Messages

Hi drmoreau ...

For the 70D, I've had success with *** Magic Lantern Nightly.2019Mar07.70D112 *** (by artur@Artur)
It's not the latest experimental build but offers RAW up to 14bit (I use12 bit), at 1832x1024 (for 16:9) or 1832x776 (for 2.35:1) aspect ratios.
It uses LOSSLESS *COMPRESSION* of the RAW, allowing for reduced data written to the card which is really neat.

A detailed post by 'VGA256' can be found here:
https://www.magiclantern.fm/forum/index.php?topic=14309.3425 ---- (see his reply #3446 dated March 03, 2020)

[Lessons Learned:]

Lessons Learned:
Reflecting on this oversight, I thought I'd record my thoughts:

● This type of mistake won't likely occur for those constantly using ML - (camera operators etc), because of continually executing the ML install procedure.(constantly treading the same ground)
● For those who 'ocassionally' dabble with ML (with maybe months in between) - these *oversights* are more likely to happen.
● So too for those (like myself) juggling not only cameras, but all things from scripting to directing, from lighting and sound through to post - it can spread a one-man operator very thin. Hence the need for copious notes. But even notes can let you down if not clearly laid out and followed. That's what happened here.

For me, notes are a constant *work in progress*. Editing them so as to be 'concise' while not excluding necessary detail. Tricky. Anyway, again I thank the 3 kind folk who helped solve this.

[Bold Red]

Yes, thanks Walter and 70MM13 - I'd totally overlooked that step:

● Launch the Firmware Update process and follow the messages on the screen.
● Once you see the green confirmation screen, restart your camera. Done.

Well of course I feel like a right idiot - I'd had it in my notes - but skimmed over it. I'll now update that part with Bold Red.

Doesn't seem be totally a Sandisk Extreme Pro issue - there's another 'variable'.
● I low level formatted the SanDisk card (using a 70D).
● Re-downloaded the ML zip folder (Nightly.2019Mar07.70D112_crop_rec_4k_mlv_lite_lossless) --- File Size = 1,918,067 bytes (zipped).
● Inserted Sandisk into PC and extracted ML contents onto SanDisk card at root level. --- Size of the 3 files = 3,406,567 bytes.
● Safely ejected SanDisk and inserted into 70D and closed the door. The camera's Red flashing light was 'brief' - not 'extended' as when loading ML.
● Switched 70D on - Live-view came on.
● Depressed the Rubbish Bin button - nothing - The live view remained, the ML page did not appear.

Now here's the 'extra variable'
● I repeated the above, but this time using a Samsung EVO Plus card - one that already had the same ML version on it -  (I checked and ML was working fine).
● So formatting over this working ML, I installed a totally 'new' ML, the same 3 files. 3,406,567 bytes.
● Inserting the EVO into a 70D and switching ON - *the exact same problem* - pressing the Bin does not bring up the ML menu - yet it was working fine a few minutes earlier.

And all 3 cameras respond the same way - they all fail to work with any NEW installs of ML, yet all 3 cameras DO work with 2 other EVO cards with ML created months back. So I can no longer say it's attributable to SD card 'type' - it's happening now on both SanDisk AND Samsung. Something has changed - I mean am I missing something? - If it's right under my nose I can't see it tbh

Thanks Walter & Levas ....
I'll do some more testing this evening and hopefully solve this. I'll post my findings or discoveries as I get closer. - thnx

Thnx

Result:

----------------------------------------------------------------------------------------------

H2testw

Writing                           Verifying
60873 MByte                60873 MByte
13:20 min                      12:18 min
76.1 MByte/s                82.4 MByte/s

Warning: Only 60873 of 60874 MByte tested.
Test finished without errors.
You can now delete the test files *.h2w or verify them again.
Writing Speed: 76.1 MB/s
Reading Speed: 82.4 MB/s
H2testw v1.4

----------------------------------------------------------------------------------------------

So ..... I'm guessing that's a good result ....
This is most strange. The SanDisk will work fine for 'normal' use - I can record normal 1920  24  ALL-I  as per Canon's settngs, - so that confirms the 70D is at least recording-to and previewing-back from the card - but it's not getting a *read* on the resident 'ML files' when pressing the Rubbish Bin - ie: pressing the Bin button does nothing - it defies logic. And it's reproduceable on 3 different bodies - exactly the same result.

