Here's the thing:  I've modified a transmap to get rid of the bangs -it's not the first transmap I've made, and yes I checked that it's black and not a shade of grey.
You'll notice on the lower right-hand corner is where the bangs were.  I've been using this modified transmap for years.

Now, with IDL during calculation I can see that the whole area of the bangs is calculated and is black with the red dots.   When it renders it's not fully transparent and the forehead as you can see is a darker colour. 

I'm really at a loss to explain this so I'm hoping someone can shed some light.  Please, before I pull out all of my hair :)

Thanks!

Have you tried plugging the diffuse value into the transparency map? Just a thought. Not sure if it'll solve your problem tho.

Laurie

Thanks for the suggestion Laurie, just tried but no, doesn't work I'm sorry to say.   This has really got me mystified.

My last guess then would be set transparency edge to 1 and then plug that into the trans map.

Wait a sec...try plugging your specular rolloff and reflectivity into the trans map too on the Blinn node. That might just be it... It's probably still picking up information to put something there since there's no map telling it to ignore the bangs area...

Laurie

No, tried both but the dark area still persists in showing through.   I'm now thinking it might be something with the mesh, I'll try something else in uvmapper.

Ok I took the .obj into uvmapper and selected those facets on the bottom right of the map where the bangs are and assigned a new material "Bangs" to them.   Then I just made that totally transparent, no map, no diffuse, no specular nothing.  
It still renders darker there!

So, now I've had it with mapping, since I assigned a new material I just used the grouping tool and selected the other materials except the bangs and spawned a new prop,  Finally got rid of that geometry :)

But I'm still curious as to why this is happening.

If you are not using GC please ignore my post.

Thanks for the suggestion madno2 but I'm afraid the results are the same.   I even tried it without any GC at all and I still get the dark forehead. 

My other solution isn't going too well either, the newly spawned prop without the bangs has a different geometry because the morphs that I copied over do not work so it's pretty useless without them . . .

If anyone has nothing to do and wants to kill time it's the Sapphire Fox Hair and I can send you the modified transmap to try it out.   I don't think GC is the problem, I think it has something to do with the specific geometry of this hair.  

I guess I'll just have to resign myself to doing 2 renders when I use this particular hair, one with IDL and one without to composite them.

they're aware of this problem (FFRender calculates transparency wrong with GIvariables invoked). 
for now, uncheck "visible in raytracing" in hair properties.
or, later, try rendering it in something else (carrara, maxwell, lux et al.)

So it's not me? Phew, that's a relief, now I can stop trying to make it work. Thank you!

I think Miss Nancy is correct that there is a known issue with IDL and transparency.
However, setting RT bounces higher than 1 (try 2 or 3 in the default render dialogue) can sometimes improve results. This will, of course, have a heavy impact on render times.
Worth a try.

Thanks carodan, I'll have to try that.  I have it set to 2 for these test renders and depending on the scene I increase it.  Anything more than 6 though will be excruciatingly slow for me.

I've done all your steps, Kalypso, in terms of blacking out the bangs part of the transMap, except instead of using Blinn I used Anisotropic, which worked well on another figure but not so great on this one. But anyway, same artefact... except not quite as noticeable over the lower part of the face, but you can see clearly where the bangs mesh stops in the front of the hair near the hairline.
I rendered this image in PP2010 (my new favourite toy!!), transMaps set to GC of 1 - which turns it off - and texture-filtering also to None, using a skydome of BB's and 3 infinite lights. Why 3? No clue... I should be using only 1 or 2 at most, but oh well. RT shadows set on Light 1.
The anisotropic reflections conflict with the burnt-in specular - wish there was an easy way to re-texture hair to get rid of that. Is there like even one hair texture resource kit that doesn't "feature" burnt-in specular? A bit like trying to find eye colour textures without burnt-in reflections. You have Buckley's of finding that, too.
In this hair:

... the specular is more spread-out and not so obtrusive so that you can do your own specular stuff and have it look reasonably natural... eventually. Still need to play with this more.

But yeah... here's the render of SF...

Same issues are you described. Oh, and that's with raytracing set to off for the hair.

Oh Robyn, thank you for going to the trouble to test it for me!  For testing purposes I just had one IBL and another light for shadows but I first came upon the problem after rendering out a final image that took 4 hours or so.  So I made another render without the IDL to salvage it.   But it bugged me and so I wasted half a day trying to figure it out ;)  At least now I'm sure it's not something I'm doing.  So I think I'll stop modifying the texture as well - as you can see that has its problems too.
I didn't use anisotropic because bagginsbill recommended against it except for very specific instances which I've now forgotten what they were.
And yeah Poser Pro 2010 rocks!  And you're right, very little hair will look good now with IDL and GC since most of it has burned in specular.  I shudder to think what they will look like in Lux.  So now I'm sleepless and playing in the Hair room since that's probably the way to go.
Now if only carodan would share some tips on growing such natural hair as in his images :) wink wink nudge nudge

Made some tests and now being confused:
see the pictures:

Sorry Calypso at the moment I don't know how to reproduce this transparency issue.

One workaround might be to simply raise the transparency value of your map to 2 instead of 1. That's not elegant but might work.

I will go to bed now, it is already early morning here.

 That is a very subtile difference, but noticeable. I do not think that using white props will reveal the issue here because if you look closely at the forehead skin color, its final color is affected by having light passing twice through a transmap (going in, and then back out).

Just a suggestion, to reveal (duplicate) the issue, you'll likely need to paste a color strip as a texture on the box primitive. The forehead's color is more red/blue now, as if a little bit of green has been subtracted.

Good luck with this!  : )

And that would explain why it usually is so obvious on hair since it's right next to the skin.    Thanks madno2 for verifying this as well and for going into such great detail.   At least in other cases we can mitigate the effect.