After reading through the first version of this post, some have asked for CPU intensive tasks such as Handbrake. Additionally, the first post didn't measure battery in a perfectly controlled fashion so this time around one of the experiments focused on that.

For these tests, I once again prepared the MacBook Air 11" 2012 with raw data sources for stressing the CPU.  This included an 8gb raw DVD source files placed on the desktop to leverage the SSD's read speed as well as placing 40gb iPhoto library.  Once the MacBook Air 2012 11" i5 was all setup, I once again cloned it to other MacBook Air 2012 11" i7 in preparation for the tests.

The tests include (a) a single cpu iPhoto database rebuild, (b) a multi-cpu Handbrake encoding, (c) a real world battery test lasting three hours of continued use, and (d) a handbrake with the i7 MacBook while in docked under a 27" apple display in the closed clam shape mode.

Part1 - iPhoto database rebuild

I have a very large 40+ gigabyte library for iPhoto (photos only as I have video's elsewhere) and it was really really sluggish.  I initially though it was due to the hard drive not able to keep up with all the media, however after copying the iPhoto library to my MacBookAir i5 I found it shocking it was as slow and sluggish with the infamous spinning colorful beach ball like cursor.  Upon discovering how to repair the iPhoto library (something I recommend everyone apply using this blog I additionally wrote ) I started to time the various repair processes simultaneously on both machines. The first repair was rather quick, about 20 seconds or so, and entailed rebuilding the permissions (the first option in the iPhoto repair dialog).  

In most cases the i7 finished a few seconds quicker than the i5 -- except for one instance where it finished a few seconds later.  I was intrigued and investigated further only to discover a running process was running and temporarily taking up about 20% cpu.  After a bit of research, I discovered the running process was spotlight indexing parts of the drive.  Clearly this process would distort the accuracy of any experiments by a small margin, so I dug further and discovered how to stop spotlight indexing temporarily with a small command line executed through the terminal app.

sudo launchctl unload -w /System/Library/LaunchDaemons/com.apple.metadata.mds.plist

To start it up again after all your tests, you can issue the load command as follows:

sudo launchctl load -w /System/Library/LaunchDaemons/com.apple.metadata.mds.plist

I'm not sure if spotlight is resource intensive to the point one looses 10 or 15% battery, but I may consider having a quick script so I can activate and deactivate it conveniently if it indeed uses valuable battery.  With spotlight disabled, the i7 continuously and predictably completed the iPhoto permission repair routing a few seconds ahead of the i5.

Main Test - iPhoto Rebuild Thumbnails

This test was straight forward.  I simply started iPhoto in repair mode (holding Command + Option while launching iPhoto) and started them both while also starting the timer on my iPhone.

 Again, as I did in the previous blog's tests, I measured CPU temperature, fan speed, overall progress and the final finish times.  The repair database process seems to be a single threaded process as only one CPU fired up to max capacity throughout the test as shown in this cpu monitoring tool called Menu Meters.

Conclusion:

This appeared to be a single cpu process for an extended period.  The i7 was able to run at full capacity as the fans never reached maximum operating speed.  From the final numbers, it seems the i7 has a very significant speed boost over the i5 with single CPU operations.

Part2 - Handbrake Encoding

To perform the hand brake encoding, I updated Handbrake to the latest version and opened the raw DVD source residing on the desktop.  By putting the DVD source on the SSD we eliminate the slow IO of an external DVD drive and maximize CPU by feeding it raw DVD as fast As the SSD can provide it. Similarly to avoid I/O bottle necks, hand brake is configured to save the encoded DVD movie on the desktop as well.

Conclusion:

Unlike the previous test, this used all available cpus and hyper threads available.  The i7 did top at 99 degrees 6500rpm fan speed and remained there for the majority of the test.  From the final numbers, it seems the i7 has a very significant speed boost over the i5.

This is an amazing speed increase and for anyone considering using this machine as a main computer and do not mind the spool up the fans during heavy use, the i7 seems to be the way to go.

