New Benchmarks Show 13-Inch MacBook Air Rivaling Its MacBook Pro Counterpart

[justfine]

But only because they cannot or are not willing to spend the money on a SSD.

If you see the Air as a cheap way to get an SSD (in a 13" Apple laptop), you are right. By foregoing processor speed and RAM you can get a 128 GB SSD $250 cheaper and 256 GB SSD $400 cheaper (though getting a third-party SSD will narrow that gap). Even if you equalise the RAM (for $100), you still come out cheaper.

The biggest problem of the Air from a performance to price ratio however is that you are stuck with one internal disk (making the 64 or 128 GB SSD combined with a 500+ GB HDD impossible, something which is close to a sweet spot performance per price wise).

Everyone is missing the point of and the market for MBAs. They are not for numbers crunching, video editing, serious Photoshopping or frankly anyone needing 500G of storage. If you need to watch the same movies over and over and want to keep them onboard then this is not for you at all. But with the SSD and 4G RAM I can run Aperture which has been slow on my first-gen (first week actually) 2008 MBA as well as Office, if I must; iLife, iWork, and a bunch of non-SuperGeek apps that "most" people need and want. This is in no way a criticism of those folks who need far, far more computing power. It's merely a reminder that everyone's needs/tastes are different and change/evolve over time.

The first computer I bought for my then small business was a DEC PDP-10, one of the first "Mini Computers". In 1981 $100K (yes, a hundred thousand dollars) bought a "distributed processing" small refrigerator sized machine with a removable 20MB (yes, that's not a typo) HDD, 3 Monochrome CRT "work stations" with built-in keyboards and modified off-the-shelf software for my type of business. Today i could do 10,000 times better for a tenth the cost. My first PC, a NEC notebook running DOS with a thrown in free copy of MS Works cost $1500 and included a 20MB HDD (yup, same as that DEC $100k job), 640K RAM and a monochromatic screen of maybe 12". I have absolutely no idea of the processor.

Today, retired and happy, my needs have changed. From a 15" Powerbook G4 to a MBA first gen 13" to the newest 11" MBA with a seemingly slower processor than my first Powerbook, but 4G (from 1G) RAM and a 128G SSD. And for less money. And, and, and. Incredible how far we've come in such a short amount of time.

 

[twoodcc]

well i'm glad to see this. next the macbooks will get a speed bump also, early next year probably

 

[lilo777]

well i'm glad to see this. next the macbooks will get a speed bump also, early next year probably

Where will they get faster C2D chips?

 

[nutjob]

Where will they get faster C2D chips?

They won't. They'll have to do an extensive redesign with discrete graphics which means saving space elsewhere. I don't think there'll be a bump on the MBP 13" for quite a while, maybe second half next year.

 

[KnightWRX]

Just weighing in here for those that think the 11 inch can "do it all". When it comes to driving a 24 inch (likely even more dramatic in the new 27 inch display), the answer is "uh, no....not exactly" and certainly not anywhere near as sharp as the 15 inch MBP's using a 330M card.

I realize this is somewhat obvious, but wanted to share a real world view.

SB

Delete your post, it is full of misinformation. My 9400 can drive the 27" cinema display with no problems, a 320M won't even sweat it. You had a configuration issue, your display was probably set as mirrored and thus you weren't pushing the full resolution out to the external monitor.

Change your display settings to extended desktop or whatever it's called so that both monitors are independant and both can use their native resolutions.

I realize this is somewhat obvious, but wanted to share a real world view.

 

[posterchild]

Love my new Air

After a few days, I love my Air. It doesn't render video fast, but for everyday stuff I do on my couch, it zips right through it. It's the computer that bridges my phone and my tower.
My eyes needed a break from my iPhone anyway

 

[wizard]

Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_1 like Mac OS X; en-us) AppleWebKit/532.9 (KHTML, like Gecko) Version/4.0.5 Mobile/8B117 Safari/6531.22.7)

Where will they get faster C2D chips?

They won't. They'll have to do an extensive redesign with discrete graphics which means saving space elsewhere. I don't think there'll be a bump on the MBP 13" for quite a while, maybe second half next year.

Hopefully the next refresh to the MBP will have a whole new generation of processors, either AMD or Intel. Hopefully late spring of next year! The potential is there for a rather large increase in performance.

I'm also hoping that Apple gets on track with SSD performance. As nice as the new machines and their SSDs are it was rather short sighted to interface the SSD via SATA which appears to be the case. Flash is already faster than SATA and notebook PCI Express slots have been around for a long time.

