R600XT to compete with 8800 GTS only

[^eMpTy^]

it's still scripted, pre-layed out, no games give you that sort of limitation during game play, why would you test under those conditions?

Well think about it this way. AnandTech, FiringSquad, Beyond3D, Hexus, HotHardware, Tech Report, etc have all used so-called 'canned benchmarks' for nearly a decade. And when has [H]ardOCP's conclusions ever disagreed with them? Sure the numbers come out a little different here and there, but the conclusions are always the same.

The term 'canned benchmark' vs 'real world benchmark' is essentially just marketing bullshit perpetrated by the [H] to make it sound like the [H] is somehow special. When in the end it all doesn't matter, the conclusions are the same. If you don't believe me, then go read the last 6 or 7 years worth of video card reviews and see for yourself.

 

[GoldenTiger]

actually my original account dated back even further, but I lost my login info and the email address it was regged on =p I started reading HOCP back when a guy named Fly did some reviews

So goldentiger why dont you tell everyone since you're all knowing what a time demo is

I think it's pretty obvious why you're even posting in these threads it's no surprise which brand you prefer

It has nothing to do with brand preference, I'm not arguing regarding that. I am arguing in regards to benchmarking methods. A timedemo, to the game, is a person running through pressing keys and mouse movements in the exact same fashion every time, with the exact same timings and amounts. Therefore, it is a "perfected" version of a supposed "real-world" run-through with *ZERO* room for variance due to human error or random disruptions.

 

[Digital Viper-X-]

That is a simpletons view.

When you record playthrough scripts, the only difference is that the script is driving, it will get the same result as when you played it the first time. The game doesn't figure out in advance where everything is going, it is the same as someone playing in real time as far as the game engine is concerned.

This is a PERFECT representation of real world play (because it is) with the advantage of being reproducible.

Complaining about canned benchmarks is a new low in whiny fan worship.

sorry when I talked about Canned banchmarks I was talking about 3DM and certain pre-installed "demos" like in Fear, or CS:Source

 

[GoldenTiger]

Well think about it this way. AnandTech, FiringSquad, Beyond3D, Hexus, HotHardware, Tech Report, etc have all used so-called 'canned benchmarks' for nearly a decade. And when has [H]ardOCP's conclusions ever disagreed with them? Sure the numbers come out a little different here and there, but the conclusions are always the same.

The term 'canned benchmark' vs 'real world benchmark' is essentially just marketing bullshit perpetrated by the [H] to make it sound like the [H] is somehow special. When in the end it all doesn't matter, the conclusions are the same. If you don't believe me, then go read the last 6 or 7 years worth of video card reviews and see for yourself.

Well-said, hence why from the outset I said (paraphrased) "While I like both, I find the canned ones easier to digest". Why I'm even bothering to post about this debate isn't because I feel the "real-world" ones are completely inaccurate, it's the high-and-mighty "yoor a fewl if u dunt beleeve ohnlee thee reel wons!!!" attitude that many posters display.

sorry when I talked about Canned banchmarks I was talking about 3DM and certain pre-installed "demos" like in Fear, or CS:Source

Only disreputable sites use timedemos that are easily known and manipulated in drivers for. That's all they are, is timedemos, but pre-installed ones (the benches you're referring to).

 

[Digital Viper-X-]

Well think about it this way. AnandTech, FiringSquad, Beyond3D, Hexus, HotHardware, Tech Report, etc have all used so-called 'canned benchmarks' for nearly a decade. And when has [H]ardOCP's conclusions ever disagreed with them? Sure the numbers come out a little different here and there, but the conclusions are always the same.

The term 'canned benchmark' vs 'real world benchmark' is essentially just marketing bullshit perpetrated by the [H] to make it sound like the [H] is somehow special. When in the end it all doesn't matter, the conclusions are the same. If you don't believe me, then go read the last 6 or 7 years worth of video card reviews and see for yourself.

While that is true to an extent, the tests I'm talking about most people don't run anymore,with the exception of 3Dmark , a recorded playthrough is fine, a PLAYER recorded play through, Fears demo, maybe not, old pre-scripted demos i wouldn't use either.

Also I prefer the Highest playable settings vs how many fps do I get at resolutions this card CANT play

another thing is I'd rather have someone sit and play the game, then tell me while It did get 70FPS the game didn't feel smooth

WHY? because I've had an expereince with a card that recorded high fps but played pretty crappy(NFS:MW on a 1900XTX) where it played alot better on a 7900GTO with lower overall fps, this is something you won't ever get a timedemo, because you need to have a feel for how the game responds to your actions as you do them.

 

[Dosomo]

You do realize that the R600 chip is ATIs top chip and it is that lack of performance, that is relegating it to battling Nvidias #2 cut down card with 32 stream processors disabled. The lettering they stick on the thing is arbitrary and driven by circumstance.

ATI didn't aim a 320 stream processor with a monster 512 bit data path at second place, they aimed for first and bungled it. ATI is only calling this their number two card because it is a screwup that can't compete for #1.

Bingo! We have a winner! It wasn't until after ATI saw that the R600 couldn't keep up with the 8800 GTX that they decided to name their fastest card the XT. This is marketing 101. Market your product against the competitor's product that it outperforms.

If it had turned out that the only nvidia cards the R600 could outperform were the 8600 GT/GTS, then trust me, the new HD 2900 XT would be branded the HD 2600 XT.

The new XT was originally most likely the "XTX with GDDR3 memory" that we heard rumors about months ago. The XTX with GDDR4 memory is now being called the real XTX, and is still not ready. By not releasing this faulty and slower XTX, they save face as best they can. And, maybe they convince some buyers to wait even longer for the new XTX. It isn't hard to figure out.

Honestly, I hope ATI gets things worked out and puts out a competitive product. Otherwise, nvidia will have little reason to release faster cards anytime soon, and the market will stay as it is.

 

[Trimlock]

Well think about it this way. AnandTech, FiringSquad, Beyond3D, Hexus, HotHardware, Tech Report, etc have all used so-called 'canned benchmarks' for nearly a decade. And when has [H]ardOCP's conclusions ever disagreed with them? Sure the numbers come out a little different here and there, but the conclusions are always the same.

The term 'canned benchmark' vs 'real world benchmark' is essentially just marketing bullshit perpetrated by the [H] to make it sound like the [H] is somehow special. When in the end it all doesn't matter, the conclusions are the same. If you don't believe me, then go read the last 6 or 7 years worth of video card reviews and see for yourself.

While this may be true or not, I'm not willing to argue it, since it seems to be an opinion and I'm going to respect it.

How ever what I absolutely love about the [H]'s review is the presentation, not only does the graphs give me a feel for which card takes certain dives but they also explain it, not just placing graph after graph of just about every freakin setting possible they can get out of the game

another thing I like about their presentation is they keep performance generally the same, if you play at that setting or not I think they were able to find a well placed middle ground and show which card (keeping around the same frame rates) can give you a better picture, instead of having to compare 10 graphs to find out which one you'd rather have

and yes, I am that lazy

 

[Zinn]

otherwise daamit is going to sink, and sink fast

have you seen their stock lately

probably IBM will end up buying them like with what happend to Cyrix.

 

[Hypernova]

Even if XTX falls flat on its face ATI will probably still win on the high end market. Most people when shopping for high end gears can't afford the GTX and since the XT is ahead of GTS by a healthy margin the sales of XT will still help ATI to the top in the market.

Arm chair speculation aside the real battle is yet to begin.