Manufacturer Product Page: TBD
Product Number: 01G-P3-1282-AR
Availability: Now
Warranty: Lifetime
Price: Approx. $700 CAD




Nvidia would like to take you on a journey the likes of which we have yet to see in the consumer graphics card world. This is a journey through the minefields that are today’s games, the peaks of processing power that come with parallel computing and the nearly endless fields of applications that will benefit from the GPU doing something other than rendering games. The brave new world we are talking about is brought to us by Nvidia’s new GTX 200-series graphics processing units which are launching today.

In recent months we have seen a gradual evolution of the G80 architecture into the G90-series with a move to 65nm manufacturing and various tweaks which has given rise to cards like the 9800 GTX. These cards have succeeded in appealing to a broad range of consumers and not only gamers due to their competitive prices are relatively high performance. If we think of the 9800 GTX as a technological evolution, then what we see here today with the GTX 200 processors can be considered a revolution akin to the first flight of the Wright brothers. This is what we have all been waiting for; a new architecture (which is still based off the unified shader concept) from Nvidia which will usher in a new age of computing for consumers. You may notice that we say “consumers” since even though they are priced pretty high into the enthusiast spectrum, these new cards will not only appeal to hard-core gamers but also to people looking for a bit more horsepower behind their Folding@Home applications and a bevy of other uses.

All of these advances have been made possible through the standardization of Nvidia’s CUDA architecture into something that is accessible to the everyday user. With this technology, a regular graphics processor of the 8 and 9-series along with the 200-series morphs into a number crunching powerhouse the likes of which cannot be matched by today’s processors. What makes the GTX 200 cards different in this respect is that they have been engineered from the ground up to take advantage of CUDA and indeed, as we will see later the possibilities are nearly endless. Thus, the term parallel processing has quickly become the new catch phrase Nvidia uses to describe the potential of their new graphics cards. What Nvidia hopes is that while GTX 200 series has considerable graphics capabilities, its other uses are numerous enough that they will appeal to a much broader customer base rather than be regarded as a pure game-playing solution.

That being said, today Nvidia will be launching two new cards: the GeForce GTX 280 and the GTX 260. Gone are the four-digit numerical names given to so many Nvidia cards of years past. Even though today is the “official” launch date for both these cards, the GTX 280 will only be available to buy from retailers the day following this review while the GTX 260’s availability has been pushed to June 26th. As with all launches, availability will be initially quite tight but as production ramps up so too will allocations to retailers so hopefully any shortages we see will be rectified in short order. As for pricing, Nvidia has quoted to us a price of $649 USD for the GTX 280 and $399 USD for the GTX 260. These prices should be a bit higher here in Canada with a non-overclocked version of the GTX 280 starting around $690 and we have not yet heard any scuttlebutt about Canadian GTX 260 pricing.

While we usually try to focus on stock-clocked cards on release date, today we have something a little different for you. In this review we will be looking at EVGA's GeForce GTX 280 1GB Superclocked card which is of course a reference design with a touch of adrenalin pumped into it for increased clock speeds. EVGA has long been known not only for their excellent customer service, Step Up program and lifetime warranty but also for their habit of releasing overclocked versions of nearly every graphics card known to man. We have been told that this Superclocked version will cost about $10 to $20 more than their stock clocked cards but as we will see later, “Superclocked” no longer means what it once did.

With all of this said and done, let’s delve a bit deeper into the new Nvidia lineup, the architecture behind this card and the applications in which it can be used.