You'll notice that the 7800 GT has four fewer pixel shader pipelines than the
GTX, and one fewer vertex shader, and it runs at a lower clock speed. On the
other hand, it has the same transistor count. How is that possible? SimpleвЂ¦the
7800 GT is a "binned" part. What does that mean? What should Nvidia do when a
7800 GTX chip fails to run at the promised clock speed reliably, or when a pixel
shader or vertex shader unit fails? It makes no sense to throw away a perfectly
usable, yet lower-performance chip. The 7800 GT chip is simply a 7800 GTX that
has had one vertex shader unit and one "quad" of pixel shader units disabled.
The chip is then downclocked to 400MHz, paired with slower 500MHz (1.0GHz
effective) memory, and sold at a reduced price. The resulting card is still
wicked fast, as you'll see in our benchmarks, but it won't set you back quite as
much as the GTX model.
XFX's cards based on the 7800 GT are very similar to Nvidia reference
designs, but with one major difference: They come overclocked.
The clock speeds for the GeForce 7800 GT are 400MHz for the core GPU and
500MHz (1.0GHz effective) for the memory. XFX sells its 7800 GT cards clocked up
to 450MHz, a significant 12.5% boost. The memory is kicked up a very small
amount, too. At 525MHz (1.05GHz effective), it is only 5% faster than the
reference memory speed. It's worth noting that the reference 7800 GT operates
cooler than the GTX, with a power consumption of about 85W instead of 100W+. The
XFX overclocked will almost surely result in an increase in power consumption,
though probably not quite as much as the GTX.
One final note about price: Originally, the GeForce 7800 GT's "suggested
retail price" was going to be $449, and for some cards it may still be. XFX has
noticed that price competition for the GTX model has driven prices down from
$599 to about $499, and no longer feels that a $449 price is the right target.
They will be selling these 7800 GT cards starting at $399.
We ran our benchmarks on a high-end gaming PC with the following
3D performance is measured with a variety of applications:
3DMark05: Futuremark's latest synthetic graphics benchmark, 3DMark05,
is a very forward-looking test. It's heavy on DX9 shaders, and runs optimized
code for either Shader Model 2.0 or 3.0, depending on your card's capabilities.
Far Cry: Crytek's shooter Far Cry is one of the most graphically demanding games
on the market. It will run optimized code paths for shader model 3.0 or 2.0b if
your card supports it. Our test runs the four demos included in the 1.3 patch,
and takes their geometric mean to produce a final score.
Doom 3: How can you review a graphics card without testing the latest
id Software game? Nvidia almost always scores better on Doom 3, whether it's because the developers optimized
more for Nvidia's architecture, or Nvidia designed with id's engine in mind.
Half-Life 2: We run a couple of custom-recorded demos of single player
play in Half-Life 2, then take their geometric mean to produce an
overall score. It's a great example of a high-end, very efficient and optimized
Painkiller: With the 1.61 patch, many of the little discrepancies with
using Painkiller as a benchmark have been ironed out. It's not especially heavy
on shaders and the like, but the long view distances and gobs of enemies and
explosions make this a good test.
Splinter Cell: Chaos Theory: UbiSoft's latest Sam Fischer adventure
uses an all-new engine, rich with detail and heavy on shaders. Splinter Cell:
Chaos Theorycan operate in a Shader Model 1.1
mode, applicable to all cards, or a Shader Model 3.0 mode for Nvidia cards. We
run both, and enable the optional high dynamic range rendering mode with SM 3.0.
Note that anti-aliasing doesn't work with HDR rendering (in this or any other
As good as these games for testing graphics performance, some of them really
don't push the latest high-end graphics cards hard enough. In particular,
Painkiller is starting to show its age and is rapidly becoming CPU-limited on
high-end cards. We have dropped UT 2004 from our testing because it is almost
always CPU limited, even on more modest-priced hardware.
We put the GeForce 7800 GT up against two of Nvidia's other high-end cards:
the GeForce 7800 GTX and GeForce 6800 Ultra. Also included is ATI's fastest and
most expensive card, the Radeon X850 XT PE. Naturally, with Nvidia's push to
greater graphics performance with SLI, and considering this is our first review
of the new GT architecture, we tested the XFX cards in SLI configuration as
well. We used driver revision 77.73 for the Nvidia-based cards, and Catalyst 5.6 for the ATI card.
benchmark 7800gt 3DMark05 results
( 7800 gt review )
Futuremark's latest 3D benchmark is the only synthetic test we use to judge
video card performance. That is, it's the only test that isn't an actual game.
Why? Simply put, we believe that it's a pretty good look at future game
graphics. The tests are well designed to mimic the type of code and content
we'll see in games coming over the next year or soвЂ”those that utilize DirectX 9
The overclocked nature of the XFX brand 7800 GT cards helps it keep pace with
the more expensive GTX (at stock speeds). It's less than 7% slower, which isn't
bad for a $100 price difference. Two cards in SLI deliver just amazing
performance, speeding things up between 50% and 80% over a single card.
