![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
3Dmark07 Vista only?
- Thread starter sam0t
- Start date
Dark Prodigy
Jawbreaker
- Joined
- Mar 10, 2006
- Messages
- 2,803
Not surprised by this at all. A little 'early',but inevitable.
Yeah i dont see why you would want the new 3dmark on XP to test dx9 with other simulation, but the new 3dmark will be multi-threaded. Finnaly something to test with Dx10 card, i have a 8600gt and ill like to check compared to a gtx 8800 what is the difference in dx10 lol
officermartinez
[H]ard|Gawd
- Joined
- Jan 21, 2006
- Messages
- 2,012
LMAO.. now I don't feel soooo bad buying dual DX10 8800gts's and Vista Ultimate x64. The newest 3dmark is going to be a nut puncher. I can see it now.. With my setup, I have a feeling my score will be LOWWWW!
Brent_Justice
Moderator
- Joined
- Apr 17, 2000
- Messages
- 17,755
3dmark07 validates your purchase of a DX10 video card?
Dark_Legacy
Limp Gawd
- Joined
- Dec 7, 2006
- Messages
- 251
3dmark05 score = 9,452
3dmark06 score = 5,100
3dmark07 score = 2,752
Variance Factor = 1.853
Divide your current 3dmark06 score by 1.853 and just skip the DX10 testing (for those of us unable to run DX10 testing natively.)
Your score is going to be 9,196 3dmarks on 07'. Still quite impressive.
3dmark06 score = 5,100
3dmark07 score = 2,752
Variance Factor = 1.853
Divide your current 3dmark06 score by 1.853 and just skip the DX10 testing (for those of us unable to run DX10 testing natively.)
Your score is going to be 9,196 3dmarks on 07'. Still quite impressive.
Dark_Legacy
Limp Gawd
- Joined
- Dec 7, 2006
- Messages
- 251
Hey guys, we should all switch to Vista so we could can go run benchmarks! Yeah! Pwn!
Great.
Hey guys, we should all switch to Vista so we could can go run benchmarks! Yeah! Pwn!
Great.
Pretty much what I was thinking.... "Yup, I'm going to buy a new operating system just so I can show all my buddies how high my 3dmark score is!!"
AltTabbins
Fully [H]
- Joined
- Jul 29, 2005
- Messages
- 20,487
Hey guys, we should all switch to Vista so we could can go run benchmarks! Yeah! Pwn!
Great.
People drop thousands of dollars on new videocards, processors and memory for that number.. I doubt vista would stop them.
Dark_Legacy
Limp Gawd
- Joined
- Dec 7, 2006
- Messages
- 251
I still don't think it's worth switching over to Vista over 3Dmark07, but perhaps for the people with DX10 cards sitting in their machines already it might be.
officermartinez
[H]ard|Gawd
- Joined
- Jan 21, 2006
- Messages
- 2,012
Absolutely not. Nothing this expensive should ever validate the end result, which is only a simple number / score. For a benchmarker, it's "fun" to reach a high number. The higher the better. For the average end user, it's not so fun. With as much money as I have tied up in my current setup, I definately see myself spending more to achieve a higher score. But... that's only a portion of the story. Personally, I like being on the "bleeding edge" so to speak.. Yeah, the rewards are marginal and amoung MY circle of friends, I guess there is a little braggin involved.. But at it's roots, it's still a hobby. My wife could care less if I score xx,xxx on 3dmark.. She could care less if I am able to max out my eye candy in games, etc..
It's just the natural progression of things.. The more involved you are in a project or hobby, the better you want it to perform..
burningrave101
[H]F Junkie
- Joined
- Sep 9, 2003
- Messages
- 11,825
The better question would be 3dmark07 validating his purchase of Vista Ultimate x64 lol. Unless he got a great price on it I don't know why anyone would buy the Ultimate version. It has nothing of real benefit. There are several benefits of running Vista over XP though. Most of the Vista bashers have never ran Vista long enough to find out those advantages. Application performance is definitely better on Vista because of the improved kernel and better memory management. Vista just isn't too great for gaming yet.
