Hacker News new | past | comments | ask | show | jobs | submit login
Watt time is left (tedunangst.com)
181 points by ingve on Dec 16, 2016 | hide | past | favorite | 161 comments



Apple should have added a Amp or Watt usage number, not removed the estimate. Similarly, most new cars have fuel estimates on their dashboards. But its not just miles to empty, its also miles per gallon. So drivers can get feedback on usage rates versus total capacity.

I imagine the argument about this, is that most people don't have any handle on what a Watt or mAh is. But people don't have a good handle on how far 25 miles is, or how much a gallon is in comparison to their tank. Think about it, you rarely see the fuel you put into your car beyond a few drips at the pump. And while you may know that its 25 miles between your house in work, if I took a random point on a map and said find something 25 miles away, with no scale, most people couldn't do it. Gallons and MPG are just arbitrary numbers that consumers have learned how to compare. If you exposed those numbers, they would get a better handling on what their device consumes, and how their activities affect it.


They still have the percentage indicator, which is presumably directly proportional to mAh remaining, and people generally have a good understanding of what percentage is.

Using your analogy, your gas gauge doesn't show the number of gallons, it shows a relative level.


His point is that Apple should show the energy consumption rate in addition to the percentage indicator, just as cars show fuel consumption as mpg and fuel remaining as a percentage.


My car tells me how many miles I have left by basing its estimation on my consumption over the past 30 miles. It has been pretty accurate.


I think you can liken the situation with macOS to switching from a gas car to a Hybrid. Gas cars have relatively predictable MPG. Hybrids don't. Your Hybrid can go from 25MPG up to 100+MPG depending on the way you drive, and it can bounce back and forth between the two constantly during a trip. Now, I have to admit, I don't own a car at all and so I don't know if hybrid cars actually try to estimate mileage remaining, but if they do, then surely their estimate will change wildly as you change how you drive (e.g. smooth highway driving to traffic jam to city driving to aggressive driving to defensive driving, etc).


I own a Prius. Fuel efficiency doesn't seem to be any more variable than the conventional cars I've driven. The distance remaining estimate seems reasonably accurate in that it goes down ~50 miles after I've driven 50 miles.


Whenever I drive a Prius with ZipCar, the MPG numbers change wildly while I drive, and even the average MPG can change by 5–10 miles over the course of a decent drive. I assume the distance remaining estimate is based on that average MPG number, but since I have seen it change by significant amounts, that means the distance remaining estimate certainly is reasonably inaccurate.

If you always drive your car in the same manner then sure, you'll probably have a reasonable estimate. But similarly if all you do with your computer is browse the web, you'll also have a predictable battery remaining estimate. But while many people do drive fairly consistently, your usage of your computer probably has a lot more variety in terms of energy impact.


That's why you average it over some time, not just take the momentary usage.


Older gauges just showed percentages, but many new cars will tell you gallons, current MPG, and an estimated distance to empty.


How far back do we have to go to get to cars that just showed percentages, cause I have literally never experienced one in my life.

Even my model year 2000 clunker has instant MPG, average MPG, miles to empty and a gas gauge with actual gallons.


My 2011 only has the gas gauge, without any gallon markers on it and numerous 2016/2017 cars I've been test driving haven't had it. It's certain classes of cars that have had it for quite a while, nowhere close to all.


My 2007 Toyota Matrix just has an analog fuel gauge that goes from Full to Empty with marks for each quarter. No gallons, no MPG calculations.


My 2003 VW Golf has a gas gauge that shows percentages - the individual markings represent 12.5%, not individual remaining gallons (it's a 14.5 gallon tank, not an 8 gallon one) My mother's 2009 Jetta has similar. Whether or not the instrument cluster implements a multi-function display that shows instant/average MPG or miles to empty, every car I have ever driven, from the 80s clunkers of my youth to brand shiny new rentals has had such a fuel gauge.

Such a style is cheaper to implement. Resistive float sensor in the tank, run a wire to a mini 90 degree needle gauge in the instrument cluster, done. Instant/Average MPG and miles to empty are additional fanciness that up until recently, came standard only on higher-end cars, or was part of an optional package that may or may not be avail on lower end brands/models.


I thought that the gallons calculation used the same cheap mechanism, just adding a trivial percentage left * capacity calculation.


My 1999 Camry has a dial that shows full, 3/4, 1/2, 1/4 and a red line.


2009 Civic LX - gas gauge with 20 squares, no mpg at all, no actual gallon number.


The only car I've ever had that showed MPGs was a 1990 Audi.

Everything else (1998 Toyota Camry, 2002 Chevy Impala, 1997 Ford Ranger, 2005 VW GTI) have only had an analog fuel gauge.


Not far. My last car was a Toyota Yaris 2001, and the display (center top) only has percentage, and it's not very fine-grained: http://i.imgur.com/ESXap6F.jpg


2005 Ford Focus, just a gauge. It has percentage fuel on the OBDII interface but not MPG estimates. Some cars might have it on the OBDII if they don't have a gauge.


You have to go back pretty far. In the wake of the 1973 Oil Crisis, most new cars had MPG displays.


