Datacolor User to User Support Group.

Try this for a test image.  Background painted with full saturation 
industrial pigments.  5d image shot in Raw, converted to 8 bit tiff 
prophoto rgb with capture one.  Reduced to 3 mb.

www.laskodyphotography.com/bluetest/bluetestimage.tif


Randy Laskody

Tom wrote:

>--- In colorvision_group@yahoogroups.com, randy <rlphoto@...> wrote:
>  
>
>>Hello everyone,  I can handle a little color shift here and there. I do 
>>90% custom portraits for a living and the rest editorial, so exact
>>    
>>
>color 
>  
>
>>is not necessary. As long as it is close and looks good.  The skin
>>    
>>
>tones 
>  
>
>>are the most important of anything. I was very impressed with PFP skin 
>>tones so far.  It was also able to correct the green blocking shadow 
>>problem of the actkinson profile that I was having also.  But I cant 
>>deal with the posterization that occurs along with the color shift. And 
>>sometimes the color shift is pretty severe.  And if doctor pro only
>>    
>>
>uses 
>  
>
>>PS tools for correction I cant see this as a global solution either.  
>>The actkinson profile doesn't have this problem, so I know it is 
>>correctable.  I believe like Xun, that  the problem lies in the 
>>algorithm.  But all I ask for is a solution. If I am at fault is some 
>>way, please help me correct it so I can go on using this product and 
>>making a living.  If this issue is not correctable,  then let me
>>    
>>
>know so 
>  
>
>>I wont spend anymore time looking for one.
>>
>>Thanks and have a good day.
>>
>>Randy Laskody
>>    
>>
>
>What if collectively we come up with a somewhat harsher test image. 
>One that stresses the portions where we are finding issues like
>posterization and blue/purple color shifts?  It may be interesting to
>keep it photographic, not mathematical like a Granger Rainbow.
>
>I'd like to have a use it for the following to test how well a profile
>works (in gamut colors) and how well the CMM engine works (out of
>gamut colors) while making clear which component is responsible for
>each.  It would also be interesting if it had gradients which stressed
>the areas where 8bit images show posterization and 16bit images do
>not.  I realize that this test is a bit unfair but ... if we keep it
>photographic ... its definately real world and exercises the real
>world issues we face.
>
>What do you think?  Is this just beating a dead horse because it has
>been done already?  How about the center containing colors which are
>normaly in gamut while the outside contains colors which stress the
>CMM engine?
>
>Its just an idea ... feel free to stomp on it .. shoot holes in it
>etc.  Thats the only way it really gets better :-)
>
>Thanks
>
>Tom
>
>
>
>
>
> 
>Yahoo! Groups Links
>
>
>
> 
>
>
>
>
>  
>