You're being dense on purpose, whilst being ignorant of the effects the drivers have or don't have based on the tested scenario. Your assertion is incorrect.
AMD: R7xx Speculation
- Thread starter Unknown Soldier
- Start date
- Status
You're being dense on purpose, whilst being ignorant of the effects the drivers have or don't have based on the tested scenario. Your assertion is incorrect.
not hardly. HOCP is proof of that. remember the 20% boost a driver set was said to give the X2. Turns out they were correct for the in game bench, but actuall performance increase was only 2-4%, SO pardon me for being a skeptic.
Mintmaster
Veteran
NVidia didn't do it last time with NV7x. They could have heavily undercut ATI on price due to G71 being so much smaller than R580, and G73 being so much faster than RV530 (and a bit smaller too).
Also, what makes you think NVidia's parts are immune to small margins when undergoing price cuts? The 9800 GTX is not cheap to make.
What evidence do you have that "flyby's, cutscenes, walkthrus" show much different relative performance figures between video cards than in game benches?
I said they could, didn't say they would. And in game benches(those built in), flybys, walkthrus and cutscenes are not actual game play. Walkthrus are by the worst, gee lets clear and area of everything so that absolutely nothing is there and walk thru and not see how the cards actualy handle the game itself. Awesome way to see how well a card can handle all the explosions, guns and debris that accurs in the game. Oh wait, its walk thru so NONE of that is taking place.
So you're basing all of this noise on something Kyle found a while ago?Interesting.
No, you're simply being dense as I said, it has nothing to do with skepticism. You didn't test it in any way, shape or form, yet you already parrot your flawed interpretation. I find that annoying, simply because the particular driver build that ended up being the 8.5s has been tested, and shown to significantly improve things(beyond what is noted in the release notes). Even with your beloved real-world whatever testing.
But that's not even the issue, the issue is having folks that don't actually know what they're talking about say things in such an absolute manner. You don't have the cards, you haven't tested the driver build, yet you're schooling everyone about the combination. Why is that?
Shown to improve things using bad benchmarking tests. Show me where any review using them says they actually played the game and giving those results of playing the game and not the results of walkthrus, cutscenes, in game benches or flybys in the graphs. Only H gives the numbers from actual game play. Sorry, but we will simply have to strongly disagree on what REAL WORLD GAMING benifits the 8.5s give. And how the hell do you know what I do and DO NOT own.
Oh boy!
Got a site that says they actually play the games in their review and use the numbers from those runs, then by all means, please share.
http://forum.beyond3d.com/showthread.php?t=48183
Like some one said earlier, you really are chasing your own tail. Its getting tiring.
Nice link to a press release with numbers from ATI/AMD, now how a review site that actualy used them actually playing games and giving those numbers. MMMMMMMMM, I can't seem to find any, so all we have is AMDs word on this, no thank you, not buying it.
Mintmaster
Veteran
You didn't answer my challenge, which just shows you how full of crap you are.
Relative performance in flybys etc. is rarely much different than in game. If GPU A is 20% faster than GPU B ina flyby, its rarely less in-game unless your CPU is holding things back (in which case GPU performance is irrelevant).
Most of us bloody do it. Do you actually think that what [H] does is an epic novelty?We all run through the games whilst testing to see how they actually play. Most current games lack in-built testing tools anyhow, and you're forced to FRAPS through them. Which is not the best thing since sliced bread. Sigh.
Google is your friend. Search for vantage hotfix schmotfix. The Vantage hotfix drivers are equal to the 8.5 Catalysts. Disregard the Lost Planet and World in Conflict numbers since those are inbuilt tests and they might hurt your feelings by being that (in spite of the improvements being verified in-game as well). The others are Frapsed-through game-scenarios.
And this is epically O/T and tiresome. You don't have the cards simply because:
a)if you had had them, you would've checked directly through your beloved in-game testing and wouldn't have spammed about how only real world testing does this and that
b)if you had them and your assertion of no real improvements being observable was true, you would've certainly made a lot of noise about it when challenged.
I've personally grown so tired of the educated forumite that adheres to whatever philosophy([H] realworld testing, site X colorful testing, site Y underwater testing) and makes blanket statements about everything under the sun based on his convictions, without even remotely considering that the possibility of him being wrong exists. Is it so hard to add an IMHO at the end?Or at least make documented postings?
Cat 8.3 VS Cat 8.4+Hotfix:
http://www.rage3d.com/articles/vantage-hotfix/index.php?p=1
Huge diferences.... and they are not Cat 8.5. With 8.5 the diference is bigger.
If you read the same article G80 had exactly the same beaviour. In the crysis stress test vs real gameplay both G80 and R680 had completly diferent results, so don´t see just to 1 side of the field.
All companies do optimization.
No, its for you to leave this thread and complain some where else.
Sigh. No. The Vantage Hotfix drivers tested there are equal to the 8.5 Cats for all intents and purposes. Can we get back to the topic of the R7xx?
Thanks for the info, but you see by that review that diferences are huge.
For example WIC improves 50% in avg fps.
The Nvidia graph whould look diferent with the Cat 8.5
- Status