The 2012 Mazda 6 doesn't show MPG, but it does calculate it and expose it over ODB II


I'm guessing Apple don't want their customers worrying about the way their computer habits affect battery usage. They sell premium hardware that "just works". Reminding the user they need to keep an eye on what they do with the device seems incompatible with that mandate.


> its also miles per gallon. So drivers can get feedback on usage rates versus total capacity.

I don't know anyone who actually adjusts their driving to try to push the mpg metric up. I'm sure there is some tiny subset of the population that does this but in general this isn't actually a useful number except to make people annoyed that they aren't getting the EPA estimated fuel efficiency.

Car manufacturers don't provide this because it's actually useful. They provide it because it's easy and people like bells and whistles.

> if I took a random point on a map and said find something 25 miles away, with no scale, most people couldn't do it.

I don't think this is a particularly meaningful observation since maps have scales precisely because it's basically impossible to know the distance between two points without a scale. This isn't a function of not intuitively grasping what a mile it. It's a function of maps being scaled. Even if you have an amazing grasp on distances it won't directly translate to the map.


raises hand

I'm not a hardcore hypermiler, but just doing little tricks (minimizing throttle/braking, "reading" traffic, maintaining an appropriate speed/distance relative to other cars, etc) makes quite a bit of difference in the commute. I have a small (tiny?) turbocharged car and can get 40-42 mpg on my commute between Denver and Boulder.


I get ~18 MPG in my Impreza, and even though my fuel mileage sucks I still take steps to improve my fuel mileage. Just because my commute is short and I'm flooring the car down the on-ramp to the highway doesn't mean I don't want to take steps to improve my mileage within the boundaries of how I want to drive my car. For instance, is it better to be in second or third gear when in slow but flowing traffic?

In the same sense, I'd like to know how my computer usage is affecting my battery life. Do the extra tabs I have open in chrome matter, or should I bookmark them and close them until I need them? Just because I value one thing over another, doesn't mean the less valued option doesn't come in to play when making a decision.


> For instance, is it better to be in second or third gear when in slow but flowing traffic?

It doesn't really matter because your gas mileage is likely dominated by things like flooring it down on-ramps.

But 3rd is more efficient in the case you describe.


I know my gas mileage is dominated by flooring it down on ramps, but my mileage is still affected by a lot of other factors. I track basically everything I do with my car, along with mileage, and small things make a surprising difference. I can detect a change in my fuel consumption based on what tires I have on my car, how old my oil is, whether my roof rack is on my car, and whether I have my bike mount on my rack.

Those effects are noticeable on my normal commute, where I travel 8 miles and spend most of the time going less than 30 mph, where aerodynamics doesn't matter that much. When I take a road trip (and thus don't have to perform any suicide merges), the difference any of these things makes is quite substantial. My summer tires are each about an inch wider than the tires I have mounted on my old rims, and switching to the narrower tires makes more of a difference in mileage than taking off the roof rack. With the narrow tires and no roof rack, I can get about 36 miles per gallon at 55 mph, and with my winter tires and my roof rack on my car I can't even approach 30 MPG.

For a tech-related example: I prefer to use my phone with the backlight turned all the way up, even though it uses an additional 15% of every full charge, but I uninstalled the Facebook app because it used 3% of every battery just running in the background.

I'm just trying to be efficient in my own way, I'm not competing on efficiency.


Turbo or non-turbo Impreza? I had a WRX (turbo) for nearly 10 years and I don't think I ever got below 25. I'm surprised at 18 - I would consider that really low. I don't know your scenario, but I would guess that you could do much better than that with just a few changes. I've never had to floor it on an on-ramp, but I haven't driven everywhere in the US.


I assume you have a manual transmission? Which engine do you have?

A huge part of why my mileage is terrible is because I basically have the worst possible commute for fuel mileage. live in Chicago, which supposedly has the worst traffic congestion in the country. My commute is only about 8-10 miles, but there is a lot of variation in speed. I can go from stopped to 65-75 miles an hour back down to stopped several times on my commute. The commute takes about 30 minutes, and I spend a lot of time idling.

Obviously that is a ridiculously low mileage for a car, even given the conditions. My fuel mileage never dips below 18 mpg for an entire tank, but it rarely goes above 20 mpg unless I leave the city. Looking at similar vehicles on Fuelly, I don't think it's that far out of the norm.

My Impreza is a 2010 OBS, with the 2.5L NA EJ25, and a 4EAT. I think what really kills my mileage is the 4-speed automatic. From what I've read, it's absolutely brutal on fuel efficiency compared to the 5 speed manual. The NA uses a lot more fuel than the turbo when just puttering along as a result of its higher compression (10.0:1 vs. 8.2:1) It's not necessarily a negative - my car is faster off the line than the turbo, especially after sitting in traffic on a hot day.

I know the WRX has a DOHC engine, whereas the 2.5i is SOHC. That probably makes some difference in efficiency, though I don't know how much.

On top of that, I almost always have my roof rack on if I'm driving in the city, and I usually have a bike rack mounted as well. Plus, I have the hatchback, which I've heard also impacts mileage, though I have no idea if that's true.

I've also replaced the somewhat narrow stock-all seasons with wider and softer tires summer tires. I run high sidewall winter tires on steel rims in the winter, which weigh a lot more than the stock alloys/all seasons.

