Potential RISC OS laptop?

Hi there, Following on from suggesting the Nokia 770 as target platform for RISC OS, I’ve come across this:

http://www.dataevolution.com/cathena%20info%202.htm

From what I understand it has a 400MHz xScale processor, and costs around $799. This would appear to be a half decent machine on which to run RISC OS would it not?

I wonder if it might be a good idea to have some kind of a poll on what new hardware to port RISC OS to, if at all?

I would think the XO-1 might be a good target for RISC OS. The target audience and design aims are similar to RISC OS’ original positioning. The unit looks like it needs a decent GUI. It is also likely to be around unchanged for quite a while.

Biggest problem is it would need an Arm emulator at the core of any port.

Hmm. What architecture is X0-1, Bear in mind RISC OS is very tightly linked to ARM.

There have been rumblings about getting a BBC Micro Emulator working on it, the problem being that the BBC BIOS roms aren’t exactly free :(

Hi,

sorry I’ve only just joined.

In my opinion, the best hardware is always going to be the simplest and cheapest, because it represents the quickest time to market. If you want a cheap & simple target system why not pick the GP2X?

http://gp2x.co.uk/

The gp2x is pretty fast and cheap (£120 or so), but because it’s open source, theoretically you can grab loads of documentation and even rewrite the kernel. It’s Samsung ARM9-based and supports external keyboards and TV displays. If there was sufficient volume I guess you could order the PCBs from GamePark; box it and attach a proper LCD. That way you get a RISC-OS based computer for around £200 including display, keyboard and mouse. Who could beat that?

There’s also a wiki for the gp2x.

http://wiki.gp2x.org/wiki/Main_Page

-cheers from jules @P FWIW I’ve been involved with mobile phone development between 2000 and 2007 doing (amongst other things) Symbian OS base-porting (getting Symbian OS to run on different, new hardware).

Garry,

Are you sure the Cathena uses an XScale processor? The other computers made by DEC all have AMD processors (check out the dectop, basically an AMD-based A9Home for less than £50:)). If it was XScale-based then I’d agree that it would make an excellent target for RISC OS.

I quite like the idea of using gp2x pcbs as the basis for a RISC OS computer, but would prefer to see work going in to targetting faster hardware. Would a gp2x computer be any more powerful than an A9? If the Iyonix is still the fastest RO computer in two years time I think the market will probably fade away. At the same time, if the RO market is to expand then we need cheaper hardware.

Ideally, I’d like to see someone mass-produce an ARM-based version of the DECtop capable of running Windows CE, Linux and RISC OS. If RISC OS could be run on a 1GHz Cortex A8 or ARM11 based sub-£200 mini-PC then a lot more people would give it a go. Failing this happening I think the best idea is to target the fastest ARM dev board or consumer device currently available – RO can always be ported on to cheaper devices using similar hardware as they become available. Have ARM released a dev board for the Cortex A8 yet? It says somewhere on their web site that there should be one available by Q3 2007.

In realtion to comments concerning the DECtop, and Windows CE:

A copuple of weeks ago I dropped some strong hints to the developers of ReactOS (http://www.reactos.org), about implementing ARM support for it. :)

The alternative for getting ARM based WindowsCE applications running on an otherwise RISC OS based system, is possibly to port portions of WINE, or license Cedega under appropriate terms. Then you would need to assign a suitable filetype allocation for Portable Executable, develop an image loader and of course work out how on earth to link Windows Style calls to RISC OS type calls. AIUI Windows handles message loops rather differently from WIMP Poll. :)

Re: Alex

I think there’s already been a project of the sort you mention aimed at getting ARM Linux programmes running on RISC OS. Success with either of these would be a massive boost to RO in terms of the software available but neither project would solve the problem of ageing and increasingly slow hardware. Without new faster hardware then we’re still going to struggle with modern multimedia apps. This is why I think developer effort is best spent in porting the OS to a new platform and preferrably one intended for another more popular purpose than running RISC OS. Without faster cheaper hardware we’re screwed.

In my opinion, RO Open should have decided on what they want to port the OS to by the time they have finished releasing the source. Then those who are willing and able to help (possibly me if things have quietened down a bit and there are non-technical tasks to be done) can focus on getting the donkey work done.