PDA

View Full Version : 13" MBA Ultimate graphics/processor compared to my old MBP




mrsir2009
Sep 29, 2011, 01:38 PM
I'm thinking about getting a MacBook Air 13" Ultimate, and giving my 2009 13" MBP to someone else in my family who needs a better laptop. So how will the processor and graphics compare to my current computer:

MBP: 256 NIVIDA graphics - 2.22GHz Core2Duo Processor

MBA: 384 INTEL integrated graphics - 1.7 i5 Processor

Thanks!



KPOM
Sep 29, 2011, 02:38 PM
I'm thinking about getting a MacBook Air 13" Ultimate, and giving my 2009 13" MBP to someone else in my family who needs a better laptop. So how will the processor and graphics compare to my current computer:

MBP: 256 NIVIDA graphics - 2.22GHz Core2Duo Processor

MBA: 384 INTEL integrated graphics - 1.7 i5 Processor

Thanks!

Both the processor and GPU are faster on the MacBook Air. Think of the 1.7GHz i5 sort of like a 3.5GHz Core 2 Duo (it isn't quite the same, but it's close, between a 2.4GHz dual core Turbo Boost plus the efficiency of the Sandy Bridge architecture). The integrated graphics are somewhere in between the NVIDIA 9400m in your 13" Pro and the 320M that was in last year's Air.

bill-p
Sep 30, 2011, 12:07 PM
I'm thinking about getting a MacBook Air 13" Ultimate, and giving my 2009 13" MBP to someone else in my family who needs a better laptop. So how will the processor and graphics compare to my current computer:

MBP: 256 NIVIDA graphics - 2.22GHz Core2Duo Processor

MBA: 384 INTEL integrated graphics - 1.7 i5 Processor

Thanks!

Processor would be faster. Slightly.

Graphics would be worse in some games and better in some others. Mostly worse.

Some games that are better are Source-based Valve game like Counter-Strike, Team Fortress 2, Left 4 Dead, WoW, StarCraft 2, and Diablo 3 if your main concern falls anywhere in between those games.

If you don't game then the graphics make no difference whatsoever.

KPOM
Sep 30, 2011, 12:25 PM
Processor would be faster. Slightly.

Graphics would be worse in some games and better in some others. Mostly worse.

Some games that are better are Source-based Valve game like Counter-Strike, Team Fortress 2, Left 4 Dead, WoW, StarCraft 2, and Diablo 3 if your main concern falls anywhere in between those games.

If you don't game then the graphics make no difference whatsoever.

The Intel HD3000 is faster than the NVIDIA 9400m, albeit slightly.

The 1.7GHz i5 is a lot faster than the 2.26GHz Core 2 Duo, not slightly faster. It is 2 full CPU generations ahead of the Core 2 Duo, can turbo boost to a higher clock speed (2.4GHz), and is about 50% more efficient per clock cycle (and even higher on multi-threaded tasks).

Xgm541
Sep 30, 2011, 12:49 PM
Processor would be faster. Slightly.


if by slightly you mean about 2x then you are correct .

nebulos
Sep 30, 2011, 01:19 PM
Processor would be faster. Slightly.

huh?

if by slightly you mean about 2x then you are correct .

oh, now i get it.

Anand's review (http://www.anandtech.com/show/4528/the-2011-macbook-air-11-13inch-review/9): includes comparison with 2009 13 MBP with 2.26GHz CPU.

here's a CPU test (http://forums.macrumors.com/showthread.php?t=1209216&highlight=garageband) i did in garageband, comparing the 2011 MBAs and my 2009 Mini with the same 2.26GHz CPU; the test used 75% of the CPU on the mini, and only 25% on the i5 13 MBA.

bill, what's up?

bill-p
Sep 30, 2011, 07:24 PM
huh?

oh, now i get it.

