New dual core CPU > older quad CPU? (for Logic)

Discussion in 'MacBook Pro' started by vladobizik, Sep 7, 2013.

  1. vladobizik macrumors 6502

    Joined:
    Aug 21, 2012
    Location:
    Slovakia
    #1
    I have a 2011 Mac mini Server with a quad core 2.0 GHz Sandy Bridge i7. I use it as my main production machine for Logic Pro X (plus everyday computer use), but I would like to get a smaller laptop in order to become completely mobile. The CPU in my mini is just fine for all tasks I need, so I just need a CPU that's not worse. For portability (and secondarily cost) reasons, I would like to get a 13" machine. I realize they all carry a dual core CPUs. If I get a top-of-the line BTO i7, for example in an upcoming Haswell MBP, I'm wondering if it can match my older quad core. Getting a MBA is also an option, but its clock speeds are quite low, I'm afraid that would certainly be a step down in Logic performance, but I'm not sure.
     
  2. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #2
    From a pure CPU processing standpoint, if you check out the Geekbench results you'll see there are a number of MBA and 13-inch MBPs that have better scores than your quad-core Mini.

    Comparing clock speed across different CPU generations is meaningless, a newer, "slower" CPU with fewer cores can easily out muscle an older, "faster" CPU with more cores.

    Your Mini is about 1/8 of the way down the page.

    http://browser.primatelabs.com/mac-benchmarks
     
  3. vladobizik thread starter macrumors 6502

    Joined:
    Aug 21, 2012
    Location:
    Slovakia
    #3
    Interesting. So you are saying I can get a high-end MBA with i7 and have no noticeable performance regression? That would be neat.
     
  4. CausticPuppy macrumors 65816

    Joined:
    May 1, 2012
    #4
    That link doesn't show any 13" notebooks with a higher score than the 2011 Mac Mini server.

    For Logic, the quad-core (even the "slow" one) will still be able to handle more tracks and more simultaneous plugins, but honestly you'd have to try really really hard to reach the limit of either CPU.

    I have Logic X and run it on my 2011 server, as well as my 13" rMBP with a 2.5GHz dual core i5.

    My anecdotal observations:

    I have actually used both right next to each other to compare performance.

    Logic starts up more quickly on the notebook, even though both systems have an SSD. My Mini has an aftermarket Crucial M4. However, my Mini also has a lot more plugins that it has to load, while my notebook has very few (only have the 128GB SSD).

    Likewise, the demo song also loads more slowly on the Mini. Much slower, in fact. Twice as long. I can't figure that out since it's loading from an SSD in both cases. Maybe the Samsung SSD in my Macbook Pro handles loading all those small files better.

    Once everything is loaded, the Mini is quicker to switch between view sets, open plugin windows, etc.

    While playing the demo song, the CPU tends to hang around 90% utilization on each system. However, the quad-core has more "headroom" since it will max out at 800% whereas the dual core maxes out at 400% (2 cores x 2 threads).

    That said, Logic does run fantastically well on my mere 2.5GHz dual core notebook; really the CPU shouldn't be a factor unless you are going to run dozens of audio tracks simultaneously.
     
  5. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #5
    You're absolutely right, I skipped down and right over the quad-core Mini to the dual-core i7, my bad. :) Thanks for catching that.
     
  6. TechZeke macrumors 68020

    TechZeke

    Joined:
    Jul 29, 2012
    Location:
    Rialto, CA
    #6
    Argh, synthetic benchmarks are absolutely useless. Based on real-world performance, this list is absolute horse sh-t.
     
  7. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #7
    And yet you offer nothing useful as an alternative means to compare general performance between models. If you're going to complain about a relatively well accepted benchmark then please provide something else for the OP to use as a general guideline/comparison when trying to understand how different models are likely to perform. Do you have an alternative?
     
  8. TechZeke macrumors 68020

    TechZeke

    Joined:
    Jul 29, 2012
    Location:
    Rialto, CA
    #8
    Actually I was typing a second response, which I would have given my opinion. However, I had to get the fact that the list is horse sh-t off my chest. Widely accepted doesn't mean much, when in the real world I get much different results. Frankly, real world should be all that matters.

    Synthetic benchmarks do things like over inflate i7 CPUs scores just because the i7 has hyper threading, which has little if any benefit in most applications. Most benefits in i7s are in the higher base clock speed more than anything.

    Now, back to the topic at hand.

    Basically, an air uses an ULV CPU. I honestly think performance-wise the OP would be better off with a 13" cMBP, since cost is an issue. Just use the external monitor the OP currently uses with the mini. He will be able to add more storage and 16gb of ram if he needs to as well.
     
