• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Conroe above 3.6Ghz is USELESS for games on current tech

Richdog said:
How can any CPU speed be "too high"? :confused:
Very easily. Too high and it can start to throttle back from the heat. That's the obvious answer. But also and probably more importantly, you have to question how they went about the overclock in the first place. If it involved increasing the multiplier but lowering the FSB then, even with the high clock speed, the overall system performance will be worse because the lower FSB cannot feed the CPU with enough data to keep it busy at all times...
 
NathanE said:
Very easily. Too high and it can start to throttle back from the heat. But also, you have to question how they went about the overclock in the first place. If it involved increasing the multiplier but lowering the FSB then, even with the high clock speed, the overall system performance will be worse because the lower FSB cannot feed the CPU with enough data to keep it busy at all times...

Oh come on nathan I was obviously talking about a stable CPU speed... not "what technical difficulties can someone experience if they clock their CPU too high." My point was Asgards comment about the clock being too high:

The Asgard said:
I would totally agree and I think 3.6 is too high. I would imagine anything over 3G's is pointless.

He made no reference to technical issues, he simply said "anything over 3G is too much". Read the context man. :p:)
 
Last edited:
I did read the context :/ What is to say that those technical difficulties aren't the cause for the benchmarks becoming worse?

At least I'm trying to provide a possible explanation instead of blindly attacking 3DMark's credibility... ;)
 
NathanE said:
I did read the context :/ What is to say that those technical difficulties aren't the cause for the benchmarks becoming worse?

At least I'm trying to provide a possible explanation instead of blindly attacking 3DMark's credibility... ;)

Blindly attacking? Sorry if you got that impression, but anyone with some knowledge of performance or benching knows 3DMARK 06 isn't a good measure of CPU performance... it's HEAVILY GPU dependant... pretty standard stuff really.

if you want to make a damning thread about CPU performance in games then you bench some... yup... wait for it... games. :p

As for the inconsistency between the score... the test platform wasn't consistent so there's your answer.
 
I know 3DMark is pretty crap for this type of thing. But the point is, if you change one variable - increase CPU clock speed, then the 3DMark score shouldn't decrease. The problem with this test is that multiple variables have been changed and so no conclusion can be made, definately not the one drawn up by the OP anyway.
 
NathanE said:
I know 3DMark is pretty crap for this type of thing. But the point is, if you change one variable - increase CPU clock speed, then the 3DMark score shouldn't decrease. The problem with this test is that multiple variables have been changed and so no conclusion can be made, definately not the one drawn up by the OP anyway.

Agree 100% mate regarding his conclusion... but his overall score did go up.

I think 3DMARL needs to be looped 3-5 times to get the average score or single runs can sometimes display inconsistencies. :)
 
Richdog said:
How can any CPU speed be "too high"? :confused:

LOL. You dont actual think that while your playing Doom 3 or the like that your CPU is bening maxed out. Most of the time I think you will find its waiting for one thing or another.

There's an interesting article in one of the mags that show what effect all the different CPU's have on games performance. This is from a Celly upwards. Conclusion , none or very little.

Most games are either refresh rate or GPU bound.
 
The Asgard said:
LOL. You dont actual think that while your playing Doom 3 or the like that your CPU is bening maxed out. Most of the time I think you will find its waiting for one thing or another.

Now we're talking about the theory that in an ideal world games make use of 100% of the CPU. Sorry but that isn't reality so we'll have to make do with increasing mhz to get more performance now i'm afraid.

There's an interesting article in one of the mags that show what effect all the different CPU's have on games performance. This is from a Celly upwards. Conclusion , none or very little.

Most games are either refresh rate or GPU bound.

What games were these, just out of interest? I presume that one article tested Oblivion, Company of Heroes and other games that are reactive to increased CPU speed, right?

Tip: never take one articles word on things, because there are many games I have experienced that are CPU sensitive, just because they didn't test games that were doesn't mean there are no games that respond to more mhz. Any gamer knows for a FACT that some games do get increased FPS wwith a beefier CPU.

