Rated: 100% (2 votes)
Loading ...What’s in a name, Nvidia?
Alex Watson
With today’s launch of the GeForce GTX 280, Nvidia is releasing not only a new GPU but a new naming system. Nvidia says that its previous naming scheme – such as GeForce 8800 GT, GTS, GTX and Ultra – had become bewildering for consumers. They got confused because it wasn’t clear that 8800-series cards should be faster that 9600-series cards, while it was never clear that the GTS suffix indicated a faster card than a GT card of the same number. Similarly people had trouble working out which card of a new generation would genuinely be an upgrade over their current card. We’ve had many emails from people complaining that their brand new GeForce 9600 GT isn’t as fast as their supposedly inferior GeForce 7900 GX2 and so forth.
Of course Nvidia exacerbated this by arbitrarily adding odd suffixes such as GSO, and by giving 8-series names to some graphics cards based on the newer G92 GPU. We ended up with a situation where the GeForce 8800 GTS line-up had three cards with 320MB, 512MB and 640MB of graphics memory, where the 512MB version was actually the fastest as it was based on a newer piece of silicon. Try explaining that to your average PC World customer.
So the new naming convention is as follows:
The prefix will roughly denote the speed of the graphics card – a GTX-class card is fastest than a GT, which is faster than a GS-class card which should walk all over a lowly G-class card.
The first number of the name denotes the family of GPU the graphics card has.
The last two numbers indicate how powerful the GPU in question is.
At launch we only know of two GPUs using htis new naming system - the GTX 280 and the GTX 260. So, as these GPUs are GTX-class, we know that these are the fast GPUs of the new range. The ‘2’ of the second part of the name indicates that they are GeForce 200-series GPUs and the ‘80’ and ‘60’ shows that the GTX 280 is faster than the GTX 260.
RESERVATIONS….
Well, I’m slightly worried that Nvidia sees the need for two sub-prime ranges of graphics card (GS and G) as typically low-end graphics cards aren’t much cop for gaming and yet are marketed as being brilliant. Imagine how many people have bought a GeForce 8200 or an ATI Radeon HD 2450 thinking they’ll be able to play Crysis, only to find the game still crawls and giving up on PC gaming to go console for ever.
Details are also sketchy on how the rest of the range will be named – if we end up with GTX 280 and 260 and GT 280 and 260, things could become confused. Nvidia is also banking on people knowing the hierarchy of prefixes (GTX trumps GT beats GS and so forth) which is by no means intuitive to the casual buyer.
We suspect that a GeForce 300-range will further confuse things as you won’t be able to tell which is the better card by price alone – a GeForce GT 380 could be the same price as a GTX 260, for example, and only benchmarks will tell which will be faster. Still, at least hardware journalists won’t be put out of a job after the naming convention change….
Bah-remember the old ATi apporach I say: More X’s =more power!
yep… until they took stock and changed their naming convention… I seem to remember at the time writing that Nvidia really really really needed to do the same… looks like they may scan these pages after all eh 
The new ATi approach makes reasonable sense when you multiply 4850 by 2 and get 9600, even if that maths analogy doesn’t work for the others in the series.
Duh…make that 9700…still you get what I mean…
What a confusing naming system! Putting the emphasis on the letters will give the wrong impression to customers. Imaging if they did that to the 9800’s. Which would sound better, GT 9800 or GTX 9600? If you believe the new naming system, it would be the latter when actually the opposite is true. To be honest I can’t see them using any other names than GTX without confusing customers.