NewLC - News - Comments

Help Required

TayyabHussain — Sat, 17 May 2008 10:29:26 +0200

How i can find the book, If any body has the link to download. Kindly send it .

Any help would be highly appreciated..

Re: Symbian and Windows on the same device - what the hell?

alh — Thu, 15 May 2008 10:38:26 +0200

I don't have any numbers, and even If I did, I probably wouldn't be allowed to say them...

But licencing cost is probably a very small part of the total cost of integrating any OS onto your platform, and fill it with sensible apps and customize it for your customers.
The engineering hours needed here are huge, for any OS. (And thats why there is a new solution every few month promising less engineering cost...)

Re: Symbian and Windows on the same device - what the hell?

KevinM — Mon, 12 May 2008 20:32:13 +0200

Other then the complexity of the various compatability issues, from the OEM perspective, anybody got any ideas on what the raw cost here per OS . How much does an OEM have to pay for having UIQ or S60 on their phone, how does that change based on volume. Who would support it and at what cost ? Any perspectives ?

Re: Penrillian uSTL now supports Series 60 3rd Edition

JavaCS — Fri, 02 May 2008 06:26:35 +0200

The two libs are both useful, and I tried to use ustl in S60 2nd Ed with some modification/ replacing, and it works well. But I think STLport/Symbian will not work because of the need of P.I.P.S.

But ustl is not very the same with STL, and no iostream etc. I think STLport may be the best choice for Symbian OS 9.x, and we can use it as a static lib, then it maybe not increase too much size of the final sis when we only use a small part of STL.

Re: Penrillian uSTL now supports Series 60 3rd Edition

Guru_Meditation — Fri, 25 Apr 2008 12:34:10 +0200

Quite so, but there's at least one relevant difference. STLport/Symbian requires P.I.P.S. that Penrillian/uSTL does not. I know that P.I.P.S will be pre-installed on the newer phones, but at the moment using STLport requires the user to install a 463 kB .sis packet that Penrillian/uSTL doesn't seem to require.

Re: Symbian and Windows on the same device - what the hell?

twmd — Wed, 23 Apr 2008 12:07:00 +0200

>> without the need to modify the existing modem software".

That was my point exactly.

Cost is the chief driver for one chip - but people often underestimate the cost of modifying and maintaining a complex platform. Software asset reuse is massively important to a big company because hardware can be swapped quicker than software these days.

On top of that. Real time programming/embedded programming is distinct engineering discipline - often clouded by the fact that many developers who have written mobile apps classify themselves as embedded programmers.

The other way of doing singe chip it is to introduce a personality layer. This essentially provides a library of mutexes, threads etc which emulates an incumbent RTOS allowing for compilation of the signaling stack and test code on top of another OS.
The disadvantage of personality layer is that you need one for each OS, and each has to be carefully verified.

Here are some examples of typical layouts:

- The first Symbian phone which I worked on had RTOS and Symbian running on a single chip (back in 1999). Series 40 devices tend to run everything on the RTOS, but "open" OSes rarely attempt to run signaling stacks.

- The iPhone is 2 chip. It has a 2.5g daughter board which could be switched for 3G without having to modify the main ARM11 cpu.

- Freescale's MXC reference design is single chip, but runs most of the signaling stack on a starcore DSP.
http://www.freescale.com/files/wireless_comm/doc/fact_sheet/3GREFDESIGNFS.pdf

When you start adding real time multimedia, virtual memory paging (by nature undeterministic) and a product which is assembled from many many partner deliveries, it takes an enormous amount of effort to verify the integrity of an entire system.

In summary. One chip, one OS is not a synonym for "best solution", like all engineering, the best solution takes into account cost, skillset, organizational factors, complexity, reuse and ROI.

I wrote a bit more about it in context here:
http://twmdesign.co.uk/theblog/?p=104

Re: Symbian and Windows on the same device - what the hell?

tote — Wed, 23 Apr 2008 11:25:18 +0200

As far as I know the reason for device manufacturers and Symbian to maintain the separation between the RTOS and the Application OS is that device manufacturers want to protect their IPRs and their home-brew GSM stacks. I'm not sure if there is a Symbian GSM stack, as such, and even if it does exist if any device vendors have taken that into use.

What I can read, though, from VirtualLogix VLX introduction is that there would be a single comms stack, which they would provide: "... Enables the use of a rich OS on a single ARM core architecture without the need to modify the existing modem software".

Re: Symbian and Windows on the same device - what the hell?

rbrunner — Wed, 23 Apr 2008 10:48:06 +0200

Yes, but the question is how it is done. I don't think you can take just any GSM stack and put it alongside Symbian on the same CPU. I would guess it must be Symbian's GSM stack, tweaked for your hardware, or at least a GSM stack that follows a lot of conventions and has to implement a lot of interfaces to play nice with the Symbian kernel.

In this way, you arrive at a GSM stack that you can't use on the same phone hardware if you as manufacturer want to offer the same phone, but with Linux in addition to your Symbian-based model. You end up "porting" (in a certain sense) your GSM stack to Linux. And to Windows Mobile. And to...

With virtualization, goes the theory as I understood it, you can put *one* GSM stack unchanged alongside *any* operating system on a *single* CPU.

Nice theory, but frankly I don't buy it. This virtualization layer adds a whole level of new complexity where things can go wrong. Maybe you might manage to save some work because there is no need for porting the GSM stack, but I think you will spend about an equal amount of work to get the whole "virtual" thing right. Plus you spend license fees for the virtual machine that you could have spent for a second CPU. Zero-sum game, at best.

Re: Symbian and Windows on the same device - what the hell?

Andreas — Wed, 23 Apr 2008 09:19:19 +0200

Symbian OS and the GSM stack can already be executed on a single processor in the new versions of the Symbian OS, if this was your point.

Re: Penrillian uSTL now supports Series 60 3rd Edition

gybrush — Tue, 22 Apr 2008 22:30:02 +0200

STL on Symbian 9.x is available from a very long time. Everybody who wanted to use STL on Symbian have already discovered http://marcoplusplus.blogspot.com/2007/05/stlport-for-symbian-os-released.html

Re: Symbian and Windows on the same device - what the hell?

twmd — Tue, 22 Apr 2008 13:53:06 +0200

Or, I've presented an alternative view here...

http://twmdesign.co.uk/theblog/?p=114

Re: Symbian and Windows on the same device - what the hell?

tote — Tue, 22 Apr 2008 12:02:36 +0200

That's what they're doing right now: changing the management. Or you would change even the new one?

Re: Symbian and Windows on the same device - what the hell?

rbrunner — Tue, 22 Apr 2008 12:01:04 +0200

Yes, what the hell. For me, this is only a sign how desperate companies already became, facing the utterly terrible fragmentation of the mobile OS landscape.

And, for Motorola, is there really no other solution to their problem "Can't decide which OS to use" other than "Bright idea, we will put them all into the phone at the same time!"

Time for new management, if you ask me.

Re: Carbide.c++ 1.3 is out

wl — Thu, 10 Apr 2008 12:43:47 +0200

cheers, it's java - to make works it smootly, u just need more powerful computer i think...
-
too bad, new carbide doesn't support old symbian (6-8.1) projects

Re: Carbide.c++ 1.3 is out

runforu — Tue, 08 Apr 2008 10:18:01 +0200

It is a big problem, it can not be disabled when your licience is expired. big foolish Nokia!!!!!