I support open source and standards. Not to mention, WinMo 6.x is rather clunky and ugly IMO.

I do like HTC's Sense UI though.

Sounds like we will need a Quake 3 test. Crank the bot difficulty up so the cpu gets a workout and the GPU's can do their thing too with the polys and textures

Whatever's faster will depend on two things..

1) degree of operating system optimization
2) hardware (and it's not necessarily a straight 1:1 compare and contrast).

I too am disappointed that the N900 has only 256 MB of ram and that Nokia has been advertising claims of 1 GB using the heavily asterisked fig leaf of a 768 MB swap file. All the previous Nokia tablets had inadequate ram for their era/use as well and frequently bogged down after more than a few apps were open. While the N900 has twice as much ram as the N810, Fremantle presumably has a much larger system footprint as well. 512 MB of ram would have provided a much bigger cushion for running applications.

I got to play with an N900 for a couple hours this last week thanks to Matt Miller. The group discussed and played with multitasking and the conclusion was that while in theory the system could have dozens of applications open at a time, the practical limit before the system starts to feel sluggish is more like 4-5.

I am also concerned with premature flash wear due to swap usage. I had a very expensive Psion Netbook Pro become an expensive brick when I flashed the firmware and it encountered bad flash blocks.

But hey, don't worry about it. The flash only has to last through the warranty period and Nokia will expect you to buy another $600 device next year. After the 770 abandonment and no clear statement from Nokia with regard to Maemo 6 support, does anyone really expect anything different?

All this talk of N900 vs HD2,iPhone,Driod is a bit weird. Its really and I mean really hard to compare these things because first and foremost the software running on all of them is very different even tho they have almost similar hardware. Its like all the motherboard reviews we all read with thousands of benchmarks using standard set of tools. But the difference here is that we don't have a standard set of tools which can be run without the different operating systems on the mobile hardware and until we can have such software that we could flash the hardware with we won't be getting any real benchmarks on the hardware. Benchmarks using applications like Quake, Office, Google Maps etc against power consumption... won't give you a real number since the underlying operating system can and will get updates later in its life cycle. So we should all try to avoid comparing these apples n oranges n grape fruits and just look at the device with the operating system as a whole and see which one suits your specific needs.

Hehe good thing if you buy it with a credit card (but not Discover) you get an automatic additional 1 year warranty. So that means your covered for two years. Then Squaretrade makes it another year.

But yeah if Nokia does that I'm switching to Android next time. Might as well familarize myself with their platform if Nokia insists on screwing with the community.

exactly nashith...different OS's, optimized in different ways...Symbian for instance runs pretty well on the older ARM 11, like the 5800, optimized well...faster processor would help a bit but not too much.

Windows mobile is very intensive so it needs a fast processor...600MHZ processors struggle on WM

To give people an idea of how much RAM is generally available:

with only the following open: photobrowser (no pic loaded), calculator, xterminal, web browser with cnn.com open there is 5MB free physical free out of the 256MB
- with no apps open, there is 30MB physical free out of the 256MB

Kind of disappointing that your forced into swap if you run more than just the web browser.

Its kinda like running Vista (uses 800MB) with only 1GB RAM. Its fine unless you run several 'large' apps (Firefox, MSWord, Excel, playing a movie etc)

For those who are not aware, virtual RAM is not 'magic', there is a cost. If there are multiple apps actively competing for the virtual RAM, the O/S is going to use up CPU cycles to page in/out from disk into RAM plus the wait time for the IO the complete.

Linux uses free memory as cache so you can't just look at memory free and determine the available memory for applications.

Jeesh! I guess the browser is like a desktop one, since it is using just as much memory as one. Does not seem too efficient. Wrong?

CNN might be the problem