Regardless, the Impreza really sucks at city mileage compared to other cars because of the added weight and drivetrain losses of the AWD. (As you probably know) It's really no surprise the 2.5i and 4EAT have been replaced by a much smaller engine coupled to a CVT.


Yes, I had a manual transmission. I assumed your comments regarding "second or third gear" implied you had a manual transmission. Automatics have come a long way in the past few years in terms of fuel mileage, but "slushboxes" like yours are probably the worst. I didn't know the Subaru automatics were that bad. I do know the new 2.0 with a CVT yields good results. It sounds like your driving pattern forces you to be in the worst fuel mileage ever (lots of time idling plus several bursts up to 75? ouch!). I assume you've explored other routes for your commute. Even something where you are driving slowly but not stopped/highway speeds should give a marked improvement.

The hatchback probably doesn't impact your fuel mileage much. I don't think the difference between DOHC and SOHC is going to have any measurable impact on fuel mileage either. Nor would I assume that a compression ratio change does that much. Your final drive ratio is probably the most significant impact. Additionally how much time you spend in "lockup" mode (where the fluid clutch is essentially disengaged and the engine crankshaft is directly connected to the driveline) is a major contributor. The roof rack absolutely impacts fuel mileage as Consumer Reports demonstrated: http://www.consumerreports.org/cro/news/2013/07/tests-show-b...


I use the manumatic function a lot in my 4EAT, because it always seems like it's in the wrong gear otherwise. I test drove a 5 speed and I hated it, cause the gearing seemed horrible for the speed limits which are really common around me.

Anyways, I take the highway to work, and I take a short stint on a highway and then surface streets on my way home. On my way home I run in to something like 40 stop signs. Pretty much every intersection has a four way stop for a few miles of my commute.

I can actually quantify how much the roof rack affects mileage, because I track whether I have it on my car or not in addition to my mileage.


On my way home I run in to something like 40 stop signs. Pretty much every intersection has a four way stop for a few miles of my commute.

That just screams "hybrid" to me. Wouldn't a hybrid or EV be the perfect vehicle for that scenario?


Does the MPG indicator actually enable anything? You can do all the things you describe without it. It's hard for me to believe that there's anyone independently discovering from their car's MPG indicator that minimizing throttle/brake usage improves mileage.


MPG indicator gives real-time feedback about what I'm doing. I could do everything without it (and did in the past) but "calculating efficiency by doing math at fill time" isn't granular enough. I don't think I could get that extra 2 or 3 MPG.


shrug None of the things you mention require constant feedback to understand. Heavy throttle followed by braking is just turning gasoline into brake wear. You know that already.


In Germany, demonstrating that you drive in a way that conserves fuel is part of the driving test. E.g. if you fail to shift to a higher gear when appropriate or shift down to brake that can be part of the reason you fail the test.


The modern Bluemotion Golfs I have driven seem to want you to drive in 6th at 30mph. This combination also results in an annoying drone coming from the engine which I can't stand. You can't accelerate in that gear. Also, I don't take instructions from an arrow on a dashboard telling me what gear to be in. That's my judgement to make based on road and traffic conditions.

The "shift up" recommendations to me seem like an attempt to replicate the ideal conditions under which the official efficiency tests are done - and we all know about /those/ figures...


Well, you're supposed to shift down before accelerating.


6th to 2nd is a heck of a downshift. It's hard for me to imagine driving around at 30 in 6th, too. The engine drone would be pretty obnoxious.


30 mph (almost 50 km/h) usually works very well on 4th in a 6-gear box, including accelerating, especially TSI engines which have a superb torque already on very low RPMs.


This in no way indicates the usefulness of the MPG indicator that most cars have. You can drive efficiently (or inefficiently) without it.


You can make assumptions about what's probably more efficient, but you can't confirm them without knowing how much gas you've consumed, and that takes several hours to justify stopping to refill the tank. Empty/full gas gauges have a pretty bad margin of error for this, compared to the ECU knowing how much gas has been pumped from the tank and/or injected into the cylinders.


This is kind of absurd. This is like saying you can't know that running sprints burns calories unless you have real-time feedback by hooking yourself up to a oxygen consumption meter.

You don't have to measure for yourself at all. The things you can do to save gas are well understood. If you generally just don't trust others' fuel efficiency guidance, I'm not sure why you'd trust someone else's mpg estimate either.


Is the air conditioner more efficient than suffering drag from open windows? Each has been commonly recommended at different times, and the correct answer depends on your compressor and body shape. The optimal cruising speed might be anywhere from 40 to 60 MPH, depending on your engine torque, transmission, and tires...


So you're trying to decide if you're better off with the A/C on or the windows down and you trust the short-term estimated MPG over the measurement you could do yourself? Fill up the car and turn on the A/C, drive an hour. Refill the car and record how much gas you put in. Turn off the A/C, roll down the windows, drive another hours. Refill and record. Now you can answer for yourself.

