The Fastest 3D Cards Go Head-To-Head

[bjornlo]

Excellent article. Lots of information. Good format. All around a very informative read. The 'net needs more articles like this one!

One suggestion for improvement was the overly narrow focus on the cards direct decendents. For example, when talking about the 4850, the assumption was that only 3850 users would be interested in it. The assumption while talking about 4870, you always refered back to ther 3870. The 260, and then it was the 8800gts. A more wide open approach and not brand specific discussion would have been even cooler. For example, I own a 8800gts 512, and have been considering a 4870. I imagine I can not be the only reader considering the "other" brand. For me it is all about the framerates, I really don't care what brand it is. Only what it costs and what I get for my benjamins.

 

[jameskangster]

First I want to say that the article itself is not bad at all. Also, I can understand why TH didn't have time to use 8.8 since it was released publicly on August 20, 2008 (Although ATI would have gladly released a beta version to TH for testing purposes). However, AMD publicly released stable Catalyst 8.7(internal version 8.512) on July 21, 2008. That's more than a month ago. It has numerous improvements (for example, CF performance increase, improved stability and performance under Vista). To be honest, most of the improvements range from 4% to 15% (In CF case, up to 1.7 X scaling). TH has rarely been unfair and/or inaccurate and they always owned up to their mistakes before, and I trust them to re-test ATI products with at least 8.7 if not 8.8 to continue to uphold their values and integrity.

 

[scelero]

[citation][nom]mjam[/nom]No 4870X2 and 1920 X 1200 max resolution tested. How about finishing the good start of an article with the rest of it...[/citation]

as much as i support tomshardware, i agree mjam.... the article says fastest 3d cards go head to head but I dont see any 4870x2. That would be a match to see!!!

 

[xrodney]

[citation][nom]Haiku214[/nom]Well the main reason why they don't have the 4870x2 and the latest drivers is simply because they made this article a couple of weeks ago. If you could just imagine how long and tedious it is to produce all these data and results. It's just sad that after finally finishing the article, a lot of new stuff has already happened(new drivers and the x2).[/citation]
Well if i remember corecly 8.7 drivers are almost 2 months here ?

 

[liubros]

The only thing I would complain is that they should using a much better CPU for the testing. All the above result are bottle throated by the testing CPU actually. I am using a MSI OC GTX280, and I am using a Q6600 OC at 3.2g. And my 16800 3dmark beat all the results, simply because my CPU is running at a higher speed. I think it might need 4g quad core to see the real value or SLI, or crossfire http://service.futuremark.com/compare?3dm06=7828994.

 

[cosmicc]

What's with all these ATI fanboys flaming the article? I agree that a better CPU, and better ATI drivers should have been used, but c'mon guys, give Tom's Hardware some credit, I don't see any of you doing extensive graphics card testing...

 

[eklipz330]

i don;t agree with anyone that says they should have used a more powerful cpu, read the title, they're trying to make a point

besides, there a plenty of benchmarks out there with more powerful cpu's... i don;t need to see another one, this is good to see on how the gpu's do on these not so great procs if any1 understands what im trying to say

 

[squiZZ]

@Haiku - yet I've had my 8.7 drivers FAR longer than 2 weeks. IIRC i dl'd them around a month ago. I really didn't expect them to post 8.8 driver scores, as you are correct and most of these tests happened prolly before 8.8 release. But....8.6?!?!?! R u Serious?

 

[scurvywombat]

I would just like to echo some of the comments above by saying that this is certainly a well-done and nearly exhaustive review. The lack of the 4870x2 and latest catalyst drivers really do damage this article's usefulness though. If I were in the market for a new graphics card, I'd skip this article because of that and check other sites' reviews. However, maybe there is a bright spot on the horizon - hopefully Tom's will redo their benchmarks with 8.8, include the 4870x2, and finally update their vga charts!

 

[customisbetter]

This is one of the best articles i have ever had the pleasure to read. The fact that it is on Tom's is a total surprise given the last few articles i have read. Keep it up Tom's!!!!!!

 

[xx12amanxx]

Wow just wow no 4870x2?...This article is useless when you dont even have the world's fastest single card! Whoever did this should be ashamed!