The SanDisk card I'm testing is:  BN2016752362G

It's as though there's *something* encoded on the card that's flagging the 70D from accessing the ML files - that's the only explanation I can think of.

Thanks Walter - I'll run some tests using your suggestion (CrystalDiskMark) - since I'm on Windows - I'll get back .... thnx

Result:
SanDisk Extreme Pro (170MB/s) 64GB SDXC UHS-1 card using CrystalDiskMark (with 'Default' settings):


                         Read [MB/s]          Write [MB/s]     
Seq
Q32T1                  84.51                       74.79

4KiB
Q8T8                    8.643                        2.785

4KiB
Q32T1                  8.750                        2.723

4KiB
Q1T1                    7.551                        2.646

I don't know what these results mean in real terms (I'm not overly 'technical') - whether good or bad ? ......

Hi All ....
I recently purchased some "Extreme Pro" (170MB/s) SanDisk cards (64GB) for my Canon 70D's.
Wierd thing is - the 70D won't enter into 'ML mode' when pressing the Rubbish-Bin button (the cards work OK for 'normal' 70D recording without ML). My Samsung EVO Plus micro SD cards have always worked with ML, so the only new variable is the SanDisk cards. And I can replicate this on 3x70D bodies.
Is there an issue with these cards? - Have SanDisk changed something other than the 170MB/s read speed - seems there's some compatability issue. I was wanting to go to SanDisk so as to have Full sized SD's rather than micros with adapters ...

Thanks.

BTW - The build I'm using is:
https://bitbucket.org/ArcziPL/magic-lantern/downloads/magiclantern-Nightly.2019Mar07.70D112_crop_rec_4k_mlv_lite_lossless.zip
I also tried the 2019 March 24th build (which also always worked on the EVO cards). Again the 70D is not recognizing ML at all. It's as though ML isn't on the card (but it definately is).

Ahh - of course. So MLVApp spits out the DNG's (* RAW* digital values), and I instruct Resolve to process those values into my chosen color space like 'Rec709' or 'Blackmagic Design'. I see, yes I was incorrectly thinking I needed to *match* (in Resolve) mlvapp's 'output files', but I see now, they don't possess 'color space' at that stage. Thanks masc for the clarification.

PS: - Just found an article entitled "Good project settings for Resolve" here: https://www.magiclantern.fm/forum/index.php?topic=17324.0 which goes into 'Color Space', - of particular interest is Post #7 by Andy600

[● No it isn't - my bad - see explanation by 'masc' below●.]

@ Dmytro_ua  "If you shoot RAW - camera profiles doesn't affect the footage".
Thanks - I followed up after your comment, finding that *any and all* changes to camera profiles alters only camera 'display' and sensor recorded data is unaffected (so only 'ISO' alters RAW recordings). So that means the 70D is outputting Rec709 in the *default* sRGB color space. ● No it isn't - my bad - see explanation by 'masc' below●. With *MLV App* making NO changes whatsoever (other than conversion to DNG), then DaVinci Resolve must be receiving Rec 709 sRGB. Is my logic correct?

