I'm probably just being a 'grumpy old man' feeling a bit nostalgic about the days of micros like the BBC and the Atari, when this sort of thing was possible, but here's a thought...

Is anybody familiar with the hardware and software inside our T816s? Anybody think it might be possible to dis-assemble the machine code back into assembly language, or, better still get the source code in C (or whatever it's been written in) so that it could be (possibly) modified and re-compiled and/or re-assembled?

If we could get hold of the original source code and a suitable compiler, and some info on the hardware, then we might have a chance of finding/fixing any bugs ourselves.

In other words, if there's no OAD coming for us from Vestel, can we help ourselves?

As an electronics engineer/computer programmer, I'd certainly be interested in finding out more about our PVRs, and it would be nice to be able to 'play around' with the firmware.

Can anybody throw any light on this?

Dave S

You might find the Rockbox project interesting (www.rockbox.org) - they have done what you say for digital music players (including the iPod) and the biggest problem for them is the special chips with no publically available docs.

I also wonder about any DRM/Copyright/Licensing type issues with the encryption used on certain channels.

It's a nice idea - I don't want to put you off, but the experiences of the MP3 players sets a certain context.

Not a chance of that. Even if you could extract the code from flash you'd still have the problem of getting the correct cross compiler for the chip used (ST5100 on the digihomes)

PVRs are complex beasts and there must be megabytes of assembly code in them. There's no way someone with no knowledge of the code structure would be able to disassemble and understand it.

You could always build yourself a PC based PVR, that way you could do all that you wanted. There are several good linux versions out there - MythTV springs to mind.

hi.

The Firmware for the Vestel units is ST20 based. There are freeware assemblers/disassemblers for this chip, which is usualy written with the intention of being used with a Linux type kernal. Since the drive on the T816s are formatted in a Linux compatable format, it is assumed that it will be also a Linux based machine. I have attempted to get details of the STM5100BUB chip inside the unit for many months. STMicroelectronics have stopped supporting the chip, and although the chip is available from Chineese electornics firms, non can supply any details, not even a spec. sheet. I therefore assume now, that Vestel might have bought up the 'rights' to the chip as well, making it impossible for easy access to it's details. I was informed by one firm, that a company called BSI Microelectronics had been given the rights to re-manufacture the chip, but would not at that time, release any details of it''s specifications.

Some attempts have already been made to disassemble the firmware, with mixed results. One thought the code was even someway 'encrypted' (?). However, someone must have managed to achieve the above, since other 'unoficial' versions of the software do appear on the web (most notably the dual record firmware version 1.0 for the previous T810 unit). Having a strong interest in the hardware architecture of the boxes myself, I have been frustrated by the lack of availability of relevant information out there. There should even be development boards available for these PVR chips, but nobody ever seems to stock them, let alone have a price for them. The firmware it'self is available if you ask Vestel for it, to download personaly to the unit via the RS232 port. After all, OADs are infrequent, if present at all!

It might be worth going through the other posts on the Digihome forum to see what else has been done - I would be extremely interested to know anymore about the hardware of the units, if more info. becomes available.

Regards,

Tim

To be honest, it's almost certain that these "unofficial" versions were internal development versions from Vestel, and were leaked either by a Vestel employee, one of the customers/partners or somebody involved with testing the software. I very much doubt the twin record firmware for the T810 was reverse engineered - it's just the first iteration of what later became the software for the T816.

Not that I'm saying that it's impossible to hack the firmware files - nothing is impossible . I'm sure there are lots of hidden variables in there that could be altered, which cannot be changed via the user interface.

Hi futaura.

Yes, you are most likely right. It is very hard to work from backward enginered code anyway.

I wonder if you can explain the new hardware revision numbers (Please see my post on Wharfedale boxes updated today). There appear to be two hardware versions now, a T825 as expected, and a T826 for the smaller HDD version. The firmware version remains the same for both at version 1.7 though.

Regards,

Tim.