Author Topic: Danne's crop_rec_4k, 5DIII  (Read 249585 times)

Bender@arsch

  • Member
  • ***
  • Posts: 110
Re: Danne's crop_rec_4k & ISO experiments
« Reply #475 on: June 25, 2020, 12:08:04 PM »
https://drive.google.com/drive/folders/1HGOqDNpQvqxeLzXa0f5Zbfzh8bbzl-GH?usp=sharing

wait 10min for uploading.

I think the Problem is, there is no .M00 file. only a MRX file.

70MM13

  • Hero Member
  • *****
  • Posts: 588
Re: Danne's crop_rec_4k & ISO experiments
« Reply #476 on: June 25, 2020, 12:12:51 PM »
danne, you were right about the autofocus lens!

problem disappeared immediately.

put cine lens back on... problem is back.

if there's no other solution, can you add the option to have overclock setting off by default for manual focus lenses?

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #477 on: June 25, 2020, 02:30:41 PM »
Ok, I see what I can do about the issue. Will test when I have my hands on my 5d3 again. Thanks.

70MM13

  • Hero Member
  • *****
  • Posts: 588
Re: Danne's crop_rec_4k & ISO experiments
« Reply #478 on: June 25, 2020, 02:39:10 PM »
fingers crossed!  :)

masc

  • Contributor
  • Hero Member
  • *****
  • Posts: 2104
Re: Danne's crop_rec_4k & ISO experiments
« Reply #479 on: June 25, 2020, 10:56:34 PM »
https://drive.google.com/drive/folders/1HGOqDNpQvqxeLzXa0f5Zbfzh8bbzl-GH?usp=sharing

wait 10min for uploading.

I think the Problem is, there is no .M00 file. only a MRX file.
Wait... you recorded with card spanning and got no .M00? That should mean some data is missing. Your file's metadata tells about existing audio channels but no audio data is existing. This was why MLVApp chrashes. I implemented another check which doesn't just trust the metadata... :D If you are able to compile, you can try the fix.
Grüße nach Dresden!
5D3.113 | EOSM.202

Grognard

  • Freshman
  • **
  • Posts: 75
Re: Danne's crop_rec_4k & ISO experiments
« Reply #480 on: June 26, 2020, 11:58:26 AM »
Hi. I'm a long time user but it's my first post here.

I've tried Danne's last build with sd_uhs module and works perfectly on my lexar 256go  633x. A relatively cheap card.
I can reach more than 40 MBs write speed (with peaks at 50mbs).

With card spanning on, in UHD mode, write speed is around 120mbs (80 mbs CF +40 mbs SD)

I'm also able to record continue 1920x804 14bits lossless on Sd card only.

I also tried 5k anamorphic and it works perfectly. No crop factor (!) almost realtime liveview and very good quality. sd_uhs is not necessary but could be usefull on very contrasted scene.
I made a short video with this mode.



I will made a more complexe short film with this mode soon.


Thank You Danne (and others A1ex, Lila.. ) for your work.

Max

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #481 on: June 26, 2020, 02:13:33 PM »
Nice footage. 5k anamorphic is undisputable the most efficient preset of them all if filming 24 or 25fps. What lens was used?

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #482 on: June 26, 2020, 03:15:14 PM »
New sd_uhs.mo version uploaded tested working with manual lenses in an automated way:
https://www.magiclantern.fm/forum/index.php?topic=23041.msg216899#msg216899
Bottom of the post. Reports if working or not welcome. @70MM13 please test with your lens giving you the error before.

EDIT: Also posted new ml builds adressing yet again custom folder function. I advice to reinstall both ml build and sd_uhs module from this post:
https://www.magiclantern.fm/forum/index.php?topic=23041.msg216899#msg216899

Grognard

  • Freshman
  • **
  • Posts: 75
Re: Danne's crop_rec_4k & ISO experiments
« Reply #483 on: June 26, 2020, 04:44:42 PM »
Nice footage. 5k anamorphic is undisputable the most efficient preset of them all if filming 24 or 25fps. What lens was used?

Thanks. I used Canon EF 24-105 L.
Around F.16 a little too closed at my taste.
I had forgotten my neutral filter but the clouds saved me.

Grognard

  • Freshman
  • **
  • Posts: 75
Re: Danne's crop_rec_4k & ISO experiments
« Reply #484 on: June 26, 2020, 05:01:53 PM »
Danne. Is it possible to make Fps overdrive usable in 5k anamorphic mode juste like the EOS M does?
Thank You.

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #485 on: June 26, 2020, 05:35:04 PM »
Which is how exactly?

70MM13

  • Hero Member
  • *****
  • Posts: 588
Re: Danne's crop_rec_4k & ISO experiments
« Reply #486 on: June 27, 2020, 03:10:02 AM »
the new build plus the new overclock module works perfectly now with manual focus lenses!

thanks!

rinski

  • Freshman
  • **
  • Posts: 50
Re: Danne's crop_rec_4k & ISO experiments
« Reply #487 on: June 28, 2020, 11:05:06 AM »

428/5000
Hello, the sd_uhs module works perfectly with my manual samyang xp 50mm 1.2 lens and with the professional lexar cf 1066x 160 MB / s and sdxc sandisk extreme pro 170 MB / s cards I can record in full anamorphic 6k 1.1 1920 x 3760 in 14 bits with 15.003 fps, indoors and outdoors, maybe 5 more minutes, and I'm wondering if it would be possible to enable 23,976 fps in 12 or 10 bits in this format for smooth movement.
Thank you.