Anand's review (http://www.anandtech.com/show/4528/the-2011-macbook-air-11-13inch-review/9): includes comparison with 2009 13 MBP with 2.26GHz CPU.

here's a CPU test (http://forums.macrumors.com/showthread.php?t=1209216&highlight=garageband) i did in garageband, comparing the 2011 MBAs and my 2009 Mini with the same 2.26GHz CPU; the test used 75% of the CPU on the mini, and only 25% on the i5 13 MBA.

bill, what's up?

Depending on the application, the SSD can help offset I/O performance and give the illusion that the CPU is a lot faster. That's what happened with the Photoshop CS4 benchmark. Unless you truly believe a 1.86GHz Core 2 Duo is almost as fast as a 2.26GHz Core 2 Duo.

If we take the Cinebench benchmark, then the advantage of faster CPU bus, faster RAM, and TurboBoost would all give the 1.7GHz Core i5 that much of an edge over the 2.26GHz Core 2 Duo. About 1.5x, but nowhere near 2x, and that's more of an unrealistic condition.

Also check your Mini to see if the CPU didn't throttle itself due to heat. I have had my Macbooks overheat and throttle itself down to almost half its actual operating frequency. If you haven't tried it yet, Coolbook is a good app to use to reduce the voltage running through the CPU and make sure it's not running too hot. It also lets you monitor current frequency and voltage if nothing else.

As a whole package, the Macbook Air 2011 is indeed a lot faster and more responsive than the Macbook Pro that the OP has, but the raw CPU performance still wouldn't be as blown out as it seems.

A good test for pure CPU performance would be to run an emulator.

KPOM
Sep 30, 2011, 11:43 PM
Depending on the application, the SSD can help offset I/O performance and give the illusion that the CPU is a lot faster. That's what happened with the Photoshop CS4 benchmark. Unless you truly believe a 1.86GHz Core 2 Duo is almost as fast as a 2.26GHz Core 2 Duo.

If we take the Cinebench benchmark, then the advantage of faster CPU bus, faster RAM, and TurboBoost would all give the 1.7GHz Core i5 that much of an edge over the 2.26GHz Core 2 Duo. About 1.5x, but nowhere near 2x, and that's more of an unrealistic condition.


I had a 2010 MacBook Air. The SSD and I/O is virtually unchanged between the 2010 and 2011. The main difference is the processor, so yes, I think I'm qualified in reporting that the 2011 is significantly faster, particularly on tasks that can take advantage of multiple processor cores.

A 1.86GHz Core 2 Duo with 6MB cache is indeed nearly as fast as a 2.26GHz Core 2 Duo with 3MB cache. The clock speed is about 82% as fast, and the additional cache makes up some of the difference.

I'm not sure why people don't want to admit that the Sandy Bridge processors are significantly faster than the Core 2 Duo chips. The Core 2 Duo is a 5 year old design. The Sandy Bridge is about a year old (and 2 micro architecture generations newer than the Core 2 Duo). With the ability to boost both its cores to 2.4GHz and hyperthread, the 1.7GHz i5 will run circles around the 2.22GHz Core 2 Duo that the OP has. Even if it's a conservative 1.5x faster, that's hardly "slightly" faster. You make it sound like virtually all the difference is from the SSD. That's a bit part of the difference when booting or launching applications, but when processing video, encoding, or doing anything else CPU-intensive, it's the Sandy Bridge chip that's making the difference.

nebulos
Oct 1, 2011, 01:29 AM
Depending on the application, the SSD can help offset I/O performance and give the illusion that the CPU is a lot faster. That's what happened with the Photoshop CS4 benchmark. Unless you truly believe a 1.86GHz Core 2 Duo is almost as fast as a 2.26GHz Core 2 Duo.

If we take the Cinebench benchmark, then the advantage of faster CPU bus, faster RAM, and TurboBoost would all give the 1.7GHz Core i5 that much of an edge over the 2.26GHz Core 2 Duo. About 1.5x, but nowhere near 2x, and that's more of an unrealistic condition.

if we're having this conversation, we understand what is disk bound and what isn't. anyways, let's forget the Photoshop test. indeed, let's look at Cinebench, which, AFAIK, is purely CPU (GPU), and so independent of RAM. (Correct me with a source if i'm wrong; I'd like to know.)

