Question about quoted 3Dmark scores.

[Mister Gin]

Is there a particular resolution you are supposed to run when posting your numbers or do people post the highest number they get regardless of resolution? I just had to ask since I just read part of the "Post your 3d mark 2001 here" thread and was curious if everybody compares apples to apples or what.

I wanted to see what my newly built computer would do so I loaded and ran the latest 3Dmark program. I changed no settings (default I guess) before running and got a score of 9989. Is that pretty good? The card is a Sapphire 8500 64MB running the latest catalyst 2.5 drivers. The picture looked pretty darn good to me.

Also, does the monitor play any role in the score? The reason I ask is because I used an old Dell 15" monitor when building and setting up the system. This system will be getting my current 19" monitor later in the week.

Ed

[Tiretool]

Honest people run 3D Mark at it's default settings. These are as follows:

1024 X 768 Resolution
32 bit Color
Compressed Textures
D3D Pure Hardware T&L
No Antialiasing

To get a true "fair" idea of how good your system's 3D performance, just run it at these default settings and compare them using the online result browser.

Alot of people will cheat by setting their resolution down to 640 X 480 and run the benchmark in 16 bit color. This boosts their score higher, but it doesn't accurately compare them to other people's systems who ran the benchmark at default settings. They're really just lying to themselves and others to make themselves look good.

To me, benchmarks are simply a tool to measure how good your system does 3D chores. Simple as that. I post my scores every so often if someone asks, but I'm not in competition with anyone. I do it for my own gratification. No matter how well your system is put together and kept up, someone is always going to get a higher score. You'd have to be up-to-the-minute with technology and always have the very latest and greatest hardware to be on top all the time. As long as I get acceptable framerates with good quality eyecandy, I'm happy. You shouldn't get caught up in the "my system's faster than your system" hype. It just burns a hole in your wallet. I found this out first hand a couple of years ago.

The score you reported looks like a very nice score. With a little tweaking you might be able to break the 10,000 mark if that's your goal. I got just over 10,000 with my setup, but I see people reporting 12,000 13,000, even 15,000 now with Radeon 9700s, the latest Athlon XP processors and Intel P4's, and a ton of DDR memory.

It all just boils down to, are you happy with the way your games play? Do you see any picture quality issues? Are your framerates acceptable? With that type of score, you're probably set for nearly any game you can throw at it.

Post your new score after installing that new monitor. I'd guess that you might see a slight difference.

[RenegadeKing]

I run at the default settings, except I change the frame buffer from double to triple. It gives a 20 pt increase or so.

Looks like a good score to me.

[Alienware_Dude]

How many of you run with Mip-mapping at 'Best Performance'?

[Mister Gin]

You shouldn't get caught up in the "my system's faster than your system" hype.

I don't plan on it.

I was just curious as to what my system would achieve. The number did drop to around 7700 when I changed the settings to 1280x1024 & 32bit. I'm sure I'll be more than happy with this card for quite some time. I'll mainly be doing video editing and watching a few DVD's but I could be persuaded into buying some games now that I know I can play them .

I'll post the score after I put on the other monitor.

Ed

[Tiretool]

7700 is still quite respectable! When I was running a GeForce 2 GTS I was hovering around 2900-3200 depending on the drivers I was using at the time. Even at this low score, if I didn't go overboard on the game detail settings, I got very playable framerates.

Let us know what game titles you get, and don't hesitate to ask about game titles before you buy them. Several of us have played almost everything currently on the store shelves so someone will almost certainly have an opinion on the worthiness of a game. Might save you some hard earned money!

[Blakhart]

Set the performance slider to aggressive, and the d3d mipmapping level to best performance. That may boost yer score, and run the bench at defaults, other than the tweaks.
For kicks, bench it with title screens disabled between tests. This may show you if your driver is cheating.

[Mister Gin]

Let us know what game titles you get, and don't hesitate to ask about game titles before you buy them.

Set the performance slider to aggressive, and the d3d mipmapping level to best performance. That may boost yer score, and run the bench at defaults, other than the tweaks. For kicks, bench it with title screens disabled between tests. This may show you if your driver is cheating.

Will do for both.

One more question.

Does the refresh rate affect the score, or does it matter what it's set at? I forgot to up this so the two tests I ran were done at 60, the lowest I believe.

Ed

[Fastfly]

With a Ti4200 I get the same scores or lower if I set the colouring or the resolution down.

[Blakhart]

In the 3dmark bench, it sets vsynch to off automaticly.

[RenegadeKing]

A different monitor won't affect your score. You could run it with no monitor and get the same score.