1.1ghz, 1.2ghz, or 1.3ghz for battery life?

[Jetcat3]

Hey y'all! I've read a lot recently about Core M processors and was wondering what would be the best choice of these 3 for battery life. Here is an interesting article @ http://www.anandtech.com/show/9117/analyzing-intel-core-m-performance

I know the 1.1ghz processor is bumped from 0.9 to 1.1ghz and the TDP raises to 6 watts. The 1.3ghz processor on the other hand has a 5.25 watt TDP I believe. Any ideas what processor would give the best battery life? Thanks!

 

[Steve121178]

Doesn't really matter about the CPU. It's what you do that really matters.

For example, if you are going to be doing tasks that consume a lot of power such as watching videos or something that taxes the CPU then battery life will take a huge hit. If you have the screen brightness more than 50% it's going to take a hit on the battery. If you use Chrome then bye-bye battery.

I'd personally opt for the 1.2GHz version if I had to buy now as it's the best chip out of the three.

 

[usernames.taken]

I'd personally opt for the 1.2GHz version if I had to buy now as it's the best chip out of the three.

I know right, the 1.3Ghz is no where near is a good as the 1.2Ghz, why on earth would Intel even bother with making a faster 1.3Ghz, or Apple even offer it, it really makes no sense, when the 1.2Ghz chip is so much better.

 

[sunshine92]

I know right, the 1.3Ghz is no where near is a good as the 1.2Ghz, why on earth would Intel even bother with making a faster 1.3Ghz, or Apple even offer it, it really makes no sense, when the 1.2Ghz chip is so much better.

hey, @usernames.taken and @Steve121178 why is the 1.2ghz better than the 1.3ghz since 1.3ghz geekbench scores are much better? i have tried googling for an answer but i can't find it. i ordered a 1.3ghz 2 weeks ago but could still change to a 1.2ghz. I'm looking for something with the best batt life and for word processing. (i travel for work a lot)
thanks!

 

[Steve121178]

I know right, the 1.3Ghz is no where near is a good as the 1.2Ghz, why on earth would Intel even bother with making a faster 1.3Ghz, or Apple even offer it, it really makes no sense, when the 1.2Ghz chip is so much better.

The 1.2GHz chip is the best performance/efficiency balance out of the 3. The 1.3GHz chip is faster but less efficient.

It's confusing for some, but the 1.2GHz chip is the one to go for. Having said that, it's really not a massive upgrade over the standard chip so it might not be worth the extra cost for a lot of people.

 

[usernames.taken]

hey, @usernames.taken and @Steve121178 why is the 1.2ghz better than the 1.3ghz since 1.3ghz geekbench scores are much better?
thanks!

Sorry you couldn't tell from my emoticon, but I was being sarcastic.

There is no reason to believe the 1.3Ghz CPU would be less energy efficient than the 1.2Ghz CPU. You are speculating.

 

[bmustaf]

I don't think that's actually accurate, the 1.3GHz is binned higher than the 1.2 and the 1.1, it's actually more thermally efficient, look at its TDP numbers.

The 1.2GHz chip is the best performance/efficiency balance out of the 3. The 1.3GHz chip is faster but less efficient.

It's confusing for some, but the 1.2GHz chip is the one to go for. Having said that, it's really not a massive upgrade over the standard chip so it might not be worth the extra cost for a lot of people.

----------

The 1.3GHz is far and away the most thermally efficient of the processors available on the rMB.

I was surprised by this (pleasantly), given I had a 1.2 for the a week or so before my 1.3 showed up. In an exceedingly unscientific and uncontrolled manner, my 1.3 has better battery life by about 45-60 mins on the "same" workload in the "same" conditions (quotes because they're never the same, but practically are, and acknowledging the batteries between the two machines maybe different, too).

But, it also makes sense when you look at this in paper and in theory:

1) The TDP/thermal performance of the 1.3 is much better - it's binned much higher

which is a good segue to:

2) The 1.3GHz are better specimens of the same microarchitecture, the exact same system, they're just the ones that had fewer defects and were better specimens inside the manufacturing variances and tolerances made: thus they clock at 1.3

which means:

3) They "hurry up and get to idle faster", that's Intel's whole design philosophy here, they do more, faster, with less power, and when they rest, they don't use much (any) more power than the 1.1 or the 1.2.

