Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Leopard Boot Speed Test: SR 2.0GHz MacBook vs. PPC 1.5GHz PB G4 12"

Squonk said:
The SR's have only come in 2.2 or 2.4 in the MBPs. I'd think that the SR would be faster. You can have a faster CPU, but if you are limited by how fast you can feed it, then the slower CPU with faster system board (within reason) would be faster.

A 2.16 is 8% faster than a 2.0GHz CPU
A 800MHz is 50% faster than a 533MHz motherboard
Click to expand...
It's obvious you're a long time Mac user. Thankfully, coming from the PC world, I probably have a better grasp on real-world performace of Intel CPU's than you do.

First off,the 2.16 and 2.33 MBP chips ran at 667mhz, not 533. Yes, the 2.2 and 2.4 run at 800. However, in the real world, the performance difference is so negligible it's not even funny.

If you want to see how insignificant a huge increase in FSB is, check out this chart:
http://www23.tomshardware.com/cpu_2007.html?modelx=33&model1=873&model2=876&chart=431

Here I've compared a desktop C2D E6700 with a E6750. The both run at 2.66ghz. The difference? The E6700 runs at 1066mhz FSB and the E6750 is a 1333mhz chip. Just scroll down the chart (you don't have to go far) and look for the two red bars. Those are the chips I have selected. In this example, the E6750 can encode a 24-second MPEG2 1080p clip to h.264 1 second faster than the E6700.

One second, for a roughly 20% increase in FSB.

Now, this is a fast chip, but I chose them because they were the only C2D's in the list that were the same clock speed but a different FSB. Let's compare something similar to what the person you responded to asked:

http://www23.tomshardware.com/cpu_2007.html?modelx=33&model1=874&model2=877&chart=431

This is the same video encoding test done with a desktop E6600 and E6550. This is the comparison I am attempting to make here:

E6600 - 2.4ghz 1066mhz FSB = MBP 2.33ghz 667mhz FSB
E6550 - 2.33ghz 1333mhz FSB = MBP 2.2ghz 800mhz FSB

What I'm saying here is I picked one chip with a slower processor speed but faster fsb and one with a slower fsb but a faster clock speed that would put us in the same situation as trying to compare the MBP 2.33 vs the MBP 2.4.

The difference in the video encoding test? The chip with the faster clock speed but slower FSB wins by 2 seconds. 1:58 vs 2:00.

Check out any of the applications. Sometimes you'll see the 6600 is on top, sometimes it's the 6550. But they're always right next to each seperated by such a thin margin you'd never be able to tell unless you pulled out a stopwatch.

So in this case, I would probably veture a guess and say that processor-wise, the 2.33ghz MBP would be slightly faster than the 2.2ghz SR-based MBP, though not by much. But the video card in the 2.2ghz machine certainly makes the purchase more worthwhile over the X1600-based 2.33ghz machine.

Remember, that in the Intel world, advertisied FSB doesn't mean a whole lot. 667mhz is really 166mhz. 800mhz is really 200mhz. 1066 is 266 and 1333 is 333. It's just that it transmits 4 times per clock cycle. That's why Intel refers to it as a 'Quad-Pumped FSB'. Saying it's 667, 800, 1066 1333 is, and always has been, very misleading, IMHO.
 
Does it matter? a 200MHz increase in speed would get your maybe 0.5 seconds off in boot time, in theory. In real-world tests, there are so many factors that contribute to speed/performance. On benchmarks, its an average rating, soemthing to go by, not something to have to get. If you score a 200 on a CPU score, but someone else with the same CPU scored maybe like 6000, then you have a problem with your hardware. But if both users scored within a few points, then its nothing.

Not every CPU/Hardware/Chip is built the same, its the same structure and design yes but theres always a better and worse one.
 
Jiddick ExRex said:
What the hell are you talking about?
Click to expand...

Bearxor did a good job at explaining it more. Thanks Bearxor!!!

To me, if a system has a 20% faster FSB, all other things being equal, you'd probably feel the difference.

Oh and thanks for the correction on the 2.16/2.33's running at 667 instead of 533.
 
I'm going to chime in with a question really quick here, completely out of ignorance from never owning a laptop nor a Mac.

Is it a good idea to keep my MBP plugged in overnight while it downloads a file while conected to the AC?

Sorry for the dumb question, just want to know if it's okay to do this ;)
 
Swift Sketcher said:
I'm going to chime in with a question really quick here, completely out of ignorance from never owning a laptop nor a Mac.

Is it a good idea to keep my MBP plugged in overnight while it downloads a file while conected to the AC?

Sorry for the dumb question, just want to know if it's okay to do this ;)
Click to expand...

Yes, it's fine. Laptops are meant to be plugged in 99%+ of the time.
 
I think constant in the proportionality equation of speed of FSB/clock speed and actual speed is a large difference.

As bearxor says the difference FSB speeds make is negligible when the clock speed of one of the processors surpasses the other by even a small margin.

take the equation y=kx where y is actual performance gained, k is the proportionality constant and x is the FSB speed

and the equation y=qz where y is actual performance gained, k is the proportionality constant and z is the clock speed.

The value for q will be greater than k as core speed makes more of a difference than FSB.

Is it not, therefore, possible to formulate an equation to instantly determine which processor is faster (theoretically) when given the speeds of the FSB and the core speed?

I think once you work out a constant when all the other specifications stay the same or you create a weighted index of all the factors of each processor specification you are able to tell which processor is faster when common sense can't help you.
 
lavrishevo said:
It it ok to leave my laptop plugged in while downloading Brokeback Mountain 2 "A Mountain of Broken Male Backs" :rolleyes:
Click to expand...

Sure why not? :) Didn't know that such film exist tho! I'm not a big fan of Brokeback Mountain, however I enjoy Bareback Mountain from Falcon Studios. :D
 
Dude, what?

ojwk said:
I think constant in the proportionality equation of speed of FSB/clock speed and actual speed is a large difference.

As bearxor says the difference FSB speeds make is negligible when the clock speed of one of the processors surpasses the other by even a small margin.

take the equation y=kx where y is actual performance gained, k is the proportionality constant and x is the FSB speed

and the equation y=qz where y is actual performance gained, k is the proportionality constant and z is the clock speed.

The value for q will be greater than k as core speed makes more of a difference than FSB.

Is it not, therefore, possible to formulate an equation to instantly determine which processor is faster (theoretically) when given the speeds of the FSB and the core speed?

I think once you work out a constant when all the other specifications stay the same or you create a weighted index of all the factors of each processor specification you are able to tell which processor is faster when common sense can't help you.
Click to expand...

No more. Please no more. ;)
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back