Hard Light Productions Forums
Off-Topic Discussion => General Discussion => Topic started by: colecampbell666 on February 07, 2008, 01:29:58 pm
-
Post your 3DMark06 score!
Mine: 310 3DMarks, 367 PCMarks.
-
Bah. If I'd run Windows on BootCamp on a Mac Pro with it's commercial grade 1.5 GB GFX Card, I'd pawn you all. Now lets scratch together the 20 grands :eek2:
-
Nah. More like 6 grand.
Still a ****load of cash.
-
3dmark is even more meaningless these days than it used to be a few years ago. :p
commercial grade 1.5 GB GFX Card
Those are generally nothing special for games (or 3dmark).
-
Why?
-
Other than Futuremark making a game and releasing it within like 2 years... 3DMark is pretty useless. I don't have scores--3DMark06 October wouldn't install on Vista for some reason or another.
-
Other than Futuremark making a game and releasing it within like 2 years... 3DMark is pretty useless. I don't have scores--3DMark06 October wouldn't install on Vista for some reason or another.
Why is it so useless?
-
I'm curious too. Why do you consider it useless? I've found it useful for getting a general idea of what a PC can do quickly.
-
1: It stresses shader power too much over fillrate. The default settings in 06 (which are the ones used in all the scores you see) are 1280x1024 with no AA, but there are tons of shader effects used. This is why, for example, the 3870 often defeats the 8800U in it despite being slower in games.
2: The current version, 06, is almost three years old, and even that is just a minor update of 05 with HDR added in.
3: It's unrealistically dependent on CPU power and multiple cores in particular, more than any actual game. This is even more so now than it used to be, since video cards have gotten much faster since it came out.
4: The predetermined camera angles (combined with the widespread usage of the program) allow Nvidia and AMD to make heavy optimizations to inflate the 3dmark scores, much more than they could do in a real game.
The only thing it's useful for is to make sure that your system is running properly and you're getting roughly the same scores as other people with the same hardware. It's not much good for comparing different setups.
-
Well, it can be useful to detect minute changes in your systems settings, for example you spend hours tweaking the memory latencys and see that your score has risen by 200 points through it! If that is what you like to spend your time on, of course. :ick:
-
I would tell you my 3dmark score, but it would require me to care enough about overrated synthetic benchmarks to download, install, and run it. But since I don't, I won't be able to mention a score that has absolutely no merit whatsoever.
-
Of course, for the 1.5 GB card, it depends on WHAT is tested. AFAIK 3DMark or whatever wants to test game situations, whilst the 1.5 GB card is optimized for high-duty rendering in applications like 3D Max, Bryce, Lightwave and the likes.
-
Of course, for the 1.5 GB card, it depends on WHAT is tested. AFAIK 3DMark or whatever wants to test game situations, whilst the 1.5 GB card is optimized for high-duty rendering in applications like 3D Max, Bryce, Lightwave and the likes.
This thread is about 3dmark, and by extension, gaming. :p
Also, those cards don't actually help with the final rendering at all. They are used for moving around in the viewports.
-
They are mostly about holding lots of big textures for the OpenGL rendering in editing, which is kind of pointless if you are running Lightwave, because it resizes all the textures for the previews anyway. So whilst it may render a 4096 texture in final, it stores a 1024 version of it in Video memory for use by the OpenGL renderer.
-
Other than Futuremark making a game and releasing it within like 2 years... 3DMark is pretty useless. I don't have scores--3DMark06 October wouldn't install on Vista for some reason or another.
Why is it so useless?
I'm curious too. Why do you consider it useless? I've found it useful for getting a general idea of what a PC can do quickly.
It has no basis on actual gaming performance. It's all about the Catalyst and Forceware drivers used, and how ATI or nVidia tweaked the drivers. That's why there's often the large rearrangement of cards between synthetic and gaming benchmarks...
-
Aaaaaahhhhh yes, I fondly remember the days that people would proudly boast online whatever their best 3dmark scores in what turned out to be a vain penis measuring contest. Damn I feel old now. :p
4: The predetermined camera angles (combined with the widespread usage of the program) allow Nvidia and AMD to make heavy optimizations to inflate the 3dmark scores, much more than they could do in a real game.
As I recall Nvidia got nailed pretty hard for doing that in '03.
-
Aaaaaahhhhh yes, I fondly remember the days that people would proudly boast online whatever their best 3dmark scores in what turned out to be a vain penis measuring contest. Damn I feel old now. :p
4: The predetermined camera angles (combined with the widespread usage of the program) allow Nvidia and AMD to make heavy optimizations to inflate the 3dmark scores, much more than they could do in a real game.
As I recall Nvidia got nailed pretty hard for doing that in '03.
It stil happens--just go to some Enthusiast forums. It's a measurement of how well you overclock too, but it has no basis on real use--thus it's synthetic.
-
Aaaaaahhhhh yes, I fondly remember the days that people would proudly boast online whatever their best 3dmark scores in what turned out to be a vain penis measuring contest. Damn I feel old now. :p
4: The predetermined camera angles (combined with the widespread usage of the program) allow Nvidia and AMD to make heavy optimizations to inflate the 3dmark scores, much more than they could do in a real game.
As I recall Nvidia got nailed pretty hard for doing that in '03.
It stil happens--just go to some Enthusiast forums. It's a measurement of how well you overclock too, but it has no basis on real use--thus it's synthetic.
But if a card has a higher score, say the Radeon XXXX had a better score than a GeForce XXXX wouldn't that correspond to a higher score ingame?
-
But if a card has a higher score, say the Radeon XXXX had a better score than a GeForce XXXX wouldn't that correspond to a higher score ingame?
Yes, when you are assuming that games behave exactly like 3DMark. Which they don't, for the most part.
EDIt: For what it's worth... I'm getting 2971 3DMarks (http://service.futuremark.com/resultAnalyzer.action;jsessionid=56BDEF4F70A12EAFE9F80131A4EFD6A1) on high performance settings (lowest quality, no FSAA etc.). CPU tests are a joke, they have about 1/15 FPS and take ages to run and the timer is slowed down so it isn't really what I'd call trustworthy test. :blah:
EDIT2: And the result comparing system SUCKS. The link to the result page doesn't seem to work. :rolleyes:
-
9864
SM2.0: 4356
HDR/SM3.0: 4713
CPU: 2281