Messages

Fantastic resolution! 3968x1800. Unfortunately the top right corner stays dead.
Frame difference, exaggerated

So, what was the trick? Was it fair?

...but that document seems to show a different kind of skipping to what canon DSLRs do

Yes. Plus we have different skipping modes, plus the pattern changes depending of the offset (in crop modes). But the variants aren't innumerable.
The task in paper was much easier. We have shifting/alternating pattern. Weights will change, as the block needs to be non-square. The idea is that we know (well, guess) so much about the type of line skipping but it's not used in processing. Current dabayering algorithms are not intended to be used on squeezed images. And all that "pattern mess". Aren't they?
I had a fantasy about such interpolation on the pre-debayer stage. So that we might fill in the missing pixels with a "simple" shifting block algorithm (like in the paper - rough but fast) and then feed it to the common dabayering processor. We can even output interpolated DNG for use in other software.

Is there anything of interest for reconstructing skipped lines?
Bayer Interpolation with Skip Mode
Are we stretching debayered data in the MLV App for now?

I used 44-2 for a long time on 600d without a problem. Cheap, easy to repair/disassemble if you need.  Great ergonomics. On a full frame it looks differently – corners are never sharp (not good for stitching panoramas, for example), the whole image seems to be much more "artistic". But you need sufficient subject.

Discussed before:
https://www.magiclantern.fm/forum/index.php?topic=19226.0

Never paid attention to this error. Everything works after it. Doesn't it?

card not formatted. So I unable to format it from any devices.

Please, do make sure that you can't format it. Try HP USB Disk Storage Format Tool on Windows or some other special utility. There was a problem with file system corruption long time ago (reported and discussed here ). I'm still trying to narrow down the search. My prime suspect is writing MLV without space preallocation, possibly writing it from Silent.mo.
Looks like this (two different crashes). You can almost read the filenames...

This month I was testing reduced analog gains on 5d2 while everyday shooting and results look great. In short: it looks like we can pull down iso100 to something about iso20-25 (upd: lets's say about 50) still having full dynamic range (up to 11.6).

Setup.
Reddeercity's builds for 5d2 + raw_diag.mo

As Alex once reported, it's reasonable to pull these DFE values from 48x down to 250 – until vertical banding becomes very noticeable. I've tried lower values and subtracted darkframes in post. Luckily vertical banding stays constant for the same ISO and DFE values. I don't know whether it is important or not, but I used CMOS[0]=0x200, 0x204, 0x208 to get the darkframes (for iso 100, 200 and 400 respectively). Either it's some physically stable bias (hard to believe that) or just the factory calibration playing against us. I keep just 3 master-darks, they work without a problem. Final images after subtraction show no FPN and very low random noise (reason 1 - analog amplifiers may generate a lot of noise themselves; reason 2 – longer exposure gives larger number of photons).

At DFE values about 100 raw_diag reports worse DR. We can see shadow clipping (posterised green shadows) in post. 111 was just my safe random choice. It turns selecting of iso 100-200-400 to something about iso 20-40-80. All look unbelievably clean. Iso800 is out of use here (it's produced with DFE values 0x8xx). Iso1600 is a digital push.
SHAD_GAIN (c0f08030) set to 0xfff disables digital push for wide apertures. Not a great change, but still useful.
 
Attaching 2  samples of processed images (shadows lifted a lot) and some raws to play with. "-df" in filenames means that I've already subtracted blacks with Pixelfixer (I like its CR2-IN > CR2-OUT approach).

Known problems.
The only way to see overexposure is looking for pink highlights. Auto-ETTR can't catch it but raw_diag.mo works perfectly.
Low analog gains may be used with Dual-ISO module (or simply patching CMOS[0]=0x343 for iso100/400). But final images are not always recognized by cr2hdr-20bit. Using –force gives processed images with striped lights. Attaching samples, didn't tried it much.
Works also with Reddeercity's video hacks (14bit raw, don't forget darkframes) but adtg_gui.mo is an overkill here – it takes a lot of resources (and power?) and a lot of time for setup. No way to save its settings. Not a run-and-gun tool. What we need to emploi low iso range is a tiny separate module, patching DFE values at start. Maybe just on/of and a choice of value for them. Unfortunately I'm not the one to write it.

Checked the same approach with 600d. It gives something, but nothing groundbreaking yet.

