• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

3dmark not good?

Performance in 3dmark does not always translate to game performance, unless every game made uses the same engine as 3dmark does. I used to do the 3dmark thing years ago, but then when I started testing games after tweaking or upgrading, I cared less about 3dmark and started caring more about my gaming perfomance.
 
because games don't run exactly like 3dmark...it's a good benchmarking tool, but who cares if you get a bigillion 3dmarks and crap fps in an actual game?
 
A "bagillion" fps would be a world record, and you could probably get money for it.
 
viperdv2k4 said:
A lot of times i hear that your 3dmark scores dont matter, only you actualy gaming does. But, i think it does because isnt it pretty much recorded gaming? If your score goes up, the better your fps in gaming. Its better than just getting a benchmark for 1 game because it does many "games".. So, why do people say that 3dmark scores dont matter?
Click to expand...

It doesn't use any game engine that any current games are using. It is not a game. So how can it be recorded gaming?

3DMark performance increases and decreases do not correlate to game performance increases or decreases.

Every game performs and acts different on hardware; 3DMark cannot tell you how each game is going to perform.
 
There is no need to have such a hostile attitude towards 3DMark, although it's not a game it ranks different vid cards with a reasonable correlation to most games. 3DMark cannot predict CPU-limited gaming performance, but it gives reproducible results and is widely used.

And yes, I've understood that there may be unresolved issues between hardocp and futuremark.
 
Higher 3dmarks dont means higher fps in your favorite game, period.

Higher 3dmarks just means higher fps in 3dmark.

~Adam
 
Because a synthtic benchmarks does not represent the game you like to play. It's great for people who want to have the best 3DMarks score but it doesn't hold any water when I look at benchmark scores on games that I play. That's my two copper.
 
To me 3Dmark is like a screen saver.

Although screen savers are a little more useful :p

I agree with the [H] for dropping 3dmark. I prefer actual game benchmarks.
 
jon67 said:
There is no need to have such a hostile attitude towards 3DMark, although it's not a game it ranks different vid cards with a reasonable correlation to most games. 3DMark cannot predict CPU-limited gaming performance, but it gives reproducible results and is widely used.

And yes, I've understood that there may be unresolved issues between hardocp and futuremark.
Click to expand...
The only thing its useful for is comparing same brand video cards performance.
And, frankly, we dont need that. thats what th enumbers and letters are for. the higher the number, the faster the card (of the same brand).
It is NOT a useful tool to guage how well a card will perform in one game engine compared to another game engine, nor is it useful to guage one brands performance agaisnt another in any application other than, well, 3dmark.
Yet, it comes off as a performance on comparison benchmark. Well, the only thing it compares performance on is itself. Rather useless, and a bit misleading, I think.
 
3dmark is a good representation that your system is running how it should under a gaming enviroment. You can look at default scores for a new card you put in and such and determine if you did it right, no driver conflicts, ect...

I think it's good in that sense.

Oh and to see how much your system is bottlenecking if any.
 
Not really, the better bet would to benchmark the games you actually play, dont ya think?
Using 3dmark as a performance tool is like measuring your penis by how high a wavelength of light it refracts. Sure, it may refract super high frequency light, but at the end of the day, that doesnt means its any better at what it's intended for than someone elses penis.
 
I'm not talking about game performance, I am talking about comparing your lets say 9800 pro to lets say...a tomshardware 3Dmark benchmark. You could be like "Gee, they got the same scores as me, my system must be running optimal." or "Wait, something is wrong here, I am getting like 500 less points then what my videocard should be." Or use ORB to compair with ppl of similar system setups.

Get what I am trying to say?
 
Yes, i do, but I dont see that as very useful. To top it off, futremark certainly doesnt market themselves that way, either.
 
Siciliano:
Yup you can compare preformance in 3Dmark...but you would be better off with comparing with the game-benches...

Again, 3Dmark is not why you got your card is it?

Terra - I don't use 3Dmark after they badmouthed [H]ardOCP...
 
Brent_Justice said:
It doesn't use any game engine that any current games are using. It is not a game. So how can it be recorded gaming?
Click to expand...

This is the kind of nonsense that always annoys me.
HL2 doesn't use the same engine as Doom3 either, but those are considered valid benchmarks while 3DMark is not?
Also, I'd like to point out that it's not 'recorded' gaming... Nearly everything is realtime, only the camera path is non-interactive (unless you have the developer version, where you can move around freely). This is far more realistic than eg the Doom3 timedemo option, where everything IS recorded (AI, physics etc), and only the graphics are being benchmarked.

Every game performs and acts different on hardware; 3DMark cannot tell you how each game is going to perform.
Click to expand...

Neither can one game about the other.
At least 3DMark tries to give a few different, common (they research the most likely scenarios for upcoming games) game tests, to give a decent overall indication of performance, unlike a game, which only says something about itself, and nothing about slightly different games.
 
