Messages

Nobody is working on it. Act like there will be no ML port for this cam ever.

There's no point to wait. It will be never be released. Live with it or change camera with larger DR.

Are either of the replies official? From someone on the dev team? Or not?

To me, it seems a completely legitimate question to ask about a road-map for a port to the 5ds and 5ds/r models. My understanding is that is considered trolling. I see it as very complimentary to the ML dev team that people are very enthusiastic about ML on a new camera, especially to have Dual-ISO.

If I was in the market for one of those cameras, the eventual release of ML would greatly help in the decision. If ML made an official statement of "No, not ever!" like they have for the 1Dx, then I would have that much less interest in the 5ds or 5ds/r.

I'm less-than-informed about crowd-sourcing, but perhaps such a campaign would be mounted to get a used 5ds for a1ex. The street price on both, but especially the 5ds, seems to have gone down drastically. That seems like it would increase the interest level. Perhaps LensRentals would be willing to give a huge discount to a1ex for a lengthy rental?

Thanks

Is the cr2hdr.exe from 2015-Nov-14 the preferred version to use with dual-iso .cr2's on Windows computers?

http://www.magiclantern.fm/forum/index.php?topic=7139.msg157485#msg157485

Reply #2667 by dfort.

5d2, Aug 19 nightly build (2015Aug19)
When I attempt to select Dual-ISO 100/1600, 100/3200 or 100/6400, I see the warning message:
"Consider using a less aggressive setting (e.g. 100/800)

The 5d2 seems to accept 100/1600, and shows 2.8 EV DR gained.

However, if I select 100/3200 or 100/6400, it seems to ignore those settings. The EV gain doesn't change. When I return to the higher menu, it continues to show 100/1600 rather than 100/3200 or 100/6400. When I "Q" back to the Dual-ISO pop-up, it does show 3200 or 6400 as the upper ISO.

Am I doing something wrong? Is this "as expected"?

IIRC, I used to be able to select 100/3200 and 100/6400 on the 5d2. I think the 6d implementation allows 100/3200 and 100/6400.

Here's a link to a DPR post with results:
http://www.dpreview.com/forums/post/56366592

> "The measured values aren't quite so high as those that Magic Lantern claimed."

I'm jumping from this ship..

Pre-ordered a7Rii in June. ML was one of the major reasons I remained semi-loyal to Canon.

Trolling? If so, sorry.

I'd be happy to read that a group of ML developers had "fork'ed" over to nex-hack.info (or equivalent project if it exists). Based on just a very quick glance at their web-site, the state of the project seems rather immature.

I've always been curious just how much Dual-ISO improved DR on my Canon DSLR's, especially my 6D.

Recently, I submitted a series of test RAW's to Bill Claff at www.PhotonsToPhotos.net. He has developed software to calculate "Photographic Dynamic Range", which seems more or less related to DxoMark and SensorGen.info numbers:
http://www.photonstophotos.net/Charts/PDR.htm

I'm rather fuzzy on just how Dual-ISO works, and I'm not all that confident I provided Mr. Claff with appropriate test RAW's. The very preliminary results are less than expected ... maybe 1.5 stops at 100/800.

If someone has the interest and time to get involved in this, that would be appreciated.

My impression is that my 6d is capable of EFCS (electronic first curtain shutter), but I'm unclear on the preferred settings.

• Is this associated with the "Silent" mode in the Canon "Drive" menu?
• Is there a way to have a delay before the shutter activates, like 2 seconds or 10 seconds? That seems mutually exclusive.
• Is there a way to use a remote to trigger the EFSC ... normally the "Drive" has to be 2 or 10 sec for the Remote to work. Or not?
• My impression is that the 6d isn't prone to "shutter shock" like many mirrorless cameras (esp. Sony A7r?), but there still is some. Or not?
• If EFCS is available, can I expect to be enough better at reduced vibration than Mirror-Up + 2 second delay? Or not worth it?
• IIRC, there is Mode-1 and Mode-2, but I'm unclear on what they are for.

