Quick Card Search:
navigation About Us Contact Us Compare Videocards Forum Links Search Cards Submit A Videocard Videocard Table Home Video Card Review Finder Video Card Superlatives Video Card Database Video Card News
Tabs   Tabs

GTX 280 vs 8800 GTX - 18 Months of Innovation

Introduction & Setup

Welcome to the very first GPU review written and produced by GPUReview.com! For now we are still purchasing the hardware that we review so we may not being reviewing cards very often until we start getting samples from manufacturers. The idea to do this review came from my desire to upgrade from my 8800 GTX to a GTX 280. So this review will be focused on the performance differences between the Evga 8800 GTX and the Evga GTX 280 in four games and two benchmarks.  

Our review "software suite" consists of:

      
  • 175.19 WHQL drivers for the 8800 GTX
  • 177.41 WHQL drivers for the GTX 280
  • Windows Vista Ultimate 64 bit
  • 3DMark06
  • 3DMark Vantage
  • Age of Conan Hyborian Adventures
  • BioShock
  • Call of Duty 4: Modern Warfare
  • Crysis
 

Our review hardware consists of (my PC):

      
 
  • CPU: Intel Core 2 Quad QX6700 @2,666MHz
  • CPU Heatsink: Zalman 9500 LED
  • Motherboard: Abit IN9 32X-MAX (8800 GTX) MSI P7N SLI Platinum (GTX 280)
  • Video Cards: Evga 8800 GTX (768-P2-N831-AR) and Evga GTX 280 (01G-P3-1280-AR)
  • Memory: 4GB (2x 2GB) Corsair Dominator PC2-8500 @800MHz
  • Hard Drive: Western Digital 250GB SATA 7,200RPM
  • PSU: Enermax Galaxy DXX 1,000 Watt
  • PC Monitor: Samsung 20" 204B LCD
  • HDTV: Sharp LC-37GP1U 37" 1080p LCD

You can see that two different motherboards. This was necessary because the GTX 280 could not be installed onto my Abit IN9 32X-MAX. For more details check out this story I wrote about the whole ordeal. The performance differences between the two motherboards are negligible.

Our testing methodology involves me playing a game and recording 2 minute sections using Fraps. I get at least 3 two minute runs and then I average the Minimum, Maximum and Average FPS numbers. The settings are kept the same for both graphics cards. 16X anisotropic filtering and 4X Multisampling antialiasing were used for all games by using the Nvidia control panel Manage 3D Settings menu. No filtering or antialiasing were used while running the 3DMark benchmarks.

Next Page »
(EVGA's GeForce GTX 280) »
17 Comments
Thursday, July 03, 2008 8:46:16 PM
YAY! *reading*
Thursday, July 03, 2008 9:21:55 PM
Steve
The Progenitor
Here is specific feedback I am looking for:

How do the graphs look? Do you like the colors I used and the green stone background?

Is anything unclear? Did you have to read any paragraphs twice?
Thursday, July 03, 2008 9:54:37 PM
I like the review, and especially the info links of the games and information (makes it convenient for people who are interested).

The graphs are great (looks 3d), coupled with the green background made it easier to see.

Nvidia really made a great product, but because of the price (and the expected price of the ATI cards), people just went on to bash Nvidia and dismissing the technology.
(This reminds me of people complaining of Nvidia rebranding 8800s into 9000s while people overlooks Ati's actions in renaming the 2900s into 3800s [a "new" generation])

It does seem that the article was written to justify your purchase of the GTX280 (well, I would do that if I got that card too =D ).
Thursday, July 03, 2008 9:58:07 PM
Duesco
Member
Alright, I read it.

Couple of points:

I don't like the charts. I mean, the 3D bars and background, that's all very nice, but I prefer the simple, white-backed flat 2D bars that I can glance at and not get distracted by colors and such.

The review was easy to understand, although I did notice you went on a tiny bit about the games in the test. Most people already know this information, and they're here for the cards, probably because they want to play these games in the first place.

There are some grammatical things that could be done to smooth out the review, but otherwise it was worded well and ran generally smoothly. I'm impressed.

(I would say you could give me the reviews to proofread, as I'm a writer, but that would seem pretentious, I think?)
Thursday, July 03, 2008 10:40:52 PM
The graphs were fine. I'm colorblind, so disregarding aesthetics, I liked the bright colors. The paragraphs were clear and informal.

