Comments on: Thurrott’ling Windows Mobile: Take 2

By: buffer

buffer — Mon, 12 Oct 2009 15:34:31 +0000

I find the MS vs. Apple discussions comparing apple to oranges. Apple devotes a lot of resources to innovation with user interface and usability for a single platform which it controls. MS devotes a lot of resources into making a standard software platform available for many hardware architectures. These are two very different paths. Apple does better at innovating UI and useability, MS does better at supporting many different hardware architectures (Apple doesn’t even try). You can knock MS for not being as innovative as Apple, but MS provides software that thousands of PC builders and smartphone-mobile computer manufacturers can make use of. I’m impressed with both companies. I’ll be most impressed with the company that can provide software that will run on anybody’s hardware, look and work great on all of it and never crash – but I’ll never see it in my lifetime from either Apple or MS.

By: cmaier

cmaier — Mon, 07 Apr 2008 13:33:46 +0000

In burst mode they are fast enough.

By: surur

surur — Mon, 07 Apr 2008 13:27:46 +0000

i agree. I just think that the bus is fast enough to make the wait no more than "barely perceptible."

The bus may be, but flash is still pretty slow, only about as slow as a HDD.

Surur

By: cmaier

cmaier — Mon, 07 Apr 2008 13:23:25 +0000

i agree. I just think that the bus is fast enough to make the wait no more than “barely perceptible.”

By: surur

surur — Mon, 07 Apr 2008 12:54:15 +0000

The hardware that detects incoming calls doubtless generates an interrupt that wakes up the processor. I doubt it relies on software on the actual processor to deal with incoming calls (that is, prior to the user interface turning on).

Yes, but calls cant exactly wait for the user interface to load, can it? Its a time-sensitive event, and if its too slow you end up with missed calls.

Surur

By: cmaier

cmaier — Mon, 07 Apr 2008 12:31:12 +0000

By: surur

surur — Mon, 07 Apr 2008 12:28:49 +0000

Doubling the RAM doesn't double the power, first of all. (unless they do it by adding a separate RAM chip).

Thats how its usually done.

I think sometimes it DOES "hibernate" - sometimes the post-slide popup seems to lag perceptibly. Given the tiny amount of memory and the width of the bus, reloading wouldn't take very long. Bigger issue is write cycling the flash.

Unlikely that they switch off the RAM chip, as the code that handles incoming phone calls have to live somewhere, and it cant all be on the radio chip, e.g. your Caller ID code has to be on the DRAM chip, and will never be swapped out, and you cant switch of just some part of the RAM chip, you have to refresh the whole thing.

Surur

By: cmaier

cmaier — Mon, 07 Apr 2008 12:20:40 +0000

Doubling the RAM doesn't double the power, first of all. (unless they do it by adding a separate RAM chip).

By: surur

surur — Mon, 07 Apr 2008 12:06:33 +0000

RAM is not a battery hog, big guy. It uses a fraction of what the processor or radio use, and, in fact, isn't even needed during standby (just dump the contents to "disk").

If you want your device to be responsive you dont want to be reading your app into RAM every time you return from suspend. I highly doubt Apple is doing it that way (although that would be a great way to use the time wasted on the slide to unlock screen).

The amount of power you use for RAM is a fixed cost, even if its only 5 mAmp/hour. That BY ITSELF will drain a 1400 mAh battery in 11 days. If you double that you drain the same battery in only 6 days. And thats all by itself, without any radio drain.

Surur

By: cmaier

cmaier — Mon, 07 Apr 2008 11:48:13 +0000

RAM is not a battery hog, big guy. It uses a fraction of what the processor or radio use, and, in fact, isn’t even needed during standby (just dump the contents to “disk”).