Keep in mind that Futuremark is a company that's out to make money, and a company that doesn't release new product is dead in the water, so I think that 3DMark07 is a natural progression.
If you don't want to upgrade to Vista, don't. However, I think that most of the anti-Vista hold outs are the same guys who ran 2000 for 3 years after XP was released.
If you don't want to upgrade to Vista, don't. However, I think that most of the anti-Vista hold outs are the same guys who ran 2000 for 3 years after XP was released.
Zardoz
2[H]4U
- Joined
- Aug 27, 2000
- Messages
- 3,251
I still find it funny that people are buying hardware to run a benchmark.
Same people who buy a V8 or a 200+ HP car to drive on public roads...
Yep. I know a few of those folks who only switched a year ago to XP.
At least chicks dig fast cars. I know you cant say that about video cards, or 3dmark benchmarks. And cars are cool, fast computers, not so much.
This has been known for several months. Im sure Ill run it, because they always look great, and its something to test an overclock with. But putting merit into their score, is something I lost long ago. Giving a false representation of how PC's compare is pretty pathetic to me.
Eh? Its an accurate representation of the 3D capabilities of the machine-how is it not? And about the car, don't tell me you guys actually stick to the speed limit? There's plenty of scope to use a 200bhp car on the roads, just don't get caught doing it
It's so 'hip' these days to diss things like 3dMark, The Inquirer, Fudzilla and such. Most people who diss these things have never produced anything worthwhile themselves in their whole lives.
3dMark gives accurate results in almost 90% of the cases. Some people try to concentrate on the 10% to make themselves look 'cool'. The same is true for the news of Inquirer or Fudzilla.
Guys, 'Coolness' is something that comes from within you. If you don't have it, you can't get it by badmouthing 3dMark. Instead, try to produce something cooler yourselves. For that of course you might have to move out of your parents' house first.
3dMark gives accurate results in almost 90% of the cases. Some people try to concentrate on the 10% to make themselves look 'cool'. The same is true for the news of Inquirer or Fudzilla.
Guys, 'Coolness' is something that comes from within you. If you don't have it, you can't get it by badmouthing 3dMark. Instead, try to produce something cooler yourselves. For that of course you might have to move out of your parents' house first.
Brent_Justice
Moderator
- Joined
- Apr 17, 2000
- Messages
- 17,755
3DMark gives accurate results in 3DMark, that's it.
That is just not true. If video card A beats video card B in 3dMark by more than 10%, it's likely to be faster in 90% of the games out there. You're just focusing on the exceptions to make yourself feel smarter than the rest of the industry.
DocFaustus
2[H]4U
- Joined
- Sep 22, 2002
- Messages
- 2,830
Heh, and as soon as you post this, the producer of video card reviews, Brent Justice, puts up a contrary post. So you immediately become one of the "hip" and diss him. You are pretty good with percentages, would you mind entertaining us with proof of the numbers you bring forth?
What have you produced?
Betrayer00GT
Limp Gawd
- Joined
- Feb 25, 2007
- Messages
- 350
lol
I switched to XP from W2K the end of this last February. Then, about a week later, I switched to Vista. I continued to run W2K until there was a reason not to (Supreme Commander required XP SP2+). I refused to deal with activation until I had to. Turned out that by the time I had to, I already had DX10 hardware. I didn't realize DX10 hardware required Vista until I tried to run the Cascades demo on XP. Sad, but true.
