6D: Enable 1/180s in M/Tv with 1/3 increments (#1 annoyance)

[Marsu42]

Disclaimer: This might sound a bit outlandish to people not using flash(es), but let me ensure you this is one of the worst regressions from 5d2->6d and one of the main complaints about the 6d from portrait/wedding photogs. If this gets solved then it'd be one big brownie point for ml.

Problem: The 6D max. x-sync is only 1/180s, but it's not mainly the mediocre speed that's the problem, but the fact that you're not able to set it if "C.Fn I-1: Exposure level increments" is set to 1/3ev. The alternative 1/2ev is not practical, the interval is simply too large, esp. since the flash exposure compensation is also tied to this.

Suggestion 1 (optimal solution): Squeeze in 1/180s between 1/160s and 1/200s by intercepting the dial and force-set it in M/Tv on 1/160s->dial up & 1/200s->dial down

Suggestion 2 (a bit more usability hassle): Tie force-setting 1/180s in M/Tv to some button like SET or another less often used button ml can intercept.

[a1ex]

Haha, I always set EV increments to 1/2, so I can count them easily.

Do you still get wheel events after reaching 1/160? (Debug->Show GUI events)

[Marsu42]

Haha, I always set EV increments to 1/2, so I can count them easily.

For camera ec this might work for some, but for flash ec and ettl metering 1/2ev is much too large

Do you still get wheel events after reaching 1/160? (Debug->Show GUI events)

Um, yes, sure, why not? Both wheels generate gui events p=54 & p=64 no matter what shutter speed is set.

[a1ex]

At the end of shutter range (30s and 1/8000) I no longer get these events, that's why I was asking.

[1%]

photo mode:
0x54 then 0x69 while in range, when at the ends only 0x54. Both wheels generate the same events. The 0x69 events have different addresses though depending on if you're doing shutter or aperture.

[Marsu42]

0x54 then 0x69 while in range, when at the ends only 0x54. Both wheels generate the same events. The 0x69 events have different addresses though depending on if you're doing shutter or aperture.

Ok, I'm new on the 6D :-p ... so as far as I understand since catching these events is possible, squeezing in 1/180s in M and Tv would be possible?

[a1ex]

Code: [Select]

#define GMT_GUICMD_PRESS_BUTTON_SOMETHING 0x54
You don't know anything more from this event (just that some button was pressed).

[Marsu42]

You don't know anything more from this event (just that some button was pressed).

You're writing this concerning the START/STOP on the 6D (which only trows a 54)? Doh, that's disappointing, I really hope some time ml finds a way to hook into the other buttons, after all the Canon fw must have some means of detecting them... but fortunately as 1% wrote the wheels also throw a 69, so that's covered by ml.

[1%]

They only throw that 0x69 when changing a setting so basically what a1ex said about 5DIII is true for 6D.

[Marsu42]

Btw: I wouldn't want to prevent anyone trying to force 1/200s as max x-sync...

... I know 1/8000s shutter seems to work but is reset by Canon, I don't know if anyone figured out if 1/180s is really the fastest speed when both curtains are open (see also the hypersync request for even faster speeds, but with black stripes).

[Marsu42]

DAMN! :-(

This is even more annoying than I thought - with 1/3ev steps enabled, I tried setting 1/180s ... and the #!$*%# Canon firmware promptly bumps it up to 1/200s!

Code: [Select]

lens_set_rawshutter(116);

One way I can think of around this would be to modify C.Fn I-1 to 1/2 EV just before settings this - is this prop reverse engineered yet on 6D, how do I set this C.Fn via ml? This is not the preferred way though, since flash ec would also jump to 1/2 EV :-(

Any idea on how to get around this and force-set 1/180s in M/Tv w/o modifying C.Fn.I-1?

[Audionut]

I wouldn't have thought 0.25EV exposure difference would be anything more then a minor annoyance.  You're only losing that exposure resolution for the ambient exposure, and only where aperture/ISO is such that exposure would require 1/180s.

[Marsu42]

I wouldn't have thought 0.25EV exposure difference would be anything more then a minor annoyance.  You're only losing that exposure resolution for the ambient exposure, and only where aperture/ISO is such that exposure would require 1/180s.

The problem is that even 1/200s of 5d3/5d2 is slow (1dx: 1/250)... 1/180s is borderline, and 1/160d really hurts - it's about the stopping power, try fill-flashing something further away that moves: You cannot use hss since that hasn't got enough power and creates blur, so you want the fastest x-sync possible.

[Audionut]

I don't have field experience in this particular situation, but the maths says there is a 0.0007ms difference between 1/180 and 1/160.