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #488 on: June 28, 2020, 01:03:48 PM »
Not possible with that high resolution.

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #489 on: July 04, 2020, 08:05:30 AM »
New builds:
https://www.magiclantern.fm/forum/index.php?topic=23041.msg216899#msg216899

- sd_uhs speed increase https://www.magiclantern.fm/forum/index.php?topic=12862.msg228659#msg228659
- sd_uhs.mo added as a module to Modules tab
- bench.mo added to Modules tab

Works pretty much the same as for eosm here. But works with 240Mhz(tested on 170Mb/s card)

EDIT: Think I fixed the File create error completely that came with manual lenses sometimes conflicting with patching.


Larouso

  • New to the forum
  • *
  • Posts: 16
Re: Danne's crop_rec_4k & ISO experiments
« Reply #490 on: July 05, 2020, 10:14:49 PM »

With card spanning on, in UHD mode, write speed is around 120mbs (80 mbs CF +40 mbs SD)


Hopefully not a very dump question - but how did you get the write values per card?

Walter Schulz

  • Contributor
  • Hero Member
  • *****
  • Posts: 8672
Re: Danne's crop_rec_4k & ISO experiments
« Reply #491 on: July 05, 2020, 10:22:59 PM »
File size / recording duration?
As in "MByte/s"...

Larouso

  • New to the forum
  • *
  • Posts: 16
Re: Danne's crop_rec_4k & ISO experiments
« Reply #492 on: July 05, 2020, 10:34:42 PM »
File size / recording duration?
As in "MByte/s"...

Sure, thought there is maybe an option to show it live per card, like the mb/s stats that you get if you kill global draw... Thanks anyway for the hint.

domasa

  • Member
  • ***
  • Posts: 114
Re: Danne's crop_rec_4k & ISO experiments
« Reply #493 on: July 12, 2020, 07:29:58 PM »
As long as these builds contain mislabeled ISO settings, I am not going to support them, sorry.
Is it planned to return this build back to Experiments page?

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #494 on: July 12, 2020, 08:22:43 PM »
Not as is and probably not a priority from a1ex atm.

Hynek_Tuleja

  • New to the forum
  • *
  • Posts: 8
Re: Danne's crop_rec_4k & ISO experiments
« Reply #495 on: July 18, 2020, 03:46:14 PM »
Hi, I have just tested Danne's build and it's incredible.

I would like to ask, if it would be possible to shot in 60fps in higher resolution (in anamorphic preset for example)?

I use the Canon HD format to shot in 60fps (1920x628 resolution) and then I scale the video to Full HD in MLVApp for slow motion. So is there a way to shot slow motion in higher resolution then Full HD with this build?

ewinemiller

  • New to the forum
  • *
  • Posts: 46
Re: Danne's crop_rec_4k & ISO experiments
« Reply #496 on: July 18, 2020, 06:22:24 PM »
Hi all,

If anyone is interested I put together some intermediate presets for 2.20:1, 2:1, and 1.85.1. I was looking for something a little in the middle of the 16:9's crop and the 2.35 aspect ratio. All are continuous on a 1000x KombuterBay without spanning or anything.

I will try to put it up on bitbucket, but have to get my head back around that process again. I don't use it enough to be comfortable or even remember the basics, and might not have time to work through it this weekend.

All changes in crop_rec.c from Danne's crop_rec_4k_mlv_snd_isogain_1x3_presets branch. Thank's Danne for your work. It was super easy to build on.

Regards,
Eric

menu definition

Code: [Select]
    static struct menu_entry ratios_menu[] =
    {
        {
        .name   = "Ratio",
        .priv   = &ratios,
        .max    = 6,
        .choices = CHOICES("OFF", "2.39:1", "2.35:1", "2.2:1", "2:1","1.85:1","16:9"),
        .help       = "Access six global ratios",
        .help2   = "only anamorphic preset\n"
        },
    };

in crop_rec_needs_lv_refresh

Code: [Select]
    if (ratios == 1) menu_set_str_value_from_script("RAW video", "Aspect ratio", "2.39:1", 5);
    if (ratios == 2) menu_set_str_value_from_script("RAW video", "Aspect ratio", "2.35:1", 6);
    if (ratios == 3) menu_set_str_value_from_script("RAW video", "Aspect ratio", "2.2:1", 7);
    if (ratios == 4) menu_set_str_value_from_script("RAW video", "Aspect ratio", "2:1", 8);
    if (ratios == 5) menu_set_str_value_from_script("RAW video", "Aspect ratio", "1.85:1", 9);
    if (ratios == 6) menu_set_str_value_from_script("RAW video", "Aspect ratio", "16:9", 10);