I have a large iPhoto library (pictures only) of about 40 gigabytes since I recently merged the various iPhoto libraries into one.  In doing so however, the library became unbearably slow to work with.  Anything I would do, including just a simple scroll through the photo library would cause the application to hang with the infamous color full beach-ball like cursor twirling.

I sincerely initially though this was to the slow I/O of traditional hard drives.  So with my new MacBook Air 2012, I decided to move my entire library to leverage the super fast SSD.  To my surprise, the performance of iPhoto was identical.  The MacBook Air is even an upgraded system with extra ram & storage, so I knew it couldn't be the disk I/O.

My search for the solution lead me to various people describing a special "diagnostic screen" of iPhoto shown below which opens when you launch iPhoto with the Command + Option keys pressed.

Since the last option below indicates to try only if the above options do not resolve the matter, I decided to start each option top down.  The first option ran rather quickly and didn't improve the performance.  The second option however did take a very long time to complete.  Even on an high performance SSD drive of a MacBook Air, it took a good hour to complete. I imagine it would take over four hours to complete with a regular hard drive and a similarly sized iPhoto library.  After completion, I can confirm, at least with my experience, iPhoto ran more smoothly with very few occurrence of the "BeachBall" cursor appearing.   Since the 2nd option provided so much improvement, I decided to apply the 3rd option to my library as well.  So far I went from a practically unusable  library to a library I was able to show to some co-workers who happening to be considering a Mac purchase.

The question many are asking is:

• If I choose the i7 will I have a noisy fan system having trouble keeping cool?
• If I choose the i7 over the i5, will I have significantly lower batter life?
• If I choose the i7 over the i5, will it be significantly hotter to touch than the i5?
• How much more powerful is the i7 and is it worth any concessions in any of the above?

 

Pre-Test Stats

Before starting the tests, I let the machines idle for a long while and then checked the stats with iStat Pro.  The results are:

MacBook Air 11" 2012 i5	MacBook Air 11" 2012 i7
Temps: 36 - HD 40 - CPU 28 - Enclosure Base 28 - Enclosure Base 2 28 - Enclosure Base 3 30 - Heatsink B 36 - Mem Bank A1 37 - Mem Controller   Fan: 2004 rpm	Temps: 37 - HD 42 - CPU 29 - Enclosure Base 29 - Enclosure Base 2 30 - Enclosure Base 3 31 - Heatsink B 39 - Mem Bank A1 38 - Mem Controller   Fan: 2001 rpm

Test #1 - FAN SPEED & NOISE

 

The results:

Outside Temperature of MacBooks:

Conclusion, the i7 gets more work done faster at a small cost in battery.  Both machines did the same work, the i5 finished with 65% battery left while the i7 finished with 60% battery left. Under stressful load, I consider this a very good feat given the performance gains achieved.  These stressful scenarios are not the norm and instead you may be working casually more often than not so I expect the difference in battery to be even lower during normal use.  As a matter of personal opinion, I didn't like seeing the i7 pegged at 91 degrees under normal room temperature while the i5 was progressively being reduced from a peak of 91 degrees down to 74 degrees near the end of the encoding.

Test #3 - Battery Drain Light Load

For the light load tests, I simply created an HTML page with an auto refresh every 10 seconds loading USAToday web site.  I turned off the brightness on both monitors to isolate the drain to the CPU as much as possible and left them alone for a few hours.

The Results:

I had to go to bed and before wrapping up; tentatively there is very little difference (1% or 2 down to 75% -- with the i7 obviously consuming a bit more).  Prior to going to bed, I started to use both macs side by side performing the same operations.  Launching a VM, browsing, refreshing, performing browser speed tests, etc. and I was able to bring the MacBooks down to a difference of 5% (i5 had 55% and i7had 50% battery left). I will have some additional updates on normal and light loads in the coming days.

