many graphics architectures as possible, but the hardware used in development always holds a more significant advantage.

With this move, AMD might also be looking to try and win back some market share, PC-side, from game consoles, where AMD itself has a commanding position. During the interview, Huynh brought up the Sony PlayStation 5 and its $499 price point. The PlayStation 5 and Microsoft’s Xbox Series X consoles both feature AMD Radeon graphics and AMD Zen processors, which means that AMD directly profits from game-console sales.

While the game consoles are a source of income from the supplied chips, AMD likely profits more on a per-unit basis from graphics-card sales. Game consoles are produced on extremely tight budgets, and it's well known that console manufacturers often sell their hardware at a loss (at least initially) with the intention of profiting later off licensed sales of software and game titles. Though it hasn't been publicly disclosed how much profit or revenue AMD draws from its Microsoft or Sony relationships, it likely makes more "per unit" off of individual graphics-card sales, particularly as AMD has more direct control over what its graphics cards cost.

Naturally, this causes AMD to want to drive sales of its graphics cards over sales of game consoles, but that's difficult to do when your best cards alone cost more than those $499 game consoles that come fully assembled and ready to game. The game consoles are not just more affordable but also friendlier and more accessible to users, as not everyone knows how to upgrade a PC (never mind how to build a PC), and today's consoles are potent enough to give you a premium, enjoyable gaming experience.

Even so, modern graphics cards tend to produce better overall graphics performance and unlock more-impressive graphical effects in games. But that helps only so much against the advantages that game consoles carry. By targeting its next generation of graphics cards at the midrange and budget segments, AMD's Radeon cards will better appeal to these gamers who want to keep costs down. That $499-or-below price point would appeal to a wider range of gamers and help to make building a PC with capable graphics more affordable. Most people who were going to opt for a game console over a gaming PC still will, but if next-gen Radeon cards are priced competitively enough, AMD might lure a few over to its camp in the gaming-PC market.

Best-Rated AMD Radeon 7000 Series Cards We've Tested

Today, $499 is closer to the middle of the graphics-card market than the high end, so I doubt AMD intends to stop selling graphics cards at this price point. However, it does suggest the company is focusing increasingly on budget and midrange buyers to grow its market share.

Budget GPUs for the Win? AMD's Done It Before

This isn’t the first time AMD has attempted this strategy—more like the fourth time!—and it has worked well in the past. Each time AMD employed these tactics, its market share increased. The last time AMD tried this was with its AMD Radeon RX 400-series and AMD Radeon RX 500-series graphics cards, the latter of which launched in 2017.

These cards were so popular that even now, seven years later, the most popular AMD graphics card in use, according to Valve’s Steam Hardware Survey, is the AMD Radeon RX 580, which launched for $229 with 8GB of RAM. At the time, these cards were selling so well that they were often out of stock for extended periods. (In fairness, the late-2010s cryptocurrency-mining boom also contributed to the Radeon RX 500 series selling super-fast.)

The RX 400- and RX 500-series graphics cards helped AMD grow its market share, but eventually, AMD fell prey to the lure of the flagships. The company has made some stellar top-end cards, so that return to form, every time AMD did it, was well-received. Still, afterward, market share tended to decline gradually, leading AMD to exit the high-end market again, eventually.

This has placed AMD’s Radeon graphics-card business in a bit of a loop that has been repeating for decades. It’s hard to anticipate what will happen this time, but every time AMD has done this, we’ve seen some spectacular-and-steady graphics cards with lasting impact, so here’s hoping the fourth time is also the charm. A new generation's Radeon RX 580 is just waiting to be born.