Canon 500D / T1i

[gwhite0992]

I am new to ML and use it mostly for bracketing. Recently I wanted to take video and to my surprise, the software crashed after a minute and a half repeatedly. (err 70) I noticed that the display has so much info that I didn't need so I used the disp button to reduce the clutter(small movie camera bottom right). The crashing stopped. Hope this is helpful  (nightly 20160ct9)
All settings were the default settings. Has anyone else has success with this change?

[thearczoro]

Yeah, I notice that error while recording with default setting.

CRASH00.log

Code: [Select]

ASSERT: FALSE
at SoundDevice\SoundDevice_CODEC.c:1080, task ASIF
lv:1 mode:20


Magic Lantern version : Nightly.2016Dec19.500D111
Mercurial changeset   : 4f72a0eabb16 (unified) tip
Built on 2016-12-19 00:14:22 UTC by jenkins@nightly.
Free Memory  : 134K + 2985K


[Applenet]

Just trying out video, and thought I'd look to ML.

I'm down at the first hurdle. The T1i firmware on my camera is 1.1.0 and on the Canon site it's 1.1.2. ML require 1.1.1, right?

What's a noob to do?

[thearczoro]

Just trying out video, and thought I'd look to ML.

I'm down at the first hurdle. The T1i firmware on my camera is 1.1.0 and on the Canon site it's 1.1.2. ML require 1.1.1, right?

What's a noob to do?

Just download ML Nightly, it's included firmware 1.1.1. PM me if you need FW 1.1.1.

[Walter Schulz]

You may want to correct this statement.
FIR in nightly: ML-SETUP.FIR, 65916 Bytes
Used to install/deinstall cam's bootflag and making card bootable/non-bootable.

[thearczoro]

You may want to correct this statement.
FIR in nightly: ML-SETUP.FIR, 65916 Bytes
Used to install/deinstall cam's bootflag and making card bootable/non-bootable.

You're correct. I got it wrong. Btw, if someone need firmware v1.1.1 you can PM me, I have a copy of it. I will send FW by email. Later @a1ex maybe update ML for 500D with FW 1.1.2. 

[Applenet]

Thanks thearczoro, but no offence, you're new to the forum, so am I. I'm not comfortable installing firmware provided by a total stranger.

Is 1.1.1 available from a trusted source? (again, don't mean to offend thearczoro). And is that my only option?

[thearczoro]

Thanks thearczoro, but no offence, you're new to the forum, so am I. I'm not comfortable installing firmware provided by a total stranger.

Is 1.1.1 available from a trusted source? (again, don't mean to offend thearczoro). And is that my only option?

That's okay, I'm just a newbie here lol. Just because I'm a silent reader with few posts over 2 years. Just helping a total stranger too, who need help. http://builds.magiclantern.fm/ all here is a trusted source with full of explanation. Do It At Your Own Risk! 

[Greg]

End of 500D support.

[thearczoro]

End of 500D support.

  Thanks for your hard work @Greg, @A1ex, and the dev team for supporting our device.

[Walter Schulz]

Happy April Fools' Day, thearczoro!

[thearczoro]

Happy April Fools' Day, thearczoro!

  Sssstt... Make it more dramatic. Lol..  Btw, any news about firmware v1.1.2?

[Greg]

I do not plan future support 500D. This is an old camera, for me it has too many limitations.

[GenerallyNerdy]

New to ML and I have recently had issues with crashing. The error it puts on screen says something about missing a sound file. I only use my T1i in video mode. It seems that if I drop the frame rate to about 15fps the issue goes away, but anything 25fps and up throws the error. HELP?

ASSERT: FALSE
at SoundDevice\SoundDevice_CODEC.c:1080, task ASIF
lv:1 mode:20


Magic Lantern version : Nightly.2016Dec19.500D111
Mercurial changeset   : 4f72a0eabb16 (unified) tip
Built on 2016-12-19 00:14:22 UTC by jenkins@nightly.
Free Memory  : 141K + 2816K

[thearczoro]

New to ML and I have recently had issues with crashing. The error it puts on screen says something about missing a sound file. I only use my T1i in video mode. It seems that if I drop the frame rate to about 15fps the issue goes away, but anything 25fps and up throws the error. HELP?

ASSERT: FALSE
at SoundDevice\SoundDevice_CODEC.c:1080, task ASIF
lv:1 mode:20


Magic Lantern version : Nightly.2016Dec19.500D111
Mercurial changeset   : 4f72a0eabb16 (unified) tip
Built on 2016-12-19 00:14:22 UTC by jenkins@nightly.
Free Memory  : 141K + 2816K

Change your display for video with minimal info.

[cbeard7693]

I just updated my 500D firmware to 1.1.2, as there is no 1.1.1 available from Canon. Will ML not work with the newest FW update?