Overall Conclusion

The overall conclusion is very subjective and different for everyone.  If you have a thunderbolt display and/or plan to use the MacBook air as your primary machine with large SSD drive, then i7 may be the way to go. If on the other hand the MacBook air is your secondary computer and you have a performance oriented iMac at home or office, then other factors such as battery and/or comfort should have higher significance.  Below are my summary conclusions based on my findings.

FAN Noise - Those wishing to have as silent of a computer as possible in most circumstances should opt for the i5.  Keep in mind however, I've been using the i7 all day today -- plugged into a 27" display and have yet to hear it spool up.  In order words, just as the test showed, under specific load, the i7 will spin up the fans within a few minutes while the i5 will take a little longer to do so -- however you need to produce load on a sustained basis for that to occur.  If you are dead set on having minimal fans then i5 gives that, but be forewarned the i5 will spool up eventually (8-12 minutes later rather than 2 minutes).

Surface & bottom chassis temperature -- both MacBooks are only off by a degree or two.  It's noticeable only if you are comparing one with the other.  I wouldn't be able to discern which MacBook I was holding in my hand under load otherwise.  If you are only concerned about temperature, there is negligible difference and should ignore this factor and move on to the next important factor.  Despite this, do read my comment on CPU temperature below under other ambients.

Battery -- Under heavy load, the i7 drew about 15% more battery (5% difference in battery level at the 60-65% level).  Given the performance results indicated a 30% improvement with i7 over i5 (25min left vs. 17min left at the 17 minute mark), the bang for the buck seems to be there.  Note, this was a heavy IO test using the optimized SSD which we all know is significantly faster than the previous model, so true performance difference between machines, as per standard benchmarks, is somewhere 15% mark.  If you do edit video or have heavy IO tasks, using say the 512MB SSD, then certainly the i7 gets you a noticeable improvement and should opt for it.  For those seeking maximum battery life, the i5 will give you some extra juice -- approximately 10-15% if used under load continuously.  Under very light load, the batter difference seems to be a negligible < 5%.  So under normal load, you can probably expect 7-10% extra battery life on the i5 vs the i7.

CPU Temp -- I did notice the CPU temperature for the i7 was pegged at 91-92 degrees under load at room temperature.  The i5 was actually lowering the CPU temperature while slowly increasing the fan speed.  Because of this, if you are working in environments where the temperature may be ambient, say out side patio on an afternoon, you may find the i7 is not able to keep the CPU cool causing the system overall to become hotter -- again only under load.  The i5 would also have the same effects, but to a lesser extent.

With Mountain Lion's imminent release any day now, according to a close friend, I'm happy to report there is evidence Mountain Lion has been optimized for improved battery life -- particularly noticeable on MacBook Air machines.   Previous to Mountain Lion, MacBook Air 11" noticeably drained battery quite quickly.  With Mountain Lion, it seems the battery life is much more respectable.  If I had to guess, I would go on a limb and say up to 25% more.

Charging would be necessary by 12noon under normal usage and it seems it lasts through early afternoon.  We'll have to see in the next few days if indeed this is a common finding upon early adopters of Mountain Lion.

UPDATE 8/7/2012 In lieu of people having battery problem with Mountain Lion: 

I didn't realize this, however as I initially started using OSX Mountain Lion I do recall making a change because my MacBook Air battery would train really fast while in sleep mode closed.  Meaning, I would close it at night with say 90% only to wake up in the morning with 75%. It seems by default the memory remains powered unnecessarily after going to sleep.  I'm not sure if this is a necessary setting for HardDisk based MacBooks, but for MacBook Air's this is certainly not the case as the computer should save an image of memory to the SSD quickly (e.g. hibernation) and then power down the system components.

I changed my power management setting manually to ensure it hibernates while powering down and ever since I have no longer had battery issues with my MacBook Air.

First, check your current power settings by running this command:

pmset -g custom

And you should see something like this:

There two sections "Battery Power" & "AC Power" along with each of the various settings for each section.  If you notice, hibernatemode for the battery section is currently 25.  This is the setting you want to have for your MacBook Air. Ever since setting this to 25, I've woken up with 100% remaining after leaving the MacBook Air to sleep for many hours.

If you look into the details of pmset command, you will find the three recommended settings as shown:

Clearly, us MacBook people, particularly the MacBook Air people or those with SSD's as storage, would want to have setting 25 and it seems it is only configurable manually via pmset.  Here is the magic command.  Run this with the command line and you should no longer have problems with battery.

sudo pmset -b standbydelay 900 standby 1 hibernatemode 25

The stand by delay of 900 seconds simply means it keeps the memory powered for 900 seconds (15 minutes) in case you return to your MacBook quickly.  This will give the instant on impression.  After 900 seconds, the memory is powered down and the machine must restore the memory from hibernation which simply means it takes an extra second or two to power up.  If you like this convenience, you may want to increase/decrease the 900 value by however many seconds makes sense for you.

Imagine if Microsoft, a similarly comparable large company like Apple, would all of a sudden gain 54% in profit. Would its stock tank 5 or 6% because they didn't sell as many licenses of Windows as economists had projected?  No, on the contrary it will rise like 10% on such profound news for typically lethargically large companies of their size.  The funny thing, this is not a one time event, this is a continual event every quarter for sometime now and will continue for a few more as the world settles into the iPad as they did with the iPhone and of course the App Store, the only place you can legitimately buy apps for it, will continue growing too.   The 4th quarter is going to be that much more of a smashing success with iPhone projections as the 3rd quarter wasn't because the sales of iPhones, which dropped in 3rd quarter due to lack of product launch in June, will fall over to 4th quarter from the iPhone 4S product launch.  So of course, when comparing Q3 2011 to Q3 2010, there is inevitably going to be some differences -- especially with the successful iPhone4 being the growth factor then.

To blatantly compare Q3 2011 with Q4 2010 in gross iPhone sales and fuss over the unsatisfied expectations has got to be the most misconstrued news I have ever heard.  This only reinforces Peter Lynch investment style of investing in what you know and you'll beat the analysts every time.  No matter which way you slice it, AAPL has 81 billion in the bank accruing at 6.5 billion per quarter, no debt, 54% increase in profit and product portfolios which are no where near becoming stagnant.  New product revenue lines like App Store, iPad, & strongly increasing revenue from existing product lines like iPhone, MacBook, iTunes (#1 digital music store) -- it's a 100% no brainer this stock will continue to push upwards for the near future.  Any drop should be considered a discounted buying opportunity.  

Oh, and this is before the announcement of their touch based iMacs which will lay slightly inclined on your desk similar to the iPad, still with a slide out keyboard for keyboard dependent apps , and any other new upcoming product line.

I'll write again when the price is at 500 and I'll reassess the situation then to see if we can expect more.  Buying now will give you roughly a 25% return from 400 to 500 within 12 months.

UPDATE 11/28/2011

Apple Stock today closed at 376.12.  It's an extreme bargain.  I just read various articles which simply justify my gut feeling.  One reported iPhone4S sales so large in Britain, it became the #1 in sales, even beating Andriod -- a feat many people thought was impossible given the diversity of Android devices at various price points.  Another article, states teens have listed iPhone & iPad with the same level of desire as cash and clothes on their holiday wish lists.  Yet, another, from a financial perspective, compares Apple stellar performance in comparison to it's current stock price.  Every financial metric used to value the stock have it as extremely undervalued.  Anyone buying at today's price will reap an easy 30% gain on the way up to 500.  Lets just hope the financial crisis doesn't bring a general recession/depression -- the only thing which could drag down the stock.

If you are a Mobile user, particularly one with the new thunderbolt or display port connections, you may be considering the Apple Thunderbolt Display or former Apple Cinema Display.  Since the iMac 27" are also usable as display via the nifty target display mode feature, this is an alternative I wanted to additionally consider.  After various forums and research I came to the following observations which I'm currently experimenting with to configure the my ultimate docking station for my MacBook Air 11".