6D 1.1.6 with Build June-19

I notice that the "Allow Clipping" shows up as an "Advanced ..." option for ETTR. Then it is "forgotten" when you shut down the camera. You have to enable "Allow Clipping" the next time you start ML and use ETTR.

Is this "as intended"? Seems like "Allow Clipping" isn't really all that advanced of a setting compared to some other choices. In any case, it seems it ought to be "sticky" and remembered from shut-down to re-start.  Or not?

Am I doing something wrong (again)? Is there a way to accomplish this?

If this setting isn't "sticky", can that be a Feature Request so that it is "sticky" and remembered?

Thanks.

[c. ML-6D116.FIR]

> 2.  Copy the following over to my SD card
     a. Autoexec.bin
     b. ML directory
     c. ML-6D116.FIR

Is there such a file as ML-6D116.FIR? I didn't notice it with the June-19 build. Unnecessary even if it does exist somewhere?

[Micro AF settings can be zeroed]

According to the Canon webpage having the firmware update, it appears that the Micro AF settings can be zeroed:

http://usa.canon.com/cusa/professional/products/professional_cameras/digital_slr_cameras/eos_6d#DriversAndSoftware
> From Canon:
Please note:
If the firmware is updated from version 1.1.3 or before, the following setting will be reset to default.
Menu > Custom Function> C.Fn II: Autofocus > AF Microadjustment> 1: All by same amount

However, that doesn't seem to have happened with the lens now on the 6d ... settings were retained (whew).

And the Build June-19 seems happy with 1.1.6, but that's just after a minute or so of quick testing (smoke test?).

I'm been quite pleased with the use of cr2hdr.exe with Dual-ISO.

Sorry if this has been asked before, but I wanted to check if the version from 16+ months ago (Dec, 2013) is still the latest/greatest?

I've seen references to 20-bit versions, and mentions of beta's that have to be compiled to use.

Also, I can't tell from the file-names in the Bit-Bucket if they have anything to do with cr2hdr.exe. Is one of those an updated cr2hdr.exe?

Some other questions:

• Does the cr2hdr plug-in to LightRoom add any functionality?
• Is anyone experiencing the Previews in Adobe Bridge and ACR to be slow and hard to use?
• In Adobe Bridge, the thumbnails have false-colors and can be rather grainy looking.
• That's also true with using the SlideShow to try to figure out which images are keepers and rejects.
• ...
• Also, in ACR, it can be sloooooow for all the "instructions" to be applied to the image.
• A normal image is almost instant when you go from one image to the next, as far as seeing what has been done to the image.
• With a Dual-ISO image, it can take 5 to 10+ seconds for the post-processing to show up.
• That seems most likely with "complicated instructions" like Gradients, Local-Adjustments, Radial-Filters, Masks, etc.
• Overall, post-processing with Dual-ISO .dng's in ACR can be very sluggish.
• Am I doing something wrong? Or is that "as expected"?

> TLDR:  Set the shutter speed to capture the motion.  Use ISO as necessary to ETTR.

Thanks for the thorough explanation. As I expected, my "rule of thumb" was less-than-informed.

BTW: This issue is being written about on the DPR web-site:
"What's that noise?"
http://www.dpreview.com/articles/8189925268/what-s-that-noise-shedding-some-light-on-the-sources-of-noise

I'm wondering if my understanding of ETTR and the resulting rule-of-thumb (ROT) is more or less correct. Or flawed?

And sorry if this has been addressed before within this 170+ post thread and 4 pages. I did read several pages and didn't notice it being addressed.

To me, if the camera's base ISO (typically 100 for Canon) isn't suitable, then ETTR may not be all that useful or appropriate. It's mostly an issue at lower light levels, like dusk, dawn, night, interiors, landscapes, etc.

This could involve a moving subject demanding a relatively high shutter speed to "freeze" the action, or not enough light (even with tripod), or perhaps too much wind for tripod and long exposure. The use of Dual-ISO is also a factor, and my impression is that Dual-ISO increases the advantages of using ETTR with ISO 100.

If ISO 100 isn't appropriate for a specific capture, then you can settle for the histogram not going all that far to the Right, or increase the ISO to get the histogram to the Right.

