From 4 to 7 hours

[leeshor]

He wasn't doing any standardized benchmark comparisons in your link. He was also using Chrome, which is (or at least when I used it) not a part of the BAPCO testing. Everyone will come up with a different battery use figure depending on configuration and usage. That's why comparison testing using a standardized app is important.

 

[jnjroach]

Much of this is because as a segment (X86 PC Computing) we are at a crossroads or transition (depending on who you believe). Win32 Applications are horrendous at Power Management, and developers used very poor tricks to speed up the responsiveness of their Applications by loading their bootstraps into RAM and continually running (often adding their bootstraps into Host Services). While the Application is running, even static binaries like PDF Editors (PDF Annotator especially) will keep CPU Cycles running in the background.

Using MUI and/or Windows Apps will greatly increase battery life as WinRT unlike .NET is designed from the ground up for Modern App Development and running on low powered devices, but they are also designed for high DPI screens.

As long as the unwashed masses (both Geeks and Normals) insist on running ChessMaster 2000 or allowing Google to run their CPU Interrupts into the ground on their shiny new Windows Tablets we won't see consistent and believable battery numbers.

We all have seen how Google manipulates the Window Kernel Subsystem to be "fast", so these numbers are quite reasonable depending on usage patterns of the user.

 

[ctitanic]

He wasn't doing any standardized benchmark comparisons in your link. He was also using Chrome, which is (or at least when I used it) not a part of the BAPCO testing. Everyone will come up with a different battery use figure depending on configuration and usage. That's why comparison testing using a standardized app is important.

Here is the thing, they said that in that same test:

The Surface 3 outlasts the competition; the Toshiba Encore 2 Write comes closest (9:03 on the same test), with the Microsoft Surface Pro 3 (8:55) clocking in just a bit behind. Our other comparison systems died hours sooner.

I really don't know how they got 8:55 out of the PRO 3 in a real life scenarios.

 

[GreyFox7]

I see a pattern here.
In one review the person says they got 8-10 using touch apps, 6-7 with Chrome, and 4-7 doing heavy graphics.
You circle 4-7 and expect that's what everyone will get all the time.

Then another review using an Industry standard benchmark posts 9:52 minutes on that test.
You argue that nobody will get 11 hours.

Both of these positions exhibit bias, fair enough, I know your point of reference.

It seems reasonable that users will get 8-10 with touch apps. I do with both Surface 2 and Surface Pro 3 and I get more running the BAPCO battery test, still I know what that tells me because its consistent. I someone says this is what I get doing *stuff* without some other point of reference it doesn't mean much because I don't really know what they did.

 

[ctitanic]

Much of this is because as a segment (X86 PC Computing) we are at a crossroads or transition (depending on who you believe). Win32 Applications are horrendous at Power Management, and developers used very poor tricks to speed up the responsiveness of their Applications by loading their bootstraps into RAM and continually running (often adding their bootstraps into Host Services). While the Application is running, even static binaries like PDF Editors (PDF Annotator especially) will keep CPU Cycles running in the background.

Using MUI and/or Windows Apps will greatly increase battery life as WinRT unlike .NET is designed from the ground up for Modern App Development and running on low powered devices, but they are also designed for high DPI screens.

As long as the unwashed masses (both Geeks and Normals) insist on running ChessMaster 2000 or allowing Google to run their CPU Interrupts into the ground on their shiny new Windows Tablets we won't see consistent and believable battery numbers.

We all have seen how Google manipulates the Window Kernel Subsystem to be "fast", so these numbers are quite reasonable depending on usage patterns of the user.

Correct, but this is not an ARM device that will force you to use metro apps all time. So PCMag battery numbers are to me a mental masturbation at best.

 

[ctitanic]

I see a pattern here.
In one review the person says they got 8-10 using touch apps, 6-7 with Chrome, and 4-7 doing heavy graphics.
You circle 4-7 and expect that's what everyone will get all the time.

Then another review using an Industry standard benchmark posts 9:52 minutes on that test.
You argue that nobody will get 11 hours.

Both of these positions exhibit bias, fair enough, I know your point of reference.

It seems reasonable that users will get 8-10 with touch apps. I do with both Surface 2 and Surface Pro 3 and I get more running the BAPCO battery test, still I know what that tells me because its consistent. I someone says this is what I get doing *stuff* without some other point of reference it doesn't mean much because I don't really know what they did.

I see your point but they are telling that the PRO3 get 8:55 in the same test and in real life only few mortals can claim to have seen such battery life in a PRO3.

 

[ctitanic]

And for the record, I use most of the time metro apps.

 

[ctitanic]

10 - 30% = 7 hours. That's what I expect to see in a S3, real life.

 

[jnjroach]

Correct, but this is not an ARM device that will force you to use metro apps all time. So PCMag battery numbers are to me a mental masturbation at best.

This is true, but like those reviewers who have been testing the ATOM based 7-8" Tablets, using them like RT give a better overall experience, the ironic part these are people who typically were very anti-RT:

• Chippy from UMPC Portal
• Daniel Rubio from Windows Central
• Brad Simms from Neowin
• Tom Warren from the Verge
• Paul Thurrott from Thurrott.com

I see your point but they are telling that the PRO3 get 8:55 in the same test and in real life only few mortals can claim to have seen such battery life in a PRO3. View attachment 5917

But that number is the average since install, which can mean if your usage patterns are like mine, you've had session that lasted 2.5-3 hours, some that lasted 4-5 hours, some 6-7 hours and a couple that lasted longer than 7....

And for the record, I use most of the time metro apps.

How many Win32 Applications do you have installed though? Usage is only part of the equation...

 

[GreyFox7]

I see your point but they are telling that the PRO3 get 8:55 in the same test and in real life only few mortals can claim to have seen such battery life in a PRO3. View attachment 5917

Ha ha, Ok Here's mine WITH *Normal* Use mostly.

Surface 2


Surface Pro 3


Actually my Surface 2 had been running a bit low lately but I investigated and found some battery suckers and pruned them so it's coming back up.

 

[ctitanic]

How many Win32 Applications do you have installed though? Usage is only part of the equation...

Correct. I would say that less than the metro apps. And I don't use them when I'm on battery.

 

[ctitanic]

Ha ha, Ok Here's mine WITH *Normal* Use mostly.

Surface 2
View attachment 5920

Surface Pro 3
View attachment 5922

Actually my Surface 2 had been running a nit low lately but I investigated and found some battery suckers and pruned them so it's coming back up.

You are the exception to the rule. You should frame those results and hang them on the wall.