Anand's Cinebench R10 single-threaded scores:

2011 1.7 13 MBA = 3770
2009 2.26 MBP = 2584
2010 1.86 MBA = 2166

these results tell you that the 1.7 is close to the 2.26, and the 1.86 isn't?

Also check your Mini to see if the CPU didn't throttle itself due to heat. ... A good test for pure CPU performance would be to run an emulator.

The GarageBand test is a CPU test. Cinebench is a CPU test.

I reran the GB test on the Mini. In fact, it did slightly better this time, using only 68% of its CPU, (down from 75%). Temp was steady at 68 degrees. Updated results, with a shot of the test in action, including readings from the istat widget can be found here (http://forums.macrumors.com/showthread.php?t=1209216). Not sure what the difference was, but I doubt there was any throttling happening in the original test, and the results are still dramatically different. For easy reference:

My GarageBand CPU test: see here (http://forums.macrumors.com/showthread.php?t=1209216)

2009 2.26 Mini 68% CPU usage
2011 1.7 MBA 25% CPU usage

Macman45
Oct 1, 2011, 01:39 AM
On the Intel Chipset will be worse than your current Nvidia setup IF you are going to game on it...Personally I don't use my Air for games so its not an issue for me.

It's very fast for copying, backing up large Video and music files created in the studio then brought home and transferred to my iMac for further work.

I've just never got the whole laptop for games thing. You have to spend megabucks to get anything that will perform even to iPad standard....Actually the iPad is a pretty good HD gaming platform and I do indulge on mine...Pretty amazing what the Devs can do with 512 Video.

bill-p
Oct 1, 2011, 04:05 PM
if we're having this conversation, we understand what is disk bound and what isn't. anyways, let's forget the Photoshop test. indeed, let's look at Cinebench, which, AFAIK, is purely CPU (GPU), and so independent of RAM. (Correct me with a source if i'm wrong; I'd like to know.)

Anand's Cinebench R10 single-threaded scores:

2011 1.7 13 MBA = 3770
2009 2.26 MBP = 2584
2010 1.86 MBA = 2166

these results tell you that the 1.7 is close to the 2.26, and the 1.86 isn't?

The single-threaded benchmark will make TurboBoost boost the 1.7 to 2.7GHz if I recall correctly, so those numbers are about right.

But that aside, it's actually quite true that Cinebench does benefit from more RAM bandwidth. Check this:

http://www.bit-tech.net/hardware/memory/2011/01/11/the-best-memory-for-sandy-bridge/8

Both Nehalem and Lynnfield can get a few hundreds more points depending on how much memory bandwidth they get. It seems small when considering that it's 12,000 points total, but... try that with about 2000 or 3000 and the scaling of a few hundreds will mean something.

The GarageBand test is a CPU test. Cinebench is a CPU test.

I reran the GB test on the Mini. In fact, it did slightly better this time, using only 68% of its CPU, (down from 75%). Temp was steady at 68 degrees. Updated results, with a shot of the test in action, including readings from the istat widget can be found here (http://forums.macrumors.com/showthread.php?t=1209216). Not sure what the difference was, but I doubt there was any throttling happening in the original test, and the results are still dramatically different. For easy reference:

My GarageBand CPU test: see here (http://forums.macrumors.com/showthread.php?t=1209216)

2009 2.26 Mini 68% CPU usage
2011 1.7 MBA 25% CPU usage


Well, now that I see your test methodology, I think it's worth pointing out that under Activity Monitor, there is a line that reads "User%", which is more accurate than taking the CPU % inside and divide by 200 or 400.

Why? Because there is no way CPU scaling is that linear. 120% may actually mean that one core runs at 80% while the other runs at 40%, and that's perfectly normal. That doesn't mean only 60% of the CPU is used. We're dealing with 2 cores, not a single processor split in two.