I get the logical appeal of the immediate feedback MPG meter, but I'm not clear how accurate these things are (vs meters on the gas pump, which are regulated to be within a tight spec). I'm also not clear how useful the MPG right now is vs the MPG over the past hour. I've seen only one car that let me reset the MPG meter and measure for an arbitrary period of time that I control, and I still have no idea how accurate it was. (Yes, I'm sure there is more than one car with this feature. I've only seen it once.)

Again, I just generally don't buy that instant (actually averaged over some amount of time that you don't control, with some accuracy that is likely not guaranteed) MPG estimates are all that useful. Are you dedicated enough to efficiency that you'll drive without A/C in the summer to see if it's better but you somehow don't care enough to bother recording actual gas usage? Who are these people willing to drive without A/C to save a MPG but too lazy to confirm that it really works? Or realistically is this one of those things where you turn off the A/C, glance at the MPG estimate and see it went up from 20 to 22, and assume that you've got a definitive answer despite not knowing anything about the margin of error for the MPG estimate?


It's… complicated.

On one hand, instantaneous mpg is seriously misleading. To me, accelerating efficiently means maximizing the amount of acceleration I get per unit of fuel—doing that does NOT maximize instantaneous mpg, but ideally finds the optimal point on the fuel-consumption-rate*time graph that you actually care about.

On the other hand, real-time feedback is probably better for the not-exactly-scientifically-rigorous way people learn, and it can be useful for noticing unexpected behavior not covered by intentional experiments.

A realtime unit-less fuel efficiency gauge in a U-Haul clued me into non-ideal behavior from the automatic transmission climbing hills. Feedback from this dumb gauge caused me to adjust my driving, which meant only my first tank of gas lasted way less time than expected.


Fair enough. You U-Haul example is reasonably compelling. I can see that being a real use case.


A large percentage of the population uses cruise control, probably without realizing that it's the single easiest thing you can do to increase your (freeway) mpg.

It's also apparently pretty common for hybrid-car drivers to adjust driving patterns to improve mpg, partly because the car gives rapid feedback - gamification in real life! - but there's probably also some selection bias, as people who buy high-mpg vehicles are more likely to be concerned about fuel efficiency.


Apple should have added a Amp or Watt usage number...

I want Activity Monitor to show how much battery each process is using.


It does. It's the "Energy" tab. It doesn't give you amperage or anything, but it tells you the "Energy Impact", which is a relative measurement of the current energy usage, and "Avg Energy Impact" which is the average for the app. Similarly, if you click on the battery menu, it will call out any processes that are using "significant" energy.


My bad. Thanks.


I started working on a little CLI battery status indicator even before I knew Apple was going to pull the time remaining feature. On macOS I like to hide the dock and menu bar but usually keep terminal open with tmux running so I decided to write something I could place in my tmux status line where I could keep an eye on my battery life.

https://github.com/imwally/battstat


I've had a python{2,3} script for, like, years, collecting advanced details using ioreg.

Sample output:

    ~> ~/Workspace/toolbelt/battery 
    Temperature = 2967
    CycleCount = 384
    DesignCapacity = 6600
    MaxCapacity = 6362
    WearRatio = 96%
    CurrentCapacity = 5886
    ChargeRatio = 92%
    Voltage = 12162
    Amperage = -1061
    InstantAmperage = 64286
    InstantTimeToEmpty = 8:34
    TimeRemaining = 5:33
    AvgTimeToEmpty = 5:33
    FullToEmptyTime = 5:59
https://github.com/lloeki/toolbelt/blob/master/battery


This is awesome. Thanks


This might work well with bitbar. https://getbitbar.com


I haven't upgraded to 10.12 yet, but this seems like a good time to remind people about coconutBattery [1].

You can't see a prediction in minutes yet (maybe they'll add one with Apple removing it), but the battery metrics have always been more useful than for me than what's built-in. You can see the discharge rate, current charge, full capacity, and original capacity though.

[1]: http://www.coconut-flavour.com/coconutbattery/


Thank you! It has "time until empty" indicator (%r) http://imgur.com/a/74qLJ


Oh, cool. I hadn't seen the menu bar option before.


I haven't used a Mac in years, but back when I had one, I did use Coconut Battery to track the degradation of my battery over time. It's a great little application.


If the core problem is supposed to be that power usage is now more erratic, how about this small refinement:

- Compute discharge rates for one minute segments, like Ted did in this post

- Store the last 10 discharge rates

- Report a time estimate based on the minimum/median/maximum of those 10 rates.

That gives fairly quick feedback after you start doing something expensive, but isolates the effect of short-lived and infrequent bursts.


It's not as if erratic power usage is a new thing, anyway. My 2013 MPB's battery life can easily vary by an order of magnitude. It'll last 10+ hours if it's sitting there not doing much, or maybe an hour with an intense 3D game. The estimate they provided up to 10.12.1 was still quite good if I was doing the same thing for a while, which I usually am.


I wrote a battery meter (BatteryBar) 8 years ago that uses a more complex algorithm to calculate an "average time remaining" rather than instantaneous. Basically, my software keeps track of how long the battery has lasted in the past and uses that to inform the current time remaining (with adjustments based on current discharge rates, etc).

I'm not sure why Microsoft/Apple haven't done something similar. I've even sold licenses to Surface Pro sales reps that they use to show off the battery life of the Surface to potential enterprise clients.