  9. vladobizik thread starter macrumors 6502

    Joined:
    Aug 21, 2012
    Location:
    Slovakia
    #9

    Thank you everyone for your observations so far.
    I however still see no clear advantage of a 13'' cMBP, since all benchmarks show its CPU only marginally better, if at all.

    Look at these GeekBench results for the latest cMBPs and MBAs. I picked the highest scoring machines for each. Incidentally, both were done in a DP of Mavericks.
    2012 cMBP:
    http://browser.primatelabs.com/geekbench2/2076507
    2013 MBA:
    http://browser.primatelabs.com/geekbench2/2276764

    Both machines have their integer and floating point scores in the same ballpark. When you look at the actual graphs, the Air even fares better in most of them.

    cMBP's clear advantage is additional storage, especially if I get an OptiBay (which I definitely would), but MBA tops that with its portability and extremely speedy SSD to my eyes.
     
  10. AXs macrumors 6502a

    AXs

    Joined:
    Sep 7, 2009
    #10
    What the dude was trying to say, which I agree with - is that this is really a hypothetical situation where these results are based on.

    it highlights 1 scenario which will NEVER happen in real life. For Example - you will NEVER be using your Mac in this manner.
    There are also a lot of other factors to account for.

    But anyways,@OP - I obviously would recommend an Air over a cMBP. It's just far superior overall, despite not being customizable (accept for replacing SSD).

    But to get an ultimate - I dont know. The base Air 13" is $1,100 - you're jumping to about $1,900 for an ultimate Air (price varies based on where you live).

    You're paying almost 75% more than the value of the 'in-store' base model to get to your Ultimate. That's kind of steep imo.

    If you're willing to spend $2000, I think you could score a good 15" Retina Pro - especially if you're one who doesn't mind refurbs, you'll get one for less than $2000. I just saw a 2013 Feb release going for $1,800 earlier today.

    of course if cpu performance is important, a 15" rpro is your dream machine. It's not as portable as an Air of course, but you're getting a dedicated graphics card, bigger better Retina screen... and much more performance that you've ever had.
     
  11. vladobizik thread starter macrumors 6502

    Joined:
    Aug 21, 2012
    Location:
    Slovakia
    #11
    I live in Slovakia, and there is no refurbished market, for Macs or for anything else, you buy either new or used. A baseline 15'' rMBP is 2,199 € here (about $2,900). But that's beside the point.
    I know performance-wise a 15'' machine with a quad core is the way to go. But since I travel a lot, and I also need my laptop for school, not just "pro" use, portability is an issue (a bigger issue than cost). That's why I'm (tentatively) deciding between 13-inchers if they can offer me just enough performance.
     
  12. Macman45 macrumors demi-god

    Macman45

    Joined:
    Jul 29, 2011
    Location:
    Somewhere Back In The Long Ago
    #12
    FWIW, I use a 27" Imac with Logic X...Main Mac is top spec I7 with 32GB RAM. I have a second installation of Logic X on my rMBP 13". I7 3GHZ 8GB RAM 500GB SSD...Works a charm for edits etc, and I would have thought you may be able to pick up a refurb if you keep your eyes on the Apple site.
     
  13. yliu macrumors regular

    Joined:
    Jun 15, 2009
    #13
    Yeah, maybe a BTO i7 in the 13 MBP will beat your mac mini. But the Mac mini is only 1 generation behind the Ivy Bridge and performance improvement is not the main thing with Haswell.
     
  14. vladobizik thread starter macrumors 6502

    Joined:
    Aug 21, 2012
    Location:
    Slovakia
    #14
    But Haswell (as well as Ivy Bridge) is leaps and bounds ahead of the Core 2 Duo powering my 2010 13'' MBP which is the computer I use for mobile scenarios now.
     
  15. yliu macrumors regular

    Joined:
    Jun 15, 2009
    #15
    Yes, I believe a BTO i7 MBA CPU would beat a Core 2 Extreme according to benchmarks :O
     
  16. Stetrain macrumors 68040

    Joined:
    Feb 6, 2009
    #16
    The 13" cMBP or 13" retina MBP will have better cooling and overall thermal headroom when compared to the Macbook Air. The 13" retina specifically has a better cooling system with the two intake vents at the front and the two exhaust fans at the rear (the cMBP and the Air only have one fan).

    The Macbook Air has strong peak performance but sustaining high CPU usage for long periods of time will most likely run into thermal limits and may cause throttling.

    I think that the 2013 Airs are a bit better about this than previous versions but it's still just a matter of physics.
     
