Reply
Thread Tools
Posts: 874 | Thanked: 316 times | Joined on Jun 2007 @ London UK
#41
Here is the Willcom page with some very slick videos
http://www.willcom-inc.com/d4/willcom_d4.html
 
allnameswereout's Avatar
Posts: 3,397 | Thanked: 1,212 times | Joined on Jul 2008 @ Netherlands
#42
Originally Posted by brontide View Post
What would you benchmark, raw CPU numbers... they are useless.
Good point, but not completely useless.

Besides, one must run the same OS, on a different architecture. Then use tools like netperf, hdbench/hdparm, memtest, iozone, bonnie++, time, and so on. You can even run real applications as test cases. Problem is, this is a difficult task, because you need very identical hardware with only the CPU being different.

Maybe if it is possible to run a VM at a certain speed, emulating certain computers completely, it might at least make the CPU part easier.

There are also still developments in the Linux kernel 2.6.2x series leading to e.g. performance improvements. Even specific to ARM.
 
brontide's Avatar
Posts: 868 | Thanked: 474 times | Joined on Oct 2007 @ Capital District, NY, USA
#43
My point still stands you can't measure usability with any micro-benchmark and when it comes down to it that's all most people care about on a mobile device. Mobility means a very complex juggling act where any 1 new feature may make or break the device ( sure you can watch 800x480 videos, but if it's only good for an hour that would suck ).

Having more mips/watt than your competitor will place you at an advantage in terms of what you can do, but it's no match for usable features. My Palm IIIc has an order of magnitude difference in computing power to the n810 and yet it could still run circles around any NIT PIM today.

So measure all you want, but the real benchmark is user uptake and satisfaction.
 
allnameswereout's Avatar
Posts: 3,397 | Thanked: 1,212 times | Joined on Jul 2008 @ Netherlands
#44
It depends on what level you are analysing, and what you are able to do with your analysis. I'd love to run a tool (PowerTOP) to figure out how many W each driver uses, and how much each application drains the battery. That way I know how long my tablet will stay alive, and what I should and should not run on which moments. This counts, ofcourse, for current applications. The lack of well-integrated PIM is a big issue for the NIT especially because many other proprietary devices do handle this pretty well (often using an open standard, iCal, no less). Or, imagine being able to run powerful DTrace scripts.
 
Posts: 874 | Thanked: 316 times | Joined on Jun 2007 @ London UK
#45
There are some size comparisons photos here, including some side by side with the N810.
http://www.pocketables.net/2008/07/w...sh-6.html#more

1.5 hour battery life is a bit of a bummer though.
 
Bundyo's Avatar
Posts: 4,708 | Thanked: 4,649 times | Joined on Oct 2007 @ Bulgaria
#46
http://www.pocketables.net/2008/07/8-willcom-d4-sh.html
__________________
Technically, there are three determinate states the cat could be in: Alive, Dead, and Bloody Furious.
 
Lord Raiden's Avatar
Posts: 1,562 | Thanked: 349 times | Joined on Jun 2008
#47
Internet tablet + Microsoft anything = Portable Evil. hehe.

No thanks, I'll stick to using open source stuff on my tablet.
 
Reply


 
Forum Jump


All times are GMT. The time now is 03:01.