Printable View
http://service.futuremark.com/compare?2k1=6653022
1.91GHz and 182fsb
http://service.futuremark.com/compare?2k1=6652987
2.13GHz and 164fsb just 58 more marks
220MHz slower and 18MHz higher fsb. Thus each MHz fsb is worth at least 10MHz cpu speed, therefore at 182fsb and 2GHz, I eat xp2700s at 166fsb and 2.16GHz :)
60 points difference doesnt prove a thing...
You can at any given time get about a 2% difference from the first run to the second.. Without rebooting, without changing a thing...
So he is well within acceptable range :)
So yet it doesnt prove anything...
EDIT: HE also said it locked up in the car chase... So if your setup is locking up it is possible his cpu is spitting out to many corrupt bits... and/or overclocked too high...
he? This is my setup LOL! and the 58 difference just goes to show that fsb rules over MHz too. oh I gain 6fps in quake3 per 90MHz cpu, thats sad. at 2GHz and 182fsb im at 331 against lostlilchild's 336 when he ran his cpu at 2.16GHz on 166fsb. 16fsb lower and 166MHz gives the same performance, so I am around xp2700 speeds at 2GHz but on 182fsb :)
Why not use SuperPI and that will tell you exactly where you are around... You cannot compare System A to System B like that... There are just too many variables...
You need to isolate the benchmark.. So SuperPI is great or Sandra :)
If you dont want to run it.. Oh well..
Run 3dmark 2003 the results are very stable unlike 2001
But that is video card only.. .He was talking about his CPU performance... :)
Your XP is probably in the ball bark of a 2400 not a 2700 :)
Wasn't this proven before?
Quote:
Originally posted by Thunderbird1GHz
he? This is my setup LOL! and the 58 difference just goes to show that fsb rules over MHz too. oh I gain 6fps in quake3 per 90MHz cpu, thats sad. at 2GHz and 182fsb im at 331 against lostlilchild's 336 when he ran his cpu at 2.16GHz on 166fsb. 16fsb lower and 166MHz gives the same performance, so I am around xp2700 speeds at 2GHz but on 182fsb :)
Yes, Lostlilchild soundly trounced t-bird at similar system specs.Quote:
Originally posted by slavik
Wasn't this proven before?
Like Big C said, Super PI or PiFast is great for testing justyour system an leaving the vid card out of it.
If you wanna do some good comparisons between systems, calculate Pi.:)
BTW, my best Super Pi time is 36 seconds.:D
You've been giving advice all this time to people about what they should buy and you are just now discovering that fsb is more important than mhz of the processor????????????????Quote:
Originally posted by Thunderbird1GHz
he? This is my setup LOL! and the 58 difference just goes to show that fsb rules over MHz too. oh I gain 6fps in quake3 per 90MHz cpu, thats sad. at 2GHz and 182fsb im at 331 against lostlilchild's 336 when he ran his cpu at 2.16GHz on 166fsb. 16fsb lower and 166MHz gives the same performance, so I am around xp2700 speeds at 2GHz but on 182fsb :)
Would you go somewhere else where they want to hear you ramble?
kirkyg
FSB is important yes, but raw speed is also VERY important.
You're not gonna get high marks without a combo of both.
But this is why fsb is more important than processor speed.
fsb is the speed of the bus that connects all components together...therefore the maximum amount of throughput a computer has is really dependent on its bus speed. The processor needs to be much faster but in many instances the processor is way too fast for the data that needs to be sent to ram agp/pci buses etc. That medium thats used to connect those parts is transfered at the rate of the front side bus. It really makes sense if you think about it.
kirkyg
T-bird, run the Sandra CPU test. It gives a PR rating.
Quote:
Originally posted by Thunderbird1GHz
he? This is my setup LOL! and the 58 difference just goes to show that fsb rules over MHz too. oh I gain 6fps in quake3 per 90MHz cpu, thats sad. at 2GHz and 182fsb im at 331 against lostlilchild's 336 when he ran his cpu at 2.16GHz on 166fsb. 16fsb lower and 166MHz gives the same performance, so I am around xp2700 speeds at 2GHz but on 182fsb :)
Yes... but that wasn't the argument! You said that your machine would totally EAT (tm) mine in 3dmark aswell as in QIII. YOu never said "oh, mine machine would score similar to your STOCK 2700+"... you said it would kill mine, and transversly, I beat you.
Of course everyone knows FSB is important... but ram MHz is just as important, as I have proved :).