Is there a background for this? I don't understand the article.


Apple removes the ‘time remaining’ battery estimate in new macOS update

http://www.theverge.com/2016/12/13/13939278/apple-macos-sier...


Well that's simply the silliest way to try and solve the battery-life issues of the new laptops.


According to Apple, the new laptops don't have a battery issue, but that the battery estimate is misleading. For example, people were reporting that macOS would show 3 hours remaining, but then the laptop would actually last 6 hours.

No idea why this came up at the same time as the hardware launch. macOS has had various background sync tasks for a while — iCloud sync, Photos, Time Machine, etc. Sierra adds some background services (specifically, "Optimized Storage") that might have a bigger effect.


That's not what they're doing. They're recognizing that the estimate has always been misleading.


Quite an amazing coincidence that they recognized this right after they released a new line of computers whose battery life failed to live up to expectations.

And quite annoying that they're "recognizing" this, when it's often quite useful and not at all misleading. I made good use of that display before, and now they've taken it away.


I also use this estimate even if it's not always right. FWIW it's still accessible in the Activity Monitor energy page.

Personally I've now disabled Apple's battery menu extra and replaced it with the one from iStat Menus because that still has a time estimate.

It's an $18 utility, but it's very well designed and has a clock menu extra that's similarly more useful than the one Apple gives you (calendar and multiple time zones in the dropdown).

iStat battery screenshot: http://i.imgur.com/bi95Fs9.jpg


And so Apple managed to make $6 off of you by removing a feature. Great.


a) I already owned it

b) It's not sold through the App Store anyway

Take your pick.


I agree; on my 2015 MBP the time remaining is extremely accurate. Probably because my behavior on the machine is very consistent.


>Quite an amazing coincidence that they recognized this right after they released a new line of computers whose battery life failed to live up to expectations.

Or was reported to have "failed to live up to expectations", by idiots who merely assumed the estimate and didn't bother to do an actual clock wall check.


So I'm being punished for the actions of some idiots I never met? That's not better.


No, I doubt they changed it to punish Mike Ash specifically.

Rather they decided to cut the fake precision (X hours YY minutes) and stick to the segmented-battery icon display.

It's like some statistics were saying there's a 45.32% chance of something, just because that's what their simplistic division of the samples gives them, despite the confidence margin being plus/minus 5 percentage points.

Instead, they now tell you it's "between 40% and 50%".

The target audience of the statistics is not "punished because some idiots took 45.32% literally". They are just shown a better value.


They could have changed it to "about X hours" if precision was the problem.


More correct but less useful isn't actually better. An estimate of time remaining is more meaningful than a percentage of battery capacity remaining, even if the latter can be provided with more accuracy.


>More correct but less useful isn't actually better.

I'd argue that more correct is by definition more useful.


Correct but inactionable is useless. It's basically undeniably correct that our universe will eventually die of entropic heat death. This isn't useful, though.

Projections about Earth's global warming are undoubtedly less correct but far more useful because they are actionable.

Correctness is not synonymous with usefulness.


And in pretty much every consumer product that you use. Ethnography is a thing.


There were many reports of tested bad battery life.


I've quite often used it to check if I can finish the movie I'm watching for example or if I have enough battery to finish writing something during the flight and so on. I'd say especially for the movie case, it's surprisingly accurate.


Didn't they have a similar reaction to the antenna issues on iPhone 4 (5?) back in the dizzy? They'd show five bars when truly, it was more like one or two? I forget the details, but this seems similar to me.


Yeah, it was the iPhone 4. Previously on their software the signal meter was tied to some specific RSSI dBm, but that phone had a poor antenna design and dropped calls at higher perceived signal. So they responded by adjusting the meter to better reflect call quality instead. You can enable raw RSSI numbers in the field test menu still.

Also related around that time: Apple pushed an update for AT&T that changed their HSPA indicator from 3G to 4G without any upgrade haven taken place.


Just trying to think back that far makes me dizzy.


The estimate has always (at least, based on my 2010 Snow Leopard MBP) been pretty good at telling you how long your laptop will last if you keep doing exactly what you're doing now.

Same as any other kind of estimate (yes, even the legendarily-terrible Windows file copy dialog). It's not misleading, you just have to understand what the numbers mean.


Uh huh...


So is there any good (hopefully free/open) third-party solution to bring it back to the menu bar on the mac?


I've been using Battery Time Remaining for a while and it has some useful features that even the old indicator didn't have (for instance, which apps are draining the battery the most).

https://itunes.apple.com/au/app/battery-time-remaining/id551...


Looks nice.

It does feel painful to give Apple 30% of $0.99 (US) in this instance.


You can also get it for free (and the source) on Github.. https://github.com/codler/Battery-Time-Remaining

PS. i especially like the notifications you can set at 100%, 15%, 5% etc battery life. very handy!


There are binaries here: (I haven't tested them.)

http://yap.nu/battery-time-remaining/


Battery Health from Mac App Store is very nice, shows a lot of information:

http://a3.mzstatic.com/us/r30/Purple20/v4/db/ec/d0/dbecd0c9-...


This looks questionable.

https://itunes.apple.com/us/app/battery-health-monitor-batte...