Also, this test just shows to me that your Mac Mini is indeed throttling itself.

How do I know that? My Macbook Air 13" gets 55% or 110% average during that "test". At 1.60GHz.

So somehow my Macbook Air with a slower processor is actually faster than your Mac Mini? That doesn't seem right.

Also I think it's worth pointing out that I repeated the test with Coolbook on a Macbook Pro as well.

At 1.33GHz, GarageBand reports that there are too many effects to be played in real time, so it cuts off right there.

1.6GHz, 1.86GHz, 1.99GHz, 2.16GHz, 2.4GHz, and 2.66GHz all yield around 110% or 55% user % in both the Macbook Air and Macbook Pro. There is practically no difference whatsoever. The only difference is that at 1.6GHz on the Air and 1.99GHz on the Pro, there is a lot less heat. 10C less, to be more precise. The Air hits 55C while the Pro hits 45C.

I think GarageBand is simply trying to optimize itself with whatever resource it has at hand rather than pushing the envelope.

nebulos
Oct 1, 2011, 06:02 PM
... it's actually quite true that Cinebench does benefit from more RAM bandwidth. Check this:

http://www.bit-tech.net/hardware/memory/2011/01/11/the-best-memory-for-sandy-bridge/8

thanks. i'll look at this.


Well, now that I see your test methodology, I think it's worth pointing out that under Activity Monitor, there is a line that reads "User%", which is more accurate than taking the CPU % inside and divide by 200 or 400.

but, we have the reading from istat as well, which gives precisely (User%)/2.

?


Also, this test just shows to me that your Mac Mini is indeed throttling itself.

How do I know that? My Macbook Air 13" gets 55% or 110% average during that "test". At 1.60GHz.

Also I think it's worth pointing out that I repeated the test with Coolbook on a Macbook Pro as well. At 1.33GHz, GarageBand reports that there are too many effects to be played in real time, so it cuts off right there.

1.6GHz, 1.86GHz, 1.99GHz, 2.16GHz, 2.4GHz, and 2.66GHz all yield around 110% or 55% user % in both the Macbook Air and Macbook Pro. There is practically no difference whatsoever. The only difference is that at 1.6GHz on the Air and 1.99GHz on the Pro, there is a lot less heat. 10C less, to be more precise. The Air hits 55C while the Pro hits 45C.

I think GarageBand is simply trying to optimize itself with whatever resource it has at hand rather than pushing the envelope.

i'll have to think about this, but thanks for running the test and sharing.

i'm not at all shocked that the 1.33 MBP choked. however, i am quite confused that your older 1.6 13 MBA would do better than the 2.26 Mini, or even survive where the 1.33 MBP could not, and that all of your other machines did the same, while the machines i tested all scored differently. (EDIT: Sorry, I misinterpreted. See below.)

:confused:

bill-p
Oct 1, 2011, 06:05 PM
Well, it wasn't an older 1.6 MBA. I simply limited the maximum operating frequency of my 13" MBA at 1.6GHz using Coolbook.

On the 13" MBA, I tested two frequencies, at 1.6GHz and 1.86GHz. The Pro has the rest, starting at 1.3GHz to 1.99GHz to 2.16GHz to 2.4GHz and finally at its max 2.66GHz.

In both cases, however, I think it is worth noting that the MBA has an SSD and the Pro doesn't.

Macman45
Oct 1, 2011, 06:12 PM
Well, it wasn't an older 1.6 MBA. I simply limited the maximum operating frequency of my 13" MBA at 1.6GHz using Coolbook.

On the 13" MBA, I tested two frequencies, at 1.6GHz and 1.86GHz. The Pro has the rest, starting at 1.3GHz to 1.99GHz to 2.16GHz to 2.4GHz and finally at its max 2.66GHz.

In both cases, however, I think it is worth noting that the MBA has an SSD and the Pro doesn't.

