Messages

FPS register values (for enabling FPS control in ML). posting it here so it's documented somewhere.

Code Select Expand


NTSC:
LV (and all non-recording modes):
       TIMER_A: 0x20f020f
       TIMER_B: 0x7e5 / 0x7e6 [changes between the 2 twice a second]
       SHUTTER (1/60): 0x3B2 (same in all recording modes too)

Recording:
1080p 30fps:
       - same as LV conditions

1080p 24fps:
       TIMER_A: 0x20f020f
       TIMER_B: 0x9de / 0x9df

720p 60fps:
       TIMER_A: 0x20f020f
       TIMER_B: 0x3f2 / 0x3f3

480p 30fps:
       - same as LV conditions



PAL:
LV (and all non-recording modes):
       TIMER_A: 0x20f020f
       TIMER_B:0x7e5/0x7e6
       SHUTTER (1/100): 0x25e

Recording:
1080p 25fps:
       TIMER_A: 0x27f027f
       TIMER_B: 0x7cf (doesn't change)
       SHUTTER: 0x1f4

1080p 24fps:
       TIMER_A: 0x20f020f
       TIMER_B: 0x9de / 0x9df
       SHUTTER: same as lv

720p 50fps:
       TIMER_A: 0x27f027f
       TIMER_B: 0x3e7 (doesn't change)
       SHUTTER: 0x1f4

480p 25fps:
       TIMER_A: 0x27f027f
       TIMER_B: 0x7cf
       SHUTTER: 0x1f4

Found a memory address that changes rapidly as i swipe my finger across the screen, so, this is the start of implementing touch buttons to ML!!

0x4D868 --> represents pin-point area on screen (x and y pixels) being touched (some mapping of high/low to represent left/right/top/bottom of screen)
0x4D810 --> value=0x11100 when screen isn't being touched, value=0x11101 when 1 finger is held touching the screen
            --> value=0x11102 with 2 fingers touching the screen
            --> value=0x11103 with 3 fingers
            --> value=0x1104 with 4 fingers! Note: only the LSB seems to be used here, other bits seem to change sometimes. but the rightmost bit always changes to match how many fingers are touching the screen.


the first address, 0x4D868 seems to map the screen like this:

top left corner of the screen = 0x0000000
top right corner = 0x00002CF
bottom right corner = 0x1DF02CF
bottom left corner = 0x1DF0000

so, lower 3 bits represent the X coordinates, from 0 to 719 (720px wide)
middle bit is always 0
upper 3 bits represent the Y coordinates, from 0 to 479 (480px tall)


Looks like touch screens won't be hard



also, I found something related to audio levels, these 2 addresses change with ambient noise, so they may represent audio levels:
0x3F240
0x3F244

Suggestion:
- long SET press: open ML menu
- L/R/U/D/wheel: navigate (like it's now)
- PLAY: open submenu
- REC: see thru the menu (e.g. for white balance)

What does REC do in photo mode?

REC doesn't send an event to gui task unless in movie mode, I tried.

I'm not sure how to implement long press but could be a good idea. the center button acts like Q in shooting mode / video mode, and set button in canon menu.

good work Coutts!

do allow other developpers working on the EOS M, why not adding a dump feature and providing a .FIR that enable the bootflag ?

Indy

I'll get working on the installer asap



I found the LV buffer addresses, so now ML graphics work (histogram and zebras for now to test).

BTW menu works and can be walked. Need to figure out how to work with the small amount of buttons we have (until later when we unlock how to use the touch screen). up/down/left/right can walk the menu, play/set either toggle a setting or open a sub menu. haven't enabled any features yet.

Nice pics!
Why the double text?
I'm agree with Roger, and if ML can fix autofocusing speed (by the way I had bad hangs after enabling autofocus in ML on 600D) I'm really tempted, a price drop will be perfect! (canon, are you reading?)

that's how it was idk. i didn't change the CONFIG_HELLO_WORLD code at all.

ML works! check the first post

The dump method we used should work on the 650D as well thanks A1ex!

We now have a full firmware dump so I'm expecting by tonight (EST) we should be printing "Hello World!"

[Update 11/25/2012:]

The EOS M port starts now! g3gg0 successfully dumped the firmware and set the bootflag, so we're all set to begin porting


--> More updates coming soon

Update 11/25/2012:







Update 12/10/2012:
Alpha1 release! Let me know what bugs you notice. Almost everything that is going to work on the M is enabled now, minus audio stuff (need more work on that). Some features won't ever work though. ie: Trap Focus - because the M doesn't send focus confirmation when in manual focus.

Update 12/10/2012 #2:
- Fixed installer (thanks Alex), no more unsafe operations (ie: try to set bootflag while in live view).



Install Procedure
Refer to Unified install page:
http://magiclantern.wikia.com/wiki/Unified/Install


Warning: I'm not responsible for anything that might happen. This is an EXPERIMENTAL build. Please report bugs in this thread.
Download Alpha 1: Here

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.

Hi,

What do you think? It is possible there will be a higher ISO value on 5d classic?

M

anything higher than I think ISO800 is "digital gain" iso, as in it's the same thing as pulling the exposure in post processing. things are already pretty noisy with ISO3200, even so: I couldn't push it any higher than 3200 (but probably can be done with lots of reverse engineering).

Can you point out the unused connections?

Unfortunately the TX chip in that package doesn't expose the JTAG pins :/

here are my pics from the disassembly (including pictures of all the chips with metal shields removed):
http://imgur.com/a/iZq4a

Well, how about we look for that JTAG port? :p

I don't know anything about jtag or how to use it, but i saw 3 connection spots not plugged into anything. I think somethings really wrong with the camera (like hardware damage or something). when I was diagnosing it would act random sometimes (one time it actuated the shutter when i turned it on). I still have no idea what happened but some stuff got pretty messed up (g3gg0 even suggested that the Digic [main processor] may be dead).

