The GeForce 8800 Ultra

Physically, the layout of the board is no different, but NVIDIA has put quite a bit of work into their latest effort. The first and most noticeable change is the HSF.

We have been very happy with NVIDIA's stock cooling solutions for the past few years. This HSF solution is no different, as it offers quiet and efficient cooling. Of course, this could be due to the fact that the only real changes are the position of the fan and the shape of the shroud.

Beyond cooling, NVIDIA has altered the G80 silicon. Though they could not go into the specifics, NVIDIA indicated that layout has been changed to allow for higher clocks. They have also enhanced the 90nm process they are using to fab the chips. Adjustments targeted at improving clock speed and reducing power (which can sometimes work against each other) were made. We certainly wish NVIDIA could have gone into more detail on this topic, but we are left to wonder exactly what is different with the new revision of G80.

As far as functionality is concerned, no features have changed between the 8800 GTX and the 8800 Ultra. What we have, for all intents and purposes, is an overclocked 8800 GTX. Here's a look at the card:



While we don't normally look at overclocking with reference hardware, NVIDIA suggested that there is much more headroom available in the 8800 Ultra than on the GTX. We decided to put the card to the test, but we will have to wait until we get our hands on retail boards to see what end users can realistically expect.

Using nTune, we were able to run completely stable at 684MHz. This is faster than any of our 8800 GTX hardware has been able to reach. Shader clock increases with core clock when set under nTune. The hardware is capable of independent clocks, but currently NVIDIA doesn't allow users to set the clocks independently without the use of a BIOS tweaking utility.

We used RivaTuner to check out where our shader clock landed when setting core clock speed in nTune. With a core clock of 684MHz, we saw 1674MHz on the shader. Pushing nTune up to 690 still gave us a core clock of 684MHz but with a shader clock of 1728MHz. The next core clock speed available is 702MHz which also pairs with 1728MHz on the shader. We could run some tests at these higher speeds, but our reference board wasn't able to handle the heat and locked up without completing our stress test.

It is possible we could see some hardware vendors release 8800 Ultra parts with over 100MHz higher core clocks than stock 8800 GTX parts, which could start to get interesting at the $700+ price range. It does seem that the revised G80 silicon may be able to hit 700+ MHz core clocks with 1.73GHz shader clocks with advanced (read: even more expensive) cooling solutions. That is, if our reference board is actually a good indication of retail parts. As we mentioned, we will have to wait and see.

Index The Test
Comments Locked

68 Comments

View All Comments

  • BloodAgent - Wednesday, January 2, 2008 - link

    It is worth it, that is all that needs to be said. If you wanna be cheap and get the lesser computer equipment that will fall short sooner then go with the GTX. I have a Intel Quad Core cpu and two EVGA-8800 Ultra's in SLI, and there is not a game out there that weakens my system. The price is a bit much, but you get what you pay for. My 8800 Ultra's are great. My buddy has a rig running two 8800 GTX's and we benchmarked them using the same version of 3DMARK and my system chewed up the benchmarks while his suffered and choked a bit across the finish line. Go with the Ultra 8800.
  • derubermensch1 - Friday, May 4, 2007 - link

    The ONLY reason to step-up to this card is if you utilise EVERY one of your expansion slots. The reason being the EVGA card has a backplate which, in most cases, blocks the adjacent slot so it is essentially a 3 slot cooler for all intents and purposes.

    The Ultra will undoubtedly overclock better but then you void that nice EVGA lifetime warranty :(

    So youre basically banking on some magic driver to come along that may take advantage of the new bios on the card, but that's a big "if"

    I was on the fence for awhile until I saw that my EVGA card does better in 1600x1200 (I use 1680x1050) so that cemented my decision to stay away. It would be cool to own such a card, just because I can, but then I realized that step-up cost would pay for an HD-DVD add-on for my 360 :)



  • derubermensch1 - Friday, May 4, 2007 - link

    I forgot to mention, don't get an Ultra or hold out on stepping up in the hopes you will be able to step-up to one of those pre-overclocked Ultras that seem to be around the corner. As anyone who tried to step-up to an ACS3 edition 8800 GTX knows, EVGA will not allow it, the program applies to reference spec'd cards ONLY, no exceptions.
  • masa77 - Thursday, May 3, 2007 - link

    I have a pair of 8800 GTXs already and both can be overclocked to 650MHz on the core and 2.1GHz on the memory, but for a boost of three frames per second or so in 2560x1600 it's pointless.

    These cards are only ideal for those who don't have an 8800GTX and even then, you might consider the 8800 GTS because it can basically run anything in 1920x1200 and below quite well. I can often game in 2560x1600 with SLi disabled and all settings maxed. The 8800GTX is a very powerful card.
  • adaofer - Thursday, May 3, 2007 - link

    There are 2 reasons why nvidia produced this card:
    1. Ati new flag ship - this card is supposed to keep the crown on nvidia side and give
    nvidia the upper hand VS Ati , it's the same story as 7800 GTX 512 it's a stepping
    stone in the road map between 8800 --> 8900
    2. price fixing between nvidia and ati : if nvidia ultra costs 830$ and Ati's new flag
    ship has the same performance why sould it cost less , this those company's are
    creating a new price level for ultra/premium cards
  • Stele - Wednesday, May 2, 2007 - link

    Almost everything significant to be said about the 8800 Ultra has already be said; it looks very much like this card's more for the clueless but loaded folks who want it as much as they'd want a Bugatti Veyron to show off to their pals.

