Open2x To Have F200/touchscreen Support?

[TrevorBradley]

I just stumbled across Open2x in the forums, I'd not heard of it before and it does sound exciting.

Reading through the Open2x wiki I don't see anything about the F200 or touchscreen support. As much as I'd love to ditch the slow firmware 4.0, I don't think I could give it up if I lost my touchscreen.

Is there any intention to get the touchscreen working for Open2x? I'm assuming this isn't just libraries but also software to tune the touchscreen.

Thanks!

 

[sbock]

A comment from the SVN:

QUOTE
More cleanup and start to support the Wintek LCD on the F200.


http://open2x.org/open2x/index.html

I'm not quite sure if that means that the touchscreen is supported but at least the devs are working on F-200 support.

Regards,
Stephan

 

[TrevorBradley]

Awesome, thanks.

I wait patiently but eagerly for it's release...

 

[DJWillis]

In case you do not read the post below I'll just state this...

-: Open2x currently does not support the F200 and any attempt to flash a DR version WILL most likely brick an F200 :-

None of the main Open2x developers (Orkie, Lithosphere or myself) have an F200 so F200 support is not going to happen any time soon.

The recent patches I have been popping into SVN are a result of be getting a working F100 again and sync'ing up my local tree (that included a few small patches built up with a pre-release F200). Sorry to get hopes up, I am not even sure the final F200 used the WinTek LCD that was in the pre-release anyway.

I have a rough idea of how the F200 differs from the F100 and some code worked out for the touchscreen (TSLib and H/W driver) but I must confess I have no desire to really put much effort into blind supporting the F200 as I don't really want to be responsible for unnecessary bricking of F200's while we get support sorted .

There are also a few subtle differences with the NAND layout etc. on the F200 that will trip us up without hardware. Not to mention the fact that the Open2x bootloader (U-Boot) does not support the F200 (I can't fix that without hardware) and that bootloader makes booting test kernels from the SD card so much easier (and as such, development quicker etc. etc.). You get the idea .

 

[sbock]

What a pity... But many thanks you make this clear, John.

Regards,
Stephan

 

[gruso]

None of the main Open2x developers (Orkie, Lithosphere or myself) have an F200 so F200 support is not going to happen any time soon.
:unsure: So F200 support would happen if one of you guys had an F200? Or is there more to it than that?

 

[DJWillis]

:unsure: So F200 support would happen if one of you guys had an F200? Or is there more to it than that?

Almost there .

F200 support would be considered if one or more of us had hardware.
I do not want to say 'happen' as these things are never certain.

The changes that GPH made with the F200 make it much less appealing for a developer and lets face it we are all developers so there is little incentive to stump up cash for a new, less capable, GP2X (from our point of view).

 

[jbrodack]

I don't care much about touchscreen support but would like the f200 to be able to work with open2x. firmware 4.0 sucks and was hoping to have open2x to turn to.

 

[TrevorBradley]

I don't care much about touchscreen support but would like the f200 to be able to work with open2x. firmware 4.0 sucks and was hoping to have open2x to turn to.

I have to admit that developing on my F200 can be a pain and that the system is crippled in terms of external access. That being said Firmware 4.0 does suck and I"d love to see a faster replacement. I actually am doing touchscreen development and a number of new apps like the ScummVM interpreter and DosBox now have touchscreen mouse support. It's to the point where I would now see the F100 as a backwards step losing a critical feature. (I've never owned or even seen an F100, so I'm a bit biased)

I'm exceptionally patient. Get Open2x stable and superior to stock firmware on the F100, then some clever person will come and add in the features we need for the F200.

 

[DJWillis]

I have to admit that developing on my F200 can be a pain and that the system is crippled in terms of external access. That being said Firmware 4.0 does suck and I"d love to see a faster replacement. I actually am doing touchscreen development and a number of new apps like the ScummVM interpreter and DosBox now have touchscreen mouse support. It's to the point where I would now see the F100 as a backwards step losing a critical feature. (I've never owned or even seen an F100, so I'm a bit biased)

I'm exceptionally patient. Get Open2x stable and superior to stock firmware on the F100, then some clever person will come and add in the features we need for the F200.

I guess I should take that as a back handed complement . I added the touchscreen support to ScummVM blind with no access to an F200 (or any GP2X at the time). Lucky for us that GPH did not mess about with the sample code for the touchscreen that much .