Hi Guys ... I'm seeking advice please - (from MLV App experts who've piped DNGs into DaVinci). Here's my procedure:
● I'm shooting 12bit lossless RAW on the Canon 70D with the 'CineStyle' profile for flat image
● The RAW files are imported into 'MLV App' then the export 'Codec' set to 'CinemaDNG Lossless' for outputting the DNGs
● The DNGs are then imported into DaVinci Resolve where I'm trying to do a LOG grade before outputting as Rec 709 (all new to me)

Given that the *Camera/CineStyle/MLVApp* combination is presumedly outputting a LOG image?, can anyone advise on the correct INPUT settings. ie: do I set Resolve's:
▪Color Space▪ to 'Rec.709'  or  'P3 D60'  or  'Blackmagic Design'   (the only 3 options)
▪Gamma▪ to 'Rec 709'  or  'sRGB'  (2 of the 'main' options)
Using a 'Color Checker' to balance shots also requires 'Source Gamma', 'Target Gamma' & 'Target Color Space' to be chosen, each with multiple options. MLV App's multiple settings of 'Tonemapping' and 'Processing Gamut', don't (I assume), come into the mix when exporting DNGs.

In any case, there's still many 'combinations of settings' overall and I've been chasing my tail for days researching and trying get the scenes to look right. I'm sure I'm missing the obvious - what is the receiving software *expecting* from MLV App's output? - And if anyone's sucessfully graded scenes in LOG using MLVapp > DR?, could you advise on settings? Thanking you.

[Thanks to all the clever folk involved - you're all thoroughly *brilliant* ....]

Nice one vga256 ....
When I saw your post here - I loaded up ArcziPL's build as well. Terrific! - now I've got sound as well.
Yes, the 'flickering' is a little annoying but it's only on the menu so no big deal. I ended up going with 1832 x 832 and was happy with 12 bit. I see what you mean about running time being dependent on the 'complexity of the scene'. If I use 'fast' camera moves it sometimes causes the recording icon to go yellow, or even red - indicating the buffer's filling too fast - but it always seems to recover and turn green again. My card's only 60MB/sec write speed, which is on the 'lean' side, but overall I still get very adequate recording times - (better than before). Thanks to all the clever folk involved - you're all thoroughly *brilliant* ....

Yeah thanks guys - so for a 12 bit approach (using Vegas) I'll go with PNG (or TIFF) exported out from MLVApp (accepting the larger file sizes) --- OR --- I'll go 10bit uncompressed out from MLV, then go with Sony 10bit YUV. It performs very well as an intermediate. -Just did a 25 generation A/B *comparative test* with Sony YUV across 25 forced recompressions - it looks fine - I'm happy ....

['Properties']

I'm using Windows, so I don't have the *AVFoundation option* for ProRes 4444 (only ffmpeg). Yes, I'm using Alexa Log.

Checking the ProRes4444 clip's 'Properties' INSIDE of Vegas, it shows 2560 x 1090 x 32 ? - which seems to imply it's 8bit + alpha?  When trying to use 'MediaInfo' to check it - I can't see any color depth reporting. All I'm trying to achieve is 12bit throughput without proxies. (wysiwyg).

Then I tried 16bit PNG sequences from MLVApp WHICH DO show in Vegas as 2560 x 1090 x 48, but even then I'm trapped.
● Rendering to almost all Vegas Codecs will result in 8 bit exports (even from these 16bit PNGs from MLVApp). (my flow is FX > render > re-import > add FX > render > reimport).
● If I *stay* PNG all the way, they double their size (Vegas turns them into 2560 x 1090 x64).
● If I export the MLVApp's PNG's from Vegas as Uncmpressed instead - it takes me back to 8 bit.
● If I export the MLVApp's PNG's from Vegas as Grass Valley Lossless, I'm also back to 8 bit.
● If I export the MLVApp's PNG's from Vegas as 'SonyYUV 10-bit YUV' - I finally get at least 10bits BUT it's YUV, and Vegas uses RGB internally so that's a needless YUV to RGB conversion every time I add FX.

I simply want to go from M/L 12bit DSLR through to a 12bit final edit in Vegas - and I'm not understanding - please, why am I having so much difficulty?

[MLV App]

I'm confused - can or does MLV App export ProRes 4444 video at 12bit color depth?  I simply need 12 bit video files exported directly from MLV App for importing into Vegas Pro 16. I don't want to use proxies or any file linking. Whatever I try, it ends up being 8 bit once inside of Vegas even though I've selected 32-bit floating point video levels in the Vegas Project Properties.

