Primate Labs Blog : MacBook Pro Performance (February 2008)

MacBook Pro Performance (February 2008)

Posted February 27th, 2008 in Geekbench, Mac OS X, MacBook Pro | Comments

Yesterday Apple released updated MacBooks and MacBook Pros both of which feature (among other things) new processors. While I don’t have Geekbench results for the new MacBooks, I wanted to compare Geekbench results for some of the new MacBook Pros against results for older MacBook Pros.

Setup

Here is the configuration of the test machines:

MacBook Pro (Early 2008)
- Intel Core 2 Duo T9300 @ 2.50GHz
- 2.00 GB 667 MHz DDR2 SDRAM
- Mac OS X 10.5.2 (Build 9C2018)
MacBook Pro (Early 2008)
- Intel Core 2 Duo T8300 @ 2.40GHz
- 2.00 GB 667 MHz DDR2 SDRAM
- Mac OS X 10.5.2 (Build 9C2018)
MacBook Pro (Mid 2007)
- Intel Core 2 Duo T7800 @ 2.60GHz
- 2.00 GB 667 MHz DDR2 SDRAM
- Mac OS X 10.5.2 (Build 9C31)
MacBook Pro (Mid 2007)
- Intel Core 2 Duo T7700 @ 2.40GHz
- 2.00 GB 667 MHz DDR2 SDRAM
- Mac OS X 10.5.2 (Build 9C31)

If you’re not familiar with Geekbench 2, benchmarks are scored against a baseline, where a score of 1000 represents the performance of a Power Mac G5 @ 1.6GHz. Higher is better.

Results

Overall Performance

MacBook Pro (Early 2008) Intel Core 2 Duo T9300 @ 2.50GHz	3228
MacBook Pro (Early 2008) Intel Core 2 Duo T8300 @ 2.40GHz	3047
MacBook Pro (Mid 2007) Intel Core 2 Duo T7800 @ 2.60GHz	3245
MacBook Pro (Mid 2007) Intel Core 2 Duo T7700 @ 2.40GHz	3107

Integer Performance

MacBook Pro (Early 2008) Intel Core 2 Duo T9300 @ 2.50GHz	2727
MacBook Pro (Early 2008) Intel Core 2 Duo T8300 @ 2.40GHz	2631
MacBook Pro (Mid 2007) Intel Core 2 Duo T7800 @ 2.60GHz	2887
MacBook Pro (Mid 2007) Intel Core 2 Duo T7700 @ 2.40GHz	2693

Floating Point Performance

MacBook Pro (Early 2008) Intel Core 2 Duo T9300 @ 2.50GHz	4579
MacBook Pro (Early 2008) Intel Core 2 Duo T8300 @ 2.40GHz	4278
MacBook Pro (Mid 2007) Intel Core 2 Duo T7800 @ 2.60GHz	4459
MacBook Pro (Mid 2007) Intel Core 2 Duo T7700 @ 2.40GHz	4380

Memory Performance

MacBook Pro (Early 2008) Intel Core 2 Duo T9300 @ 2.50GHz	2412
MacBook Pro (Early 2008) Intel Core 2 Duo T8300 @ 2.40GHz	2279
MacBook Pro (Mid 2007) Intel Core 2 Duo T7800 @ 2.60GHz	2387
MacBook Pro (Mid 2007) Intel Core 2 Duo T7700 @ 2.40GHz	2253

Stream Performance

MacBook Pro (Early 2008) Intel Core 2 Duo T9300 @ 2.50GHz	1889
MacBook Pro (Early 2008) Intel Core 2 Duo T8300 @ 2.40GHz	1739
MacBook Pro (Mid 2007) Intel Core 2 Duo T7800 @ 2.60GHz	1971
MacBook Pro (Mid 2007) Intel Core 2 Duo T7700 @ 2.40GHz	1818

Conclusions

At first glance the results seems counter-intuitive; shouldn’t the new MacBook Pro @ 2.4GHz be faster than the old MacBook Pro @ 2.4GHz? Isn’t the Penryn better than the Merom? It turns out the answer is, “not necessarily”.

The processor in the new MacBook Pro @ 2.4GHz (T8300) has less L2 cache than processor in the old MacBook Pro @ 2.4GHz (T7700) which explains the slight drop in performance.

However, when you consider the new MacBook Pro 2.4GHz is the base model, while the old MacBook Pro 2.4GHz was the mid-range model, the small drop in performance comes with a significant reduction in price. Plus, the new Penryn processor uses less energy than the old Merom processor, which means a cooler laptop with increased battery life.

Joseph

I would love to see some SSE4 tests and see how they compete. Do you have any applications that can use SSE4? Anandtech.com says we should see 40% increase in speed when using SSE4.

Also, I'd love to see some battery comparisons to see what we are gaining. DVD playback and just idling along. Penryn is supposed to use almost nothing when idling. So, this should boast tremendous gains when web browsing where you stop and read for a while.

Steve

A good article. However, it is not accurate to make the "significant reduction in price" argument in this case as the previous Merom 2.4 model can be had TODAY (and TODAY is all that counts) for $1744 and the Penryn model is still $100 more at $1844 (Amazon prices) -- so that's $100 more for what may be a slower computer that has pinch capability. That is a question to ask: is this capability worth it? That and the bigger question: is the increased battery life appreciable -- or not? $100 is nearly what a hi-cap battery runs, or covers most of that cost, so this point may be moot as well.

McDave

I can't check from work (page blocked) but do the Geekbench tests look at higher functions like video encodes etc. Something which engages the SIMD otherwise it's a bit like testing a car engine with the turbo switched off

McD

Ren Calderhead

My question is this. Why make this release if they are going to do it again in June. There is a huge investment (cost) in each release unless there is just chip replacement (therefor little additional cost) and this is the base for totally new architecture! Or is this some kind of ill conceived attempt to keep up with Intel or who? HMMMM? All of this hoopla and stuff around the Intel core duo & we are still stuck at 3ghz.

What is really going on here?

Nick

It will be interesting to see what gains there are in SSE4-aware applications, since the Penryn is the first CPU with these new instructions.

michael

Thanks for doing these!

I'd love to see comparisons between a base model pro and a macbook. I am having a hard time deciding between the two. ...

Terry

I'll pay the $1999 for the 2.4 ghx model with slightly reduced performance but much lower heat and an hour longer battery life. It's a no brainer as far as I am concerned. I'm putting 4 GB of ram in it an ordering a 320 GB drive from OWC. That drive is $187 and is a much better deal than paying $500 for the middle MBP with a slightly larger drive and slightly increased speed. if the model in June is better I'll eBay this one for very little loss and buy the new one. No problem.

Primate Labs Blog