Test raw files - https://app.box.com/s/d7i3qj09pc300ffb73sgrtnj7totjc7i

Simple registering here - https://developers.canon-europe.com/developers/s/login/SelfRegister
gives you enough information.
Latest CCAPI supports EOS 250D, EOS RP (firmware 1.1.0 is required), PowerShot SX70 HS (firmware 1.1.0). Usual SDK support Eos R from version EDSDK v3.9.0, EOS RP – from EDSDK v13.10.0, EOS 250D – from EDSDK v13.10.20. These are consequent releases.
There were rumors that you can complete registration, fill the profile and get access to it after awhile.
If somebody needs version 2.14 (models up to EOS Kiss X70 / EOS 1200D / EOS REBEL T5 / EOS Hi / EOS M2) – it's available in the net.

Detailed answer
https://www.magiclantern.fm/forum/index.php?topic=9702.msg93292#msg93292

Some people are really greedy (https://www.vid-atlantic.com/lensshop/helios-mod-lens). $30 is worth a try.
Working aperture is good but, I guess, closing it kills the visual effect.

$160 looks like enormous price for such a thing. Helios 44-[any number or letter here] looks rather stylish as it is (sharp center, twirls on the corners). But it's always very cheap, not to mention ones with the aperture stuck. Replacing aperture with cardboard doesn't seem to be a complex task. You'd better find somebody willing to buy it nearby

As for the effect... I don't think it looks like "high quality thing". Bokeh shows horizontal lines – reflection of the ellipses on the cardboard aperture. Why do we need them? I'd prefer black velvet. If you try a cardboard aperture in front of another lens – doesn't it look better? Or just the same?

I say it as usual cinema watcher. I've never used anamorphic lenses myself. But I know standard Helios 44-2 enough.

By the way, iso research build doesn't work on 5d2 (2019Mar23.5D2212.zip). It can't power off the camera – both leds stay on for long (not 2-3 sec, as described – long enough to panic) [I needed some time to get the card door trick].

Another finding, which makes me exited. After reading the nice article from RawDigger developers (https://www.rawdigger.com/howtouse/iso-is-seldom-just-digital-gain) I ran into checking my own files from 5D2. With such a histogram it takes just several minutes to know, that the most interesting ISOs were 160, 320 and 640 – with clear signs of data division. All others look like pulled up (with periodic gaps in the histogram). That just proved what everybody told (but I'm a hard believer).

But that's not the end of the story. I continued checking files from my archive and found a batch of files with ISO100 and ISO50 (which is supposed to be complete fake) – with absolutely smooth histogram – no gaps, no spikes. How could it be so? It was hard to reproduce them: finally I grasped it. With EMF chip attached (telling f1.4) or an AF lens I get gaps in the histogram. Without electrical connection to the lens – histogram is nice and smooth. Peripheral lighting correction is disabled, so I can't guess what kind of correction is applied here. Should we avoid it?
It needs much more testing in the terms of noise and impact on real images. Taking a break now. Maybe, I'm completely wrong...


Continued:
I've checked results with Canon 50mm lens:
At f/1.4 – each 13th value in the histogram has a gap. It becomes less noticeable at closed diaphragms.
f/1.4 - each 13th
1.8-25
2.8-31
2.2-43
2.5-55
2.8-84(85)
3.2-70
I can't imagine any fair reason to use pull up at open apertures. Just maybe their lenses lacked light at open apertures and canon gays are simply cheating to satisfy specs? Looks rather probable.
In any case it would be nice to disable this pull up if somebody finds it.

...SD-to-CF adapters which are crap and not supported

By the way, I've used it for months on 5D2 without a fault (several 64Gb SD cards + cheapest adapter from aliexpress). Sometimes it's worth a try.

It's more like a simple description for the topic about "CMOS/ADTG/Digic register investigation on ISO" and the great "sensor update". BMC has no "Dual ISO" at the same time, just a sensor limited with two native ISO's.

Absolutely (sorry for the messy post).
So the stripes come from the underexposed low ISO layer. That's just a usual noise pattern (like in pushed up darkframe) exaggerated while scaling. I've blended them manually, using low-iso layer for extreme highlights only.

I've just checked another approach: MLVProducer with DualISO processing (high-iso option). The result looks perfect.

Thanks a lot. No more evidence needed ) That's a kind of  picturesque sunsets, I've seen many times, but mostly in the cinema. There isn't a lot of them in Saint-Petersburg.
Best wishes and – by the way – PM me, if you ever need help digging for lenses here.