7800gt benchmark Painkiller
7800 gt review )
Painkiller exhibited some odd irregularities with its first attempt at a
benchmark mode, but the 1.61 patch has addressed those. It is a reasonably
impressive game with a good benchmark mode, but it simply doesn't tax graphics
cards like it did when it was released.
The results here are a little odd. The 7800 GT actually runs Painkiller more
slowly than the 6800 Ultra. What's more, SLI mode isn't really any faster than a
single card. We double-checked all our settings and the configuration of the
test machine, and everything was in order. This is just an odd aberrationвЂ”a
repeatable result that isn't at all what we expected.
7800gt benchmark Half-Life 2 and Doom
3( 7800 gt review )
Valve's Half-Life 2 is one of the more graphically advanced games available
today, making fairly heavy use of shaders in some areas. It is well engineered
and optimized to make the most out of almost any graphics card.
e 7800 GT delivers much better Half-Life 2 performance than the 6800 Ultra when
you crank up the resolution and add AA and AF (anisotropic filtering). It's a
little disheartening to see that this 20-pipeline "next generation" card from
Nvidia really only runs the game a few frames per second faster than ATI's
16-pipeline "old technology" card. But then, ATI has always done very well in
If Half-Life 2 is said to be ATI's best game, then Doom 3 is Nvidia's. Again,
the new $400 card pushes well past the old one, when you run the game at very
high resolution with AA and AF. During normal gameplay, Doom 3 is limited to
60fps, and a single GeForce 7800 GT achieves that and more at all but the
highest test settings. Even then, 50fps is no slouch. Of course, with two cards
in SLI mode, you can average well over 60fps all the time.
7800gt benchmakr Far Cry and
Splinter Cell: Chaos Theory
( 7800 gt review )
With vast outdoor scenes flush with vegetation, complex water, and a fair
amount of normal mapping, Far Cry pushes most graphics cards to the limit. The
1.3 patch ships with four built-in demos. We ran all four and take their
geometric mean to produce a final score.
Far Cry continues to impress us with its ability to push graphics hardware to
the limit. The real benefit of XFX's overclocked 7800 GT again becomes apparent
at the highest resolutions and with AA and AF enabled, but it can't match the
7800 GTX. SLI offers huge improvements here.
A new patch to Splinter Cell: Chaos Theory enables Shader Model 2.0 support,
letting ATI cards perform HDR rendering, parallax mapping, and all the other
effects that used to be reserved for the Shader Model 3.0 path. We run the
benchmark in both SM 1.1 mode and SM 3.0 with HDR. Anti-aliasing is incompatible
with HDR rendering, so we run our two test resolutions with and without AF
Now that a new patch enables us to get an apples-to-apples view of performance,
we uncover some interesting things. With HDR lighting, parallax mapping, and the
other Shader Model 2.0/3.0 features, the ATI card drastically outpaces the
GeForce 6800 Ultra it originally competed with. Nvidia's ability to deliver its
new architecture first has given it a big performance advantage in the 7 series
cards. There are a couple of render states where XFX's 7800 GT card actually
outpaces the 7800 GTXвЂ”chalk that up to its overclocked nature.
Final Thoughts/Should You Buy One?
The GeForce 7800 GT is a good product, overall. Though the price is more
affordable, it's still definitely a high-end card. The group of gamers willing
to spend $400 on a video card is still rather small (let alone those who will
spend twice that to run in SLI mode). Having said that, the price/performance
ratio of the GT model seems a bit easier to swallow than the GTX, especially
when compared with its original $600 price. By disabling a single vertex
pipeline and four pixel pipelines, Nvidia is able to offer a chip that still has
fantastic capabilities and speed at a reduced price, while making better use of
some of the chips that, when they came off the line, didn't quite pass the test
to go on 7800 GTX cards. It's a better deal for everyone. It also gives them a
bigger leg-up on ATI, which has not announced or previewed, let alone shipped,
the first card in its next-generation R5xx family.
Check out other
coverage of 3D graphics here.
The fact is, 7800 GTX cards have dropped to about $500 now, and several
vendors are offering overclocked models for a bit more. For the GT model, XFX is
offering great value by coming in at $400 with an overclocked card that does
video in/video out, and comes with a decent game bundle. If you're in the market
for a $400 video card, it's going to be hard to find a better performer than
this one or a card with more complete features and software
xtreview is your : Video card - cpu - memory - Hard drive - power supply unit source
we would be happy to answer for your question . if you have suggestion or comment
regarding this review our support would be glad to help just join our forum and ask u will get the best answer
to discuss check our forum section :-)