Not a single of the word "menu" in the description page, even. And the UI is kinda..all-over-the-place, to be honest.


Sure, iStat Menus is great.


Seconded on iStat Menus. Despite its cost, it shows a ton of useful statistics, including moving graphs of CPU activity, network IO, storage IO, etc.. It's easy to customize the display so you see just what you care about.


It's also $18 and closed-source. You might have wanted to disclose that since I mentioned free/open. Just saying.


Years ago (probably 2009) I wrote a small utility which polls the SMC to get current and voltage readings and from hardware. Wrapped it up in a simple app which shows realtime power consumption in my menu bar.

Didn't do much with it since the SMC keys differ for pretty much every piece of hardware Apple create and there is practically no information so it requires some investigation for each.


Can anyone on the latest update of macOS confirm whether the time prediction is also gone from Activity Monitor > Energy at the bottom?


The estimate is still there, you can also get if by running `pmset -g batt`.


I don't think it's fair to suggest Apple removed the battery life indicator because they don't know how to make one.


It's a joke. Of course they know how to make one. They had one in 10.12.1.


> sysctl -A | ag watt

Gives nothing. Why it's not shown? rMBP late 2013, macOS 10.12.2.


The author is using a Thinkpad running OpenBSD. I don't think macos or linux kernels expose battery info as kernel state.

Also, thanks for using ag. :)


> I don't think macos or linux kernels expose battery info as kernel state.

Bummer.

> Also, thanks for using ag. :)

Thank you for making it ;)


They removed it to cut down on complaints. Most people wont get the advertised 10 hours, and now most people wont notice.


Or:

Most people will get the advertised 10 hours, and now people won't think they're not because of an inaccurate estimate.


I guess it depends. Seems like people are getting more that 10 hours out of it when used for light tasks.


197.45836985100788781720 is grossly superfluous precision is it not?


The point is that those estimates aren't actually accurate. That they mislead the end user so they don't belong in primary UI.

This article is nice to show that you can get the data from the command line (you can still get it elsewhere in macOS too) but the snark at Apple for saying they can't do it when it's in fact a deliberate design choice just makes the author look uninformed & petty.


Oh come on now, you can't be serious.

The fluctuation, if it exists at all, has real meaning in how people are managing their system.

It also gives an indication to the health of battery

They could have showed a range, or a smoother average, or a prediction based on previous use.

They could have a graphical interface illustrating what programs are causing the fluctuation. Many ways to improve something that's already perfectly adequate.

A "people are stupid, let's tell them less" approach is one of the ones making software so profoundly terrible and increasingly less useful.


And this is why you don't design UI for major companies.

Yes, the data has value to some users: which is why it's still available.

It's completely misleading in the general case however which is why it's gone from the general UI


A personal attack?!

Being a protector of the hypothetically imbecile user is why some software has been rushing full speed to lowest common denominator design.

This toxic culture strives to corral users and remove features.

These are the same projects which seem to be getting slower and buggier with each release. It's making me absolutely abhor the industry I work in and I'm ashamed of such workmanship.

I don't know when it became so fashionable to try to excel at making terrible software.


I will actually take this. If you are talking about useless information for most users what about the battery percentage? What does that 45% gives a layman user? They basically will try to estimate in their head how long until they need to charge and you are saying its more user friendly for a user to guess what a percentage (or Wh) means than for the OS with the actual information of the current & historical powerdraw to estimate and help the user?

I thought computers were suppose to this sort silly computations for you.


It was optional and off by default. Why wasn't that good enough?


Presumably this was their thinking: either the number is accurate, or it isn't. If it is accurate, then we can have a discussion about whether or not we should show it by default. In this case, you would probably decide to show it (or give the option to show or hide it), because it's a number that's highly valuable to the end user.

However, if the number IS NOT accurate, then don't even give the user the option to see it. It's just going to be misleading.


The problem with this thinking is that "accurate" is not a binary property.


Is the percentage then accurate? Hardly.


Are you implying that major companies can never do wrong? I don't care whether UI is designed by big or small company, usability is what matter and what we discuss here.


They were sufficiently accurate enough to be useful for a couple of decades. Are we supposed to believe that the release of 10.12.2 marks some watershed moment where they suddenly became useless?

No, it's just BS. Apple doesn't want people paying attention to it anymore, so they removed a useful feature just to avoid complaints.

The "Apple no longer knows how" bit is obviously a joke. I agree with petty, but it doesn't make them look misinformed.


They were pretty terrible all through those decades. They're second only to Windows "time remaining" dialogs as a joke.

That's a terrible argument.


Not at all. They assumed a consistent workload, but given that assumption, they were quite good. Programming in a coffee shop, if the estimate said three hours, I'd get about three hours. If I fired up a game and the estimate dropped to 40 minutes, I'd get about 40 minutes. It failed if I took the estimate for one and applied it to the other, but, well, don't do that.

I'm getting rather annoyed at people telling me the estimate was useless anyway. I used it!


> I'm getting rather annoyed at people telling me the estimate was useless anyway. I used it!

Indeed, but I think the argument is more: at best it was a current workload time indicator. I'm still on 10.11 but think I might just make a battery power indicator of my own. Been wanting a better one for a while anyway.


