In the Cryptocurrency Era, Approach Used GPUs With Caution

We’ve written a number of articles on the current lousy state of the GPU market and what you can practically do about it, ranging from optimization guides for existing cards to whether AMD’s Ryzen 5 2400G can serve as a plausible low-end gaming solution to tide you over if you need a new card. But there’s another option available — the used GPU market. Does it provide a reasonable alternative?

The short answer is: Maybe. But you’re going to want to be extremely careful before you pull the trigger.

The problem with buying a used GPU off an unknown seller is that you’ve got no idea what the card’s operating conditions were. While this has always been true, cryptocurrency mining puts unique stresses on GPUs compared with conventional gaming.

First, cryptocurrency miners typically run 24/7, while most gaming GPUs don’t. Gamers also tend to understand and prioritize good cooling, while cryptocurrency miners tend to maximize performance first and worry about cooling later. Four GPUs that are perfectly capable of cooling themselves individually can run extremely hot when put loaded into a chassis at the same time. GPU miners may also have overclocked a GPU or tweaked its voltage for stability. While this is also a risk when buying a used gaming card, again, we’re talking about the difference between 24/7 operation and occasional use.

eBay offers certain protections to guard against being sold defective equipment. But just because hardware works the day you receive it doesn’t mean it won’t fail in a month or two, or display erratic crashes when loaded in a very specific fashion. We recommend buyers looking for a used card prioritize owners who specifically state that the GPU has not been overclocked or used for cryptocurrency mining.

Still, this advice relies on the original owner being honest. What can we do to avoid relying on the owner’s honesty?

Simple: We can test the card. But in order for the tests to be effective as diagnostic criteria, you’ll need to test your current GPU first.

It’s common to see people recommend benchmarks like Furmark for GPU stress testing, but I’m going to disagree with that assessment. AMD and Nvidia both scan for thermal viruses like Furmark and typically prevent them from pushing GPUs as hard as possible. Instead, I’m going to recommend you run your tests in games you already own, either by looping built-in benchmarks or with play-through tests.

The following slide show is from our RX 580 versus GTX 1060 test. Of the games listed, all of them except for Doom have built-in benchmark modes. One game, Metro Last Light / Metro Last Light Redux has a benchmark you can loop for n number of tries.

Company of Heroes 2 is the RTS sequel to the acclaimed Company of Heroes and it remains downright stingy with its performance. The R9 390 remains basically tied with the RX 480, but the RX 580 pulls ahead on all fronts, only surpassed by the much more expensive GTX 1070.

Metro Last Light Redux is an updated version of the original Metro Last Light. Our tests put a heavy load on all GPUs by using supersampled antialiasing. Here, the GTX 1060 still leads the RX 580, with the R9 390 delivering the fastest AMD performance overall.

AoTS was the first DX12 game on the market and remains an impressive demo of what that API can do. The RX 580 has just enough advantage over the RX 480 to take the 1080p benchmark and tie with the GTX 1060 at 1440p. The R9 390 is a bit faster in 1080p than RX 480, but ties it at 1440p.

In Hitman, the RX 580 performed extremely well, winning by a whisker against the GTX 1060 and leaving its own predecessors well behind. The R9 390 is a whisker ahead of the RX 480, but the difference falls into the margin of error.

Shadow of Mordor is one of the best-regarded RPG/action hybrids of the past few years. Nvidia cards typically perform better than AMD hardware in this title, so it's not surprising to see Team Green win overall. The RX 580 is faster than both the RX 480 and R9 390, and effectively tied with the GTX 1060 at 1440p.

Rise of the Tomb Raider is the second game in the rebooted Tomb Raider franchise. Here, the RX 580 again pulls well away from the RX 480 and the R9 390 and noses past the GTX 1060 to overall victory. The R9 390's additional ROPs (64, compared to 32 for the RX 480 and RX 580) don't give it any help here.

We've moved from Total War: Rome 2 and to the fantasy-themed Warhammer title with DX12 support. The performance gaps between the RX 480 and RX 580 are quite small here, and the R9 390 actually ekes out a win at 1440p compared to either AMD card. The GTX 1060 enjoys a very small advantage at 1440p, but the R9 390 very nearly ties it.

DiRT Rally is the most recent rally game by Codemasters, and is somewhat newer than the Autosport we've used for our recent CPU tests. Here, Nvidia scores some of the biggest wins we've seen in this review, vastly outpacing any of the AMD cards. The RX 580 does turn in the best performance overall, however, with frame rates somewhat ahead of the R9 390.

We tested Doom at 1080p in both Vulkan and OpenGL. The results aren't all that surprising — AMD is well behind Nvidia in OpenGL, but ahead of the 1060 in Vulkan. The R9 390 can't quite beat the GTX 1060 in that API, but the RX 480 ties it and the RX 580 pulls ahead.

To start this process, you’ll want to download the same driver you’ll use for your new card (this does assume that your current GPU if still compatible with modern driver releases — if it isn’t, omit the initial driver removal and reinstallation steps on your current GPU). Uninstall your old driver and then run a utility like DDU (Display Driver Uninstaller) from inside Windows Safe Mode to absolutely remove all trace of a previous driver install. Next, install the new driver from a standard boot of Windows. With the new driver in place, fire up your game or benchmark of choice and start playing. Loop tests or game for 15-30 minutes, and make sure you test more than one title. It’s absolutely possible for a dodgy card to be stable in five titles and badly unstable in the sixth, so you’ll want to test a selection of games.

What we’re doing here is establishing a baseline for system stability. When you receive your used card, uninstall your GPU driver using the instructions above and reinstall it alongside the new card. Run the same benchmarks and gaming checks again. If your “new” card is stable and your PSU is appropriately specced for your GPU, you can be reasonably confident that the card will perform properly thereafter. This process doesn’t eliminate the chance of a longer-term failure, but it does offer a test sequence you can perform fairly quickly to ensure a GPU isn’t broken out of the box.

Also, always make sure to check current used prices against current new prices. With the speed at which GPU pricing can change these days, that’s the only way to make sure you won’t end up paying more for a used card than a new one. Also, tip of the hat to Tom’s Hardware, which has some additional tips on evaluating whether a GPU is safe for purchase.

Continue reading

Early Adopters of Apple M1 Macs Should Be Cautious About Compatibility

Apple's new MacBooks and Mac mini have made waves, partly thanks to the new silicon inside of them. Apple's new ARM ecosystem, however, is not without its growing pains.

First Impressions of the Galaxy Fold Are Cautiously Optimistic

Samsung's Galaxy Fold represents a truly new method of building displays into smartphones. Whether you like the idea of a foldable device or not, the option to own a functional model at all feels like something out of sci-fi.