Hey y'all! I've read a lot recently about Core M processors and was wondering what would be the best choice of these 3 for battery life. Here is an interesting article @ http://www.anandtech.com/show/9117/analyzing-intel-core-m-performance

I know the 1.1ghz processor is bumped from 0.9 to 1.1ghz and the TDP raises to 6 watts. The 1.3ghz processor on the other hand has a 5.25 watt TDP I believe. Any ideas what processor would give the best battery life? Thanks!

 

[Steve121178]

Hmm, interesting. I still don't think the 1.3GHz chip is worth the extra cost for what you are getting.

 

[bmustaf]

Well, that is certainly a very personal choice, you may certainly think the extra cost isn't worth it for what *you* are getting, but a mere $150 more for 45-60 more mins of battery life (not to mention 100 MHz on the low end and 300MHz on the high end) is a drop in the bucket in terms of a value proposition for a lot of other folks.

I had both, both in return policies of the retailers I got them from (BB and AAPL) and the 1.2 went back, no brainer (for me), but I understand if the $100/$150 means more to others than 45-60 mins of battery life/a 10% performance boost (actually when you look at the benchmarks the delta in performance is almost exactly in line with the delta in cost...). But, again, that's just me which neither makes it right or wrong for anyone but myself.

That's why AAPL markets these options, they understand their market very, very well, but suggesting "[you] still don't think the 1.3GHz chip is worth the extra cost for what [I am] getting" is pretty ridiculous/monotheistic. It's kinda like when people thumping a Bible tell others they're wrong for "not believing", gross. Ironically, no such thing as one-size-fits-all in nearly anything, IMO.

Hmm, interesting. I still don't think the 1.3GHz chip is worth the extra cost for what you are getting.

 

[Music Lover]

The 1.3GHz is far and away the most thermally efficient of the processors available on the rMB.

But, it also makes sense when you look at this in paper and in theory:

1) The TDP/thermal performance of the 1.3 is much better - it's binned much higher

Hi, interesting!
Sorry but what is binned and TDP? (for a non-computer guy like me)
Where can I read more about this in an easy way? The link in the first post is a bit too technical for me.

PS
Have a 1.3GHz ordered, soon to be delivered

 

[deleder]

Doesnt it sound very unlikely that the 1.3 ghz will go 45-60 minutes longer? If it does we need someone to do a proper test...

 

[ValSalva]

Hmm, interesting. I still don't think the 1.3GHz chip is worth the extra cost for what you are getting.

I'd agree. If one is buying the rMB for processor intensive tasks then they're likely purchasing the wrong computer.

 

[Zapdoc]

Similarly for MBA 13 inch is i5 more efficient than i7 for 2015 models and does i7 give better battery life than i5 ?

Thanks

 

[ziggy29]

Almost all users will have to consider four factors:

* cost
* performance
* power consumption/ battery life
* thermal/heat generation

Of course, each individual will have their own weightings and rankings of relative importance to them, but almost all users will factor all four of these into the decision.

 

[bmustaf]

I'm not sure what a good resource for that is that is any less technical, but here's an analogy that may gloss over some nuances and make electrical engineers (and car lovers) cringe, because like any analogue it has its challenges, but it relates the concept I think:

You're Ferrari. You make amazing engines to very high tolerances. Despite the sophistication of your manufacturing process, each engine that comes off your line has certain variances that make it unique. Some variances just make them fail and they get sent to rework or recycle, of course, but some variances actually make them faster (e.g. they're more perfect of a specimen of the theoretical design), they have the perfect storm of every single moving part properly machined exactly and that they all fit together just right.

Most engines off your production line don't lie in this 5-10%. They also don't fall in the 5-10% that just fail or are so, so bad they don't pass basic quality control at the end of the line, they fall somewhere in the middle 80-90%.

So, as an astute manager of this business, you have one design and assembly line for engines for a model of car you are selling, but for a certain price premium, you offer your customers the option to buy the engine that tested at a higher horsepower for an added cost as an option.

That's binning, and it's done in the processor, RAM, and microprocessor world *all the time*. Now, it's not to say Intel makes only one chip, hardly, not even close to true, they market a whole diversity of chips they in turn sell to system manufacturers, but within the Core M line (the processors that go into the rMB), it's really only one design. The 1.1s are the low end of the assembly line, they work, they're nice chips, reliable, but they test the slowest of the processors Apple has chosen. The 1.3s test the highest.

The TDP rating is just how much heat the chip will put off when running a "representative" workload (e.g. under normal use). The "lower" (better? In the eyes of the beholder here) the TDP, the less heat that is put off, which generally (but not always) relates to less power consumed, too.

This is related to binning, in the sense that for the same work, the higher-binned (better performing) processors will put off less power and also consume less power.

Hi, interesting!
Sorry but what is binned and TDP? (for a non-computer guy like me)
Where can I read more about this in an easy way? The link in the first post is a bit too technical for me.

PS
Have a 1.3GHz ordered, soon to be delivered

----------

The i5 and i7 are substantially different enough processors that they're not simply binned, they're actually different machines.

The i7 is rated to consume more power, too.

This is the opposite of the 1.3 v 1.2 question here, where the 1.3 is faster b/c it's simply thermally more efficient, but on the other hand the i7 is a substantially different device that actually draws more power and gives off more heat,

Similarly for MBA 13 inch is i5 more efficient than i7 for 2015 models and does i7 give better battery life than i5 ?

Thanks

 

[usernames.taken]

Hmm, interesting. I still don't think the 1.3GHz chip is worth the extra cost for what you are getting.

And some people don't think the extra cost of buying a Mac (vs a PC) is worth the extra cost for what you are getting either. Thankfully we don't live in a world where we get to choose what is of value to us rather than having it forced upon us.

 

[deleder]

And some people don't think the extra cost of buying a Mac (vs a PC) is worth the extra cost for what you are getting either. Thankfully we don't live in a world where we get to choose what is of value to us rather than having it forced upon us.

Absolutely not. The question is if you would notice any difference at all if you were in a blind test. I am sure not many would distinguish the 1.1 and 1.2 from each other, or 1.2 and 1.3.

 

[Zapdoc]

Good explanation -

Thanks that is a good explanation and reflects what I found with i5 vs i7 general

I'm not sure what a good resource for that is that is any less technical, but here's an analogy that may gloss over some nuances and make electrical engineers (and car lovers) cringe, because like any analogue it has its challenges, but it relates the concept I think:

You're Ferrari. You make amazing engines to very high tolerances. Despite the sophistication of your manufacturing process, each engine that comes off your line has certain variances that make it unique. Some variances just make them fail and they get sent to rework or recycle, of course, but some variances actually make them faster (e.g. they're more perfect of a specimen of the theoretical design), they have the perfect storm of every single moving part properly machined exactly and that they all fit together just right.

Most engines off your production line don't lie in this 5-10%. They also don't fall in the 5-10% that just fail or are so, so bad they don't pass basic quality control at the end of the line, they fall somewhere in the middle 80-90%.

So, as an astute manager of this business, you have one design and assembly line for engines for a model of car you are selling, but for a certain price premium, you offer your customers the option to buy the engine that tested at a higher horsepower for an added cost as an option.

That's binning, and it's done in the processor, RAM, and microprocessor world *all the time*. Now, it's not to say Intel makes only one chip, hardly, not even close to true, they market a whole diversity of chips they in turn sell to system manufacturers, but within the Core M line (the processors that go into the rMB), it's really only one design. The 1.1s are the low end of the assembly line, they work, they're nice chips, reliable, but they test the slowest of the processors Apple has chosen. The 1.3s test the highest.

The TDP rating is just how much heat the chip will put off when running a "representative" workload (e.g. under normal use). The "lower" (better? In the eyes of the beholder here) the TDP, the less heat that is put off, which generally (but not always) relates to less power consumed, too.

This is related to binning, in the sense that for the same work, the higher-binned (better performing) processors will put off less power and also consume less power.



----------

The i5 and i7 are substantially different enough processors that they're not simply binned, they're actually different machines.

The i7 is rated to consume more power, too.

This is the opposite of the 1.3 v 1.2 question here, where the 1.3 is faster b/c it's simply thermally more efficient, but on the other hand the i7 is a substantially different device that actually draws more power and gives off more heat,

 

[Music Lover]

bmustaf, thanks for an excellent answer to my question, highly appreciated !!

 

[icymountain]

I have been researching this topic a bit, as I am preparing to order one of these small machines.

Is there a link where I could find some more quantitative data on this issue ?
(I am trying to determine whether it is worth it for me, and would rather do so on the basis of as solid a comparison as possible)

 

[callea]

More battery life doesn't always mean more efficiency.
It could means just more throttling.

 

[Ilya Alex]

bmustaf, tnx for a comprehensive answer!
Totally agree

 

[tbirdparis]

I have been researching this topic a bit, as I am preparing to order one of these small machines.

Is there a link where I could find some more quantitative data on this issue ?
(I am trying to determine whether it is worth it for me, and would rather do so on the basis of as solid a comparison as possible)

The best technical overview I've found about Core M is pretty much everything I've read at Anandtech - the review of the rMB itself and also an article they published on Core M itself:

http://www.anandtech.com/show/9117/analyzing-intel-core-m-performance

But even then, it remains pretty hard to get a clear and direct answer on which one is better in any significant ways and exactly by how much, if at all. Maybe this is because there just isn't a clear and direct answer?

For what it's worth, my reasoning came down to this: I was going to get the 512GB storage model no matter what, and this is where the biggest jump in price happens. Once I'd decided the 256 was ruled out for me, adding on an extra 150 bucks or so (depending on which country you're buying in) felt anecdotal. Not a whole lot more to spend for a machine that I was going to have to order CTO anyway because I wanted a keyboard (British english) which isn't available in stores where I live. If I were able to just walk out of an Apple Store with a 512/1.2 with the keyboard I wanted, I might have been more tempted to just get one and be done with it. But since I was waiting anyway, I thought why not.

I read around some more, found some discussion that because of the Core M version in the 1.3 (it uses the M-5Y71), this model was supposedly slightly cooler-running and a little more power efficient. This seemed counterintuitive - I thought that going for the slightly faster one should chew through the battery a bit more. But apparently not. I still don't know how true this is, but from my own experience of a couple of months ownership so far, I would have to say that my battery life seems to be at the better end of what people have been mentioning around here. I mean at least I'm not getting anywhere near the disappointing results some people are talking about. But who knows, this could just be totally down to the luck of the draw of the quality of the battery I happened to get and nothing else.

The other thing I also kept forgetting and then remembering again was that the difference between 1.2 and 1.3, when you look at it like that is actually pretty misleading. Every time I thought meh what's the big deal, it's .1 of a GHz difference, I'd then remember that actually the turbo speeds are a little more different: the 1.2 jumps to 2.6 max and the 1.3 to 2.9. So I figured OK, when just using the machine to do ordinary things I'm sure they'd feel exactly the same. But the times when you might want a bit of juice - any task when the CPU has to knuckle down and work, at that point it's no longer 1.2 vs 1.3 but 2.6 vs 2.9. Again, for 150 bucks extra this felt like a reasonable deal. And ultimately, all of the geek bench results I've seen support at least the fact that the 1.3 is technically faster even if it's not flagrantly noticeable.

Pretty sure I'd have been just as happy with a 1.2, but I don't regret having forked out a tiny bit more. If the battery life had sucked though, then I would have regretted it. But as I said above, I haven't seen enough substantial info to suggest that I could be sure it was the processor if the battery life was bad as opposed to having just gotten a dodgy battery..

 

[icymountain]

Thanks a lot, tbirdparis, it helps!
I also feel that I need 512Gb, so I may follow more or less the same reasoning as you did.

 

[tbirdparis]

Thanks a lot, tbirdparis, it helps!
I also feel that I need 512Gb, so I may follow more or less the same reasoning as you did.

No prob, you're welcome. Even if you end up going for the 1.2, I was definitely glad I went for 512GB because I really hate feeling cramped with internal storage. It certainly does push the price of the machine up to something less easy to swallow.. but for me it made it much more useful, especially seeing as you can't upgrade later.