Trust me man....I've been 3D Markin for a long time. Yes FSB is very important but so is raw speed.Quote:
Originally posted by kirkyg
But this is why fsb is more important than processor speed.
fsb is the speed of the bus that connects all components together...therefore the maximum amount of throughput a computer has is really dependent on its bus speed. The processor needs to be much faster but in many instances the processor is way too fast for the data that needs to be sent to ram agp/pci buses etc. That medium thats used to connect those parts is transfered at the rate of the front side bus. It really makes sense if you think about it.
kirkyg
Believe me, I understand how a bus works.;)
I also think we should have more Pifast competitions.:)
Yep that would be fun, and I'm sure there would be only one person here who would go OTT about the results. :) On the subject of him, this is just another thread of his that really pissed me off in so much that he is telling us something that most of us already know for a fact, but also totally skewing them too. :mad:Quote:
Originally posted by vertices
I also think we should have more Pifast competitions.:)
Does this hold true in rela world applications?Quote:
Originally posted by vertices
Trust me man....I've been 3D Markin for a long time. Yes FSB is very important but so is raw speed.
Believe me, I understand how a bus works.;)
Of course it does.Quote:
Originally posted by Mancora
Does this hold true in rela world applications?
If all that mattered is fsb then we'd be fine right now with a 200mhz CPU at 1x200.
Raw speed matters. Just run some Super Pi, or folding@home. Raw CPU speed is just as important as fsb.
Hey lostlilchild your forgot one thing...
Eat(TM) is a registered trademark of Thunderbird1Ghz and cannot be used without the trademark label (TM).
Actually 3dmark is not great for proving much...it does show how your computer will run in a simulated game environment but not much else. There is alot more to take into consideration...but 3dmark is usually very responsive to fsb change...if the processor is of reasonable speed id rather have 30 more mhz fsb than i would 200 more mhz from the processor.Quote:
Originally posted by vertices
Trust me man....I've been 3D Markin for a long time. Yes FSB is very important but so is raw speed.
Believe me, I understand how a bus works.;)
kirkyg
Again with these flame hungry threads..
So guys, hows the whats your plans for the weekend?? :D
raw MHz matters in specific apps, obviously superPI and whatever does not use the video card or the ram, just the cpu for number crunching. Games use everything and games also love fast ram and lots and lots of ram bandwith this is why an increase of 1MHz to the fsb gains me more than 10MHz increase on the cpu. 33fsb is equivelent to 333MHz, thus xp3000s run 200fsb for 2.2GHz while xp2700 at 166fsb and 2.16GHz. 200fsb and 1800MHz is just as fast in games as 166fsb and 2.16GHz or stock xp2700 speeds :)
wow, 3dmark taught you all of that? Gotta hand it to futuremark, they sure know how to make people "smart." 33fsb = 333mhz...I like that. 1mHz on the FSB = 10 raw mHz, I like that also.Quote:
Originally posted by Thunderbird1GHz
raw MHz matters in specific apps, obviously superPI and whatever does not use the video card or the ram, just the cpu for number crunching. Games use everything and games also love fast ram and lots and lots of ram bandwith this is why an increase of 1MHz to the fsb gains me more than 10MHz increase on the cpu. 33fsb is equivelent to 333MHz, thus xp3000s run 200fsb for 2.2GHz while xp2700 at 166fsb and 2.16GHz. 200fsb and 1800MHz is just as fast in games as 166fsb and 2.16GHz or stock xp2700 speeds :)
True, FSB does make things better...what's funny is how you just realized that now. You're too funny...you should make your own tech site or write articles for [H] or something. Brilliant evaluation!
Quote:
Originally posted by Thunderbird1GHz
raw MHz matters in specific apps, obviously superPI and whatever does not use the video card or the ram, just the cpu for number crunching. Games use everything and games also love fast ram and lots and lots of ram bandwith this is why an increase of 1MHz to the fsb gains me more than 10MHz increase on the cpu. 33fsb is equivelent to 333MHz, thus xp3000s run 200fsb for 2.2GHz while xp2700 at 166fsb and 2.16GHz. 200fsb and 1800MHz is just as fast in games as 166fsb and 2.16GHz or stock xp2700 speeds :)
On the contrary, it isn’t superpi that is the exception here it’s 3Dmark. 3Dmark was never designed to be a CPU benchmark, and isn’t particularly representative of the type of work a CPU will do in game.
While games are often more sensitive to memory performance then may other apps, they still scaled anywhere between .6 and .7 with CPU clock rate the lat I checked (I.E. a 10% increase in clock rate yields a 6%-7% increase in performance.) This in indicative of a task that is still primarily CPU bound. I am of course referring to low resolution here, since hi image quality benchmarks are more often GPU bound.