I effectively skipped XP, and I wear it proud as one of the seminal achievements in my life (
). Just like I wear my 13 second 1/4 mile time from my '00GT (>200HP V8), which I drive only on public roads until I can afford to have two different cars I like, one for racing and one for commuting. That will be sometime after I absorb the cost of a Kenne Bell supercharger getting installed in the Mustang. maleficarus
Gawd
- Joined
- Nov 17, 2004
- Messages
- 744
I agree. 3dmark is a solid benchmark. I think it always has been. [H] took the stance a few years ago now I think to not use 3dmark in any of there testing. They seem to think it isn't real world which I have to laugh at. If you chug in 3dmark 99.9% of the time you will be chugging in any new game that will come out. Seems real world to me. If you kick ass in 3dmark 99.99% chance you will be kicking ass in ANY game that comes out or is out. If that isn't a reflection of real world then what is?
People started having a hard time with 3dmark at around the time that NVIDIA and ATi started optimizing driver builds specifically for 3dmark. Some felt this was a bad thing but what these people seem to forget is NVIDIA and ATi would release dozens of quake3 optimized drivers just to increase the time demo scores even if by a few frames. Sure it is a real game but 3dmark is a real game in a sense. It is another thing to play around on your PC. Absolutely no different then running time demos. From what I saw that started happening is as soon as future mark started approving only certain driver builds some got upset over this and then started dissing the benchmark altogether.
[H] Posted up an article a few years back about 3dmark and real world benchmarking all of which I didn't agree with at all. Fact is 3dmark is one of the most popular and widely known PC benchmarks in the industry. It is here to stay whether sites like [H] like it or not...
Mark my words...[H] will be forced to use it sooner or later especially with DX10 as they won't have a darn choice! =)
Zardoz
2[H]4U
- Joined
- Aug 27, 2000
- Messages
- 3,251
You are playing with numbers, 3DMark is a tool to compare systems to systems running the same tool. This gives you somewhat of an out look of how one system will match up with another using the same tool. however you have to take in account real world games, they use different engines and use a system differently for one game to the next.
To put it in simple terms lets call 3DMark a game. It has it's on engine it on set of code and how it works. would you not want to use other games to see how well it compares to 3DMark? I mean quake1 runs a lot faster then quake4, but you have to take in account that they are different engines, I hope you understand.
That's the same for 3dMark too. One test is shader intensive. One test is HDR intensive. Another is geometry intensive etc. etc. You can choose to run the test you want depending on what you're trying to measure.
It may not be completely accurate, but it's accurate enough to get a feel of how a card performs compared to others.
The opposite, [H]'s style of benchmarking... (I want to stress that I'm not flaming, I'm just giving my opinion)... is taking a few games and trying to find out the best playable settings and resolutions, is not a good idea. There are tens of thousands of games. No matter how many games you test, you still can't give us any idea about the card's performance because there are many more games people play and you haven't tested.
I agree with this somewhat. I have personally found [H]'s reviews to be less useful when trying to get a feel for general performance. When they list all the settings at which (in their opinion) a card is playable in some game it's sometimes impossible to determine a particular architecture's strengths and weaknesses. There are too many variables and too little room for interpretation left to the reader. Telling me how one card performs in one game at specific settings is no better than 3dmark and arguably worse since it's not comparable to anything.
I've found myself reading [H]'s stuff less and less as the info there just isn't very relevant or useful to me. In its defense if you are looking at specific games and specific hardware and don't want to fiddle around with settings yourself the reviews here might be helpful.
To me buying a 200+ HP car to run on public roads is more like Buying a GTX to game at 1280x1024, perhaps overkill but not a total waste.
A better analogy would be people buying a powerful car Just to stick it on a dyno.
To keep with the car analogy (without wanting to go too far o/t) when comparing [H]'s testing methodology with other sites, the others just stick it on a dyno and say 'X wins because it has most raw HP' whereas the [H] takes it for a proper test drive.
I'm not dising 3Dmark, but to me it's more usefull as a diagnostic to see things are working as they should, than a guide to how the next big thing will run on my system.
For the time it takes to run, I'd personaly like to see a 3D Mark score as a footnote to the review proper, I know it's available elsewhere, but it'd be nice to have it under one roof.