3DMark Vantage Soon

Agreed.

I beg to differ here. When XP came out people were complaining left and right that their games ran faster under Win9x. They said that XP was too slow, took too much memory (128 MB) and many didn't like the new shell (too colorful! too different!). Sounds familiar?

Driver quality for XP drivers at the time of launch was poor, much poorer than Vista drivers. Due to that XP bluescreened. A lot. In that days many people claimed that XP could never, ever, ever supersede Win9x as the predominant gaming OS. I don't remember how long it took XP to pull ahead of Win9x in terms of 3D performance, but it finally happened and XP became the predominant gaming OS.

Now when I compare XP and Vista at launch, Vista clearly wins. I think I could really be called an early adopter, because I installed Vista in december 2006 before it was even launched officially (the joys of having a msdn subscription). I had some driver issues at first, but most were resolved even before the launch. After that I only had two major issues (one caused by a bad driver, the other was resolved in a hotfix) and a couple of minor ones (most resolved by hotfixes in the last year). BTW, with XP you mostly had to wait for the service packs for problem fixes because Microsoft wasn't as forthcoming with hotfixes as they are now.

Currently I run Vista SP1 RC1 on a couple of computers here and since then I had zero issues. SP1 will probably come out this month (release notes are out already), so the chances are good that most of issues people have will go away and Vista uptake will increase even more. Nobody will be downgrading any more.

When it comes to games I remember that situation with XP was much worse than it is with Vista now. I remember that people would complain that new games would crash frequently under XP, the graphics were corrupted (due to broken drivers, back then Win9x and XP drivers usually had different code bases), or they were horribly slow to the point of unplayability. It's not that bad with Vista. New games usually run flawlessly unless they run out of memory (e.g. Witcher), just a bit slower. Some other issues have already been fixed by hotfixes (DX9 memory, SLI). With Vista's new driver architecture there is a good chance, that 3D performance finally will surpass XP, once drivers are optimized.

There is really no point in hanging on to XP.

 

Sweet! thats exactly what I wanted to hear

 

Any DX10.1 goodies in there?

 

I think the question on everyone's mind is, Do thread participants get a free copy for spreading the word to all of the INTERNETS? I have a 7900GTX that can skip every test!

Just a quick question, hope this doesn't trip an NDA. Do you guys recieve early hardware from THE companies to test your product on?

 

The fundamental fallacy of comparing Vista and XP is that XP as an OS been tweaked for over how many years? 5? 6? with 2 service packs.
I would say MS did a poor job and cheated all users IF NOW VISTA is FASTER than XP, an OS received so many tweaks and optimizations.
By comparing these two...any meaningful insight? Maybe. Any indication of how Vista will perform on future computers? No, just like when XP released.

Back to topic, looking forward to this!
Can have some thing to try out my Geforce 8800M GTS besides Crysis.

 

You can have the free version!

But a pro version is much welcome!

 

I don't know what it is, but seeing links to buy graphics cards on the site doesn't seem quite right to me.

 

Are links to CPUs or Systems on the site any better?

 

The whole idea of buying video card based on results in synthetic benchmark is bad IMHO.
I do think that "HardOCP way of evaluating" is more consumer oriented than all these synthetic/timedemo benches.

 

It might appear to be more consumer friendly, but as has been said, using gameplay only shows you how it will perform on that title. If you're not playing/don't intend to play that game, the conclusions that are drawn are not necessarily applicable to your own usage. This is compounded by the fact that the, labor intensive method (yes its still a method ), used by [H] limits them to testing very few titles.

I'm not suggesting the results used at [H] aren't as good as others, just that I wouldn't take them in a vacuum. I'm glad they are doing what they are doing, but I wouldn't want every site doing it.

 

I don't know, it just feels wrong... like there's a conflict of interest there or something.

 

http://r800.blogspot.com/2007/12/3dmark-vantage-brief-test.html
This was judged as fake originally in December, however as we saw later, the space shot at least is legit even if it's from older build, it's 100% definately from the same test as officially released pictures.

Now, new pictures surfaced: http://en.expreview.com/?p=260#more-260
There's no pictures of "Jane Nash" test, but the name suggest that at least the picture with the girl at R800-blog could be legit aswell, what about the rest? And more importantly - could the text be right aswell, could the 2nd test be limited to DX10.1 cards only?

 

The greenish feature test looks like some particles rendering with vertex tex fetch.

 

That site does help move hardware. People get too caught up in the final score and buy only to increase it. I know, I used to do it myself.

 

And presumably making a profit on the side, which is why it doesn't feel right to me.

A benchmark should be created to be objective. Throw money into it and imo that objectivity disappears.

Seems to me that I've argued this before.

 

And who do you think would pay the massive amounts of money a benchmark of these proportions costs to make, then?

 

http://shop.futuremark.com/shop/