MLV App  >  Export 12bit Video  >  import into Vegas 16

Thanks ......

[@KirbyLikes525]

Hey, thanks very much for your quick replies!

@KirbyLikes525 .....
Yes, I'd purchased 3 identical 'Samsung EVO Plus 64GB micro SD cards' from ebay at a really good price. But I was careful they weren't fakes - and I'm certain they're genuine. So because I'm 'new' to Magic Lantern, I started thinking it was either my 70D or the SD card at fault, so I posted the above query. Then I discovered that the *other 2 cards* were a lot better, they were consistently showing write speeds of 39.3 to 39.5 MB/sec. So I did the following:

● swapped the faulty card into several other adapters, (the problem remained)
● cleaned the copper contacts of the the mico-SD card using pure alcohol. (the problerm remained)
● ran a 'speed' test on it usng 'the free tester H2testw'. It showed a consistant 63.5 MB/sec write speed (the card was fine - it's rated at 60B/s)
● decided to remove M/L and again format the card in the camera, then reinstalled M/L (the problem remained)
● decided to format the card in the PC instead, this time doing a FULL (slow) format, rather than a 'quick' one. Reinstalled M/L and tried again.
● Voila! - the card returned a write speed of 39.3 MB/sec just like the others.

Conclusion: I'm thinking that every time I fill a card with video, I'll dump the files to my PC, then do a proper *full* format of the card, then re-copy the M/L files to it again, and re-initiate M/L on the camera for that card, ready for next use. It's a bit of extra work, but it may be worth it for the sake of predictable recording times.

Kirby, - it turns out that the 10/12 Bit RAW video experimental is the build I'm actually using! - yes and I love it !!!! --- Hey - and you're getting some really great recording times on your Canon! - better than on the 70D. I've tried 10bit and saw a definite improvement. I had a play with 12 and 14 as well. I found the 14 bit is too demanding on the 70D (for Full Frame - and I'm not that keen on using 3x crop). And I do like the 12 bit - though again it's still intense on data. At the highest available resolution of 1832 x 780, I can get around 25 seconds of recording time (that's 2.35:1 using 24fps) - I'm very happy with that since 25 seconds is ample for most shots. Hey Kirby, I'll be keen to check out your project for sure!



@yourboylloyd ....
Thanks for that link - yes, I'm wondering whether this idea of fully fornatting the SD card is maybe the answer. I'm still too new to say. I read online that when you do a quick format, it only erases the 'adresses' to where your data is located on the card - but that it doesn't erase the data itself (until you record over the next time). I'm thinking this *may* affect the future 'writes' causing a bit of a bottle neck. I also read that some SD cards (like the Samsung EVO I'm using), are 'optimized' for continuous (sequential) writing. That makes me think that a fresh, properly formatted card is maybe best each time it's re-used. All I know is that it worked in my particular case, and I'm so glad it did because I was close to throwing the card away thinking it was faulty ....

Greetings ..... Could someone kindly tell me why this *read-out* (please see pic) - is sometimes high (like 39.2 MB/sec) and at other times low (like 32 MB/sec) - or anywhere in between. It seems as though every time I turn on the 70D, my expected recording time can vary considerably. eg: A RAW file might sometimes yield 17 seconds or 25 seconds or even 30 seconds, yet the 'settings' have not been altered. It's just that the currently estimated 'write speed' varies so much. It's the yellow message that reads (for example) "Expect 694-780 frames at 38.3MB/s"  (please see pic).

Thankyou 

https://www.dropbox.com/s/s22ehidafouslw5/Write%20Speed%20changing.jpg?dl=0

[@IDA_ML]

@IDA_ML ....
Ah ... OK - that may explain things too. My Display is 1920 x 1080. For quality, I've been upscaling the 70D shots to 3K for adding CGI later (which tends to be 'sharp'). Then I'll finalize at 1920 x 817 and/or 3840 x 1634. Since I'm 'monitoring' at 1920 x 1080 it's not ideal. (Can't afford a proper broadcast monitor). Even so, I *feel* that it's more of a 'moire' issue than aliasing. Maybe they're linked? - don't know. - Thanks for your thoughts!

