• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Why do people live and die by 3DMark to bench their systems?

NFS4

NFS4

No Lifer
I swear, people cling to this benchmark like Anna Nicole Smith on a chicken sandwich
rolleye.gif
People get a 900 point drop and they think the world is gonna end. They get a 20 point increase and they act as though the gain is like a gift from Allah.

I just don't get it
rolleye.gif
 
Oh my god man, you must have missed my many, many 3DMark rants in the video forum. I HATE 3DMark, and people do nothing in video but whine about their score being too low, and then they get all defensive when I snap at them. People are idiots, they don't udnerstand that it isn't an accurate benchmark, that game scores are sooo much more accurate, and frankly, who really cares??? I for one don't sit at my desk and watch 3DMark all day. If you have great UT2k3 scores, but your 3DMark is low, does it matter? Of course not! Back in the day, people used the excuse that there really wasn't a good D3D benchmark around. Well, now there is, so there are no excuses. 3DMark sucks, always has sucked, and always will suck, but people will still treat it like it is all there is to the speed of a computer.
 
3DMark is still a good tool for testing the stability of a system. It's fairly accurate though... The score only seems to differentiate +-30 from a central score. However, I don't worry if I lose a few hundred in it for the sake of stability, or whatever the reason may be.
 
Yes, it does make for an excellent stability test, that is all I use it for. However, it is not an accurate way to compare systems. I have a Radeon 9700, the fastest video card out, there are tons of people out there with GF4 Ti4200's that probably have higher 3DMark's than me, because they sit there messing with their settings until 3DMark is maxed out. Gues what, load up a game, and I will blow them away in speed AND quality.
 
I agree. Drives me crazy. Every time a new nV driver release comes out, there will be 10+ posts with nothing but how much 3DMark went up or down. "I lost 50 3DMark points. These drivers sux0r". Not one word about visual quality, nothing about actual games that matter.
 
I don't really use it to benchmark, I just like to use it to give me a good idea of what the hardware can do as a team.

Bill
 
Originally posted by: NFS4
I swear, people cling to this benchmark like Anna Nicole Smith on a chicken sandwich
rolleye.gif
People get a 900 point drop and they think the world is gonna end. They get a 20 point increase and they act as though the gain is like a gift from Allah.

I just don't get it
rolleye.gif
Click to expand...

since when does society make sense?
 
I use it once in a while when people wanna see what their new fancy GFX card can do, but haven't got any games handy, but that's it.

I've never installed it on my current box, I see no reason to, all my games work fine without performance problems, so I couldn't care less if my 3DMark score was 50 or 50000.
 
IMHO, 3d Mark is more of a stablity test. If you can run it 6 hours without any massive slowdowns or system stalls. You've got a great rig.
 
Hehe yah, I find it funny that some people seem to think 3dmark is their favorite game.
rolleye.gif
They spend more time running 3dmark in search of the perfect OC/tweak, and forget the whole reasoning behind the need for speed. As mentioned above, it is a good tool though to test stability, particularly GPU overclocking. Artifacts and crashes occur quickly if your OC isn't stable.

Chiz
 
Well,

I understand that the industry needs tools to compare different chips, and easily understandable rating for them. I will not comment whether 3Dmark is the best tool for this, but thats obviously why it was made.

That individuals tweak the sh!t out of it to improve their scores... well
rolleye.gif



-DaFinn
 
I would ask the same question about those bloody Sandra benchmarks! They are less than useless. The ALU and FPU benchmarks are worthless and the RAM benchmarks are "broken". I'd much rather people used STREAM for that. There is a compiled version of STREAM floating around called WSTREAM.

There is a reason why STREAM is an industry accepted benchmark and Sandra isn't...
 
lol do you know what I think is the FUNNIEST thing about 3dmark 2k1 SE? The only people complaining about the benchmark are the ones who feel their score is out of whack. I'm certainly not complaining and whining...Its a tool, I use it. Enough said...why cry and complain about it? Its a free world, if it bothers u... DON'T FVCKING USE IT ...and stop crying about the other thousands of people who do choose freely to use the benchmark.



....because they (3d mark users) sit there messing with their settings until 3DMark is maxed out. Gues what, load up a game, and I will blow them away in speed AND quality.
Click to expand...


whiney voice/ "Oh...I hate sitting here messing with my settings....what a stupid benchmark!!!" end whiney voice/

I'm guessing if we load up the same game, you will have your hands full blowing anything away in the speed and quality dept. 😕
 
fkloster, fkloster, fkloster. There is nothing wrong with my score, it is right along where it should be, in accordance to all the reviews of the 9700. There is a difference between tweaking your settings, and messing with stuff that doesn't give you any real world gain, just a few extra 3DMark points. Don't try and argue with me over the subject of 3DMark, you won't win.
 
Oh yea, forgot to mention, I've been arguing against 3DMark since the 3DMark 2K days, I had a Voodoo5 back then, what was notorious for its poor 3DMark score, and mine was abnormally high, like in the low GTS ranges. So I don't think my problem with 3DMark is my score.
 