    Even with the supposedly enhanced overclocking headroom, it's unlikely that it'd be able to justify the massive premium - it can only reach 684 MHz core while remaining stable, which is certainly not $300's worth more than the Foxconn's 630 or EVA's 626. That it can reach even higher (702/1728 as Anandtech found) is not much use since, to maintain stability, one would have to spend even more on top of the large premium to replace the stock HSF with a better one - it would have been a little different if, for example, the extra money gives us a superb cooler that allows us to reach 700 (and stay there) out of the box.

    Right now there's the 8500, 8600 and 8800. There's a very conspicuous and inviting hole for an 8700 with 64 SPs, 12/16 ROPs and maybe 550/1190/1000 MHz clocks at the $250 mark. A pretty sweet spot, imho. Nvidia really ought to have something there... or perhaps that's an ace in the hole to be pulled out depending on how AMD's lineup performs. Besides that, hopefully Nvidia manages to iron out any kinks in the transition to 65nm fast, as these GPUs could sure use the lower TDPs. On the other side of the fence, hope AMD bucks up soon... they've dropped the ball quite a bit lately and R600's already a whole 6 months behind the 8800.
  • cornfedone - Wednesday, May 2, 2007 - link

    Nvidia is no doubt just trolling to find gullible fools with more money than technical knowledge. Only a fool would cough up this kind of money for a video card. Don't be surprised if these outrageously over-priced cards become symbols of cluelessness for the well healed but technically challenged fanboys. There's a sucker born every second.

  • Mumrik - Wednesday, May 2, 2007 - link

    I don't get it... it's not April 1st...
  • TA152H - Wednesday, May 2, 2007 - link

    Why is everyone whining about this card? I mean, you realize that NVIDIA is in the business of making money, right? You also realize that no one from NVIDIA will come to your home and hold a gun to your head and make you buy it, right? Wow, a company that wants to make money is greedy! Too funny.

    In actuality, they are probably just looking for an identity and want to go more upscale. Is it worth it? Maybe to someone, and now they have the choice. High end CPUs have always been a bargain for some uses, even though they cost a lot more and offer only a few percentages more performance. The reason is simple, for some companies the cost for employees is so much higher than the cost of equipment, even a $1000 is trivial if it makes people more productive or gets answers faster. I have been in these situations before, and they aren't willing to overclock so they buy the best and don't really care about the cost. I'm just not sure if a video card applies to many of these situations, since shooting space aliens isn't a viable business strategy. But there are probably some pursuits that benefit from this besides kiddies and their games. For them, this might make sense.

    Comparing an overclocked card with a stock, pre-production card is kind of silly too. I mean, it's fine for the review, but when people extrapolate from this that they should get other card and overclock it, they ignore that this new card might overclock better, so again, it's not a matter of getting the same thing for less, but of getting less performance for less.

    AMD probably realizes the kiddie market is too small like Intel has and prefer to go with better mainstream products. They don't need to be in that business anymore, since it's a much bigger company and they can expand in different, more meaningful ways (like Fusion). NVIDIA has big problems, the merger between AMD and ATI has made their life a whole lot more difficult in that chipset market, and I don't know anyone with an IQ over 70 that buys a chipset from a penny-ante company like NVIDIA when they can buy a real Intel chipset for their Intel processors. Intel chipsets have always been the best, if not in performance then in reliability and support.

    So, NVIDIA is looking like the odd man out, especially with IGPs slowly creeping up. Maybe they need to reinvent themselves as a high end company, as ATI goes a bit more mainstream. Good luck to them, I'll be surprised if they're not belly up in a few years, or really bought by another company, especially if processors continue to gain GPU functions as seems the current path.

    Does anyone else think it's a little funny how everything goes back to the way things were? I mean, there are math coprocessors again! Now, it seems like all the graphical stuff will be done by the processor again pretty soonb (as in the days of the dumb frame buffer video cards). I wonder when the DOS prompt will be the interface for the next generation operating system. Google isn't that far from it, although it's not an OS.
  • DerekWilson - Saturday, May 5, 2007 - link

    Just one issue with what you said.
    quote:

    Comparing an overclocked card with a stock, pre-production card is kind of silly too. I mean, it's fine for the review, but when people extrapolate from this that they should get other card and overclock it, they ignore that this new card might overclock better, so again, it's not a matter of getting the same thing for less, but of getting less performance for less.
    You can look at it like an overclocked card, but the fact is that the EVGA card we tested and stock 8800 Ultra cards compete on the same level. This isn't a card we bought and then overclocked, this is a retail product that EVGA sells at higher performance than stock 8800 GTX hardware. It's got a lifetime warranty at the clock speeds they set at the factory. As will their stock 8800 Ultra card which won't perform any better.

    I understand the argument that comparing the 8800 Ultra to an 8800 GTX that we overclocked ourselves is silly. But people that don't want to overclock their hardware who are in the market for a high end graphics card will only see a negligible performance difference with a huge price gap between EVGA's 8800 GTX KO ACS3 and any stock 8800 Ultra. It absolutely is a matter of getting the same thing for less.

    Overclockers may see some value in the 8800 Ultra, but even the businesses you mentioned should look at the EVGA 8800 GTX KO ACS3 and the 8800 Ultra as competitors on equal ground. Sure a company might not mind paying more money for more performance, but it is hard to understand why a company would needlessly pay more money for somethign that offers the same performance as a competing part.

    I do agree that a lot of people are making a bigger deal out of this than they need to. The market is still driven by supply and demand afterall. If people are smart enough to realize that NVIDIA has priced their hardware too high, the price will come down. We certainly won't pass up the opportunity to educate people about the fact that the 8800 Ultra is an incredibly bad value with other retail graphics cards performing the same and costing much less.

Log in

Don't have an account? Sign up now