Have you tried mounting the flashes closer to the subject and triggering them wirelessly?  You could dial down the flash power considerably, gain much better stopping power and extend run time.
Zip tie it to the back of a tree or something.  Or leave it in frame for effect/clone it out in post.

[Marsu42]

Have you tried mounting the flashes closer to the subject and triggering them wirelessly?

Thanks for the suggestions, but you don't want to try this with wild horses that are constantly running around :-p ... really, unfortunately there aren't alternatives in many situations such as wildlife & event. Canon, bless them, know that and that's why their 1dx has 1/250s x-sync while the 5d3 only has 1/200s (and many photogs are still crying about the 1d1 with 1/500s x-sync)

With  the argument "it's just a little" you get from 1/250s which is the least safe bet to a little lower 1/200s to a little lower 1/180s to a little lower 1/160s ... why not go for 1/125s, it's just a little lower after all :-p ... and with a tele lens, the 1/focal length min shutter speed guideline is there for a reason, even a little lower shutter will have an impact summing up all small difference it makes.

I really hope 1% or a1ex have some input about how to modify C.Fn.I-1 which would be some sort of fix...

[Audionut]

I didn't think a1ex was keen on changing properties.

From what I have read, the guys trying to flash illuminate birds from 13klms away (yes an exaggeration), use a snoot with a reflective inside.  This has the advantage of increasing the zoom factor of the flash and giving you a higher GN.  Enough flash light on subject will help with the stopping power.
I've personally never had problems with motion @ 1/160, but I always have enough stopping power with flash.

If you're trying to flash illuminate running horses 10klms away with an 8mm focal length (yes another exaggeration ), you should really think about other means though. 
I can't see how increasing the sync speed to 1/180 while retaining 1/3rd stop FEC is going to net you more keepers. 

If you have enough power to FEC, half a stop isn't that large.  Best case, 1/3rd stop FEC is only ever going to be at most 0.25EV more accurate then 1/2 stop, best case.  If the flash exposure is anything other then at that 1/3rd stop exactly, the accuracy becomes even less.

I'm just being difficult as usual though 

[Marsu42]

I didn't think a1ex was keen on changing properties.

Never the less, some C.Fn props are there and can be changed.

From what I have read, the guys trying to flash illuminate birds from 13klms away (yes an exaggeration), use a snoot with a reflective inside.

It's called better beamer - for my lenses, my 600rt flashes with the 200mm reflector are sufficient and don't have any clutter in front.

I've personally never had problems with motion @ 1/160, but I always have enough stopping power with flash.

Depends on the focal length you're using.

I can't see how increasing the sync speed to 1/180 while retaining 1/3rd stop FEC is going to net you more keepers. 

Of course it is going to net more keepers, the question is: How much... and as I wrote, even 1/180s is very short, so everything helps... if there wouldn't be any keeper increase at all with a faster shutter speed, you could just take all pictures with 1/60s x-sync. The fec isn't about keepers, but about proper ettl exposure - 1/3 fec has a large impact with ettl, 1/2 is often too large.

[1%]

You should have get/set cfn fuctions, if not they are easy to find. Nothing wrong with changing properties to valid values... I tried all sorts of stuff to crack 1/200-1/8000, it always changes it before the shot.

[Marsu42]

Now that we have get/set cfn I think I'll manage to squeeze 1/180 between 1/160 and 1/200 by changing exposure level increments to 1/2 just for this one shutter speed - that's for Tv/M modes.

For Av mode, there already is an option in the Canon fw (flash menu) to limit the shutter speed to either fixed 1/180s or 1/60s-1/180s - but the prop (if it is a single prop) isn't in the 6d headers yet. Question: How do I find the location of this prop or setting and manage to change it from Magic Lantern?

[darkstarr]

im in the same boat so +1 for this feature!

also keep my exposure increments at 1/3 and 1/160 shutter speed sucks lol

[1%]

Question: How do I find the location of this prop or setting and manage to change it from Magic Lantern?

The easiest thing for me is to start a dm log and change the setting then see what props are changed in the log. After that you can browse them and see whats inside.

DM being broken on 7D is what is holding me up from trying out the AF points stuff.

[Marsu42]

The easiest thing for me is to start a dm log and change the setting then see what props are changed in the log. After that you can browse them and see whats inside.

Thanks, I hope I'll come around to try this sooner or later.

im in the same boat so +1 for this feature! also keep my exposure increments at 1/3 and 1/160 shutter speed sucks lol

There is one workaround I recently discovered: In M and Tv mode you can select 1/3ev eli (exposure level increments) and force the camera to go 1/160s->1/180s if you *don't* set the flash to hss mode. This is still inconvenient because you have to decide about a shutter speed range, but it's better than constantly switching eli in the camera.