Author Topic: Danne's crop_rec_4k experiments for EOS M  (Read 212912 times)

critix

  • Contributor
  • Member
  • *****
  • Posts: 183
Canon 1300D, 500D, EOS M, EOS M2

roogii

  • New to the forum
  • *
  • Posts: 5
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2151 on: February 06, 2020, 07:04:18 PM »
Is it possible to shift the position of the 3x crop mode anywhere on the sensor? I have a schneider 6-66  super 8 lens that doesn't cover even the 4.6x crop, for some reason I'm not central in the image circle though.

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7368
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2152 on: February 06, 2020, 07:50:06 PM »
Check post the recent two months. A user did just that.

2blackbar

  • Hero Member
  • *****
  • Posts: 507
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2153 on: February 08, 2020, 10:14:06 AM »
That super 8 lens might cover 3x crop mode on 66mm range, maybe not in 1920x1080 , maybe in 720x*** mode , i did not tried any super8mm lenses because they obviously wont cover full area , you need cctv lens for 1/2'' sensor at least or 16mm film camera lenses.
I do have personalised ML build for canon M with centered mode for my cctv lens BUT, i have 2 adapters for CCTV mount(c mount) and one is slightly shifted , other is centered more ( or vice versa) so not all adapters are equal.With that being said you still can do micro adjustments to crop by moving i left/right with arrows onscreen in 2.5k mode which helps in some cases.

tomcarrolls

  • New to the forum
  • *
  • Posts: 6
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2154 on: February 09, 2020, 07:23:22 PM »
Hi guys! So my EOS M is doing great with 10 bit Raw, unless I record with an external monitor, then this happens (purple artifacting, screen tearing): https://drive.google.com/file/d/1ijDakNgj66t5jPWaaaI2wCg5DEqXyoaF/view?usp=sharing

I tried turning on VGA mode or whatever it's called, no luck. Will try turning off histogram and other overlays, see if that helps.

tomcarrolls

  • New to the forum
  • *
  • Posts: 6
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2155 on: February 09, 2020, 08:13:59 PM »
Okay the only way I could get it to work well was to switch my in camera FPS to 24. It was on 30 before this. I think that was the problem.

roofusj

  • Just arrived
  • *
  • Posts: 1
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2156 on: February 10, 2020, 08:52:08 AM »
Hi everyone. I have a problem using canon eosm and magic lantern latest BUILD "crop_rec_4k_mlv_snd_raw_only_2019Dec08.EOSM202"
I am using lexar 128gb 1667x sdxc memory cards.
When i am trying to shoot raw videos 2.5k or even 1920*1080 camera doesn't record a video it stops every time at the beginning because of the high bitrate > card seems to be isnt able to record a bitrate higher than 16 m/s. But after a several attempts to shoot at the speed 16-20 m/s card starts to record a higher bitrate about 45-50m/s but never at the beginning. Does anybody has this issue with eosm and lexar or other cards? Thanks for answers!

I'm using a Lexar 1667x 64 gb V60 card and i have similar behaviors from it. At first it won't record more than 3 or so seconds in 5k rewire mode. Then after 3 or 4 failed attempts it starts recording just fine and as long as i don't turn it off it will keep going like a champ almost getting to 60 MB/s with constant recording. It works fine in mcm rewire.

Mythic Cinema

  • New to the forum
  • *
  • Posts: 33
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2157 on: February 10, 2020, 07:32:46 PM »
x5 Full Spectrum Canon EOS M (Blue) | x1 Full Spectrum Sony a7S III (Soon)

alexeym

  • Just arrived
  • *
  • Posts: 1
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2158 on: February 11, 2020, 09:58:13 PM »
I'm using a Lexar 1667x 64 gb V60 card and i have similar behaviors from it. At first it won't record more than 3 or so seconds in 5k rewire mode. Then after 3 or 4 failed attempts it starts recording just fine and as long as i don't turn it off it will keep going like a champ almost getting to 60 MB/s with constant recording. It works fine in mcm rewire.
Just about a week ago I tried Lexar (I think it was 1800x or something very similar to yours) and experienced exactly the same behavior.
A bit of remedy was to use "Warm up card" option in menu. I think 256MB size worked for me.
It improved the issue a lot, but occasionally the card was still not able to start recording at higher speed. Also boot up time was much longer due to "Warm up" setting.
I also tried Samsung Evo with very similar issues.
Eventually I returned the card and ordered Sandisk Extreme Pro 128GB. Never had the issue since.

