questions about degrunge technique

[mountain]

Hi all,

I have some questions about the degrunge technique published by byRo, http://retouchpro.com/tutorials/?m=show&id=213

I have a few points i can't really understand.

first the 3 layers method..

after 3 layers are done, it mentions to turn off and on to check the 'grunge'..these grunge layer is supposed to show the 'details' in between blurred and high-pass (sharpen) details, right?

so in the tutorial it mentions to put a 'reveal all' layer mask..and paint black but watch out the outline..i don't really understand this part.

does it mean to paint black over the layer mask of 'grunge layer'..so the blurred layer under it will smooth the skin?

and about the second part, quick degrunge method..
it mentioned some like ctrl+I, to invert.. and then 'hide all' layer mask and paint white.. I am quite confused of this part also, does it mean i can just skip the ctrl+I, and just use 'reveal all' layer mask and paint black?

and in this quick degrunge method..there is only original images under..if we had layer mask on it and brush away the grunge layer, it is the original skin that's revealed (in the 3 layers method, it is the blurred layer under the grunge layer), so it sounds like only sharpen the details without smoothed the skin.

I hope someone can help me to understand it..thanks!

[Boneappetit]

Hi mountain: As for the 1st part I don't know exactly how to explain it to you.. If you do the quick method the first one is not necessary, they wrote it there so we can technically understand it. "Hide all" and "Reveal all" are counterparts... When you do a hide all (black) mask it means you want to locally apply the effect, for instance, on the skin or texture. The black mask hides everything, and a white brush reveals the parts you want to reveal.

When you do a "reveal all" (white mask) what you want to do is bring back details from the last estate of your image before the actual effect, (a black brush will bring that back). So, to invert the layer is necessary in this process...

Watch out for the outlines, means that you don't want to apply the degrunge effect on the edges, (eyes, mouth, chin etc) because it looks pretty messy and unnatural.

Hope this helps.

[mistermonday]

does it mean to paint black over the layer mask of 'grunge layer'..so the blurred layer under it will smooth the skin?

Basically the answer is yes. Think of this technique as separating the image into detail defined by edge radius. The Blur layer isolates and protects all of the edges in the image whose radius is 5.1 px or greater, and it blurs everything lower than 5.1. The detail layer isolates finer detail edges at 1.7 px and below. The grunge layer represents all the edges between 1.7 and 5.1. Linear Light blend mode at 50% opacity acts like a mathematical addition. therefore when you add the 3 layers GB + Grunge + HP your image looks just like the original. Adding a Reveal All mask to the grunge layer has no effect but painting black on the layer mask effectively blocks out / eliminates the grunge in that area of the image. In doing so, you are blurring the image in that area thereby "degrunging" it. The degree to which you degrunge is directly proportional to how much black you paint on the mask. Usually you start with a low opacity and blur only the the extent that you need to.

and about the second part, quick degrunge method..
it mentioned some like ctrl+I, to invert.. and then 'hide all' layer mask and paint white.. I am quite confused of this part also, does it mean i can just skip the ctrl+I, and just use 'reveal all' layer mask and paint black?

No, it's not the same. The two layer method just further takes advantage of math. Inverting a layer is mathematically equivalent to -layer. This leaves you with a degrunge layer on top of your original which looks blurred until you add the Hide All mask so you just see the Original. By painting on that mask with white, you are allowing the degrunge layer to show and thereby blurring the original in those areas.
Regards, Murray

[mountain]

hi thank Boneappetit and Mistermonday, i realise that if you get near to the outline, it will have strange glowing and plastic looking...

but it seems the quick way is lacking the control of the first method, which one can control the opacity of details and blur..

should we usually remove the obvious thing like bumps or pimples before we degrunge? or it should do it after?

[mistermonday]

Originally Posted by mountain

hi thank Boneappetit and Mistermonday, i realise that if you get near to the outline, it will have strange glowing and plastic looking...

but it seems the quick way is lacking the control of the first method, which one can control the opacity of details and blur..

should we usually remove the obvious thing like bumps or pimples before we degrunge? or it should do it after?

Mountain, the two methods produce the exact same result but the one uses 3 layers and you paint black on a white mask, the other method uses only two layers and you paint white on a black mask. In both cases you vary the opacity and / or the flow of the brush to have exactly the same precise control. If the two methods work differently, then perhaps you are doing something different in the process of creating the two layer method. ByRo only used 3 layers as a means to illustrate the concept. There is no reason not to use 2 layers as it is quicker and gives you a smaller file size.

Usually it is better to remove the large blemishes like scars, pimples, etc because those are textures /areasyou want to get ris of completely as opposed to smoothing.
Regards, Murray

[Boneappetit]

Originally Posted by mistermonday

Mountain, the two methods produce the exact same result but the one uses 3 layers and you paint black on a white mask, the other method uses only two layers and you paint white on a black mask. In both cases you vary the opacity and / or the flow of the brush to have exactly the same precise control. If the two methods work differently, then perhaps you are doing something different in the process of creating the two layer method. ByRo only used 3 layers as a means to illustrate the concept. There is no reason not to use 2 layers as it is quicker and gives you a smaller file size.

Usually it is better to remove the large blemishes like scars, pimples, etc because those are textures /areasyou want to get ris of completely as opposed to smoothing.
Regards, Murray

Agree 100% Murray... Couldn't be more clear than this...

[danielstagner]

This may be a dumb question, and I realize this thread has been closed for a while, but isn't this just blurring the skin? I thought that was frowned upon in retouching.

[mistermonday]

Yes, it is a form of controlled blurring. Used carefully and in the right places it will have a subtle effect while saving time but it can be visible. It is a trade-off between speed and quality so it depends how high end a job you are doing.

[kav]

Originally Posted by danielstagner

This may be a dumb question, and I realize this thread has been closed for a while, but isn't this just blurring the skin? I thought that was frowned upon in retouching.

You're correct. I never found it necessary. There are better ways to approach the things that it does tend to hide. To me it was something that was popular due to heavy sharpening and bad color interpolation of the really old digital cameras. When I look at most images, much of the skin texture that is there can stay. For big stuff it's not like such a technique would fix it anyway. The rest can be evened out. Your eyes lie to you sometimes and make things much rougher than they really are in the early stages of working on an image.