Whist I can see your optimism, in all the years I have kept Open2x ticking over as a project it has never once been inundated with help. The reason it has got as far as it has is down to the tireless work of a few rather then any large push from many. I am just honestly not sure how many devout kernel hackers there are in the smaller F200 userbase .
I would love to see F200 support for my own sake of completeness but I am just not sure of the practicalities of somebody coming in to do it (and the reasons are clear why none of the current team are doing it). People also seem to forget that the touchscreen is not the only major hardware difference between the F200 and F100.

 

[JakeK]

I know this is going to sound strange, but can someone explain to me in simple terms what Open2x is? I read about it and all I can see is mumbo jumbo . I hear about it a lot and would just like to know exactly what it is.

 

[TrevorBradley]

I guess I should take that as a back handed complement . I added the touchscreen support to ScummVM blind with no access to an F200 (or any GP2X at the time). Lucky for us that GPH did not mess about with the sample code for the touchscreen that much .

Whist I can see your optimism, in all the years I have kept Open2x ticking over as a project it has never once been inundated with help. The reason it has got as far as it has is down to the tireless work of a few rather then any large push from many. I am just honestly not sure how many devout kernel hackers there are in the smaller F200 userbase .
I would love to see F200 support for my own sake of completeness but I am just not sure of the practicalities of somebody coming in to do it (and the reasons are clear why none of the current team are doing it). People also seem to forget that the touchscreen is not the only major hardware difference between the F200 and F100.

Just to terrify the uninitiated, what exactly would be involved for a new developer to the project to introduce F200 support? Is it more than just:
A) Touchscreen drivers
B) Touchscreen application for tuning
C) Flagging certain functions the F100 has off?

 

[DJWillis]

I know this is going to sound strange, but can someone explain to me in simple terms what Open2x is? I read about it and all I can see is mumbo jumbo . I hear about it a lot and would just like to know exactly what it is.

It is several things. To quote the Open2x Wiki

"Open2x is an umbrella project to bring open source, community friendly, solutions to the GP2X Linux handheld system. The project consists of a number of complementing sub-projects that work together to deliver things such as the Open2x toolchains, kernel and firmware.

The project is very heavily centred on its SVN repository but the project is actively working towards milestone public releases. "

In this context people are talking about the Open2x kernel and firmware. This is basically a replacement for the firmware GPH ship on the GP2X. The main points to note are that it uses a more recent Linux kernel, has newer drivers and is optemised in ways the GPH firmware is not.

The other key thing Open2x put together is the toolchain and libpack. This is used by a large number of developers to build and test there applications.

Just to terrify the uninitiated, what exactly would be involved for a new developer to the project to introduce F200 support? Is it more than just:
A) Touchscreen drivers
B) Touchscreen application for tuning
C) Flagging certain functions the F100 has off?

Ok,

This would be a very high level check list I would think. The only way to know for sure is to do it .

Touchscreen drivers - Check
Touchscreen calibration app - Check (Both these are not that difficult, in fact if somebody was interested I think I have most of this done)
F200 specific kernel build rules/config - Check
Kernel hacking to build an F200 specific kernel with touchscreen support (and no NetChip support)
F200 specific variant of the ROOTFS and changes to our build system - You maybe able to get away with one unified ROOTFS with work but you would need both an F100 and F200 to test really
F200 install utility
Clean, consistent way (in code) to tell F100 and F200 apart (something in the environment maybe?) - Not just checking for a touchscreen device .
Small changes to account for the subtle changes to the NAND layout
Numerous little changes here and there (#ifdef style stuff) in the Open2x support apps

Most of that assumes your already familiar with hacking the Linux kernel, have a little experience with YAFFS file systems and don't mind reflashing your GP2X every time you want to test a new kernel or ROOTFS change (fairly sure fire way to brick it ).

If you wanted to speed up development the other major task would be to merge the F200 changes to U-Boot (the bootloader) with the current Open2x bootloader so you have things like SD/SDHC booting on the F200. This is a fairly low level task and would require somebody who is familiar with jTAG (debugging and unbricking) and is not afraid to mess up there unit (I went through several F100's playing with the bootloader, some of them are still not unbricked to this day).

It is all very possible and could be done fairly quickly by the right sort of person. The problem is, I have yet to meet a person with an F200, loads of free time and a desire to try the above . If you think you are that person do get in touch B).