Brent_Justice said:
you could do the same thing with a game
Click to expand...

With at least 3 games, since 3DMark has 3 different game tests (and then a bunch of feature tests, to see what the strong and weak points of your hardware are... How many games can do that?).
 
3DMark scores give a representative ranking of different brands vid cards performance in D3D games in cases where the gameplay is not CPU-limited. So it's more useful than a screensaver, it's easily installed and used, and it's free.
 
jon67 said:
3DMark scores give a representative ranking of different brands vid cards performance in D3D games in cases where the gameplay is not CPU-limited. So it's more useful than a screensaver, it's easily installed and used, and it's free.
Click to expand...

I have seen 3D sreensavers that I found more pretty, easily installed and free...

Terra - And the "free part" isn't quite true..you have to pay to get FULL functionality...
 
Terra said:
I have seen 3D sreensavers that I found more pretty
Click to expand...

I doubt that, since 3DMark is at the cutting edge of what is possible with today's 3d accelerators. I have yet to see a screensaver that does the same.
This smells of petty bias.
 
Actually, 3dmark05 does little to try to be pretty, at all.

Saying '05 is pretty is like calling a can of compressed air a piece of art.

~Adam
 
PRIME1 said:
It's pretty useless.
Click to expand...


Wow. Thanks for going around all the valid arguments presented in the last page. You, my friend, are ignorant.

3d mark is just as good as any other benchmark, especially since it's benchmarks show different possibilies. If you want to know if your computer will do better with doom 3, use a doom 3 benchmark. But if you want a *general* answer, i believe 3dmark to be more rounded.

And if you say that 3dmark is pointless, that it doesnt represent gaming with the same benchsmark demos being run, look at any other game's benchmark. They benchmarks only represent performance during a specific scene of a specific game.
 
SystemERRor said:
Wow. Thanks for going around all the valid arguments presented in the last page. You, my friend, are ignorant.

3d mark is just as good as any other benchmark, especially since it's benchmarks show different possibilies. If you want to know if your computer will do better with doom 3, use a doom 3 benchmark. But if you want a *general* answer, i believe 3dmark to be more rounded.

And if you say that 3dmark is pointless, that it doesnt represent gaming with the same benchsmark demos being run, look at any other game's benchmark. They benchmarks only represent performance during a specific scene of a specific game.
Click to expand...

And still one game-benchmark gives a better result regarding real-world performance in games...

Terra...
 
Scali said:
I doubt that, since 3DMark is at the cutting edge of what is possible with today's 3d accelerators. I have yet to see a screensaver that does the same.
This smells of petty bias.
Click to expand...

Yeah I love the SM3.0 support in 3Dmark 05 :rolleyes:

Terra...
 
I just figured you use 3DMark as a tweaker. You run it, get the score, then go to bios and adjust the timings or tweak something in the general driver options, then run it again to see if the score got any higher without crashing. Don't you guys use it like that?

If 3DMark doesn't cut it. Maybe you guys can suggest a better general benchmarking utility?

Terra said:
I have seen 3D sreensavers that I found more pretty, easily installed and free...
Click to expand...

Can you suggest some? Maybe PM me some links? I've already got a few nice ones but it doesn't hurt to get more :p Marine Aquarium and DesktopDreamscapes are my favorite :)
 
SystemERRor said:
Wow. Thanks for going around all the valid arguments presented in the last page. You, my friend, are ignorant.

3d mark is just as good as any other benchmark, especially since it's benchmarks show different possibilies. If you want to know if your computer will do better with doom 3, use a doom 3 benchmark. But if you want a *general* answer, i believe 3dmark to be more rounded.

And if you say that 3dmark is pointless, that it doesnt represent gaming with the same benchsmark demos being run, look at any other game's benchmark. They benchmarks only represent performance during a specific scene of a specific game.
Click to expand...

Ignorant? Maybe it is you that label applies to. First to understand my previous post please go look up sarcasm in the dictionary. Second, this very site no longer uses 3Dmark because it is not comparable to the results achieved from using actual games. You don't even have to pay to use these games for benchmarks because most (if not all) have demos you can download and run to benchmark.

I only care how a card performs in ACTUAL games, not screen savers. 3Dmark only covers DirectX performance and it does not use things like 3DC, SM3.0, HDR, etc.
Not to mention some DirectX games seem to perform better on NVIDIA cards or on ATI cards so a lot of performance is based on how the game is written and each game is different. So since the 3Dmark "engine" is not used in any games it's not going to tell you sh** about game performance.

Hey if it makes you feel better to know that your card got 345,789 points on 3dskidmark05
you must be easy to please and maybe you don't even need to play games.
 