 

[Proximon]

I realize this is the result of a great deal of work, and I can find some good information here and there. However, you should have discarded results that were clearly useless and inaccurate. By lumping all results together without identifying the erroneous items, you only invalidate the entire work and render it useless to those who really need it, namely the average consumer.
This kind of undertaking should be an ongoing project that is continually updated. Not a one-time picture made with your thumb over half the lens.

 

[Guest]

CrossFire and SLI, Nvidia's GeForce GTX 260 and 280, AMD's Radeon HD 4850 and 4870--they're all here in a battle royale, with comparisons to previous-generation boards that'll help you decide just how much graphics horsepower your next machine needs!

The Fastest 3D Cards Go Head-To-Head : Read more

 

[am71]

Thank you to the testers. Seems you did the best with what you had. Do you really think these results are valid on your benching platforms? Really?

Anyhow, I don't need comparisons of VGAs. Anyone who has bought more than one video card knows that price pretty much determines performance.

I would like to see comparisons of whole systems with the same video cards. I want to know how much difference there really is between a high end Intel and AMD system in real world apps and games.

 

[xx12amanxx]

Wow what a dissapointment1 The world fastest single card the 4870x2 is missing and on top of that they are using old 8.6 drivers which are not even coded for the 4k series!

Shame really a great article that shot straight into the toilet!

 

[njalterio]

Excellent article!

This is why I keep coming to Tom's Hardware. Recently there have been some strange articles, but this is a great one.

I think I will be sticking with my 3870 and this is why:

1) I don't play Crysis.
2) 30+ FPS is fine for me.
3) Good thing I got an X38 motherboard, as I will just add another HD 3870 in a couple months (along with a better monitor).

 

[Guest]

They Do Not Do These Tests Over Night!

But maybe a retest with the 4870x2, after Nvidia's Big Bang 2 driver release would be in order. See how the new drivers hold up, see how Nvidia's SLI in a window/dual screen works out.

Possibly Tri-Sli as well.

2 4870x2s vs 3 280 gts, whose king of the mountain!

 

[Guest]

Not including the 4870x2 is horrendous - no matter how recently they reviewed it. This is supposed to be a "comprehensive" article on the "fastest" cards?

...right

 

[goonting]

its a confusing article...hehehehehe.... it counters the previous benchmarks of tomshardware

 

[goonting]

They just want to confuse the un aware buyer into buying crap....

 

[goonting]

my hd3870 is faster than hd4870 hehehehe they want to trick buyers to buy the wrong card

 

[TheGreatGrapeApe]

Stop re-printing reviews from the German site.

This article is 24+ days old (published on the German site Aug 4th, review however many days before that!), and shows its age by the glarinngly outdated drivers, and issues, and it's lack of the HD4870X2.

Either translate and re-publish reviews within 72 hours, update them with fresh tests, or else don't re-publish them ever.

It simply makes the site look bad by having a terribly out of touch article on the front page and one that permanently shows it to be out of line with what every other site is showing with updated drivers from both companies.

In this case, no news at all would've been far better than regurgitating an out of date review.

 

[dannyaa]

The way they add points up at the end is not a good reflection of actual performance. The 4870 beats or dominates the GTX 260 in most tests - except mass effect. Where test after test, the GTX 260 is about 20-some fps ahead of the 4870 for whatever reason (poor driver optimization? Other UT3 games work much better with the 4870). Anyway, because of this, the GTX 260 ends up with over 100 extra FPS in the final comparison...

When in reality, if you took out Mass Effect, the 4870 would be the CLEAR winner over the GTX 260.

That said, even with Mass Effect, the 4870 is still very close to the 260... surprisingly.

 

[NightLight]

First of all, good riddance to gill bates...

What is up with knocking on everyone's review lately... I think Tom's hw should begin a big usercleanup.

Imho, both nvidia and amd have good cards out there. The difference in visible detail is at a minimum. I switched to the nvidia camp beacause they are more stable, and no driver issues as opposed to amd, with lots of trouble. That alone is worth the extra $ ...

 

[wh3resmycar]

"Nvidia’s disadvantage is high noise, but for that price, we’re willing to live with it."

im sorry, whats this? you're paying a premium for a card that'll sound like a vacuum cleaner? for that price i'll be expecting more.