Automask - show actual mask rather than what is painted

When using the Automask you always see what is painted as a mask but I would like to see the actual mask that is used. This would make it much easier to see what is masked and for making fine adjustment.

This request is one more dupe - to a very long standing issue.

Unfortunately, “Auto Mask” generally does not work in a reasonable way at all. :frowning:
Also see Shortcut to observe the actual auto-mask?
(and particularly the links I provided in that thread).

In fact, it is really frustating that this issue apparently is completely ignored by DxO for several (more than four) years now, despite of a fix being promised for almost the same time…

2 Likes

Yeah, I tried Auto Mask for the first time yesterday and assumed I had missed something. I could tell no difference between it and a normal brush. What is it supposed to be doing?

Auto brush is hard edge aware and there is no feathering.

Mark

1 Like

Keep the smaller inner circle strictly inside the area to be masked and make sure to paint the whole area. Allow the larger outer circle to extend over the edges and the actual mask will stop at the edges, though you can’t see it until you make a big adjustment(e.g. Lum -100). A change in this behavior is what this thread is all about.

3 Likes

Ok that worked pretty good on a quick test. It would need some cleanup, but pretty close.

Yes, you can go back and forth between the Automask and the eraser to touch it up.

1 Like

We’d need an option to show either the mask painted OR the mask implemented - not only the latter … otherwise, it would be very difficult to fine-tune the mask (with Eraser & Repaint) if we could not see the mask we had actually painted.

(That’s perhaps the reason for the current implementation)

John M

2 Likes

It’s on my “list” :+1:
Showing the actual mask should be the way to go.

Steven.

5 Likes

@John-M, I respectfully disagree, the current mask shown has no relationship to the actual mask used, it only shows where the outside circle was drawn but you have no way of telling where the inner circle was. The actual mask used would show the area that is masked and you will be able to see very clearly see if it went outside the edges or did not reach the edges you want to detect.

For editing you can either remove areas you don’t want masked and also add to the areas you want masked.

1 Like

Great to hear :slightly_smiling_face:

1 Like

Any idea about when this will (finally…) be fixed?

At the moment we don’t have any timeframe to give you.

Steven.

Well, I hope it is soon and not have us wait for PL6.

When @StevenL indicates that they do not have any timeframe, that most often means it is on a longer-term unscheduled list of future enhancements, and will probably not be available until at least PL 7 or an even later version. If it was going to be in PL 6 or before, they would have a timeframe for it. PhotoLab 6 should be released in 3 months during the 3rd week in October.

Mark

1 Like

I agree to Mark (except that it’s not about a future enhancement, but a necessary fix to a broken feature), and that’s one reason why I am really frustrated about the behaviour of DxO in this concern.

Remember that this bug has been promised to be fixed “soon” back in 2018, which is four years (!) ago meanwhile. And still they don’t even have any timeframe for the fix.

I would never dare to behave that way against my own customers…

2 Likes

I fear there is a tendency for, at best, very slow reaction to many problems. This at many times is when they create problems by only part implementing something like a part implemented mask, or in my pet annoyance, remove focusing distance from Sony E lenses in Windows but not in Mac (and with this there is just a refusal to respond on when this will be rectified (bearing in mind it works in the Mac version!)). The bigger problem is the refusal to even respond to problems and that I fear happens a lot at least here there are some , if not very positive input from DxO on masks.

I share in feeling frustrated by not having a simple (e.g., keystroke) way to determine what the effective (not “painted”) mark is. It seems to me this is an integral part of the tool; it’s absence leaves the masking process incomplete. In effect, “well, you can sling paint, and something will happen, but what the results will be is something of a surprise”. Obviously, practice and experience reduce the chances for unexpected results, but having better control over any tool is to the good.

Thanks to rrblint for the suggested work-around.

2 Likes

Totally support this feature request …