PRIME1 said:
3Dmark only covers DirectX performance and it does not use things like 3DC, SM3.0, HDR, etc.
Click to expand...

Uhhh... 3DMark uses SM3.0 if you want, you can select the shader version manually.
It also has HDR (just look at the bloom effect).
It doesn't use 3Dc because it uses an antialiasing algorithm for its normalmaps that is not compatible with 3Dc. It's not too far-fetched to think that games will use similar algorithms in the future, and therefore will not use 3Dc either.
And it does use DST/PCF when available.
So your argument is pretty flawed.

Not to mention some DirectX games seem to perform better on NVIDIA cards or on ATI cards so a lot of performance is based on how the game is written and each game is different.
Click to expand...

Some do, some don't. What's your point?
Also note that some games have shader replacement, which makes the comparison to 3DMark's scores flawed.

So since the 3Dmark "engine" is not used in any games it's not going to tell you sh** about game performance.
Click to expand...

That is utter nonsense. An engine is just an engine. As long as you have similar engines, that perform similar shading on similar workloads, you will get valuable information on performance.
 
Scali said:
This is the kind of nonsense that always annoys me.
HL2 doesn't use the same engine as Doom3 either, but those are considered valid benchmarks while 3DMark is not?
Also, I'd like to point out that it's not 'recorded' gaming... Nearly everything is realtime, only the camera path is non-interactive (unless you have the developer version, where you can move around freely). This is far more realistic than eg the Doom3 timedemo option, where everything IS recorded (AI, physics etc), and only the graphics are being benchmarked.

Neither can one game about the other.
At least 3DMark tries to give a few different, common (they research the most likely scenarios for upcoming games) game tests, to give a decent overall indication of performance, unlike a game, which only says something about itself, and nothing about slightly different games.
Click to expand...

Excellent post! Check out this (very) recent EB article for some interesting 3dmark05 info - http://www.elitebastards.com/page.php?pageid=10599&head=1&comments=1
 
PRIME1 said:
If you want to know how well you card performs using games, run games on it.

If you want to know how will your card peforms using 3dMark run 3dMark.

I like to play games, so that is the only measure I will accept. I don't "play" 3Dmark so it's results tell me nothing.

http://www.hardforum.com/showthread.php?t=816850&page=1&pp=20
Click to expand...

Every game is different. They use different engines. It has become clear to most people reading this thread that you are the kind of person who refuses to look at other points of view even in the face of overwhelming logic and your own fatally flawed arguments.

Some people want to see how much the changes they make to their system affect performance. They want to see if the new clock speed of their GPU core will cause artifacts. Now they can benchmark in a game, but it means nothing if a game gets patched or if someone doesnt own the same game. Furthermore, some people want more data then just a single number for FPS results.
 
trinibwoy said:
Excellent post! Check out this (very) recent EB article for some interesting 3dmark05 info - http://www.elitebastards.com/page.php?pageid=10599&head=1&comments=1
Click to expand...

This article is really nothing new. Basically it is saying that 3dmark has a feature only nvidia can use, one that helps those cards, inferring that the company must be slanted towards nvidia. However, consider most of the new games released (Far Cry, Half Life 2) Each of these games is connected with either nvidia or ati. Some features that the games include are only usable by cards of one company. But hardocp focuses of 3dmark more than anything else.

Kyle just seems to have some kind of hatrid towards 3dmark, and his articles express that. Did his computer get a crappy score? We'll never know. But unlike the vendetta against a certain gaming console, this one i just dont get.
 
SystemERRor said:
Every game is different. They use different engines. It has become clear to most people reading this thread that you are the kind of person who refuses to look at other points of view even in the face of overwhelming logic and your own fatally flawed arguments.
Click to expand...

First go look up logic in the dictionary. OK, good. Now if you want to know how well your card will play games then LOGICALLY you will use games to test your card. And (as you pointed out) every game IS different that's why no synthetic benchmark will be very accurate in judging true game performance (especially with many games using OpenGL instead of DirectX)

However if you want to use it as a stress test for an overclocked or modified card.... I agree it should work just fine for that purpose. However changes in your 3dmark score may not actually mean changes in real world game performance.
 
PRIME1 said:
If you want to know how well your card performs using games, run games on it.

If you want to know how will your card peforms using 3dMark run 3dMark.

I like to play games, so that is the only measure I will accept. I don't "play" 3Dmark so it's results tell me nothing.

http://www.hardforum.com/showthread.php?t=816850&page=1&pp=20
Click to expand...

I don't play games, so those benchmarks mean nothing to me!?
What crap is that? I use 3DMark to figure out which card gives me the most bang for the buck as a developer. It does that better than any game can.
Who cares if I don't actually develop with 3DMark itself? It tells me what I need to know.
 
You must log in or register to reply here.
Back
Top