2017 Pro - m3 smaller battery than i5?

[sberti]

I'm thinking of upgrading to the new (2017) Surface Pro. I have a i5 Pro4, but the main reason is to get better battery life.
Does the m3 have any better battery life than the i5? The specs list no difference.
I might assume it does, but then I see that the m3 is 20g lighter than the i5. Does that mean the m3 has a smaller battery?
I need to get through a whole 8 hour day of note-taking using the pen in OneNote.

 

[GreyFox7]

There must be a 20g difference between the m3 and i5 packages because AFAIK the have the same battery. There probably isn't much difference between battery life either amazingly, well there wasn't on the SP4 with Skylake it remains to be seen how Kaby Lake does. I think engineers are still learning how to optimize Core m chips but one day they will get it, maybe this will be the day.

 

[Exhile]

The m3 consumes 4.5W of power while the i5 consumes 15W of power. Lower power consumption means longer battery life.

 

[GreyFox7]

The m3 consumes 4.5W of power while the i5 consumes 15W of power. Lower power consumption means longer battery life.

Counter intuitive isn't it. In the battle for energy saving there is a concept known as the race to idle. The theory Race to idle hinges on the idea that the quicker you get to idle and sleep the more energy you save. Clearly the i7 uses more energy so there is a point beyond which it doesn't work. That core m and i5 are close to the same on the SP4, the slower, use low power for longer and faster, use higher power for shorter strategies are in equilibrium.

That may lead to the Kaby Lake core m taking the lead if it's a little faster than the Skylake core m but we won't know until actual usage reports come in.

 

[sberti]

Thanks for the replies.
If there might not be too much difference in the battery life between the i5 & m3, I'm leaning toward the m3.
Does using the pen, instead of the keyboard, increase power consumption?

 

[MickeyLittle]

I would think that if you use the backlight on the keyboard that would use more battery than not. What would be interesting for you to test would be to type something from a book for an hour and monitor how much battery you used doing that and then use the stylus and write for an hour using the same application and see how much battery that used.

 

[GreyFox7]

I would think that if you use the backlight on the keyboard that would use more battery than not. What would be interesting for you to test would be to type something from a book for an hour and monitor how much battery you used doing that and then use the stylus and write for an hour using the same application and see how much battery that used.

There's a free app for that, Battery Bar.

 

[sberti]

Thanks for the suggestion to test it and the tip about Battery Bar.
If/when I get the time to test it, I'll post my results.

(Sorry for the typo in the thread title. I don't know how to edit it.)