Try playing Company of Heores, Oblivion or Dawn of War on a celeron... then compare it to a Core2. Difference is huge. Hell, the difference between my Opty 175 @ 2.7ghz and my Core2 @ 3.5ghz was very noticeable in COH, the minimum framerate shot up.
 
Last edited:
NathanE said:
Very easily. Too high and it can start to throttle back from the heat. That's the obvious answer.

That's the stupid answer :rolleyes: Obviously he was referring to a normal scenario. @ Asgard - it's also obvious that given a normal scenario and ceteris paribus, a faster CPU cannot hurt framerate. Asgard seems to be arguing that above a certain point extra CPU will make no difference. Well it will always make SOME difference, so this is rubbish also.

Come on people use some common sense.
 
Richdog said:
Tip: never take one articles word on things, because there are many games I have experienced that are CPU sensitive, just because they didn't test games that were doesn't mean there are no games that respond to more mhz. Any gamer knows for a FACT that some games do get increased FPS wwith a beefier CPU.

I don't base my info on any articles I base it on my job backed up by 23 years of software coding much of which is multi threaded. What do you base it on ??

I said most!

Oblivion is more CPU dependant than most.
 
Last edited:
The Asgard said:
I don't base my info on any articles I base it on my job backed up by 23 years of software coding much of which is multi threaded. What do you base it on ??

I said most!

Oblivion is more CPU dependant than most.

hey, whip us some multithreaded benchmark programs that i can try out :D
 
Concorde Rules said:
FS9 and FSX, 3D Mark 2005.

Supreme commander will be CPU limited.

Any CPU intensive tasks...

Not had any experience with FS9 or FSX and the last time I looked 3D Mark 2005 was not a game :D

Like I have already said "most games". With regard to the future with games I don't know how to use a crystal ball like you obviously can ;)
 
The Asgard said:
I don't base my info on any articles I base it on my job backed up by 23 years of software coding much of which is multi threaded. What do you base it on ??

I base it on practical experience building lots of computers and playing lots of games since I was about 12. I would hzard a guess that all of your 23 years of software coding and playing with multi-threaded apps don't help that much with regards to the last couple of years of changes in CPU power and their effect on modern games, which is the specific subject at hand. :)

I said most!

Oblivion is more CPU dependant than most.

So are many RTS games, RPG's etc... the point i'm making is that games that react to CPU speed are not limited to a select few. It applies to a fair emount modern and popular games, even if it's a small percentage in the grand scheme of things. :)
 
Sorry, but my CPU in EVERY game I own is at 100% ALL THE TIME.

I game at 1600x1200, X1900XTX.

Therefore its using all the CPU.

Even if it isn't using all the CPU, at some points it WILL, and you have the power there.

The suggestion that "Conroe above 3.6Ghz is USELESS for games on current tech" is on the most part, ****.

If there was no benefit, we wouldn't overclock to the limits, just enough to get max performance, but we don't we get as much as possible.

BF2 is some cases its CPU limited, 64 players, big map, lots of vehicles, physics etc, then you are CPU limited...
 
Faster CPU is useful for general Windows usage. Imagine a chip at twice the speed of the X6800 overclocked to the max - you'd notice a difference in everything you did.
 
Concorde Rules said:
Sorry, but my CPU in EVERY game I own is at 100% ALL THE TIME.

I game at 1600x1200, X1900XTX.

Therefore its using all the CPU.

Even if it isn't using all the CPU, at some points it WILL, and you have the power there.

The suggestion that "Conroe above 3.6Ghz is USELESS for games on current tech" is on the most part, ****.

If there was no benefit, we wouldn't overclock to the limits, just enough to get max performance, but we don't we get as much as possible.

BF2 is some cases its CPU limited, 64 players, big map, lots of vehicles, physics etc, then you are CPU limited...

Check the FPS of BF2 at 1600x1200 at 3gigs then try it at 3.6 gigs i suspect it will be exactly the same. Try it ?
 
Back
Top Bottom