8800GTX vs. GTX 280 Benchies @ 2560x1600, 1920x1200, and 1680x1050.

SpeedyVV

Supreme [H]ardness
Joined
Sep 14, 2007
Messages
4,210
As a follow up to my user article on Video Card upgrades here, I benchmarked the XFX GTX 280 XXX at stock and compared it to the results of the same setup except with an EVGA 8800 GTX also at stock.

Benchmarking details including approach is on the article linked above.

I hope this will help those looking to upgrade from an 8800 GTX or to an GTX 280.

image006.jpg
 
Not much a huge performance jump when it comes to World In Conflict for all resolutions, but still nice anyway.
 
Nice comparison. It's easy to see that the GTX 280 really shows it's value over the 8800GTX at 2560x1600. At that res it seems to, on average, nearly double the performance of the 8800GTX. There are jumps in performance at the other resolutions but, from what I can see, the biggest jumps are at 2560x1600.
 
I think the GTX280 clearly shines in higher res situations, as for the top end cards I think Nvidia did it right(except for the 9800gtx). Now as for the mid range cards I think they screwed up with all the dam model #'s.
 
The 280 beats the pants off of the 8800. Where is the surprise? The 8800 was a great card, but it's two generations old at this point.
 
The surprise is that the 8800 IS STILL a great card.

It is 2 generations old, i can still play many games with max eye candy up to 1920x1200.

What that gives you is probably another year of great gaming, so that you only need to upgrade next year.
 
Very nice, may I ask how you measured in each of these games? Also a suggestion, % difference in a fourth column next to the 280 averages would've been a nice touch.

Nevermind, just skimmed through your original post and found your methods. Nice job.
 
Very nice, may I ask how you measured in each of these games? Also a suggestion, % difference in a fourth column next to the 280 averages would've been a nice touch.

Nevermind, just skimmed through your original post and found your methods. Nice job.

Great suggestion, I will add that to the spreadsheet.
 
The surprise is that the 8800 IS STILL a great card.

It is 2 generations old, i can still play many games with max eye candy up to 1920x1200.

What that gives you is probably another year of great gaming, so that you only need to upgrade next year.
The 9 series doesn't really count as a generation.
 
Added percent difference for averages as suggested by Sabrewulf165
 
I really like the work you did here, but I'm not getting the % average. maybe I'm reading it wrong, but let's use Grid for example in the first chart - 8800GTX = 19.5 GTX280 = 41.2 and you have the delta average as 52.7%. Wouldn't that be over 100% improvement? Maybe I'm just dumb this afternoon.
 
uhmm.. I had % Diff = (280 - 8800) / 280. So % diff over 280. I think you are correct, my bad.

It should be (280 - 8800)/ 8800 or % diff over xisting 8800. Will fix.

EDIT: Fixed
 
Question Speedy. Did you underclock your 280 to reference speeds? Because your "stock" XFX 280 XXX comes with a huge overclock from reference speeds. If not, then a better "Apples to Apples" comparison would be a highly overclocked 8800 versus your overclocked 280.
 
Johnny78, I see where you are comming from, but this is not an Apples to Apples comparison.

As the 1st post states, this is a follow up to the upgrade article.

Therefore I am showing the diff between what I had, and sold, to what I bought and am currently running.

Now I am running the non-reference XXX card at stock for the benchmarks, because that is how I benchmarked the 8800 GTX.

I paid a premium to get the OCed version of the GTX 280 because I believe that they are "chosen".

Well actually it was not much of a premium because it was on a nice Sale but still higher than reference.

Hey, I would not expect you to underclock your nice Ultra Superclocked. It is what it is "Super" ;)
 
Johnny78, I see where you are comming from, but this is not an Apples to Apples comparison.

As the 1st post states, this is a follow up to the upgrade article.

Therefore I am showing the diff between what I had, and sold, to what I bought and am currently running.

Now I am running the non-reference XXX card at stock for the benchmarks, because that is how I benchmarked the 8800 GTX.

I paid a premium to get the OCed version of the GTX 280 because I believe that they are "chosen".

Well actually it was not much of a premium because it was on a nice Sale but still higher than reference.

Hey, I would not expect you to underclock your nice Ultra Superclocked. It is what it is "Super" ;)


Cool, more than anything i just wanted some clarification by what you meant as "stock". I was actually seriously considering getting that XFX 280 XXX. Looks like a really solid performer. However after a lengthy thought process, I'm about 95% sure I'm going with that OC'd Diamond X2. Either way, thanks for the graphs. Very informative. :D
 
I dont usually play games that I cant play at native resolution.

My old 8800GTX definately was not going to play Crysis at 2560x1600.

So, nothing to compare to.

Mind you, folks seem to enjoy these mini benchies, so I might do a follow up comparing my current stock system, with a fully OCed system.

If I do that, I might get some new games.
 
Back
Top