Is this 3dmark06 looking good?

[roadrunner197069]

@ dagger : Do us all a favor and leave, and never come back: until you know something.

 

[invisik]

And how did you scientifically calculate the 6-9fps difference? Of is it just a "feeling." 😛

People who get low benchmarks always say it doesn't mean much. If you can't reach the grapes, they must be sour!

The point is, if you don't admit it's broken, it'll never get fixed. Foolish pride won't do you any good. Find out what's wrong, get it fixed, and then boast about it when it's running like clockwork.

Anyway, look at the first 2 categories (sm2.0 and sm3.0 scores) instead of the cpu score or overall score for a more accurate bench of the graphics card. And keep in mind cpu performance do matter in real life (duh).

No some of my games have built in test and i noticed an increase in fps unless those fps r false then i agree with u.
=]

 

[TheGreatGrapeApe]

Chillax boyz and move on.

 

[omahagtp]

Win Vista SP1
Gefore 9800gtx stock at the moment
2gb DDR 600
AMD Opteron 170 dual core at 3ghz

Got 10243 for 3dmark06 free edition. Seems a bit low but thinking its cause I aint using a quad core cpu yet. This good?

Not bad, as others have mentioned you definitely should look into a CPU upgrade.

For reference, my e8400 dual core at 3.6 GHz and 8800gts 512 (oc'ed at 770/1050) gets 15,800 in XP and 14,500 in Vista with the 3dmark06 trial software.

 

[Crazy-PC]

The improved 45nm process will not create a significance difference among processors with the same architecture. If clocked at the same speed then the performance will be very very similar. The operating system is the killer in this scenario.

Best,

3Ball

Do you mean Dagger using Vista while mousemonkey use XP? If it is the case, this is still normal understanding and no surprise.

 

[dagger]

Do you mean Dagger using Vista while mousemonkey use XP? If it is the case, this is still normal understanding and no surprise.

Yep, it makes a big impact on cpu score. That's why for graphics performance, you can look at scores of the first 2 categories (for results of the two shader models), instead of cpu score or even overall score. Although obviously cpu performance matters.

3dmark is the most authoritative among benchmark out there. If you don't think it's belivable, there's no other benchmark to believe in. People who say it has nothing to do with system performance are the same ones who got very low scores and claim it's the benchmark that's broken. It's the foot-in-mouth approach.

 

[LukeBird]

Not bad, as others have mentioned you definitely should look into a CPU upgrade.

For reference, my e8400 dual core at 3.6 GHz and 8800gts 512 (oc'ed at 770/1050) gets 15,800 in XP and 14,500 in Vista with the 3dmark06 trial software.

He doesn't need a new CPU unless he wants to shout how many 3DMarks he can get!
My system scores just over 13k, but with a decent quad, it should easily hit 16-18k I'd guess.
Do I care, absolutely not, because in game it's fine. 3DMark is synthetic and it is skewed massively to giving higher scores with a quad over a dual!
In the real world, there is barely going to be a difference between a dual and a quad especially in (most!) games. OP don't worry about it, it's a fine score unless you want more 3DMarks!

 

[basketcase]

3dmark is the most authoritative among benchmark out there. If you don't think it's belivable, there's no other benchmark to believe in. People who say it has nothing to do with system performance are the same ones who got very low scores and claim it's the benchmark that's broken. It's the foot-in-mouth approach.

The point is, what matters is the real world performance. How games actually perform is more important than what 3DMark shows you for a score. For instance, the OP is getting a little over 10k. If his gaming performance is good, it doesn't matter if he ups the score to 15k. In most games (except Crysis), unless you are running an extremely high resolution like 1900x1200, anything over 10k in 3DMark06 will most likely mean your games will run perfectly fine. That isn't to say that there is no correlation between your 3Dmarks and how your games perform. But, there is a point where it doesn't make a noticeable difference. An example, you get 150 fps in CS:S. Is it worth it to upgrade your CPU and Mobo (among other parts) to get your FPS upto 200? No, that would be a waste of money. It is all in the eye of the beholder.

So, the question is: If you are happy with gaming performance, is it worth it to get a new CPU (and most likely a motherboard) to increase a synthetic benchmarks score.

Now, I am not sure if the OP's games are running good or not, I am just generalizing.

 

[omahagtp]

He doesn't need a new CPU unless he wants to shout how many 3DMarks he can get!
My system scores just over 13k, but with a decent quad, it should easily hit 16-18k I'd guess.
Do I care, absolutely not, because in game it's fine. 3DMark is synthetic and it is skewed massively to giving higher scores with a quad over a dual!
In the real world, there is barely going to be a difference between a dual and a quad especially in (most!) games. OP don't worry about it, it's a fine score unless you want more 3DMarks!

Well, he did wonder why it was low, and that is the definitive suggestion as how to raise it.

Don't quote me about quad vs dual cores. I have a dual, and do not, and did not, suggest a quad.

Unless you think I'm bragging about 3dmark scores. Which I am not. Only a reference point. You don't see me with a list of my junk in my sig.