If there is nothing wrong with your score and it is right in line with the other 9700 scores what is your bitch Deeko?

1) 3dmark is a tool: True or False?

2) You live in a free world: True or False?

Anyone @ anytime can tweak settings while running any app and gain or lose different results....why is 3dmark any different?

If you answered 'True' to both questions, then let it rest...
 
My bitch is that I am a frequent poster in the video forum, where until I started thrashing people in there for it, there would be 10 posts a day of "my 3DMark is too low" or "how can I get an extra 50 3DMark points?" You ask them for game benchmarks, if they'd actually run them, generally they would be right where they should be. Yet they still whine and whine because they want more points, which won't necassarily give them real world gains. You get sick of it after awhile. 3DMark is NOT always an accurate tool, two systems with identical game performance can have different 3DMark scores, and people don't understand that, and they should.
 
It is such an ancient benchmark tool . . . 2 years old.

Frankly, I am sick of all the old "scenes" and am really looking forward to DX9 and 3DMark2003 so we can start comparing scores all over again.

😛
 
Originally posted by: Deeko
My bitch is that I am a frequent poster in the video forum, where until I started thrashing people in there for it, there would be 10 posts a day of "my 3DMark is too low" or "how can I get an extra 50 3DMark points?" You ask them for game benchmarks, if they'd actually run them, generally they would be right where they should be. Yet they still whine and whine because they want more points, which won't necassarily give them real world gains. You get sick of it after awhile. 3DMark is NOT always an accurate tool, two systems with identical game performance can have different 3DMark scores, and people don't understand that, and they should.
Click to expand...


Ahh, I see. I get sick of that too. U and I are on the same page I think...
 
My bitch is that I am a frequent poster in the video forum, where until I started thrashing people in there for it, there would be 10 posts a day of "my 3DMark is too low" or "how can I get an extra 50 3DMark points?" You ask them for game benchmarks, if they'd actually run them, generally they would be right where they should be. Yet they still whine and whine because they want more points, which won't necassarily give them real world gains. You get sick of it after awhile. 3DMark is NOT always an accurate tool, two systems with identical game performance can have different 3DMark scores, and people don't understand that, and they should.
Click to expand...
I still don't see the problem.
Your actually complaining that somebody uses 3dmark? Ha Ha

I suppose you'll say that UT2k3 is a good benchmark. A good benchmark for what?
 
Originally posted by: Dug
I still don't see the problem.

I suppose you'll say that UT2k3 is a good benchmark. A good benchmark for what?
Click to expand...

A good benchmark for how fast UT2k3 runs on your computer. 🙂

No, seriously, what serious reviewers do, run 3DMark2001, UT2k3, Quake, etc., that gives you a feel for how fast the system is.
But that is complicated, and so people just run 3DMark2001. It's not the worst thing to compare score, no, but it's obnoxious to argue every single point.
For instance, I find it also funny when some reviewers (not on anandtech, of course), claim you should get MoBo A because you get 3 fps more than any other board.
SO???? I run with vsync enabled, so it doesn't affect me (I do run at 85 Hz though). It's the same story, really.

One thing I do like about 3DMark 2001 - I overclocked my 8500LE to retail levels, and the increase was marginal (like 700 points). So I decided to leave it at default values for most of the time. 3DMark 2001 is good for such comparisons.
 
Yes...Deeko and the 3dmark police have done an excellent job of cleaning up the video forum. I too have an insanely high 3dMark score (almost 12000 on a gf4 ti4200), but I don't take much stock in it. If you want to max your score out, you have to do all kinds of goofy stuff to your settings that improve performance but degrade IQ. Also, people OC their cards to a point that may not be stable just to get a high 3dmark score, regardless of artifacts or tearing. Of course they don't mention that when they post their scores :|

As for the UT2k3 benchmark, it is useful b/c it shows REAL WORLD results of gaming performance using the latest game engine available (and no, I don't consider D3a to be the latest b/c its still not optimized). I think the point mentioned above and definitely covered in the past is that 3dmark does NOT measure real world performance in proportion to the scores it produces. For instance, someone may score 10000 in 3dmark and another person scores 9000, @10% difference in 3dmark. But in UT2k3, the difference may only be 60fps vs 63fps, a difference of around 5%.

Chiz
 
Why do people live and die by 3DMark to bench their systems?
Click to expand...

This is a hardware forum and people here like to try to make their systems run as well as possible, 3DMark is a good indicator.

You should know this of all people - but hey... ya know...
 
Yes, vigilante justice, the best way to keep a forum friendly and informative
rolleye.gif


If people are concerned about their 3DMark score, let them post. You don't have to read the posts. There might even be some people offering helpful advice floating around the forums somewhere (god forbid...)

I know Deeko, why don't you start flaming people who post Troubleshooting questions in HT, or the endless string of people in Networking who cant share files but havn't quite gotten round to reading the FAQs.

As I see it, there are no excuses for flaming people, especially when they're posting reasonable questions. And questions about 3DMark scores are reasonable, whatever you say.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top