Well, that's my argument. A current workload time indicator is really handy to have.


Agreed, just trying to apply the principle of kindness to their arguments.

What I want for a battery indicator is more a dual battery indicator/workload indicator. Aka, lets say some process goes nuts for 10 seconds, I'd want some sort of "hey this thing started up over here and is using energy like a fiend" alert. But if I were on power then I'd say only do that after 30 seconds or whatever. That and also keep historic power use and tie that into current use and times and it would seem like you could have some really stupid ML/linear regression to figure out: the user normally runs a compile every 4pm, and its 4pm and he's not on power so that means it'll likely use XMah which means based on the power available he's got about... N minutes of power remaining.

Still seems a stupidly annoying problem to solve with so many variables.


>They were sufficiently accurate enough to be useful for a couple of decades.

Citation needed.

>Are we supposed to believe that the release of 10.12.2 marks some watershed moment where they suddenly became useless?

No, it just marks the moment when they were removed. They had been useless for some time now with modern power-saving processors.


Citation: about two decades of personal experience using portable Macs on battery power.

The estimate is quite useful on my 2013 MacBook Pro, which I'm pretty sure uses a modern power-saving CPU. Correction: it was quite useful. Now it's gone.


Was it actually useful? I think I can manage the battery on my tablet and iPhone just as well as I can on my laptop. Neither the tablet or iPhone estimate how long the battery will last, and somehow I do just fine. That tells me that the time estimate is just noise that doesn't actually contribute to the utility of the machine.

I'm OK with removing it since it was never really accurate for me. The time estimates did give me anxiety from time to time. What's interesting is since removing the battery percentage label from my phone, I have less battery anxiety there and it "seems" to last longer since I'm less concerned about the actual percentage remaining.


Of course it was useful. Why would I be complaining about its removal if it wasn't useful to me?

I don't do just fine with my iPhone. I have no real idea how long it'll last. I know that if it's nearly full, it'll last "a while" but I can't judge with any accuracy when I'll need to charge it. My approach is to just plug it in at every reasonable opportunity, which ensures it doesn't quit before the end of the day.

I don't fault Apple for that. My iPhone rarely sees a consistent workload, so I don't see how any sort of useful battery estimate could be done there.

But my computer is a different story. I usually do the same thing with it for hours at a time. Under those conditions, the estimate was both accurate and useful.


>Of course it was useful. Why would I be complaining about its removal if it wasn't useful to me?

Because people make cargo cult attachments all the time?


This is just rude. Reading these threads a lot of people have made very specific, good arguments as to why this feature was useful to them _for years_. No matter what you think of that feature, you can't just condescend people because you are not accepting reasonable and logical arguments. I too found this feature useful for _years_. Please be respectful of other peoples' opinions. Surely you can tell a lot of people use this feature for _a_ reason.


>This is just rude.

It's also true. "It was useful to me" doesn't settle the matter of whether it was actually useful or not.

>No matter what you think of that feature, you can't just condescend people because you are not accepting reasonable and logical arguments.

"It was useful for me" is not a logical argument.

A logical argument would start by checking whether the time estimate matched the remaining time under different usage scenarios.


I already covered that:

"But my computer is a different story. I usually do the same thing with it for hours at a time. Under those conditions, the estimate was both accurate and useful."

Can we now move past the part where we assume I'm an idiot, and get on with the conversation?


>I usually do the same thing with it for hours at a time. Under those conditions, the estimate was both accurate and useful.

Why, in the world, would this matter, in a world where people do all kinds of different things with their computers?

In what world does it makes sense to keep a display only for people who "usually do the same thing with it for hours at a time"?


Work is usually consistent. If you're working in Xcode or Mail or Excel for long periods, the number will be accurate.

I guess Apple thinks nobody uses their computers for work anymore? And you think that's reasonable?


Right, and that must be what's going on here, as opposed to all the Apple defenders who clearly are doing it on an entirely rational basis.


Or, you know, we've been burned by BS "2 hours remaining" only to have the laptop re-adjust that behind our back and disrupt our work way before.


I never had that happen unless there was some major change in the work my computer was doing, in which case I'm grateful to know about the change.

That's probably what I dislike most about losing this UI tidbit, actually. Number of hours to empty is good, but knowing that my energy usage has silently gone through the roof is really handy.

We're both just talking about our personal experiences, of course. But your position requires nobody to have had my experience, whereas my position is compatible with people having had your experience. I fully acknowledge that the estimate was not perfect and wasn't useful for everybody. But it was for me.


>That's probably what I dislike most about losing this UI tidbit, actually. Number of hours to empty is good, but knowing that my energy usage has silently gone through the roof is really handy.

There's a really handy Activity Monitor window then, with energy usage details over time.


>Correction: it was quite useful. Now it's gone.

Don't worry, there's all kind of aftermarket crap menubar apps to put it back in.


Citation: I repaired Apple products from the G3 iBooks up to G5 pro towers. Battery life indicator was even in OS9


How are they not accurate? I'm watching a movie (a fairly common thing to do, I'd say) - the movie has 45 minutes left. I check the battery indicator if I can finish it or do I need to go get the charger. If it's telling me I have more than 50 minutes left I don't get the charger.