  17. yosemit, Sep 8, 2013
    Last edited: Sep 8, 2013

    yosemit macrumors regular

    Joined:
    Jul 19, 2013
    #17
    You are right. The 2013 Airs, however, are much better than 2012 Airs. Battery life is 12 hours vs. 7 hours, which means the power rate ratio is about 7:12, so the performance of the new Airs is much less thermally restrained than previous ones. It matches my experience with an old Air and a new Air.

     
  18. AXs macrumors 6502a

    AXs

    Joined:
    Sep 7, 2009
    #18
    Umm, the cMBP has 2 fans because that's what it takes to keep the system running without overheating. You can twist it around to say it has a 'better cooling system' because it has 2 fans, but what you're really saying is that the cMBP just gets much hotter.

    I'm currently at 1195rpm, and cpu 44degrees C. This includes running regular essential apps, and Safari with a 1080p video on youtube that has been playing for 2 hours now on an external monitor via hdmi.

    When I'm just browsing, using office... cpu is at 32degreesC.

    What's the min fan speed for a pro? 2000rpm? minimum temperature arnd 50degrees? arnd 60 degrees for surfing?

    Next you're going to say that the Pro has a better cooling system than the iPad, which is conditionally true - since iPads don't have fans in them.
     
  19. TechZeke macrumors 68020

    TechZeke

    Joined:
    Jul 29, 2012
    Location:
    Rialto, CA
    #19
    Ok, got it.
     
  20. Stetrain macrumors 68040

    Joined:
    Feb 6, 2009
    #20
    But we've already established that the processors are very similar in performance. Being in a larger chassis with more air flow with the same performance means lower temperatures, I'm not sure where you're getting your information from.

    Also I said that the cMBP has one fan, the one with the improved cooling system is the rMBP:

    http://www.anandtech.com/show/6409/13inch-retina-macbook-pro-review/12
     
  21. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #21
    The 13" cMBP has one fan, it's the 15" and 17" that have two.
     
  22. AXs macrumors 6502a

    AXs

    Joined:
    Sep 7, 2009
    #22
    You guys are so missing the point. okay, the rmbp is the 1 with the 'improved cooling system'.

    Improved Cooling System = System that easily gets hot.

    That's the whole point, it's how you state it.

    Like I said, in terms of cooling systems - the iPad must be considered LOUSY right?

    No, because even though it gets hot it doesn't need a fan. The same way the MBA only needs a very lightly powered singular fan. That's because it has a superior overall temperature control and requires very little power to run - which in turns produces less heat.

    Next you'll say a monster gaming pc with gaming PC with liquid cooling system is the most superior cooling system... which again, would be accurate - but irrelevant because it requires high cooling in the first place because of how hot it runs.

    THAT is the point.

    The End.
     
  23. Mr. Retrofire macrumors 601

    Mr. Retrofire

    Joined:
    Mar 2, 2010
    Location:
    www.emiliana.cl/en
    #23
    iThink™ that it is not logically logic that Logic runs logically faster on two cores. That’s my logic.
     
  24. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #24
    It's hard to understand your point when you don't even know the layout/config of the hardware that you're trying to make the point about.

    The rMBPs don't have an "improved cooling system" as a result of the intake vents. The rMBPs have those intake vents because the system doesn't have the necessary ports/openings like the cMBP.

    The cMBP pulls air in from multiple locations, including space around the expansion port openings, and as a result there's no need to have additional intake vents built in to the side or bottom of the case.

    The rMBP is also packed much more tightly and the vents are needed to get air in to the correct side of the logic board.

    The design created certain airflow requirements, the cooling is not "improved" over any other MBP, IMO.
     
  25. Stetrain macrumors 68040

    Joined:
    Feb 6, 2009
    #25
    What? More thermal dissipation capacity with the same amount of peak performance means less heat buildup, or the same heat buildup with greater sustained levels of performance.

    An iPad has a much lower level of performance. It generates less heat, so it needs less cooling. Having more cooling with the same amount of heat generation results in a cooler running system that can sustain peak performance for longer.

    A monster gaming PC has a lot more performance and therefore has to dissipate a lot more heat. It uses a bigger cooling system so that it can continue to run cool despite the higher performance.

    As I recall, what the OP is actually asking is which machine will give him more CPU performance.

    The Macbook Air uses processors with a 17W TDP. That roughly means that the cooling system should be able to dissipate a continuous 17W heat load.

    The 13" MBPs use processors with a 35W TDP. That roughly means that the cooling system should be able to dissipate a continuous 35W heat load.

    Being able to dissipate that extra heat means that the 35W machines can give you more sustained performance than the 17W machines. The 17W machines might have similar burst performance in a benchmark but that doesn't mean that they can sustain that performance.


    A machine with more thermal capacity is not a machine that runs hotter. If the cooling system can dissipate that extra generated heat, it means you get more performance without a hotter machine in your lap.
     

Share This Page