Is everything when it comes to doing these type of benchmarking exercises but as you point out...Putting the MBA and it's SSD up against anything with a motorised HDD and there will only ever be one winner. SSD is bound to win. My MBA still amazes friends who moan that their MBP's won't perform that way.

You can't compare the two. Well...You can but the end result will alway's be the same. No moving parts= more speed.

nebulos
Oct 1, 2011, 06:17 PM
Well, it wasn't an older 1.6 MBA. I simply limited the maximum operating frequency of my 13" MBA at 1.6GHz using Coolbook.

On the 13" MBA, I tested two frequencies, at 1.6GHz and 1.86GHz. The Pro has the rest, starting at 1.3GHz to 1.99GHz to 2.16GHz to 2.4GHz and finally at its max 2.66GHz.

In both cases, however, I think it is worth noting that the MBA has an SSD and the Pro doesn't.

Okay. Sorry, obviously I was confused and jumped to the wrong conclusion. (I didn't think there was ever a 1.99GHz C2D!)

Why would my Mini throttle itself, but not your MBA?

Are you verifying your various underclocked CPU speeds when running the tests?

I don't think the SSD is super relevant, but, yes, the recorded audio loop is sitting on the disk, so perhaps it comes in somehow.

Do these results make sense to you?

----------

Is everything when it comes to doing these type of benchmarking exercises but as you point out...Putting the MBA and it's SSD up against anything with a motorised HDD and there will only ever be one winner. SSD is bound to win. My MBA still amazes friends who moan that their MBP's won't perform that way.

You can't compare the two. Well...You can but the end result will alway's be the same. No moving parts= more speed.

this depends on the task. some tasks are (more) CPU critical, others are (more) 'disk' critical.

just look at tests done on both, the stock 2011 MBAs (SSD) and the stock 2011 MBPs (HDD). obviously, they are not all better on the MBAs.

bill-p
Oct 1, 2011, 06:43 PM
It can throttle because running at maximum frequency bumps up voltage, and thus... heat.

I have seen that happen a lot with Core 2 Duo Macs. Even my 2010 Macbook Pro had that after a few months of purchase... until I installed Coolbook and reduced the voltage.

Take, for example, my 2010 MBP 13". At stock settings, it can draw 1.2v at 2.66GHz. If left alone, it automatically throttles itself to 1.6GHz at 1v. But using Coolbook, I can run 2.66GHz again with only 0.975v. Considering most Macbooks wouldn't be used for intensive tasks like encoding videos on a regular basis, I guess that's why Apple feel 1.6GHz was an okay compromise.

And I think the result makes sense if I were to look at the Core i results on your post. All of them hover around the same amount of CPU usage, which is 100% or so. Except for the 2011 MBP.

That is to say, we don't know for sure which frequency those Core i CPUs were running at. But they all fit between the 1.6GHz to 2.66GHz with or without TurboBoost. The 2011 MBP, on the other hand, can TurboBoost to above 3GHz, so if it was at that frequency, that would explain the low CPU usage.

nebulos
Oct 1, 2011, 07:35 PM
hmmm ....

nebulos
Oct 1, 2011, 08:49 PM
It can throttle because running at maximum frequency bumps up voltage, and thus... heat.

but my temp and fans were at 68C and 1500rpm. ?

Lock
Oct 1, 2011, 11:10 PM
OP, the biggest difference will be from the SSD, but the i5/7 CPUs are also a significant improvement over the c2ds. It must take a lot to impress bill-p, because there is definitely more than a "slight" improvement - I might apply that to a 5-10% improvement, but this is more like 25-100+% improvement depending on the c2d.