But when you increase the ISO, noise increases and the tonality quality decreases. It gets to be a trade-off of whether ISO is the priority, or shutter speed.

I think with an ISO-less sensor like Exmor's, you'd tend to favor ISO 100 and whatever shutter speed that results in, within reason. With Canon sensors not being ISO-less, it's less clear-cut what to do when you can't use ISO 100.

Or not? I find I learn more when someone "corrects the error of my ways".

It seems unlikely to show up again. I use the older 5d2 less than my newer 6d. Also, it was with an somewhat older Nightly Build ... although the related code would seem unlikely to have changed.

I would suggest thinking about changing how the log###.log file is written out. I don't know what the issues are, but I have some doubts that you really want a null byte after the CR-LR \n.  For my situation (using Windows), that made it difficult to cut/paste the contents of log000.log.

Nightly build: Jan 28, 2015

If it is of interest, here are the contents of:
CRASH00.LOG

Code Select Expand

ASSERT: 0
at KernelDry\KerSem.c:306, task Fread
lv:0 mode:3


Magic Lantern version : Nightly.2015Jan29.5D2212
Mercurial changeset   : 703ee626326d (unified) tip
Built on 2015-01-28 23:10:03 UTC by [email protected].
Free Memory  : 128K + 3991K

and ...
log000.log
... difficult to copy/paste ... has null codes embedded in log000.log on almost every line ...
I can try again and make more of an effort to maybe replace null codes if it would help.

Code Select Expand

58th (of 3189 lines)
1524: 13169.205 [LVCFG] PROP_EFIC_TEMP 9

The camera had only been on several minutes at the most ... and not in LiveMode.

If helpful, I can open a "Bug Report".

It does seem like a possible bug to have the format of log000.log have lines with the format of ...

Code Select Expand

null-code contents \n
null-code contents \n
.....
null-code contents \n


For the past several days, the links to download Nightly Builds for the 6d, 5d2, and 600d have been broken.
Win 7, IE 11.

It does appear to work with Chrome.

When I click on the "Download" button for Jan 28, 2015, I see a message about: Oops! This link appears to be broken

Am I doing something wrong?

> Nightly Builds seem to have stopped for Canon MkII and MkIII...is there any reason for this?

Just noticed "fresh" nightly builds from mid-January. Thanks, development team!!!

Thanks for the advice, has really helped me out! How do you handle ETTR in low light conditions with dim lighting? Usually I can't expose to the right in those conditions, wondering what the standard thing to do in that situation is?

Why not able to use ETTR? If the shutter speed would be too slow to hand-hold, would a tripod help? Is the subject static, or moving? Dual-ISO can sometimes help with very high contrast situations.

With ML, the BULB is enhanced to easily provide much longer exposures than 30 seconds, without having to hold the shutter down, or use a remote control.

I've done ISO 100 ETTR at f8 during full moon night captures that are several minutes long (but the moon will move if it is in the frame). However, Auto-ETTR can sometimes have trouble deciding on the exposure with really long exposures.

I'm also curious why there haven't been new nightly builds since early Oct, 2014. Any reason that can be shared with end-users / non-devs? Is there some thread with an explanation? Some other link to get the "latest / greatest"?

Is the delivery mode of "rolling releases" no longer rolling? Are the devs going back to a delivery approach of Pre-Alpha --> Alpha --> Beta --> Release Candidate --> Release? (which I would find preferred ... YMMV ... my experience has been that the "rolling releases" worked much better with more stability than I expected)

You will find nightly build has been stopped about 7th/8th, Oct. 2014 for all cams.

FWIW and OT: forum no longer seems to support "quotes" from earlier threads with the author and time stamp. Am I doing something wrong?

http://www.catb.org/~esr/faqs/smart-questions.html#idp54032560

Sorry I submitted what you apparently feel was an non-smart question. I didn't submit a bug report on an intermittent problem that I couldn't recreate, but I was trying to see if anyone had seen that, or perhaps to watch for it.

Post your settings (the delta menu).

I don't have the specific settings from when the problem happened, as my priority was to get the camera back to doing the time-lapse with intervalometer.

