Is Nvidia’s GTX 690 Priced Too High?

Not if you’re a fan of supply and demand…

Nvidia world iconNope. But what’s more, it should actually be priced higher. And while that may not be what you want to hear, we can assure you that whether retailers are charging $1000 or $1200 Nvidia has priced the GTX 690 to sell.

There are a number of different ways to look at this equation. Let’s start by coming at it from a performance perspective. Looking at Nvidia’s own review we can see that they feel that the GTX 690 can offer above 90 percent of the performance of a pair of GTX 680’s in SLI. This level of performance is confirmed by looking at either [H]ard|OCP’s or Anantech’s review of the GTX 690.

So let’s relate this level of performance to a price point. The GTX 680 has an MSRP of $499, but if you really want a GTX 680 right now you’re probably going to have to pay a bit of a mark-up on that price, or sign up for some in-stock auto notification e-mails and be prepared to spam F5 at certain times of day. Thus acquiring a GTX 680 is not exactly a trivial matter. Investing the time, and quite possibly the extra money, needed to acquire a pair of them probably won’t be worth it to you over nabbing at single GTX 690.

Now let’s pretend we’re Nvidia for a moment. We have a product that received good, if somewhat tepid, reviews, coming on the heels of a product, the GTX 680, which was roundly praised across the tech sector. As it stands the GTX 680 is still sold out. So if you can’t supply enough chips to meet demand for a product that uses a lower bin split and half the GPUs per unit, how can you expect to supply something like the GTX 690? Simply put, you can’t.

But while Nvidia can’t do anything in the short run about the total quantity of GTX 690’s, and for that matter GTX 680’s, available, it can do something about the demand for them, by changing their prices. Basic economics holds that as you increase the price of a product, demand for that product decreases because people just aren’t willing to pay that much for it. Thus if Nvidia really wanted to end the currently rather agonizing Kepler shortage it would bump the price of its GTX 680 up by $50-$200, which is something online retailers and AIB vendors alike are already doing, often tossing in improved cooling solutions and mild overclocks to boot.

And the same is true of the GTX 690. Nvidia could have reasonably priced it above $1200, like the listing from Amazon below, and it still would have sold quite quickly.

So really what we’re getting at here is that this is an artificial shortage of Kepler based GPUs, and it’s happening because Nvidia needs to, at the very least, appear to be price competitive with AMD’s more mature, and judging by the stock found at retailers, higher volume 28nm GPUs.

The only two events on the horizon that are going to ease the pain of this shortage are the passage of time and the launch of the GTX 670, which Nvidia has been stock piling for sometime. Browsing through the GPU section of our Forums you can see that there is a pretty active discussion going on about Kepler in general and the GTX 670 specifically. Looking at TweakTown’s preview from last week we can see that GTX 670 isn’t all that much slower than the GTX 680, and lines-up pretty well with AMD’s HD 7970. Admittedly, TweakTown’s game selection uses TWIMTBP games pretty heavily, but to get a general sense of how these cards are going to end up being positioned they’ve done a good job.

It will be interesting to see if the situation changes at all when the rest of the reviews go live on the 10th, but for now you can look forward to a new graphics card from Nvidia that might actually stay in stock for more than a day. Keep your fingers crossed.S|A

The following two tabs change content below.
Thomas Ryan is a freelance technology writer and photographer from Seattle, living in Austin. You can also find his work on SemiAccurate and PCWorld. He has a BA in Geography from the University of Washington with a minor in Urban Design and Planning and specializes in geospatial data science. If you have a hardware performance question or an interesting data set Thomas has you covered.