 

[dagger]

The point is, what matters is the real world performance. How games actually perform is more important than what 3DMark shows you for a score. For instance, the OP is getting a little over 10k. If his gaming performance is good, it doesn't matter if he ups the score to 15k. In most games (except Crysis), unless you are running an extremely high resolution like 1900x1200, anything over 10k in 3DMark06 will most likely mean your games will run perfectly fine. That isn't to say that there is no correlation between your 3Dmarks and how your games perform. But, there is a point where it doesn't make a noticeable difference. An example, you get 150 fps in CS:S. Is it worth it to upgrade your CPU and Mobo (among other parts) to get your FPS upto 200? No, that would be a waste of money. It is all in the eye of the beholder.

So, the question is: If you are happy with gaming performance, is it worth it to get a new CPU (and most likely a motherboard) to increase a synthetic benchmarks score.

Now, I am not sure if the OP's games are running good or not, I am just generalizing.

Then why did OP spent extra on 9800 when 9600 run games just fine? It's all for a little futureproofing. Those extra spent should come out to mean something.

And as I pointed out, 3dmark has cpu and graphics components, if you think the cpu part is skewed, which is debateable, then just look at the graphics scores. To say the whole benchmark is worthless just because you got a low score is childish.

 

[basketcase]

Then why did OP spent extra on 9800 when 9600 run games just fine? It's all for a little futureproofing. Those extra spent should come out to mean something.

And as I pointed out, 3dmark has cpu and graphics components, if you think the cpu part is skewed, which is debateable, then just look at the graphics scores. To say the whole benchmark is worthless just because you got a low score is childish.

Did I say it was worthless?

And there is a bit of a skew related to quad cores. It gives a significant gain in Marks for having a quad core. Which wouldn't be an issue if games were doing the same. But, most games currently available do not really get that much in gains from a quad core CPU, when compared to a similarily clocked dual core. In the future, yes, more and more games will utilize more cores better. But, until then, it is skewed. And that being said, it doesn't make 3Dmark worthless. It just makes understanding what real world performance a little bit harder to judge. Not impossible, but you need to read into things a little more. For instance, you could have a e6600 cpu at 3.0 ghz and a q6600 at 3.0 ghz. The quad would obviously score higher in 3DMark06. But, in real world scenarios, they would (in most games currently available) get ALMOST the same performance.

 

[dagger]

Did I say it was worthless?

And there is a bit of a skew related to quad cores. It gives a significant gain in Marks for having a quad core. Which wouldn't be an issue if games were doing the same. But, most games currently available do not really get that much in gains from a quad core CPU, when compared to a similarily clocked dual core. In the future, yes, more and more games will utilize more cores better. But, until then, it is skewed.

If you think that way, faster dual cores are also skewed, since few games can saturate cpu usage, and gpu is usually the bottleneck. But let's not debate this, just look at the graphics component if you think it's more important.

 

[basketcase]

If you think that way, faster dual cores are also skewed, since few games can saturate cpu usage, and gpu is usually the bottleneck. But let's not debate this, just look at the graphics component if you think it's more important.

There is a difference. Most game currently available are not written to fully take advantage of 4 cores. But, 3DMark06 is. So, that is where the skew comes into play. CPU saturation and GPU bottleneck is a different factor that is not really skewed in 3DMark, as if it is doing that in 3Dmark, it is likely doing the same thing for your games (and vice-versa).

 

[dagger]

There is a difference. Most game currently available are not written to fully take advantage of 4 cores. But, 3DMark06 is. So, that is where the skew comes into play. CPU saturation and GPU bottleneck is a different factor that is not really skewed in 3DMark, as if it is doing that in 3Dmark, it is likely doing the same thing for your games (and vice-versa).

Compare e8400 with 8600gt and e4400 with 8600gt on a typical graphically intensive game, you'll get similar if not identical fps, which is because graphics is the bottleneck, not cpu. It'll show up on benchmarks though.

Basically, when some other cpu beats yours in benchmarks, it's not fair. But when you beat others in the same benchmarks, it's fair?

Life isn't fair. Sometimes you get the longer end of the stick, sometimes you don't. Let it go already.

 

[basketcase]

Ok, I see you are not reading what anyone is saying, so I will stop.

 

[Mousemonkey]

Do you mean Dagger using Vista while mousemonkey use XP? If it is the case, this is still normal understanding and no surprise.

Oh, there's more to it than that. 😗

For reference, my e8400 dual core at 3.6 GHz and 8800gts 512 (oc'ed at 770/1050) gets 15,800 in XP and 14,500 in Vista with the 3dmark06 trial software.

Same CPU and clock speed same OS GTS rather than GT and still lower, have you sussed out the 'booster' yet?

 

[aznguy0028]

@addiarmadar

if you're getting the q6600 soon, and have a bit more money left over, why not go for the q6700? recent intel price slash = heaven x) its 299 on newegg (doesn't include tax + shipping yet) and might be cheaper elsewhere, such as directron 289 no tax, free shipping im thinking bout replacing my q9450 with the q6700, 400fsb x 10= 4ghz sounds really nice!