So far for me, it has worked every single time that I can remember.


It doesn't matter if they aren't perfectly accurate. Neither is the battery "percentage"; it's a constructed number that usually has an unintuitive relationship with the actual physical state of the battery. The point is that it's a useful number, as it answers the question "approximately how much longer can I keep doing what I'm doing right now?".


This. I never expected the battery meter to be perfectly accurate, just to give me a rough idea of how much time I have left before I need to start worrying about looking for a power outlet. I find this more useful than a percentage.


>The point is that it's a useful number, as it answers the question "approximately how much longer can I keep doing what I'm doing right now?".

If it's not accurate, as you admit, then it doesn't need to be shown with precision it doesn't actually have (X hours YY minutes). The segmented battery icon still shows the same information in a more fuzzy way that fits the accuracy (or lack thereof) better.


No, the percentage tells me nothing about how long I can expect to keep doing what I'm doing. I haven't memorized the power consumption profiles of every task I use my laptop for.


>No, the percentage tells me nothing about how long I can expect to keep doing what I'm doing.

At least it misleads you LESS than the time estimate.


Go ahead and round it to nearest 15 or 30 minutes. But showing raw battery charge doesn't have the same info about whether my workload is going to drain that in 2 hours or 5 hours unless I pull out a stopwatch to help teach my computer to count.


Percentage is pretty darn accurate actually. Mostly because it's based on remaining charge not fluctuations in load like time.

You can measure this yourself with any lipoly batteries you have at home. It's quite reliable


> Percentage is pretty darn accurate actually. Mostly because it's based on remaining charge

This is incorrect on both counts. It's based on remaining available energy, which is actually quite hard to calculate.

There are two ways you can estimate the remaining energy in a battery under load, and neither of them are very accurate. You can measure the voltage, the relationship between which and remaining energy depends on a huge number of factors including effective internal resistance (which changes with battery age, temperature, and across batches), particulars of the chosen chemistry, physical profile of the cells, current draw, etc. This is probably what you're talking about when you suggest measuring this "at home". It's not what any decent quality laptops use. It works OK at equilibrium though.

The other approach is dead reckoning. This involves estimating the discharge profile of the battery (which isn't as simple as "how much charge" the battery holds) and then integrating a non-linear function of the measured current draw based on the discharge profile. This works OK, but you need to combine it with occasional voltage-based recalibration to avoid integration error drift. That usually happens during charging.

Either way, not as accurate as you think, and you have to make assumptions about future current draw. If you give me a AA battery, it's literally impossible for me to tell you how much energy you're going to get out of it. It can easily vary by a factor of 20 depending on how you're going to use it.

Once you have a rough estimate of the remaining energy (which, again, is different from remaining charge), you can use that to formulate a percentage. But for human factors reasons, the displayed percentage almost never correspond to the actual remaining safe power. Most products stretch out percentages near 100% and 0% to give an impression of longer battery life and to give the user plenty of warning to find a charger.


Percentage is a dynamically constructed estimate too. It's more accurate because the underlying variables tend to change more slowly and predictably, but the actual charge levels corresponding to "0%" and "100%" aren't fixed, and the calculations do sometimes drift noticeably from reality.


Incidentally, this is why you'll sometimes see an improvement in battery life if you do a full discharge. It's not that it's good for the battery itself (quite the opposite, in fact), but it gives the estimator a better view of how the battery is behaving, so it can better estimate the charge level at any given point.


>Mostly because it's based on remaining charge not fluctuations in load like time.

It should display "remaining charge" then, not absolute hours and minutes.


Most people aren't educated enough to know what math operations to do in order to convert and get a time estimate.


Most people don't understand that "1 hour remaining" will actually be "20 minutes remaining" if they start some heavy gaming either...


I find the time estimates quite useful. They're not 100% accurate, but they give a good idea of "If I keep doing whatever task I'm doing right now, I have X hours left".


I think the apple bashing was sarcasm and your comment is sillier than the article. It wasn't the primary UI, the primary UI just shows the percentage. Its useful for a lot of user to know some expected runtime given their current usage and now that is inaccessible for a whole lot of regular user. The only thing they needed to make it not confusing to the user was to change 'Time remaining' word to 'Estimated time remaining'.

The only thing apple's change did was remove a useful feature from users and maybe make it harder to notice battery performance regression (not that I'm sure if there was any with the latest macbook).


The battery indicator menu in the main menu bar isn't primary UI? That's a novel argument.


Its not what people look at first glance. You get the battery percentage as default. You get this when you actually want an estimate. The wrong thing apple did was calling it 'Time remaining' and not calling it in an estimate.

If Microsoft or some Linux distro did this for user complaint I believe you would have been hard pressed to find anyone defending it. I'm typing this on a Macbook pro and while I think this is the best device for me for a lot of reason there so many idiotic Apple decisions out there that I have seen excused away that I'm frankly tired of it.


Teds aren't very accurate either, haha

Time remaining... 364 Time remaining... 327 - 37 Time remaining... 326 - 1 Time remaining... 361 + 35

Ted, i suggest you remove it too, and not mislead your users :D


or maybe the author has a sense of humor which you lack




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: