Q: Battery life dropped considerably on Mountain Lion.

I posted in the early days of this thread.  But i wanted to chime in again because the AWFUL battery life is just KILLING me.  I never thought much about my MacBook's battery.  In fact, I simply never ran out--maybe a small handful of times since my first Unibody MBP back around 2008/2009.  I work in different school systems and this is my incredibly busy season and I just can't believe how bad the battery is.  The laptop gets smoking hot, fans blasting, and I'm seriously getting like two or three hours MAX battery life.  I don't know what to do.  Yes, I upgraded my SSD.  Yes I replaced my DVD with a 1TB HD.  Yes I run lots of apps.  But under Snow Leopard (oh, how I miss you!) battery life was seemingly endless with the SAME APPS and the SAME drive configuration.  I would go back to SL in a second but my addiction to iCloud makes that incredibly hard to swallow and all my clients are now on Lion or ML.  Sorry for the rant.  But in short, THIS *****.  I paid nearly $3,000 for this laptop less than a year ago, and this is what I get? Apple called me weeks ago for info on my situation.  Never heard anything after that.  10.8.1 makes no difference at all.  I called support and they pretended they had never heard of this issue.  What an insult.  I'm so disillusioned with Apple right now.

End of rant.

hi dansmacbook, rant away as far as I'm concerned - same rubbish from apple here in sydney- officially the apple line is that there is no problem with ML - its always a problem with something else, or some nonsense along the lines of 'teething problems with all upgrades" - off the record an apple geek said they're machine was stuffed with mountain lion -

it really is insulting - apple must think you and I, and everyone else with ML problems are morons - we've all had machines doing ok with snow leopard or lion, then with mountain lion drastic overheating, fans going mad, battery fading - the one thing in common is mountain lion -

the next most annoying thing is apple apologists who keep on saying their machines a fine  with ML (hooray, but this forum is for those of us who are having problems) or who kep on telling us that we're in that margin of error where things just don't work (gee thanks for that useful information but we've paid $$$$$ for machines and software that apple told us was the "best experience yet" - so even if we are in the margin of error - apple needs to 1. admit the problem and 2. a sorry for the hassle would be nice 3. fix the problem or 4. officially advise those of us with ML problems that we need to downgrade back to Lion or Snow Leopard if they don't have a fix very soon)

how hard is it for apple to do the right thing ? and why is apple leaving us in these pages to try and fix it ourselves - on the basis that apple staffers can read and write - why not post an official reply from apple addressing our problems, telling us when the battery fix is coming, of if it isn't coming soon, what we should do ? why is that so hard for apple ? either all of us  with a ML problem are making it all up, for the fun of it, or there is a real ML problem -

apple I know you're reading these pages - will you please post a reply telling all of us why you haven't lifted a finger to help us by advising officially when the battery fix is out, or if you don't know, then officially what your advice to us is  in the mean time - in case you haven't noticed, people are buying ML and reporting the same issues I wrote about 50 pages ago - these new ML users are all having the same hassles and very single one is reinventing the wheel over and over again -

like microsoft apple show every sign of  putting profits ahead of you dansmacbook, and me and everyone else whose machines have been stuffed thanks to mountain lion....or real hassle of trying to figure the problem out, then having to downgrade to Lion ...

rant over - but I challenge apple to do the right thing buy all of us - post a reply apple

Well I'm doing an experiment on my mid 2009 MBP. In looking at the "pmset" command, there's a description of the types of hibernate modes that can be set:

     hibernatemode = 0 (binary 0000) by default on supported desktops. The system will not back memory up to persistent storage. The system must wake from the contents of memory; the system will lose context on power loss. This is, historically, plain old sleep.

     hibernatemode = 3 (binary 0011) by default on supported portables. The system will store a copy of memory to persistent storage (the disk), and will power memory during sleep. The system will wake from memory, unless a power loss forces it to restore from disk image.

     hibernatemode = 25 (binary 0001 1001) is only settable via pmset. The system will store a copy of memory to persistent storage (the disk), and will remove power to memory. The system will restore from disk image. If you want "hibernation" - slower sleeps, slower wakes, and better battery life, you should use this setting.

By default, hibernate mode is 3 - keeps power to memory to quickly wake from sleep. I switched it to 25, and put it to sleep at 50% power. I'm going to see what it is in the morning. If there's little to no drain (as I am predicting), my guess would be ML has changed how it powers the memory - somehow giving more power to memory than Lion did. As a contrast, my older MBP that only has Lion on it barely loses power while asleep overnight, with the same (default of 3)  hibernate mode. Depending on what happens, I'll change it to mode 0 and leave it asleep until I get home from work.

Beisarius wrote:

 

Hey adams,

 

Your math is impeccable and indeed it will not be a linear graph... So you can choose the lesser of two evils- Lion.  If you are out of warranty - I think you are- I would wait 200 cycles provided you get acceptable hours (4 minimum). When that battery requires servicing, go at the Apple Store and have the battery replaced and used exclusively with Lion. If the new battery works fine, then keep it. If not, your choice to keep using at the Lion consumption rate, or return it for refund.

 

Are you in ML presently, what is your machine again, age? Any peripherals, mouse, blootooth? 4% overnight is indeed a lot for 2011/2012 units. How much time do you get when surfing? Unfortunately batteries are inside on these, so not too easy to open and install them outside Apple's store.

 

Next year, if Apple still uses Ivy Bridge, might buy the next OS- which would have been tested on it. If Intel changes architecture, and Apple follows, I will not dare install a new OS on the current hardware. I kinda like this 7-8 hours even if it will go down a bit over time.

 

 

 

Chris

Hey Chris,

Late 2010 MacBook Air, 1.6GHz, Core 2 Duo, 4GB RAM, 120GB HDD.  I'm back on 10.7.4 Lion now.  No peripherals and I did not reinstall any Apps.  Yep, it's out of warranty. 

With normal use, some web surfing Mail, etc., nothing heavy (not even photo editing), screen at 50% with WiFi on it gives me about three to maybe four hours now.  Used to be 6 hours or more.  Thought I might try a clean install of ML one more time just to see if it makes any difference.  It still bothers me that ML (or something associated with it) caused this and I can't get back to square one with a clean install of Lion though.

BTW, I open both Coconut Battery and Battery Health and take screenshots every time I start and stop my MBA.  I closed the lid at 9:56 a.m. today...battery was at 53.5%, 2:26.  I just opened it up (11:30 p.m. PST).  It's now at 38.5%, 1:48.  So in just over 12 hours in sleep mode the battery dropped a full 15%.  I close all apps and have "Wake on network WiFi access" disabled (doesn't have the "Nap" feature) and BT off but WiFi on when I close the lid.

15% during 12 hours of sleep is the worst I've seen so I opened Console and found line after line of activity between the time I closed the lid and opened it again.  Most are grouped by time.  Every hour, sometimes more often there is network and kernel activity.  I'm not sure what it's up to.  I'm too tired to review it now, but in the morning I'm going to see what's going on while it's supposed to be asleep.

I turned WiFi off, closed the lid and will see how things look in the morning.

Cheers. 

jt519 wrote:

 

Well I'm doing an experiment on my mid 2009 MBP. In looking at the "pmset" command, there's a description of the types of hibernate modes that can be set:

 

.....

Well my suspicions are confirmed. My "Lemon" MBP did not drain during the night when set to hibernate mode 25. In fact, I put it to sleep at 50%, and it woke up at 55%. So. That tells me that ML is definitely using too much power while asleep to keep the memory "refreshed" in order to wake up quickly. I'm going to try and set it to mode 0, and see what that does. Also, it has done nothing for the abnormal power drain while awake. It's already dropped 2% in about 2 minutes of typing this. So. My MBP "defective".. No.. ML "defective" More likely yes.

Hey Adams,

Just tested this one for sleep, so overnight it dropped 1 to 2 %. As for functionality, some things i noticed too. mozilla firefox takes about an hour off. Difference betweem Chrome and Safari much smaller. Web is also more demanding- flash so on.

Looking at how my performance stats changed right after 10.8.1, very reasonable to assume ML tweaked some settings. So at this point it is one of two things, battery or logicboard. If you did a full clean Lion install, my hunch is that were you to change your battery, you would perhaps jump back to 6. Reason I am saying that is I documented that reverting Lion still gave me poor battery life until I experiemented with replacing batteries. So if the ML algorythm, somehow, permanently affected your battery chip and calibration, that would explain why reverting has worked for some, not others. If you replace the battery (and you can try it, while recouping your current one), then the new one gives you 6, just found your culprit. if the new one still gives you 4, take it back and have your old one put in.

As the batteryuniversity site explains, if a battery is 'shocked' by sudden demands and drain- ie ML on your system, and it cycles a few times, then it may indeed be damaged and loose health. May say 93% but no longer perform at 93%. Check your drainage. What is your amperage in Power? i found it consistent across systems. if ML says it drains -900mAh, it is, -1200, it is, -2000, it is.

So if yours says -1100 range, I bet you it is your battery. it does not have 6 hours worth of juice even though the meter says 93%. that number is artificial.

On two older systems I also experimented with three generic batteries. Both ML and Lion just tore them up. I then placed in (after resets) brand new Apple batteries, manufactured last 6 months, and they have been flawless. I bet ML was also  optimized for  newer Apple brand batteries to reduce copyright issues. think the new iPhone 5 dock, speculated to arrive with a built in copyright shake.

So the question here is, how many people had the ML issue, did clean installs, resets + new batteries and still had the issue?

if it is battery, it still means hardware- be it lemon batteries. Ironically, first batteries were invented in Egypt using lemon acid...

Would just like to add mine to the list of 2010 Macbook Pros affected by the ML upgrade. Have yet to quantify but battery life is clearly shorter, with it lasting a couple of hours compared to an average of ~4 on Lion.

jt519, thanks for conducting a formal experiment which confirms ML as the problem versus defective batteries.  Nice Job!  I think this will put some people's minds at ease, as long as Apple can come out to address the issue before this ML issue adversely results in the battery health drop that some people are experiencing.  I hope your finding helps the Apple engineers track this one down.

Also, this disciplined approach sort of makes some other posters look silly, biased, or both, when they assert that ML is not the problem, especially when you consider that most people were not experiencing the problem before their ML update.

hey jt, interesting experiment.

As per my previous posts, 2008-2010 units are not necessarily defective. ML is not meant to run a 2009 as it is a 2011/2012 macbook. It is not meant to give same discharge as Snow Leopard. Yet same default ML settings on other macbooks- such as the two I tried in a row, and sleep drain is negligible. Just by the age of your machine, i would expect a default 20 to 30% extra drain due to chipset architecture difference and how tasks are optimized on ML for Sandy/Ivy Bridge.

By not enabling Power Nap, I evidently altered the hybernate/sleep settings- no doubt hence my 8-9 hrs longevity. But your 2009 has ML yet not running Power Nap as yours does not come with it. Do not believe your ML is defective, just your unit not compatible with ML. if ML was defective it could be recreated on most new machines, yet I cannot do it nor can most people. ML on older machines drains faster than a firehose, but I take that for OS demands on the hardware. So clear cut assumption on my part: you can't find 2008/2009 macbooks with ML that have SL type life.

10.6.8 is not the same thing as  10.6, 10.7.8 neither than 10.7, each of these variants modified substiantially these OSes. So ML 10.8.5 will not be quite the same as 10.8.0. I believe that trying to run ML on a 2008/2009/2010 with Lion type results will loose time and battery health on your part. Am also sure that were you to put new batteries on your 2009, with ML, results would be the same. If they are not, then you found your culprit.

Your best power management configuration is 2009+10.6.8 or 2009 + 10.7.8. You are welcome to prove me wrong. however, i would use new batteries if reverting to SL / L.

So your experiment, in my view, did not prove a bit ML is defective, but, like on my older machines, not at all optimized or meant to run on them as it does on newer macbooks- which I also concluded. I stick to my observations, ML runs as advertized on the vast vast majority of 2011/2012 units, and again, you are welcome to go buy one and prove me wrong and return it if found defective. Easy test. As we speak, Apple sells thousands of macbooks per minute, hundreds a second, and they are not failing at all. odd ones (like my faulty trackpad MBA), will still come along.

My percpetions were that ML was awful, but the stats I listed in previous posts told me that it cannot be so. Went ahead, ordered, two MBAs last 10 days, not a battery or ML performance glitch. Some users, whom have it, got it right away. But then again it is a minority of purchasers. So agree with your findings just not how you interpret your data.

Chris

I read in this page http://guides.macrumors.com/Apple_Notebook_Battery_FAQ that it is quite normal for a battery to drain 1% per hour while the mac is sleeping - which is what I am seeing on my 2012 MBP 13". So i'm not sure how you can get 1%  battery drain overnight Beisarius. Did you mean 1% for the entire night? Or 1% per hour on average?

I have also installed iStat Pro, and I'm seeing 94% battery health, 20 cycles. Not sure if this is normal though.

I have finally reverted back to Lion last weekend, after doing a clean install. Battery seems to work ok now. Uptime shows 19%, and battery level has dropped around 4% since this mbp was turned on. If my math is correct, that is equal to ~ 7hours.

hey neo,

Umm, quite simple: that forum is outdated and is not talking 2012 units. Getting 1-2 percent over 7-10 hrs sleep, 3% a day, and a projection is about 30 days sleep / standby time. Which is exactly what Apple claims. Battery Health equally lists it, see my pics from yesterday, bottom figure. The Standby 36: is days:h:min, after calibrations and usage will likely be 30. So if I shut the lid on day one, at 100%, it will take over 30 days to drain to zero. One of the reasons i waited and got a 2012 unit, the longevity.

2008 2009 with SL give me 1-2% per hour drain when in sleep, consistent with the post. hope I answered. Link is outdated and no bearing on 2012 macbooks, batteries and power management.

Hi Beisarius...

Can you point me to where Apple claims the 30 days standby? I couldnt find it in the MBP 13" product page...

Will be a good argument to get a replacement battery / unit if it turns out that I have a faulty unit. (as I have reverted to Lion, and it didnt help)

Thanks in advance...

Dont worry... found it...

http://www.apple.com/macbook-pro/specs/

thanks!

Beisarius wrote:

 

hey jt, interesting experiment.

 

As per my previous posts, 2008-2010 units are not necessarily defective. ML is not meant to run a 2009 as it is a 2011/2012 macbook. It is not meant to give same discharge as Snow Leopard. Yet same default ML settings on other macbooks- such as the two I tried in a row, and sleep drain is negligible. Just by the age of your machine, i would expect a default 20 to 30% extra drain due to chipset architecture difference and how tasks are optimized on ML for Sandy/Ivy Bridge.

 

...

 

 

So your experiment, in my view, did not prove a bit ML is defective, but, like on my older machines, not at all optimized or meant to run on them as it does on newer macbooks- which I also concluded. I stick to my observations, ML runs as advertized on the vast vast majority of 2011/2012 units, and again, you are welcome to go buy one and prove me wrong and return it if found defective. Easy test. As we speak, Apple sells thousands of macbooks per minute, hundreds a second, and they are not failing at all. odd ones (like my faulty trackpad MBA), will still come along.

 

Chris

Sorry, but I'm still not buying that. Sure ML may be more optimized for the newer hardware, but if Apple had an issue with it running on older hardware, they would not have let it run on older hardware.. case in point my older MBP. Since it doesn't have the graphics chips / drivers necessary for ML (which I still don't get.. there really isn't much more eye candy.. but lets not go there), I can't even install it. And that's fine. It's an older unit, it runs great under Lion. ML is advertised to have different/better/updated power management - go watch the WWDC session on Power management and see what they've added over the past OS's. I'm not surprised that Apple didn't quite fully test "just new enough" systems to make sure that power drain, especially while sleeping, isn't an issue.

Again, the issue of power draining while awake and being used - approx. 1% per minute, is very similar to what others are seeing. The fact that the CPU is not taxed, the fans are not running all-out, and it's not getting blistering hot, tells me that ML isn't taxing the system and sucking up juice. If the system, just sitting there, was doing all sorta of crazy drawing, fancy effects, etc. that would use CPU / GPU, then yeah, I could see why it would use more power. But it will drain while sitting at an empty Finder screen, with no apps running... and even at the log in screen, with no user logged in. All those fancy ML - power hungry features really aren't being used at that point.

The issue of sleep drain is completely different. I've shown that just by chaging the option to not keep power to the memory, the excessive drain is eliminated. In Lion, with the default sleep setting, there was similar negligible power drain while sleeping. While asleep, there isn't any "high performance ML" features running - it's asleep! That tells me that in ML, something was changed to alter the power profile while the system is asleep. I'm running another test now, using a different sleep setting (0, instead of 3 or 25) to see what difference, if any, that makes. If 0 keeps the the battery from running out overnight, I can live with the extra second or two of wake up time (gives me time to get a drink )

Continuing my experiment with the pmset command. After about 4.5 hours using the hibernate setting of 0 ("default on supported desktops. The system will not back memory up to persistent storage. The system must wake from the contents of memory; the system will lose context on power loss. This is, historically, plain old sleep."), I lost 8% power. Which makes sense. That mode does not explicitly say it will remove power from memory, like mode 25 does. I suppose I could leave it at mode 25 - that state will do a fill hibernate when the lid is closed, then will have to restore from disk upon waking. Not sure if I want to go that far, but if the drain keeps up while asleep, I may have to. Of course, just using it ***** power like crazy too. Oh well. Hopefully that helps someone.