Long-shot?

[Massman82]

Developers,

Tell me how wrong I am!

It would seem as if doing a dual-boot between Pocket PC and the Tungsten C would be possible.

Here is why:

Same processor as many PPC's (400mhz)
Higher resolution - The PPC OS would use part of the screen
Same memory - 64MB
WiFi - Toshiba's have this.

Ok, so you would have to write some drivers for the screen, speakers, and voice recording but would that be all? Would anyone like to try?

Ayasin?

[CliePet]

> It would seem as if doing a dual-boot between Pocket PC and the Tungsten C would be possible.

Many things are possible, but not necessary practical or legal.
The rest is just software ;->

Two show stoppers:

Each operating system needs a lot of ROM. I haven't seen specs on the size of Tungsten C's ROM (size, how much used), but I suspect it is not enough.

Pocket PC has to be licenced on a device specific basis (and pay a per-device fee). If someone went through the hassle of writing PocketPC drivers, they couldn't legally redistribute the binaries that make up the bulk of the Pocket PC software [without cutting a deal with Microsoft that is]

[ayasin]

Also there are practical market issues. Why would you want a dual boot handheld? This would mean doing a _Hard reset_ every time you wanted to switch OSes.

Amir

[Unregistered]

Originally posted by ayasin
Also there are practical market issues. Why would you want a dual boot handheld? This would mean doing a _Hard reset_ every time you wanted to switch OSes.

Amir

I see. Would be awesome to have if there were not such problems. Both OSes have their advantages/disadvatages!

[Token User]

Originally posted by Unregistered


I see. Would be awesome to have if there were not such problems. Both OSes have their advantages/disadvatages! I think what you really want is something more akin to a WINE emulator for PalmOS ... the only advantage that PPC has over PalmOS is the range of games that are available (does that in itself say something about the target market??). Of course, games are a problem in their own right - as they often are the least standard code written (eg. to tweak the extra 3 FPS durng optimisations they write directly do register).

As as was said, nothing is impossible, and while it would be a very cool thing to do, if you need to do it, you have obviously bought the wrong PDA.

YVMV.

[ThomasK]

Hi all,

in principle you can have a dual boot, it's a little bit like the booting of Windows or LINUX on a PC - both operating systems have their advantages.
But time you need for writing the PalmOS for a PocketPC device or otherwise is immense!

The first (and mostly not solveable) problem is in getting a detailed hardware specification of the device. In case of SONY it's the killer of the project...

[nevarDeath]

Yeah, PPC has PalmOS emulators, but that opnly because they emulate certain devices, and PalmOS devices were freiggin slow before...in fact the POS emu for the sharp zaurus runs quite well....I think it'd be awesome, but ow, I'm trying to think, and I'm not even sure where you'd start...

ack, I wish I knew more about this kind of thing, that sounds like it'd be a fun project...I'm sure that M$ would be all over your *** in no time flat, but if it was done, it could be distributed through Warez easily enough...M$ can't touch sites in some countries :)

[nevarDeath]

Originally posted by ThomasK
Hi all,

in principle you can have a dual boot, it's a little bit like the booting of Windows or LINUX on a PC - both operating systems have their advantages.
But time you need for writing the PalmOS for a PocketPC device or otherwise is immense!

The first (and mostly not solveable) problem is in getting a detailed hardware specification of the device. In case of SONY it's the killer of the project...

well, if you had enough time on your hands and the right tools, I couldn't think of any reason you couldn't map out a schematic of the device, then look to the manufacturer of chip companies to find out what 'mystery' chips do...I duno, i've only taken a year oif electronics, so if I'm wrong, someone please correct me

[ayasin]

Originally posted by nevarDeath


well, if you had enough time on your hands and the right tools, I couldn't think of any reason you couldn't map out a schematic of the device, then look to the manufacturer of chip companies to find out what 'mystery' chips do...I duno, i've only taken a year oif electronics, so if I'm wrong, someone please correct me

Sony uses some gate arrays so without knowing how they programmed them you're pretty much out of luck. Additionally this is not a trivial task (the circuit board is multi layer). Also this still doesn't address the issue of having to do a hard reset. Even the most basic things (memory models for example) are incompatable between the two OSes.

[Token User]

Originally posted by nevarDeath
ack, I wish I knew more about this kind of thing, that sounds like it'd be a fun project...I'm sure that M$ would be all over your *** in no time flat, but if it was done, it could be distributed through Warez easily enough...M$ can't touch sites in some countries :) I read this, and your following post about reverse engineering the internals of the Clie based on a circuit board layout and immediately came to the conclusion that you are completely clueless in this area.

POSE (Palm Operating System Emulator) came out of an app called co-pilot that was gnu licensed. That base code is out there, and this is what the versions available on Linux (including Zaurus) and PPC are based upon.

Truth be told, running Palm apps on a PalmOS 5 device is a much better model for what you are talking about. To the end user, it appears as a native app, but in fact all Palm apps written for PalmOS 4.x and lower use PACE (Palm Application Compatibility Environment) to emulate an original Palm. This PACE app, takes calls to the original PalmOS libraries, and redirects them to the new PalmOS 5 equivalent.

If you want to run Palm apps on a PPC, reverse engineering the PACE app would be a better place to start, redirecting PalmOS library calls to equivalent PPC library calls. Not an easy task, but more doable than reverse engineering a multilayer PCB with programmed logic circuitry.

As for running a PPC app on a Palm ... the only apps I can think of that you might want to do that with is some of the games.

If you really want to run PPC aps - then perhaps you should have chosen a PPC ...

[ThomasK]

Of course you're right. ;)
But the things I wanted to figure out are the special features of the SONY devices (i.e. the build in camera). Because these features may be worth the two (or more) OSes. I think there are some of these features at the PPC or LINUX devices.
And these features will not be supported by any PACE or so - you have to have the interface of the hardware to integrate. So we stand at the point described above.

[Token User]

Ahhh, I get it, so you want to take the superior, feature rich hardware of the Clie and knobble it by running PPC or Linux on it.

FWIW - I think that would be a legendary hack, but highly unlikely.

Given the limited SDK Sony released supporting the "advanced" features of the Clie NX/NZ/TG series, even porting the cool feature from a PCC or Linux platform would prove problematic (but far more likely - especially following the CF Driver example).

[iebnn]

It will not happen. And don't say that I'm just like the people who said the CF driver wouldn't happen, because this project is on an entirely different level.

And regarding a PPC emulator for Palm OS... you all seem to agree that the main thing you'd want are the games... If they already run somewhat slowly on a PPC, they're going to run like slugs under emulation.

As for "porting" WinCE to a Palm device..... why would a developer want to do this? It would take countless hours to do, and there would be many legal issues especially if he's going to sell it.

And what about the games that directly access screen memory? PPCs have a smaller resolution. The PPC OS wouldn't work with 320x320 screen hardware. The PPC OS isn't open source, and I don't think it's possible to write software for it that will take control of screen blitting (even when the OS is booting). So emulation seems a lot more practical, but what's the use? Would it really be worth the developer's time? Games wouldn't be playable (way too slow), and what other software is there that would really make it worth writing an emulator for PPC OS on the Palm? It would be faster to recode those apps for Palm OS than it would be to code a PPC emulator anyway.

[nevarDeath]

Originally posted by Token User
I read this, and your following post about reverse engineering the internals of the Clie based on a circuit board layout and immediately came to the conclusion that you are completely clueless in this area.


you're right, I have very little knowledge about emulation or reverse engineering hardware. I wasn't aware of the complexity of the PCBs and such, So since I didn't, I put 'someone please correct me if I'm wrong' and someone did :)