@masc ....
Yes, I used the f:2.8 (Tokina) 'wide open'. Thanks though, it's good to know wide-open is best. That'll work well with my plans. I've recently purchased ND filters, and intending to shoot everything at max aperture - love that shallow depth of field. (btw - the only vision I've put on this forum so far are (links) to a kitchen stove showing a motion problem, so when you refer to what you see in a 'video', I'm wondering if maybe you're thinking of Andrews video upload?) Wow - I'd not tried that Chromatic Abberation control. That's absolute magic! I'd seen it there but thought nah - the lenses didn't need CA correction. Amazing, and it works *really* well on 14 bit test shots - (was getting magenta & green fringing on edges) - This setting's totally removed them. Thank you!

[@ masc]

@ masc My 'fastest' lens is the Sigma 18-35mm at f:1.8, but these M/L tests I've been doing (indoors) have been on a Tokina 11-16mm f:2.8 lens. As you mentioned, I've used 1/50 sec as my shutter speed, because of shooting at 24fps. (I *think* M/L gives me 1/48sec?). I'm not sure what filter Andrew used for his video, but I've been doing tests with 'AMaZE' which I'm finding really nice. - It may not be the sharpest, but I think overall it looks the most natural.
Oh - and when you say "Color moiree can be filtered by MLVApp" - which setting are you referring to? (please) .....

@ilia3101 ..... "Always use the highest resolution you can - using a lower resolution simply crops the full image down."
Ah ... that makes sense - I was wondering why the M/L 'calculator' kept increasing the 'crop factor' readout when I was clicking down through the frame sizes. Thanks!

[@ ilia3101]

@ ilia3101 - thankyou - I'm just discovering things as I go ....

@ andrew_dotdot  Hey Andrew! .... thanks so much for your video, I really enjoyed it ! - It's so nice to see such intimate views of creative skills at work, most enjoyable. These are the 'hands-on' arts and crafts that are sadly *missing* from my life - my own doing of course - since I'm a movie making addict - lol.

But yes - I see exactly what you mean about the aliasing and moire - thanks for that ...... You know, though such shots might be 'objectionable', but there's a low percentage overerall. I'm wondering whether it would be possible to make 'changes' during the shoot like 'removing' such zipped clothing (hit and miss, I know), OR to somehow deal with these shots in post production by slightly blurring the affected region - granted, this would be easier on a 'static shot'.

Andrew, my 'problem' is I'm heavily invested in the 70D's. I have a mix of both APS-C and vintage Full Frame lenses all bought for this model. (I've actually got 3 x 70D bodies, all purchased second hand, and all in really good shape with low shutter counts). So I'm not sure I could make the change to other gear - it's taken me several years getting to where I'm at ...... This is a real dilemma .... Having 'tasted' Magic Lantern's visual quality, I don't think I could go back to filming compressed/lossy formats.

I've been experimenting with different horizontal resolutions between 1600 and 1832 pixels but the 'moire' always seems to be there on *certain shots*. Anyway, I guess it's a matter of how much I'm going to let this be a road block - (which it somewhat is) - since I like 'perfect shots' as much as most of us do .......

Thanks again for your kind assistance - I'll *keep on trying* - maybe there's an answer somewhere .....
- Steve -

[FIRST WAITING 5 seconds AFTER opening the SD card DOOR]

Greetings .... I'm 'new' to magic lantern, though I've been reading the forums on and off for a couple of years ... The following account *may* be of help to some who are starting out ...

Naturally I was fearful, worrying about possible camera damage. Trying to figure out *which version* of M/L to load was a nightmare and I pondered this for a long time. Would it be the December 24th, 2018 nightly build? That *seems* rather dated now. Then again, would I risk a more recent 'experimental' build? Decisions, decisions. Well, I finally chose the experimental version of March 24th 2019. Why? Simply because it offered the ability to record in 10, 12 or 14 bit RAW - straight from the 70D's sensor. Now I must say, I'm NOT an overly 'techy' sort of guy. I'm 'visual', and I prefer to keep things 'simple'.....