Despite that, I wouldn't bother upgrading yet - I would have waited for better graphics if I'd had the flexibility. The Air's SSD and the new CPUs mean that the GPU is now the weak link in these macs. Hopefully this will be addressed with the next update.

nebulos
Oct 2, 2011, 12:09 PM
Cinebench and Memory

it's actually quite true that Cinebench does benefit from more RAM bandwidth. Check this:

http://www.bit-tech.net/hardware/memory/2011/01/11/the-best-memory-for-sandy-bridge/8

Both Nehalem and Lynnfield can get a few hundreds more points depending on how much memory bandwidth they get. It seems small when considering that it's 12,000 points total, but... try that with about 2000 or 3000 and the scaling of a few hundreds will mean something.

i don't know if we're looking at the same numbers; i see 17000's for Nehalem and Lynnfield, with the speed mattering most to the Lynnfield. the results for the Sandy Bridge are strange, as the highest score happens with the slowest memory, and bounces in between.

in any case, the biggest difference is for the Lynnfield, which increases from 17189 @ 1.333MHz to 18098 @ 2.133MHz.

18098/17189 = 1.053, so we're talking about a 5% difference.

if we 'scale', we do so precisely by assuming this percentage stays the same, of course. so the relative difference is the same. the absolute difference will just get smaller with smaller scores though. (5% of 1,000 is much bigger than 5% of 200.)

anyways, it seems memory is doing something, but not too much.




My GarageBand Test

I'm interested in making sure I understand this test and the results that I and others get. So far, I'm very confused by your results. You've also raised questions about my results, or at least their interpretation. I'd like to settle this if possible.

i. Measuring CPU usage:

Well, now that I see your test methodology, I think
it's worth pointing out that under Activity Monitor, there is a line that reads "User%", which is more accurate than taking the CPU % inside and divide by 200 or 400.

Why? Because there is no way CPU scaling is that linear. 120% may actually mean that one core runs at 80% while the other runs at 40%, and that's perfectly normal. That doesn't mean only 60% of the CPU is used. We're dealing with 2 cores, not a single processor split in two.

As I pointed out above, though I did not use it originally, in my recent second run of the 2.26GHz Mini, i used the istat widget to monitor things like temperature and fan speed. this also gives a CPU reading, which coincides exactly with User%/2 (for the total two cores of the Mini).

if this doesn't satisfy you, please explain why.


ii. Throttling?

Also, this test just shows to me that your Mac Mini is indeed throttling itself. ... It can throttle because running at maximum frequency bumps up voltage, and thus... heat.

As I pointed out above, istat is telling us where my temp and fans were at, and these, as well as CPU readings, were quite stable during the test.

I think this is evidence that the Mini was, in fact, NOT throttling, no?



iii. Your Results

My Macbook Air 13" [2010 1.86GHz] gets 55% or 110% average during that "test". At 1.60GHz [running Coolbook].

So somehow my Macbook Air with a slower processor is actually faster than your Mac Mini? That doesn't seem right.

Also I think it's worth pointing out that I repeated the test with Coolbook on a Macbook Pro as well.

At 1.33GHz, GarageBand reports that there are too many effects to be played in real time, so it cuts off right there.

1.6GHz, 1.86GHz, 1.99GHz, 2.16GHz, 2.4GHz, and 2.66GHz all yield around 110% or 55% user % in both the Macbook Air and Macbook Pro. There is practically no difference whatsoever. The only difference is that at 1.6GHz on the Air and 1.99GHz on the Pro, there is a lot less heat. 10C less, to be more precise. The Air hits 55C while the Pro hits 45C.

Again, I don't understand your results.

First of all, it's a fact that audio effects like amp simulators are primarily CPU-intensive creatures.

My results are all different for different CPUs; the 2011 MBAs we tested all have SSDs and 4GB RAM. (I doubt disk has anything to do with this test, and Samsung vs Toshiba even less.)

Your results seem not to change when underclocking your CPUs (until you go too far). But also, your 2010 1.86GHz MBA does better than my Mini.

I think you've tried to explain this with throttling, but again, I don't see the Mini throttling. I think you tried to explain that your underclocked test being the same means GarageBand was working relative to the available resources, but then what about my results on the 2011 MBAs?

:confused::confused::confused: