The problem with Windows "apps" is that they are made inefficient and un-optimized by default, because they were targeted at much more powerful processors, and as processors became even more powerful, they didn't care about increasing the performance or at least keeping it at the previous level (and not slower).
Developers never really cared about netbooks, and these devices, either with ARM processors or with Atom, are only at that level. So most Windows apps will struggle on these machines, while most apps on iOS or Android will be a lot faster on the same processors, because their low-end target was something like a 600 Mhz ARM11 CPU.
The point is that you can get away with inefficient code on desktop Windows, which is invariably caused by your favourite application framework rather than glib developers (not) inserting sleep calls everywhere.
Very knowledgeable opinion sir. This is indeed true. Us as windows developers are required to add a sleep statement every few lines of code so as it not make it appear responsive.
A sleep() statement?! Novice! That won't even busy-wait! I like to loop around for a few thousand times writing nonsense to dummy variables. I earned those time slices - no way am I going to just give them back to the OS scheduler.
That's not what I meant of course. What I meant is that apps on Windows tend to be a lot more "bloated"/full of features/graphics compared to the much leaner mobile apps.
Also, most devs probably build for the average performance, so say something like an old Core 2 Duo laptop, and if their app there has a 10% CPU utilization, it might have 30% CPU utilization on Atom, or more, which means on ARM chips, especially the mid-end to low-end ones, the same app will use the CPU a lot more. Apps designed for mobile are meant to use 5-10% of those ARM CPU's from day one.
My understanding however from the "Windows 8 Doesn’t Want Your App. Try Again Later" post was that as part of the approval process, MSFT does automatically test the apps in a low performance environment, and if they don't meed specific benchmarks there (example in the post was startup or shutdown time of 2 second, but I can't remember exactly and am on mobile), they will reject the app and tel you to work on it.
So if that's really the case, maybe they are selective with it. Maybe they don't apply the same rigor to their own apps, which would be a shame since you'd think they have to set an example.
One could also go with the premise that the "conquer world market" option actually disappeared a year ago, when Apple released the product, then improved on it with almost no competition in the game.
If Microsoft thinks that it can release a product with software that's not ready for release on just price parity with the market leader, they're delusional.
Apple spent the better part of 2011 (ie, last year) proving that iOS doesn't need a daddy OS around anymore - why is a USB host feature in a tablet meaningful for users? So that I can upload my picutres? Why not I just stream them to your photostream? or push them into your google drive or dropbox?
When the PlayBook was released, people also slammed it because its Mail client was terrible (you had to have a BB phone). The internet was sure that it wasn't even worth a look as Android tablet sales would take off real soon now.
And all the independent developers have had a chance to see what makes a fast program on a Windows-based ARM device?
I think as people begin programming for Windows with an actual RT instance to test on (for example, the Surface), apps will become faster and more efficient. Personally, I haven't seen a problem with this as of yet; but, I haven't used that many apps in the store, just Netflix and MetroTwit, FreshPaint and a couple of others.