On a side note. It seems to me in 5K mode, 12 bit changes resolution. 14 or 10bits have different one. Is it a bug or I'm still missing something after reading this whole thread.

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7368
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2159 on: February 12, 2020, 04:16:20 PM »
On a side note. It seems to me in 5K mode, 12 bit changes resolution. 14 or 10bits have different one. Is it a bug or I'm still missing something after reading this whole thread.
Changes to keep continuous recording.

rh224

  • New to the forum
  • *
  • Posts: 2
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2160 on: February 17, 2020, 09:27:39 AM »
Just about a week ago I tried Lexar (I think it was 1800x or something very similar to yours) and experienced exactly the same behavior.
A bit of remedy was to use "Warm up card" option in menu. I think 256MB size worked for me.
It improved the issue a lot, but occasionally the card was still not able to start recording at higher speed. Also boot up time was much longer due to "Warm up" setting.
I also tried Samsung Evo with very similar issues.
Eventually I returned the card and ordered Sandisk Extreme Pro 128GB. Never had the issue since.

I just experienced similar issues with one of my Samsung Evo cards and my M for the first time today. They've been rock steady for continuous recording in 14/12bit when used with both my M and my 6D (Use MCMRewire @ 1736 x 976 on the M). Today, it recorded two clips of about 2:30 on the M just fine and afterwards 5 out of about every 6 runs would give me a max record time of 4-5 seconds (even dropping to 1600 x 900 @ 10bit) and randomly I'd get rock solid continuous recording. Power off/on, pull battery, reset ML settings, reset camera settings, turn on Card Warm Up, no difference or pattern I could see.

It was a family outing and I hadn't planned to shoot much at and just happened to have the M and card with ML, so I couldn't do much testing. I plan to do some more troubleshooting when I have time  It's possible the card is going bad, but it's probably only the 4th or 5th time I've used it. I should probably just quit being stubborn and buy some Sandisk cards...

Also, I've got two Lexar 64GB 1000x cards that have never worked right in my M with the SD overclock.  They work very well when used with my 6D however. Never figured out why.

P.S. This is my first post here, though I'm no stranger to ML.  Long time listener, first time caller sort of deal... I can't post with out a shout out to the community here and the incredible knowledge, talent and accomplishments pushing the limits of what this amazing little camera can do. Thank You!

Capsulepill

  • Just arrived
  • *
  • Posts: 1
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2161 on: February 17, 2020, 09:10:52 PM »
i'm having some trouble with the latest version of dannes ML build for eos m (JAN-18-2020). i'm pretty new to using magic lantern.
but everytime i try and press the rec button to start a video recording, it doesn't do anything, the last build i used allowed me to use the trigger. but the newest build doesn't, or i'm not aware of the setting to put it on.

i don't know if i'm doing something wrong, but if anyone could help i would appreciate it!

[UPDATE] I figured it out guys haha. if your also having the same kind of trouble that i wrote down here
in order to fix it, you have to go to the canon functions menu, and In Custom function menu: C.Fn-7 Release shutter without lense: on

this allows you to use lenses not recognised by canon camera systems, like the vintage lenses im using.

Bernie54

  • New to the forum
  • *
  • Posts: 8
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2162 on: February 18, 2020, 12:02:27 PM »
How to use Manual mode and JPG on EOS M with magiclantern-Nightly.2018Jul03.EOSM202?
When using this build on EOS M with FW 2.0.2 JPG-shooting in M mode is disabled if modules are loaded but switched off.

Issue description:
When cam is set to M mode and shutter is pressed half, the display is cleared and the ML message "Set picture quality to RAW" appears. When shutter is pressed full no shoot is taken and cam returns to default display. If mode is set to RAW a silent full res picture is taken - even if this is set to OFF - and the cam hangs up and must be restarted. Only if ML is loaded without modules the cam works like expected.

Used ML settings:
Serveral modules like adv_int, file_man, mvl_lite, silent, ... are loaded but all ML functions are set to OFF, except some overlays .

Expected behavior:
Just shoot JPGs in manual mode when modules are loaded but the corresponding functions are set to OFF.

Any hint is appreciated.

cmh

  • Contributor
  • Member
  • *****
  • Posts: 138
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2163 on: February 18, 2020, 11:35:58 PM »
It's probably the same shutter bug that I got my silver EOS M. Try turning it off and on two or three time real quick (while the led blink right before the display is on).

Bernie54

  • New to the forum
  • *
  • Posts: 8
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2164 on: February 19, 2020, 02:14:19 PM »
@cmh: Thank's for the reply.