I did attempt to recreate the issue yesterday, but it behaved.
Auto ETTR   OFF
Dual ISO   OFF
Advanced Bracket OFF
Intervalometer   ON, 1 minute
   Take a pic every   1 min 0 sec
   Start trigger   Half-shutter
   Start after      10 sec
Bulb Timer    OFF

FYI: A week or so ago with an older Nightly Blind Build, the Intervalometer was mis-behaving after I'd used the Advanced Bracket with 5 steps. After I disabled Advanced Bracket, the camera still insisted on taking 5 captures during each capture of a TimeLapse.

What seemed to be happening was that the Settings (which I think is written to internal memory on the 6d) weren't being updated. I couldn't get the Advanced Bracket to not interfere with normal operations, even though it was Disabled. The Menu showed "Advanced Bracket" OFF, but the 6d was taking 5 captures each time the Intervalometer counted down to zero.

The "fix" seemed to be to re-install the Nightly Build. (or it might have been an updated Nightly Build, sorry, I don't recall the specifics). My uninformed speculation was that the internal memory that held the settings was put back in a sane state. Or not?

The "common denominator" of the two issues may be the Intervalometer.

Another speculation might be that my process of installing a fresh Nightly Blind Build may be flawed.

Lately, I've seen intermittent problems and possible "bit rot" with the "blind nightlies" ... ML on my 6d seems to be getting less and less stable, especially the intervalometer.

My practice is to watch for fresh nightly blind builds for my 6d, cross my fingers, and install.

I use the 6d a fair amount in "time lapse mode" with the Intervalometer for once a minute captures. A typical Manual mode setting would be ISO 1600, f5.6, 1/50th with EF 35mm f2 IS on tripod for an indoor non-action event (groups of singers and speakers). It will start off working fine, while I'm using my T3i / 600d with 200mm f2.8 (=320mm) for closer views on individual performers.

I'll go see how the 6d is doing on the other side of the "crow's nest", and it will have decided to use ISO 100, so the exposure is completely off for many captures. That's happened several times, but seems intermittent. It's almost like it was using a previous setting for Dual-ISO, or Auto-ETTR, although both of those are OFF.

Or it might change both the ISO, and then the shutter speed, like it was briefly in Av mode instead of Manual mode, so the exposure is OK, but images are blurred due to too slow shutter speed.

Anyone else seeing that? I don't yet have enough repeatability to make a bug report. The times it has happened, I've got two or even three cameras going, and my priority is to try to get the 6d operational, and then get back to the 600d.

Have you looked at http://www.rawdigger.com/howtouse/rawdigger-histograms-overexposure-shapes ?
To determine the clipping point, you can shoot a curved shiny surface, like a bell, a spoon, a ball from a bearing (I happen to like DATACube, as it also contains a useful light trap for black point) and study the histogram of the rough selection around the specular highlight in RD.

I'll read the entire three series of articles, and expect to learn a lot. Thanks.

I used to have a Datacolor SpyderCube, and found it very useful. I did manage to misplace it, and came up with a do-it-yourself substitute with "black trap" and a stainless steel spoon for spectral highlights.
http://www.dpreview.com/forums/post/54142816

With my 6d at ISO 100, RawDigger treats values of 13,236 and above as that channel being blown.

With Dual-ISO and 100/800, RawDigger treats values of 57,539 and above as blown.

Values in a "black trap" go as low as (0,0,0,0).

Is this "as expected"? My impression is "yes", but I wanted to check for sure.

OK, I'll put that cube next to a light bulb and lock everything down like the last test.

Any time I've used that cube I've gotten blown highlights and pure black.

The Datacolor SpyderCube is one of the neatest little devices I've come across. Overpriced, however.

BTW, it isn't difficult to make your own:

• small box, like what smaller Canon lens come in
• cut out hole, and paint interior black for the "black trap"
• maybe put in black unmatched sock
• put stainless steel spoon next to it for spectral highlights.
• gray gaffers tape is decent for gray card

http://berean.zenfolio.com/dual_iso/h101028e1#h101028e1

http://berean.zenfolio.com/dual_iso/h101028e1#h3e0c323b