As some of you already know, I bricked my 5dc yesterday while testing some new boot code. I tried to diagnose and fix it but unfortunately it's out of our reach for reviving. The bootloader appears to be toast, so canon firmware doesn't boot and no bootloader = nobody to boot autoexec.bin files. Basically we can't do anything since we can't run autoexec.bin files. It essentially acts as if it is soft bricked (like many early 5dc testers experienced), except there is no fixing it - pulling the battery has no effect. As a result, the 5dc port is now on hold until further notice until someone with a 5dc can take over development.

I disassembled the camera to see if anything was fried, but I didn't see anything. I'm sorry everyone, this is just a case of bad luck (I wasn't even trying anything dangerous, it just bricked). It was fun while it lasted (and my 5d too, i just bought it in February..). Now I just have this new paper weight:



So... who will take over now?

update:
got it put back together fine. it was actually easier to do than I thought, just takes a while (it's like a puzzle kind of). it turned on once, but no buttons worked, either did the screen. I noticed there wasn't any led activity when i hit the power button either, which says that neither the bootloader or firmware work.

after that initial power-up, I couldn't get it to even half-boot like that first time. It's back to being a paper weight now.

[View all pictures from my disassembly here:]

Well sadly this morning while trying some cache_hack code, my 5dc appeared to be bricked. Basically it acted like there wasn't a battery in it despite the fact the battery was charged. So, I began disassembling to see if anythings noticeably wrong. Here's some of my findings.

View all pictures from my disassembly here:


(Full-resolution 5mp photos: Here )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
--> Note, this is a very good PDF to have if you are going to be disassembling a 5d classic:
          https://dl.dropbox.com/u/33161628/ML/Datasheets/EOS_5D_Parts_Catalog.pdf
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

TX19A (MPU) Chip



Toshiba TMP1942CZU
CH4-6311-000
0707AA JAPAN

Datasheets:
BCE0023_catalog.pdf
TMP1942CZU.pdf




K213 Microcontroller



Toshiba T6H82CG-104
JAPAN 0548HAL
021725

Datasheets:
** Refer to BCE0023_catalog.pdf linked above **




Digic / SDRAM



Canon DIGIC 2
4x 512Mib SDRAM (256mb total)
021725

Datasheets:
IC-ON-LINE.CN_edd5116adta-5cli-e_4529134.pdf


[ ..more coming soon ]

I used the 350d project a good deal to get the 5dc port off the ground. Probably wouldn't be hard to port now that 5dc / 40d are in the ML tree and CONFIG_VXWORKS exists. don't own a 350d though sorry.

Coutts, u r a hero!

For the 5D Classic, would you please add the False color feature as I find it very handy to check portrait exposure
when using TTL flash under backlight situation. It is amazing! The exposure would be spot on and much less
post-production work   Tks!

I'll look into that when I get time

By the way, beta 5 will be coming this week with updated boot code. g3gg0 and jplxpto made the cache_hack boot method work on the 40d, so I'll be adopting that to the 5dc (and fix some bugs I found over the weekend using ML).

Used the 5dc ML beta 4 last night for the first time in real-world application.

Shot about 1100 pics at a concert last night (with Steve Aoki there, cool!). Experienced quite a few bugs, so I'll be fixing those this week. Look forward to a beta 5 coming soon!

You shouldn't have claimed you had an idea on how to hack the M well before canon releases an update (although the same was done on the 1100D)

yea now canon is making a patch for us

i'm itching to get my hands on this camera

http://www.canonwatch.com/canon-eos-m-shipping-in-the-usa-delayed-to-end-of-october/

So looks like I'll be waiting until then to get my hands on this camera..

We have a 5D3 now and start on that since the 650D/t4i requires a firmware update to get started.

nonsense
I should be getting my EOS M within the next week, and I plan to have ML started on that well before canon releases a firmware update.


plan:
- use 5d3 dump (both cameras have Digic5) to map the bootloader functions for reading / writing the bootflag (like the method I had to use for the 5dc).
- blindly find the matching functions in the EOS M bootloader (all i need to do is find the LED address [brute force method]).
- check a few lines near cstart and verify that the boot code is identical to the 5d3 (expecting it to be).
- apply boot code to EOS M like it's done on the 5d3


It's like a challenge / game for me

not sure about DryOS but in VxWorks like the 5dc there are functions to simply pause a task, or halt it. Maybe that's less destructive than deleting the task.

>I'm getting the EOS M here in a few weeks, so once that comes I will be diverting my time to porting ML to that instead of the 5dc.

Nooooooooo!     

Please don't leave us for good. We need you! By the way, does the 5DC Port will be merged into the maintrunk of ML? Of course with less functionality...

it's been merged into the main sourcebase since june
https://bitbucket.org/hudson/magic-lantern/src/9dbfdc26c53f/platform/5DC.111

I even might; got nothing to do all weekend...

Take time to research what you are doing, this isn't something I would just dive into / start running code. That's a fast way to brick your camera.

It's taken me 2 years to learn what I know now, with countless nights without sleep (pulling 40+ hours sometimes without sleep working on ML). If you are serious about helping, I will help you as I don't have the time to work on ML every day now. it would be nice to have someone else working on the 5dc.

I'm getting the EOS M here in a few weeks, so once that comes I will be diverting my time to porting ML to that instead of the 5dc.