Meanwhile I got rid of the issue with a fresh reinstall of magiclantern-Nightly.2018Jul03.EOSM202 to the SD card.
Because I had tested another very experimental build on the same SD card before, my assumption is that some config stuff was left over that conflicts with the nightly build. But I don't have a proof for that and the shutter bug thing might be a root cause too.

ArchGalileu

  • New to the forum
  • *
  • Posts: 2
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2165 on: February 22, 2020, 09:34:58 PM »
 Hello all

Great work Danne ;)

In 5k anamorphic I can't see the overexposed zebras, any advice?

Best regards
ArchGalileu

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7368
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2166 on: February 24, 2020, 08:03:41 AM »
Not really. Turned them on? I have zebras turned off upon first install of magic lantern.

keel

  • Guest
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2167 on: February 24, 2020, 05:42:40 PM »
Can I get a quick breakdown on "rewire".

Specifically this 5k vs 5k rewire, or more info on rewire in general.

I assume both are strictly for anamorphic lenses.


Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7368
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2168 on: February 24, 2020, 06:43:55 PM »
Start readig from the download post. It is expected for newcomers.

roogii

  • New to the forum
  • *
  • Posts: 5
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2169 on: February 24, 2020, 10:35:29 PM »
context: my preferred mode is mv1080p MCM rewire x3 crop, 1736x1158 max resolution 3:2. The full aps sensor rainbow aliasing is too much, and I have perfect alias free 5D3 for big sensor stuff anyway. I like having proper real time preview of the real framing. I treat the MagicM as my 100£ ikonoskop, incredible.

Danne:
1. is it only your build that gets the increase in vertical resolution in mv1080p MCM rewire 1158px vs 1024px in the main website build?
2. in mv1080pMCMrewire x3crop, is it 1:1 pixel sensor readout without any skipping/ binning?
3. if i enter 2.5k mode and set 1736x1158, am i theoretically getting exactly the same output as mv1080p set at 1736x1158 x3crop?
4. what is the difference between 2.5k 1:1 centred mode and 2.5k 2520x1418 >> is it just extra vertical available pixels otherwise the same?
5. if mv1080pMCMrewire x3 crop is 1:1 sensor just like 2.5k mode, is there anything stopping increasing the resolution of the x3 crop within the mv1080 mode whilst retaining realtime full framing preview?
6. I also use your 5d3 build, i like using the same full 3:2 1080p mode, and if i setup clear overlays at halfshutter for example, the magic lantern bars disappear and I get full 3:2 realtime preview. On the M the magic lantern bars disappear leaving black bars in its place, with a 16:9 preview still. Am i right in thinking that the 5d3 actually uses a 3:2 preview through the buffer whereas the M is streaming a 16:9 native preview? or would it be possible to clear the overlays on the M to reveal a 3:2 liveview.

2blackbars >> is that you synthsundry? So you enjoy C lens too, did you play around with moving the framing in the standard mv1080pMCMreire x3 crop mode?

Oh...Danne, I have four of these M cameras and I assume there must be some kind of hardware variation, because using all same settings and build in mv1080MCMrewire, 2/4 of my bodies corrupt the image when changing into x3 crop, video >>>



Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7368
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2170 on: February 25, 2020, 12:15:38 AM »
I think your glitch is fps related. Either canon fps mode or fps override or set 25 fps is enabled causing the issue in x3 crop.

keel

  • Guest
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2171 on: February 25, 2020, 03:59:49 PM »
Anyone ?

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7368
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2172 on: February 25, 2020, 06:51:14 PM »
Anyone ?
Try back track a few pages. If you pay attention you´ll soon see that there´s a download post referred to on a regular basis. Find that page and start reading from that post and onwards. Most of your questions will be answered really fast.
You don´t mention any info about what build you are using or where you found it? Sources needed.

roogii

  • New to the forum
  • *
  • Posts: 5
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2173 on: February 26, 2020, 07:30:09 PM »
Ok that got it, so that's what set 25fps is for - matching the Canon menu.  The 2 other bodies are fine being mismatch set though.

Other questions too boring? I read the whole thread,    those questions didn't seem answered at least on my passing and in my head.

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7368
Re: Danne's crop_rec_4k experiments for EOS M
« Reply #2174 on: February 26, 2020, 07:55:22 PM »
Ok that got it, so that's what set 25fps is for - matching the Canon menu.  The 2 other bodies are fine being mismatch set though.
Set 25fps is not to match canon menu.