First, lets start by laying down some facts as ultimate the facts which are more pertinent to your situation will have higher weight in deciding which direction to choose for your display needs.  Below are some pros / cons when considering the use of each: 27" cinema display, 27" thunderbolt display, iMac 27" 2010 display port model, iMac 27" 2011 thunderbolt model.

• DisplayPort has wider compatibility with a somewhat simplified spec when compared to thunderbolt
• The Cinema Display & iMac 27" (2010 model w/ displayport), can be connected to any display port compatible device or digital display capable of adapting to displayport
• The thunderbolt display would provide expandability with the ethernet, firewire, & additional USB ports (USB 2.0 though)
• Running Windows in Bootcamp currently will not work in target display mode iMacs using thunderbolt, does however in cinema and thunderbolt displays
• When using target display mode, the ports on the back of the targeted iMac are not transferred to the connecting MacBook like they are in the dedicated displays
• an iMac serving primarily as a display for a MacBook would have a double duty of being a workstation for whomever needs it while the MacBook is not docked
• the typically more powerful iMac is available to be used for high end duties such as video editing or gaming

So, interestingly enough though, no such hub existed at the time, however Belkin recently showcased their upcoming Thunderbolt Dock, with the same ports as the Thunderbolt display, making such a configuration a possibility.  Since thunderbolts can drive display port monitors, using such a hub/dock with a mini displayport display should have straight forward compatibility.

The next hurdle would be the accessibility of connected devices.  No one wants to disconnect a device from one computer to then reconnect it to another -- let alone multiple devices.  The Cinema Display and Thunderbolt display both redirect the devices to the connecting MacBook via their USB & thunderbolt connections respective, however a connected iMac in target display mode doesn't.  

So far I was leaning towards the one of the stand alone displays, however if a USB hub switcher existed where I can switch a device from one computer to another at a push of a button, this would allow for what I would call the ultimate docking station for a Macbook.  So with this new device, in theory, I can leave my iMac 27", say, rendering a video off an external USB data drive while I can be working on the MacBook using another USB drive.  Or a more typical scenario, I can temporarily redirect the time machine drive to the MacBook, let it do is incremental back, and then redirect it back to the iMac -- all with a single push of a button and without interrupting other USB devices who can be connected to either computer - WOW!.  After searching forums after forum, I found one small niche company who makes such an adaptor.  The device allows for 4 USB devices to be connected to 2 computers, mutually exclusive to each other and one computer at a time.

I bought this USB device for a little less than $50 and was off to try this new setup.  Although the configuration is not complex, I decided to diagram a possible configuration.  It's not too clear, but in essence:

• The iMac & Air are connected via displayport cable.
• Both computers are connected to hub
• Port A of hub has time machine drive
• Port B of hub has my main data drive for video, etc.
• Port C has a simple USB hub where I attached my keyboard, mouse, has an extra vertical port for plugging in a flash drive from time to time.

With this configuration, I can easily move time machine from one machine to the other with a push of a button.  Or similarly, my main data drive can switch from one to the other.

One benefit over the external displays is the ability to plug in external speakers to the iMac 27.  These speakers are permanently connected to the iMac and work as any normal set of speakers with the addition of working with the MacbookAir when in target display mode.  The current cinema & thunderbolt displays do not have this plug and as a result you need to plug in the speakers to the MacBook every time you dock if you want to use them.

In summary, I have what I consider the ultimate MacBook docking station.  As soon as the Belkin thunderbolt hub/dock arrives, this will change a bit, however it will be just as good or better than the thunderbolt display.

I was intrigued at how fast the battery drained while running bootcamp on my MacBook Air 11" (2011 model).  There are various blogs discussing how Windows is not SSD friendly, particularly Vista, however I felt this would have been resolved with the latest version of Windows and/or latest service packs.

After various trials and errors, I achieved reasonable battery life by turning off Windows Search which has indexing running in the background.  Simply turn off the service called "Windows Search" and change it's startup to either "Manual" or "Disabled".