I've read several review sites and the are all fairly scientific. Of them all, [H]ardOCP is the best because I feel like I will get the same results.

And that is where your strong point lies. You did this review in your living room - not in a sterile lab. You have the feel of a normal guy comparing his old system against his new upgrade and sharing his experiences and results. There's something trustworthy about that - I feel like I could do the same thing and have the same success, even if my numbers end up being slightly different (horrors!). If I want hard stats I'll go to Tom's or Anand (or the comparison page), but what I want to read is how it works for another guy like me. That's exactly what you did. You may want to include even a few more subjective statements about what you liked and didn't like, how the game looked, etc. to emphasize the 'real' nature of your review.

You've got something here. I'd look forward to more!
Friday, July 04, 2008 1:58:38 AM
They are simple and easy to understand, but as WhistleHeat says, I don't think they have undergo all the quality control steps and results. For this type of information you are presenting you would have to standarize a system and keep using it for all your future tests, so that the results hold some value and work as an indicator, but not be conclusive. For that, a higher level of testing and the six sense of a gamer apresiation must be in the equation.
But I think the site its getting better, also the graphs appear to be giving the finger to the reader... or its just me?
Friday, July 04, 2008 2:17:24 AM
@evil...

Did i forget to brush my teeth? oh no did i show up in court without my pants on again?

relax, its good!

I think you should cram a little more into your conclusion, touch on all the data you've collected so far (all be it, you need more cards, but thats $$$), answer the question who should buy it and who shouldn't. Get in any outstanding frustrations as well as any oustanding positives! Longer conclusion would be better.
Friday, July 04, 2008 2:22:33 AM
I agree with MrWizard. More conclusion, that's what I meant by "subjective statements," although, you could also sprinkle a few of those throughout.
Friday, July 04, 2008 3:10:06 AM
OMG YOUR LIKE THE ONLY ONE TO ADD A SECTION FOR OVERCLOCKING :):):):)!!!!!!!
Ya I'm obviously excited about that part, becuase I overclock EVERYTHING, thats a great review, to bad you can't do 2 or 3 way SLI, but I totally understand, GTX280 = 2nd morgage, lol. Awseom man, just awsome.
Friday, July 04, 2008 4:15:46 AM
wyz135
Senior Member
Here's my suggestion: Why not have a part in this site, which have all the reviews made by you, or maybe us (I made a Radeon HD3850 review and I know it's awful).
Friday, July 04, 2008 4:36:57 AM
yeah I'm thinkin in the new gpu review you should have a section for user reviews. I (like, i suspect, a large portion of this site) wouldn't mind running my own review site. Having gpureview as a place where I can post my stuff to see what you guys think before going seriously public would be awsome.

Bring on the user review section!
Friday, July 04, 2008 12:01:10 PM
Mike
GPUReview Founder
Hmm, user review section...interesting.
Friday, July 04, 2008 3:33:55 PM
aliquidparadigm
Senior Member
You put up a user review section, I can provide at least... (looking through records) 3 reviews. 4 if I can possibly get one of my 3870s' fans to cooperate. And, assuming I'm not being lazy, I can actually write quality reviews.
Friday, July 04, 2008 4:47:02 PM
You could add two sections to the Review Finder tab: Official GPUReviews and User reviews. That way you could turn that into a nice central review part of the site.
Saturday, July 05, 2008 1:09:48 AM
Yeah, or you can Outsource reviews, like Nvidia and AMD with their GPU's if you don't have thet much time or resourses. Of course the supllier have to get something, maybe an advertizing in your site.
Saturday, July 05, 2008 10:32:37 AM
Radiator
Senior Member
158 degrees Celsius under load O_O ? Or Farenheit ?
But anyways , nice review .
Tuesday, July 15, 2008 9:53:22 AM
yay! Mike, did you write this review? I'm happy to see it! I was really wondering about this sort of sciency 'living room' test, as I've never gotten the performance people seem to get in a lab setting.
Add Comment
Name:

Comments:
This is a moderated comment list. All posts are subject to my personal approval. Please limit your comments to the subject of the article. Don't (for example) start asking questions about your video card here, you'll get a much better response in the forums.

Anonymous posts are allowed, but I encourage you to sign up for the forums and post comments under your name.
none   none