Toshiba TX19A discussions on nikonhacker.com

[Indy]

Hi,

Just to let you know, Nikon hackers are discussing about the TX19A chip here:
http://nikonhacker.com/viewtopic.php?f=2&t=214
http://nikonhacker.com/viewtopic.php?f=2&t=167
this chip is also in our camera body.

ML Wiki page is here:
http://magiclantern.wikia.com/wiki/Tx19a

IDA Pro edition (with MIPS support) is required for TX19A. Starter edition does not support MIPS

could we add AF micro-adjustment on xxxD bodies if we understand TX19A code ?

Indy

[ilguercio]

Faking a lens would be even better.

[nanomad]

Provided we can patch tx19 code, we can do pretty much whatever we desire

[ilguercio]

Is IDA PRO coming soon on A1ex's machine?

[nanomad]

Good question, I have lost track of the IDA status to be honest

[1%]

Better yet, full I/O control.

[Indy]

on the paper, "yes we can" patch TX19A, because:
- TX19A is updated : http://magiclantern.wikia.com/wiki/Update_records#SRecord_files
- we can create .FIR files

but it is very risky : bricking probability is very high

Indy

[ilguercio]

on the paper, "yes we can" patch TX19A, because:
- TX19A is updated : http://magiclantern.wikia.com/wiki/Update_records#SRecord_files
- we can create .FIR files

but it is very risky : bricking probability is very high

Indy

What do you think, could you theoretically upgrade the firmware from one camera to another?

[nanomad]

I see little to no point in that, most cameras have different sensor / AF pixels arrangement

[1%]

Maybe overclock SD mhz somehow? But if you kill TX19 then camera is dead to the world, no IO will happen.

[Indy]

What do you think, could you theoretically upgrade the firmware from one camera to another?

there is no camera difference, one update is applicable to all camera of the same model.
we can patch one TX19A record in an update, repackage the update file, update the camera, thus pushing our TX19A patch.
this is theory, in practice we'll have to understand TX19A code, which is not easy.

[nanomad]

I think he meant camera as in model 

[a1ex]

That would infringe on Canon's copyright => no

[1%]

Not that much in the firmware to risk flashing that first camera and have it not work. Maybe between same models like 1100D and 600D... but then the screens are different and you'd lose that. Better off altering I/O and making fake lenses... would put that whole dandelion chip making industry out. I bet canon has it like openwrt where you compile the modules from the source tree needed for each camera. We'd have to make custom firmware to cross flash and we don't have the source and plus the copyright issues on top.

[ilguercio]

That would infringe on Canon's copyright => no

Yeah, i know, it meant feasible in a potential way, law aside.

[g3gg0]

alex and i have talked about buying the pro version for tx19a.
tx19a is interesting, but far beyond ML scope.
also patching is too risky imho and i am not sure what the advantages will be.

so we decided to not buy a pro version but use the money to buy starter editions for some devs.

[ilguercio]

alex and i have talked about buying the pro version for tx19a.
tx19a is interesting, but far beyond ML scope.
also patching is too risky imho and i am not sure what the advantages will be.

so we decided to not buy a pro version but use the money to buy starter editions for some devs.

[Indy]

That would infringe on Canon's copyright => no

if we distribute an binary patch against a canon update, we do not ourself distribute Canon code...

[ilguercio]

Well, since this is a common chip to be investigated what if you team up with them? Both of you have the interest in acquiring informations.

[jplxpto]

http://www.toshiba.com/taec/components2/Datasheet_Sync/134/26203.pdf

[a1ex]

if we distribute an binary patch against a canon update, we do not ourself distribute Canon code...

I'm not that sure - the end result would be a modified version (derivative work) of Canon code, and some users might be tempted to post it online.

[ilguercio]

Have you tried contacting them?  I know that hardware is different on a Nikon but since the chip is the same there is a big piece of work that can be done together, probably?

[coutts]

the TX19A chip is still alive on my 5d, could be used for testing if someone wanted (i can send the 5d to you). though, the firmware doesn't boot so you can't test much.

[coderat]

Have you tried contacting them?  I know that hardware is different on a Nikon but since the chip is the same there is a big piece of work that can be done together, probably?

Guys, we have TX19A fully simulated in Emulator. I would not buy IDA for TX19A - it is very expensive and has bugs. Also a problem to generate correct code for TX19A. But you can try our TX19A disassembler, just download NikonEmulator 2.50 and look in startDtx.bat and use something like

startDtx.bat -wstructure -wparameters -wmemory -wshift -wli -wquestion -wdmov -wbz -wret -waddress -whexcode -x firmware.dtx.txt -o firmare.asm firmware.bin

For syntax of firmware.dtx.txt you can execute "startDtx.bat -m?" or "startDtx.bat -w?" or look inside our a640m010100.dtx.txt.

Cheers,
coderat

[ilguercio]

Anybody gave a look at their forum?