So, I got to thinking that this 'cropped' (3x or 5x - whatever) was toooo much of an inconvenience. I just wanted 'simple'. Soon I realized that the largest video 'frame' size M/L would allow me to capture (without the 3x), was 1832 x 776 pixels. That's based on a 2.35 to 1 aspect ratio. So not quite full HD width. But I'd read that you still get great results after up-scaling, so I continued in hope. Well, really the rest is history. At the end if the day, all I was after was a *film look* from the 70D - with it's amazing focus system. Based on this goal, here's how I proceded:

■ Normally, (without M/L), I'd go with 24 fps (like 'real' film), using a fixed aperture of 1/50th of a second (roughly 'twice' the frame rate).
■ Also I find that an ISO setting of 640 - is as high as I can get away with, without introducing too much visual noise. It also happens to be an *ideal ISO* for shooting indoors. In a room with bright natural or artificial light, you get a perfect exposure at f:2.8. Neat.
■ I followed M/L instructions, loading the (3 files) onto a fast SD card, (which I'd formatted first). I used a Samsung EVO 64GB SD card, 60MB/sec write speed.
■ As mentioned, I went with the March 24th, 2019 M/L 'experimental' build. I chose to record at 10 bit, which (I thought) was enough of a quality boost without stressing my camera too much. A more demanding 12 bit capture would mean reducing my frame resolution.
■ 10 bit recordings would give me 1832 x 776 resolution, (set for 2.35:1 aspect ratio), resulting in 40.6 Megabytes of data being sent to the SD card each second. This is the bottleneck of the 70D, since it's limit is 40MB/s throughput to the memory card. 40.6 MB/sec is *just over* the limit, and a test yielded just under 4 minutes before the camera stopped recording. (the M/L 'display' showed that the camera's internal temperature had risen from 26° to 32° during recording). 4 minutes would be ample for most shots, and *if* a longer duration was needed, the frame resolution could be lowered. Reducing to a 1792 x 762 resolution gives unlimited recording time, (the on-screen counter being 'green' this time, rather than amber). I only went to 7 minutes - way long enough, (this time the temperature rose from 31° to 41°). This is to be expected - the internals are working harder than usual.
■ After recording, I removed the SD card but FIRST WAITING 5 seconds AFTER opening the SD card DOOR. The card is then slotted into a PC.
■ Using the free program called "MLV App" (a 'RAW to Video' converter from Magic Lantern), just go to 'File' > 'Import MLV' & navigate to (open) your file.
■ In "MLV App"s right-side pane, select 'Fix Bad Pixels' > ON (This does *wonders* for hiding those bright 'hot' pixels).
■ Under 'Debayer Algothithm' I went with 'AMaZE' (which is the default and to me gave the best result).
■ To sharpen or not to sharpen. *Maybe* to just '20' - a tiny boost, to get a little more 'percieved' resolution. But with a sharp lens, I'd leave this at '0'.
■ Set RBF Denoise Chroma to desired level. Around the 40 to 60 range worked best for me.
Bear in mind, this run-down is for a 'basic approach' to quickly get your RAW files into a standard (editable) video file. By default this will output your video as 'REC 709'. You *could* do extra fine tuning with MLV's built-in 'level' and 'color' controls (or do that later in your NLE). You could also choose a 'log' profile for better grading control. You could even output as 'Digital Negatives' (DNG's) for maximum control in software like 'DaVinci Resolve'. But for now - back to basics.
■ EXPORT: Click the "Cogs" icon at the top, choose 'Codec' > 'Uncompressed AVI' and 'V210 10bit' (or ProRes 4444 for smaller file sizes).
■ RESIZE: I entered 3664 x 1552 - exacly twice my recorded resolution. I wanted the best I could get straight out of the gate. But you leave this unselected if you're happy with the camera's recorded resolution.
■ I used 'Smooth aliasing' 3 pass. It does a fantstic job smoothing jaggy edges and reducing the high ISO noise. (Long render times though, and if you don't want (nice) Motion Blur at this stage then DON'T select it).
■ Click the folder Icon left of the 'Cogs icon', enter a File Name & Location and hit SAVE. That's it! - Editable files processed from the RAW data. Now it's onto the editing and color grading.

