It would be a huge understatement to say that no one is happy with the prices gamers pay for graphics cards today. Sure, Nvidia’s new GeForce RTX 3050 and AMD’s new Radeon RX 6500 XT sport suggested price tags of $250 and $200, respectively, but as you know, you’re not touching those cards right now for such a low price during the current GPU shortage.
Who benefits, where the money goes, and the true meaning of MSRPs is not what we are discussing today. Instead, in the interest of lazy Fridays, we wanted to take a quick look at what GPUs are marked up for compared to their MSRP, and how much performance you’re getting with the high costs included. We also wanted to show what you’re missing out on when you pay exorbitant markup prices, showing you how much next-gen DDR5 memory or fancy Apple monitor stands you could have bought if graphics cards were sold at reasonable prices. instead of.
Real-World GPU Prices vs. Performance
But first, the performance comparison. We searched popular hardware retailer Newegg for GPU models that were in stock and in new condition and came up with the most affordable list price we could find. All cards indicated that they would be shipped directly from Newegg, but Newegg did not necessarily sell them. Like other retailers, Newegg will sell you a product that is actually being stocked by a partner store. Like any fashion item, prices fluctuate a lot, so today’s prices may or may not exactly match what we found this week. Prices may also be different at other retailers or on user selling sites like Ebay.
We then compare the public figures in the 3DMark Time Spy GPU test with the marked price. While Time Spy is not based on a game engine, it is a respected and popular synthetic benchmark. It’s also very consistent and almost everything is GPU based, so the CPU and most other components don’t count. You can see the notes we generated below.
The lowest profit margin we found this week was the much-criticized AMD, but no Highly marked Radeon RX 6500 XT, a card that sells for “only” 31 percent off its $199 MSRP. The worst is the Radeon RX 5700, which is out of production but obviously still in high demand with a margin of 329 percent profit over its $349 MSRP. Nvidia’s GeForce RTX 3060 is also a hot item, at 207 percent over its $329 MSRP. The surprise “deal” is AMD’s Radeon RX 6600 , which is “only” $131 more than its $329 MSRP, for a 40 percent markup. If you have an extra $1,500, you’d only pay 50 percent of the Radeon RX 6900 XT’s markup, which would actually be a bargain. Well, stealing as in “is still an armed robbery of your wallet.” But that’s better than the soak other GPUs would give you to walk out the door with them.
But how much performance are you getting after markup? To get a general idea, we analyzed the public 3DMark Time Spy scores and represented how many “3DMarks” you get per cent. Not surprisingly, the cards that are handled the hardest give you the worst post-mark return percentage. The GeForce RTX 3060 costs you 11.5 cents for 3DMark. Despite being stupidly fast, the GeForce RTX 3090 is priced so high on the streets that it actually gives you terrible relative performance, costing 13.9 cents per 3DMark point.
Before you put too much value on this chart, you should know that 3DMark Time Spy is not the end of all in graphics testing. You’ll get different performance depending on the benchmark or the game, obviously. But what is killing the GeForce RTX 3090 is its sky-high price of $2,800.
Another interesting data point jumps out from this chart. PCWorld GPU reviewer Brad Chacos points to the generally lower penny per 3DMark demanded by most AMD 6000-series cards. Overall, they’re a bit more forgiving than Nvidia’s GeForce cards and even older AMD Radeon RX 5000 cards. There may be brand equity costs associated with that, or simply its lower demand from miners. Chacos points out that most Ethereum mining benefits from higher memory bandwidth and with the 6000 line, AMD was able to stick with smaller memory buses (and thus lower overall memory bandwidth). low) as its radical Infinity Cache feature on GPU makes up for it in games. . That Infinity Cache, however, doesn’t help much with Ethereum mining, so it again points to the problem with GPU demand today largely at the feet of miners.
GPU brands vs. DDR5 and Apple
So how much is that profit margin costing you in missed opportunity? We decided to see how many monitor mounts Apple could buy based on the markup on the MSRP of the cards. Why Apple Monitor Stand? Because Apple surprised even its own hardcore fan base when it released a $1,000 monitor mount. Yes, a monitor mount for $1,000. The Radeon RX 6500 XT’s markup ($61) would give you 0.06 of an Apple monitor mount. However, if you could buy a GeForce RTX 3080 Ti at its MSRP, you could have taken the $933 you saved and almost bought an entire Apple monitor stand!
Since few people can relate to a $1,000 monitor stand, we decided to see how much next-gen DDR5 memory you could buy if you didn’t have to pay ridiculous graphics card prices. When it was first introduced, the price of DDR5 was similarly exaggerated, but lately it has come back down to earth and is so cheap that if I didn’t have to pay the hefty markup on a GeForce RTX 3050, I could have bought a 5.35 DIMM Crucial DDR5 /4800 8 GB (or 40 GB and some changes) of the latest generation ultra-fast RAM. For the markup that a retail GeForce RTX 3080 Ti will get, you could equip the systems with 96GB of DDR5.
Gordon, one of the founding fathers of intense tech reporting, has been covering PCs and components since 1998.