This brings up the question of just what signals that new card slot actually has going to it. I'm wondering if it is infact PCI-E with SATA lines tacked on. It would be really nice to see third party SSDs for Apples new slot.

Next year should be. Very interesting; new processors, new SSDs, new GPUs and hopefully new ports. Very much a new generation of MBPs.

 

[nutjob]

Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_1 like Mac OS X; en-us) AppleWebKit/532.9 (KHTML, like Gecko) Version/4.0.5 Mobile/8B117 Safari/6531.22.7)



Hopefully the next refresh to the MBP will have a whole new generation of processors, either AMD or Intel. Hopefully late spring of next year! The potential is there for a rather large increase in performance.

I'm also hoping that Apple gets on track with SSD performance. As nice as the new machines and their SSDs are it was rather short sighted to interface the SSD via SATA which appears to be the case. Flash is already faster than SATA and notebook PCI Express slots have been around for a long time.

This brings up the question of just what signals that new card slot actually has going to it. I'm wondering if it is infact PCI-E with SATA lines tacked on. It would be really nice to see third party SSDs for Apples new slot.

Next year should be. Very interesting; new processors, new SSDs, new GPUs and hopefully new ports. Very much a new generation of MBPs.

Don't hold your breath. The MBP has got tricky combination of space/heat issues, cost issues and battery issues. If they change processors they have to increase the size of the motherboard to improve the graphics performance which means less battery and more heat from the extra chip needed, which increases cost. I don't see them jumping to AMD chips, why would they bother upsetting Intel when they obviously get a good run from them? But anything's possible, especially if they can't get decent graphics any other way. They may wait until Intel comes out with decent integrated graphics... which might be a while. A step up to an i3 is only going to deliver about 10% increased performance. I don't think the interface is the limiting factor on the Apple SSD, their performance is below the latest SSDs on the market.

Either way I don't see Apple rushing into updating the internals of the MBP 13.

 

[revelated]

A lot of posts bemoaning people that are calling out these guys for not doing apples-to-apples. "it doesn't apply!!!" Guess what? It does.

You can't measure processor and memory performance alone for one reason: virtual memory. You could have the fastest RAM and the fastest processor, but a slow hard drive is going to nerf your results every time. This is because as applications and/or processes eat up the RAM, your machine is forced to page certain data to a virtual memory cache whose speed is entirely dependent on the speed of the drive. If you compare a 5400RPM drive to a SSD with equal RAM and equal processors, you will see the one with the SSD showing higher RAM performance, because the virtual memory performance would be superior.

So no, in my opinion, this benchmark is null and void. In order to feasibly test true performance of RAM, you have to include equal drive types. Processor depends on its cache size, mostly.

 

[manu chao]

This is because as applications and/or processes eat up the RAM, your machine is forced to page certain data to a virtual memory cache whose speed is entirely dependent on the speed of the drive. If you compare a 5400RPM drive to a SSD with equal RAM and equal processors, you will see the one with the SSD showing higher RAM performance, because the virtual memory performance would be superior.

Not if you have enough RAM, and all kind of RAM is still faster than the fastest SSD. Thus upping your RAM will give better performance than adding a SSD for the scenario described by you.

 

[nutjob]

A lot of posts bemoaning people that are calling out these guys for not doing apples-to-apples. "it doesn't apply!!!" Guess what? It does.

You can't measure processor and memory performance alone for one reason: virtual memory. You could have the fastest RAM and the fastest processor, but a slow hard drive is going to nerf your results every time. This is because as applications and/or processes eat up the RAM, your machine is forced to page certain data to a virtual memory cache whose speed is entirely dependent on the speed of the drive. If you compare a 5400RPM drive to a SSD with equal RAM and equal processors, you will see the one with the SSD showing higher RAM performance, because the virtual memory performance would be superior.

So no, in my opinion, this benchmark is null and void. In order to feasibly test true performance of RAM, you have to include equal drive types. Processor depends on its cache size, mostly.

Guess what? You're wrong!

Of course you can measure CPU and memory performance alone, you just avoid using virtual memory. But more generally any given benchmark that combines the performance of several components is meaningless. The question is how do you weight the performance of each component? You can't, it's different for each application. Only simple benchmarks like the various disk IO benchmarks, integer and floating point performance of the CPU, graphics benchmarks, each taken alone, have any meaning and can be usefullu compared between machines.

 

[mdude85]

Where is the 2.66 Ghz MBP?