Uncompressed files are large, so I convert them to an intermediate codec inside Vegas, which reduces them to about 1/3 their size. A more economical approach would be to scale them to 1080p directly from inside the *MLV App* if that's the desired delivery resolution. I'm still experimenting, here's some comparisons (all set for 2.35:1),

10 bit recordings:
1832 x 776 (largest FF) = 40.6MB/sec of data (Amber display - buffer's fed 0.6MB/sec above what it can write to the card)  test = 3m 50s
1792 x 762 = 39.0MB/sec of data (Green display - buffer not being choked)  Unlimited record time

12 bit recordings:
1832 x 780 (largest FF) = 49.0MB/sec of data (Amber display - buffer's fed 9.0MB/sec above what it can write to the card)  test = 25 seconds
1728 x 736 = 43.6MB/sec of data (Amber display - buffer's fed 3.6MB/sec above what it can write to the card)  test = 1 minute
1632 x 694 = 38.8MB/sec of data (Green display - buffer not being choked)  Unlimited record time

14 bit recordings:
1832 x 776 (largest FF) = 56.8MB/sec of data (Amber display - buffer's fed 16.8MB/sec above what it can write to the card)  test = 14 seconds
1728 x 736 = 50.8MB/sec of data (Amber display - buffer's fed 10.8MB/sec above what it can write to the card)  test = 22 seconds
1520 x 644 = 39.1MB/sec of data (Green display - buffer not being choked)  Unlimited record time

Note: The resolution sizes shown here are what I was able to select on the 70D. They differ slightly to what's shown in the M/L RAW Calculator https://rawcalculator.netlify.com/calculator_desktop , which indicates '1824' (not 1832) as the maximum available width.

Hoping this helps or inspires some-one ..... And I'd be amiss if I didn't say a big 'thank-you' to the amazing M/L Gurus who've achieved such incredible results from these Canon cameras.

Images of the issue found here:

1st pic - Frame capture of a 'normal' unaffected frame
2nd pic - Frame capture of a 'blurred' frame where the intermittant/cyclic problem occurs:

https://www.dropbox.com/s/29z8ujzsd03v4jz/02%20Normal.png?dl=0
https://www.dropbox.com/s/96ticexdxu7f9n1/02%20Blurred.png?dl=0

Note that the merging/blurring is random and sudden during footage replay.

Hi Guys ... I'm 'new' to both Magic Lantern and MLV App. Congrats to all for your amazing work ...
I've done some testing with the M/L 10/12-bit 'RAW' release on a 70D and created uncompressed files via the MLV App with 'good' results - until ....
my 3rd test which was a few seconds of a 'kitchen scene' and here that I discovered (after 'converting' via MLV App as usual) an unexpected result.
There was a 'cooking stove' in the back-ground with 6 identical circular gas controllers (all horizonally in line), and which, once the footage was converted, manifested a horizontal 'blur' on these dials on *some* frames, as though the encoder was confused as to which dial was which and trying to somehow compensate.

After further testing, (and one by one removing ALL *settings* that I'd used), it came down to the ●Smooth Aliasing● on the Export Setting. When selected, (and in ANY mode of selection), it will evoke the aforementioned problem of random (horizontal) blurring of the 6 identical circular gas controls, as though one dial was being 'confused' with the adjacent dial.

If I knew how to post an image of this anomoly I would (maybe someone could). I'm hoping it's not just me - ie: 'operator error', but I've really been rather methodical with my approach and feel that this may well be a legitimate problem. Thanking you for your precious time ....

~ Steve ~