In LVINFO_UPDATE_FUNC

Code: [Select]
    if (CROP_PRESET_MENU == CROP_PRESET_1x3)
    {
        if (ratios == 0x1)
        {
            snprintf(buffer, sizeof(buffer), "anamorph 2.39:1");
        }
        if (ratios == 0x2)
        {
            snprintf(buffer, sizeof(buffer), "anamorph 2.35:1");
        }
        if (ratios == 0x3)
        {
            snprintf(buffer, sizeof(buffer), "anamorph 2.2:1");
        }
        if (ratios == 0x4)
        {
            snprintf(buffer, sizeof(buffer), "anamorph 2:1");
        }
        if (ratios == 0x5)
        {
            snprintf(buffer, sizeof(buffer), "anamorph 1.85:1");
        }
        if (ratios == 0x6)
        {
            snprintf(buffer, sizeof(buffer), "anamorph 16:9");
        }
        if (ratios == 0x0)
        {
            snprintf(buffer, sizeof(buffer), "anamorphic");
        }
    }

in calc_skip_offsets

Code: [Select]
        case CROP_PRESET_1x3:
            if (ratios == 0x1)
            {
                skip_left       = 206;
                skip_right      = 62;
                skip_top        = 60;
                skip_bottom     = 30;
            }
            if (ratios == 0x1 && set_25fps == 0x1)
            {
                skip_left       = 248;
                skip_right      = 110;
                skip_top        = 60;
                skip_bottom     = 42;
            }
            if (ratios == 0x2)
            {
                skip_left       = 206;
                skip_right      = 62;
                skip_top        = 60;
            }
            if (ratios == 0x2 && set_25fps == 0x1)
            {
                skip_left       = 248;
                skip_right      = 110;
                skip_top        = 60;
            }
            if (ratios == 0x3)
            {
                skip_left       = 268;
                skip_right      = 124;
                skip_top        = 60;
            }
            if (ratios == 0x3 && set_25fps == 0x1)
            {
                skip_left       = 360;
                skip_right      = 118;
                skip_top        = 60;
            }
            if (ratios == 0x4) 
            {
                skip_left       = 340;
                skip_right      = 196;
                skip_top        = 60;
            }
            if (ratios == 0x4 && set_25fps == 0x1)
            {
                skip_left       = 432;
                skip_right      = 190;
                skip_top        = 60;
            }
            if (ratios == 0x5)
            {
                skip_left       = 402;
                skip_right      = 258;
                skip_top        = 60;
            }
            if (ratios == 0x5 && set_25fps == 0x1)
            {
                skip_left       = 488;
                skip_right      = 246;
                skip_top        = 60;
            }
            if (ratios == 0x6)
            {
                skip_left       = 426;
                skip_right      = 282;
                skip_top        = 60;
            }
            if (ratios == 0x6 && set_25fps == 0x1)
            {
                skip_left       = 512;
                skip_right      = 270;
                skip_top        = 60;
            }
            break;
           

Danne

  • Developer
  • Hero Member
  • *****
  • Posts: 7702
Re: Danne's crop_rec_4k & ISO experiments
« Reply #497 on: July 18, 2020, 07:09:09 PM »
Nice. Will check it out when I have the chance.
Ongoing code atm:
https://bitbucket.org/Dannephoto/magic-lantern_dannephoto_git/src/master/

I piggy back on a git account but still include hg commits working with my compiler tool:
https://bitbucket.org/Dannephoto/ml_compiler-git/src/master/
After commits I push all hg and code changes to the git account. It's all temporary. Bitbucket are deleting all mercurial sources. Bad move on their part.

ewinemiller

  • New to the forum
  • *
  • Posts: 46
Re: Danne's crop_rec_4k & ISO experiments
« Reply #498 on: July 19, 2020, 01:11:45 PM »
Since the "where we're landing" question doesn't seem to be answered yet (or at least I couldn't pick it out of the end mercurial thread), I just forked your git and stuffed the changes here https://bitbucket.org/ewinemiller/magic-lantern_dannephoto_git/src/newpresets/.

GullRaDriel

  • New to the forum
  • *
  • Posts: 49
Re: Danne's crop_rec_4k & ISO experiments
« Reply #499 on: July 20, 2020, 08:46:05 AM »
I got some problems while trying to make a raw video 3k + fps override @2.5:
ML ASSERT:
RAW_IS_IDLE
at mlv_lite.c:969 (measure_compression_ratio), task shoot_task
lv:1 mode:3

shoot_task stack: 1c2498 [1c2600-1c0600]
0x000B6A18 @ 956f0:1c2590
0xUNKNOWN  @ b6a6c:1c2578
0x00B3B854 @ b3d710:1c24d8
0x0006A01C @ b3b91c:1c24c8
0x00069878 @ 6a084:1c2498

Magic Lantern version : Nightly.2020Jul15.5D3113
Mercurial changeset   : 8f38ee536de0+ (crop_rec_4k_mlv_snd_isogain_1x3_presets) tip
Built on 2020-07-15 19:47:31 UTC by danne@danands-MBP.
Free Memory  : 213K + 2445K

Did not stumbled upon this before, so I don't know if it's related to your build only or not.