[Walter Schulz]

Please read installation instructions: Top of page -> Downloads -> Download Nightly Builds -> Select your cam.

[thearczoro]

I just updated my 500D firmware to 1.1.2, as there is no 1.1.1 available from Canon. Will ML not work with the newest FW update?

You need to downgrade first to FW v1.1.1: http://pel.hu/down/v111-t1i-500d-x3-win.exe

[dfort]

Knock knock, Hello? Any 500D users still out there? @a1ex - have you managed to brick yours? We're making some progress with 10bit/12bit raw recording on Digic IV cameras that use LVState which includes the 500D. If one of you could run a test build to get the SRM_BUFFER_SIZE for this camera we could give it a try.

Test builds are on my downloads page:

https://bitbucket.org/daniel_fort/magic-lantern/downloads/

Look for this file:

find-SRM_BUFFER_SIZE.2017Dec06.500D111.zip

Limited time offer. I prefer keeping my downloads page free of obsolete builds.

[Danne]

Got two of these babies atcwork. Will test tomorrow.

[a1ex]

Ran it in QEMU - it flashes too fast (you have replaced the printf with a bmp_printf, right?), but it's 1AE followed by some zeros. That should give around 27 MB, so it's 0x1ae0000.

NotifyBox will redraw the message in background, should it get erased, while bmp_printf just prints it once.

[dfort]

I was using bmp_printf because of the control over where to print on the screen. I moved it around so it wouldn't interfere with the other overlays but I suppose that on some cameras they still clash. Switched over to NotifyBox and gave it a timeout value of 10000 so it should stay on the screen long enough to read it.

Posted a new build on my downloads page.

Off topic--how did you shoot a silent still in QEMU? According to the documentation:

What does not work (yet):

• LiveView (WIP, very hard);
  
• Still photo capture (WIP - the capture process itself works);

[dfort]

Got two of these babies atcwork. Will test tomorrow.

Try the new and improved find-SRM_BUFFER_SIZE build I posted and see how it works on the camera.

By the way I made these builds so anyone who can install ML can help find SRM_BUFFER_SIZE for their camera. The way it works is like this, check consts.h file for a camera that has SRM_BUFFER_SIZE working (look in either crop_rec_4k or raw_video_10bit_12bit branches) and you'll see a hint at how to find it:

Code: [Select]

#define SRM_BUFFER_SIZE 0x1AE0000   /* print it from srm_malloc_cbr */
The srm_malloc_cbr function lives in src/exmem.c - I just added that last line that uses NotifyBox to print the SRM_BUFFER_SIZE onto the screen:

Code: [Select]

static void srm_malloc_cbr(void** dst_ptr, void* raw_buffer, uint32_t raw_buffer_size)
{
    if (!srm_buffer_size)
    {
        /* we can't tell how much to allocate; the allocator tells us */
        srm_buffer_size = raw_buffer_size;
    }
    else
    {
        /* it should tell us the same thing every time */
        ASSERT(srm_buffer_size == raw_buffer_size);
    }
   
    /* return the newly allocated buffer in the output variable */
    *dst_ptr = raw_buffer;
   
    /* announce it's done */
    give_semaphore(srm_alloc_sem);
   
    NotifyBox(10000, "SRM_BUFFER_SIZE 0x%x", srm_buffer_size);
}

@a1ex - I didn't replaced the printf with a bmp_printf, because there was no printf there to begin with. I didn't use printf in my builds because that involves using the console and I wanted to make this as easy as possible to use.

[a1ex]

To get this value, I didn't even have to take a picture - just went to the Free Memory dialog, which does a test allocation.

Taking an emulated full-res capture with the vanilla silent module is doable, but requires a gdb script (or a custom build) to fake the LiveView status. One such test is ran on the nightly builds (see this screenshot and this animation, or this and this for 500D, although not all models can run it that well) and another one (with a minimal codebase) is ran with the QEMU test suite (qemu-frsp).

In the crop_rec_4k branch there is a printf commented out.

[dfort]

To get this value, I didn't even have to take a picture - just went to the Free Memory dialog, which does a test allocation.

Interesting. The new build with NotifyBox is working great with the Free Memory dialog trick.

Thanks for posting those QEMU tips. Hum, a gdb script (or a custom build) to fake the LiveView status? So much to learn here.

In the crop_rec_4k branch there is a printf commented out.

That figures. I did it using the raw_video_10bit_12bit branch and only printed out the SRM_BUFFER_SIZE. Looks like there's more information in the crop_rec_4k branch code.

Code: [Select]

//printf("srm_malloc_cbr(%x, %x, %x)\n", dst_ptr, raw_buffer, raw_buffer_size);
Now if I could only get all this stuff for the Digic IV 10bit/12bit stuff working in the crop_rec_4k branch. Back on topic -- and that includes the 500D.