PDA

View Full Version : MacBook Pro Core 2 Duo Performance


MacBytes
Oct 26, 2006, 11:56 PM
http://www.macbytes.com/images/bytessig.gif (http://www.macbytes.com)

Category: Benchmarks
Link: MacBook Pro Core 2 Duo Performance (http://www.macbytes.com/link.php?sid=20061027005637)
Description:: A performance comparison between the new MacBook Pro (with a Core 2 Duo processor) and the old MacBook Pro (with a Core Duo processor). Also compares the performance between 32-bit and 64-bit performance on the new MacBook Pro.

Posted on MacBytes.com (http://www.macbytes.com)
Approved by Mudbug

interlaced
Oct 27, 2006, 12:34 AM
Hooray! Benchmarks! Just what I wanted to see.

Rocksaurus
Oct 27, 2006, 12:39 AM
"Moving from the Core Duo to the Core 2 Duo means 32-bit MacBook Pro performance is up 10% without an increase in processor clock speed. Thatís impressive"

I don't get what all the Core 2 (mobile) hype is about. 10% is the speed increase you gain from doubling the cache... Which the Core 2 Duo did (4 MB vs 2 MB)... And don't say 64-bit.

Analog Kid
Oct 27, 2006, 03:22 AM
"Moving from the Core Duo to the Core 2 Duo means 32-bit MacBook Pro performance is up 10% without an increase in processor clock speed. That’s impressive"

I don't get what all the Core 2 (mobile) hype is about. 10% is the speed increase you gain from doubling the cache... Which the Core 2 Duo did (4 MB vs 2 MB)... And don't say 64-bit.
I'm with you-- I couldn't understand all the moaning from people wanting Merom. It's just not that big of a deal. Granted, it looks like there are certain operations that give you more than 10%, so some folks will be pretty happy.

Anybody out there understand why Core2 choked on a couple of these? Stdlib allocation jumps out as a major bummer...

[edit: oh, and note to benchmarkers-- if you're trying to show the advantage of B over A, normalize to A, not C...]

Swarmlord
Oct 27, 2006, 09:15 AM
Hmmm, these benchmarks are making me feel a bit better about getting the Mini with Core Duo sooner rather than later.

twoodcc
Oct 27, 2006, 11:28 AM
nice benchmarks. about what i thought the difference would be

wyatt23
Oct 27, 2006, 12:42 PM
wow. i was really expecting a much bigger increase in performance. oh well... does it get way better battery life? if not, i may have to try and grab an "old" mbp.

Bill Gates
Oct 27, 2006, 01:50 PM
I'm with you-- I couldn't understand all the moaning from people wanting Merom. It's just not that big of a deal. Granted, it looks like there are certain operations that give you more than 10%, so some folks will be pretty happy.

Anybody out there understand why Core2 choked on a couple of these? Stdlib allocation jumps out as a major bummer...

[edit: oh, and note to benchmarkers-- if you're trying to show the advantage of B over A, normalize to A, not C...]
Yes, but look at the 64-bit performance! In may cases it's leaps and bounds ahead of the 32-bit Core Duo. Once Leopard is released, more of its potential will be realized.

n8236
Oct 27, 2006, 04:05 PM
I dunno what the author was talking about, 10% speed improvement is NOT "impressive" imo. It's hardly a double digit bump, if not a single digit bump if it had a % of error margin.

As for the 64 vs 32 bit, why da Hell would u cross compare a C2D running 64bit to a C1D running 32bit? It makes no sense to me. I can't think of any real works situations like that. Wouldn't comparing a C2D to a C1D on either 32 then 64bit make more sense? o.O

Anyways, 10% is rather not "impressive." In a non-scientific way, if the encoding of video took took 5 mins on a C1D, the C2D would only save u 30 seconds.

Bill Gates
Oct 27, 2006, 04:09 PM
I dunno what the author was talking about, 10% speed improvement is NOT "impressive" imo. It's hardly a double digit bump, if not a single digit bump if it had a % of error margin.

As for the 64 vs 32 bit, why da Hell would u cross compare a C2D running 64bit to a C1D running 32bit? It makes no sense to me. I can't think of any real works situations like that. Wouldn't comparing a C2D to a C1D on either 32 then 64bit make more sense? o.O

Anyways, 10% is rather not "impressive." In a non-scientific way, if the encoding of video took took 5 mins on a C1D, the C2D would only save u 30 seconds.
A Core Duo is a 32-bit chip. It can't run 64-bit code, plain and simple. It won't be able to take advantage of 64-bit optimizations in Leopard.

Regarding the 30 second savings for that benchmark, 30 seconds for several files can yield a decent saving in time.

n8236
Oct 27, 2006, 04:11 PM
A Core Duo is a 32-bit chip. It can't run 64-bit code, plain and simple. It won't be able to take advantage of 64-bit optimizations in Leopard.

Regarding the 30 second savings for that benchmark, 30 seconds for several files can yield a decent saving in time.

I see, gotcha. If that's the case, will it mean that the C2D will run much better than the C1D w/ Leopard?

Bill Gates
Oct 27, 2006, 04:13 PM
I see, gotcha. If that's the case, will it mean that the C2D will run much better than the C1D w/ Leopard?
Depending on how much Apple optimizes it, it's certainly possible.