Toshiba Gigabeat T

[LambdaCalculus]

This thread is for discussion of porting Rockbox to the Toshiba Gigabeat T.

I just received one from saratoga, and would like to get a porting effort going (and *hopefully* stick to it! )

The Gigabeat T is largely the same as the Gigabeat S (or Gigabeast) internally, with the same i.MX31 CPU and Wolfson DAC. However, it's flash-based as opposed to HDD-based.
There's a firmware update available on the GigabeatTInfo wiki page. Its layout is similar to
other Windows Mobile-based Gigabeats (S and V). However, what I need to look for is if it has
a similar digital signature exploit as the S did.

If there is, then porting should be a snap, as we would need a bootloader and tweaking the beast
code to use flash instead of ATA. Oh yes, I also need to see how to access the flash, of course.

[scharkalvin]

What ever happened to your Rio Karma?

The Gigabeat T sounds interesting, though I don't recall ever seeing
any on Ebay.  I've seen LOTS of F's and a few S's though.

[LambdaCalculus]

Good news on the Gigabeat T port: thanks to toffe and saratoga, we can pretty much confirm that the digital signal exploit from the Gigabeast is indeed present in the Gigabeat T, and also that the firmware and data partition structure must also be the same.

We'll be investigating further.

[IrQ]

What ever happened to your Rio Karma?

The Gigabeat T sounds interesting, though I don't recall ever seeing
any on Ebay.  I've seen LOTS of F's and a few S's though.

look in this: http://page5.auctions.yahoo.co.jp/jp/auction/e95381523

by the link 401 model (4GB)
802 model is selling only in Japan shops; http://page17.auctions.yahoo.co.jp/jp/auction/v110959011
802 differs from 401: 8GB memory and WiFi builtin. In original firmware (Windows Embeded) it used only to connect to japaneese analog youtube with anime.

P.S. Sorry for my English.

[LambdaCalculus]

I've seen those players as well, but currently I'm concentrating on the MET400 (the US model, as that's what  I have). So in essence, when a port is completed, it will not have WiFi support, unless someone wants to donate a Japanese Gigabeat T model.

The best thing to do is to look at a firmware update for the MET802 and see if there's still a digital signature exploit on it as well. The exploit is what got us Rockbox on the Gigabeat S series, and may very well get us Rockbox on the T series as well (as they share much of the same hardware).

[v1rus]

Hey Lambda.

I happen to have one of these players. If there is anything I can do with mine (tests, etc.) let me know.

most of my "hacking" experience involves arduino based automation of photography rigs, so I imagine I am not much use, but I am just putting it out there.

[LambdaCalculus]

Currently, I haven't had very much of a chance at all to sit with the Gigabeat T again and try to do something with it. I work two jobs, have a relationship, am studying, and much more to do.

When I can actually get a little "me" time to sit down, I'll work on the Gigabeat T. No promises until then.

So in short, no more "When will it be done?" questions, please.

[v1rus]

Hey, the last thing I mean to do was rush you. I was just checking if you might have had some grunt work I could help with.

Take your time. I am in the same boat as you with regards to my free time. It's rough.

In the mean time I will try to get my hands on an arm chip and poke at it.

[LambdaCalculus]

In the mean time I will try to get my hands on an arm chip and poke at it.

The first and foremost thing that needs to be done is to see if the same (or a similar) digital signature exploit is present on the T's firmware. As I stated before, there was a signature exploit on the S's firmware that allowed us to inject custom code, and from earlier findings, there may be one as well on the T, but I haven't had much time to go over my disassembly of the T's firmware in order to confirm it.

If someone wants to help out by doing an arm-elf-objdump of the T firmware and going through it, please do. I'll do the same when I can actually sit down and do so.