It's really fascinating.
I'm curious to know: that strange contour of the falling sun – was that real stratification of atmosphere, or that's a kind of dual-iso artefact?

So... You are right. Continue with pulling and rotating.

Sorry to bother. Please clarify: is 0.7m a number which you read on the scale, or you have measured actual distance with you film camera (or the proper DSLR adapter)? Or just both?
If you are absolutely sure, then yes – rubber gloves and the smallest drop of liquid silicone grease (but be careful with the lens) will make the trick... But I'm not so sure)

"this one twists out" – it locks the front lens. Leave it as it was. Try to avoid sliding and rotating the front lens. Some people use fridge or smartphone vibration to settle the lens down, then tighten collar. Don't use any force. If you get terrible lack of sharpness – it means the front element is a bit off.
The branded collar and the filter thread should be one piece, I can't believe you ever rotated them separately (I can be absolutely wrong, naturally). But this part does nothing about focusing.
"bottom of this collides with the top of this" – yes, at this position the lens is focused to the infinity or even behind it.
So now you have 6 (or something alike) variants to screw these parts together. Screw it in to the end, screw a bit back as you need to set the synching thumb to its place (on the photo). Then check the focus with the camera (don't rotate the ring while mounting, just check whether it's infinity or not). When you find the lucky position and focusing works from end to the end – it's time to look at the distance scale.
Does it show the wrong numbers? If so, find some small screws in it, release them. Rotate the scale to the correct position and fix it. If so, you are done.

Have you disassembled it in two parts? Looks like it.
What you have inside of the screw helicoid now is multiple thread i.e. there is more then one way to screw the lens part in. So just draw some marks and check all the possible combinations. But there may be a kind of thumb (like that http://album.foto.ru/photo/2921784/ - possibly skewed) and parts of aperture control, which prevent the lens block from screwing in to the end. It's just a guess, I've never saw your lens.

I used to like such superresolution tricks many years ago with a digital compact camera and PhotoAcute. It took a long time to understand that it's not superresolution, though it's much wider dynamic range. Camera sensor needs  a strong anti-aliasing filter, otherwise R, G and B pixels will reflect absolutely different sharp details of the subject. So the image is optically blurred just above the sensor - to get the needed resolution. Some astrophoto enthusiast try to scratch out microlense layer with the rgb pattern but the effect isn't as great as may be expected (pixel sensitivity drops much without the microlenses).

But we still get some good results averaging layers. We are sacrificing resolution (fine detail) while alignment and image rotation, but the noise falls dramatically. Just in case - the image averaging scripts for Photoshop are something good but not perfect. There is a number of averaging approaches. I'll refer to MaximDl manual - https://diffractionlimited.com/help/maximdl/Combine_Method.htm - a was a great pleasure to read this book even not using the program itself.

I've packed logs and settings. http://disk.karelia.pro/crfYi80/broken3x-12bit-bug.zip
In short - files from the build stay unaltered but all modules vanished. That's what cart reader sees. While previous crash I saw some alien characters instead of the filenames. Windows can do nothing with them but format.  The second time I've just copied the Modules\ back – and everything works with the same settings.
What camera sees when  the error occurs:  the message about 5D2_212.sym and no Global draw overlays. You can shoot, can't call ML menus. It's interesting, what was before that, but I can't roll my mind back.
There was no intensive card writing, the cards are OK (SanDisk Extreme Compact  SDCFXSB-128G-G46 and smaller SDCFXSB-064G-G46). No video recording, only lazy taking pictures in Live view. ISO1000 is a bad setting, but it doesn't stop taking stills or working live view. Another bad setting is leaving mlv_lite, mlv_rec and mlv_snd enabled together. That's a cruel way to get 10bit raw with sound. And that's not a sound of explosion : ) Is there any legal way to do it?
In fact 10-12bit is a fascinating feature for 5d2 and I'm watching  every reddeercity's  post with great attention. I'm not afraid of multiple crashes now, I have to accommodate with the limitations - to avoid them in the field.
Maybe that's not very useful information, it's hard to trace it from actual user error to the final result. My only question is: why ML rewrites modules on shutdown? and what for?
To end with something good – I'm actually charmed with ML for a decade.

The card was in all the time. As well as battery. It's something build specific, I guess.