Colour Space bug in CS5 12.0.4

[Colcam]

Are any of you having trouble with prints coming out a bit flatter in oomph or wrong colour balance? I have been chasing the problem for a while now and it seems so are many others around the traps. In sepia photo's which I produce quite a few of, the colour difference is from a brown to a slight greenish tint.

I have been having a great deal of trouble with colour instability since recent updates to CS5 also. Am running XP64, CS5 12.0.4, Camera Raw 6.4.1. I run Adobe RGB 1998 workflow from camera to printer.
Seems to be totally random result of some files returning an sRGB colour space and some returning AdobeRGB 1998 when converted to jpegs for printing. Same result for scanned images too, so it is not the camera. The 32bit version appears to be worse, but the 64 bit version does it too.
EXIF details in Bridge show Adobe rgb but this is not the case when viewed with a hex editor. Some files are showing up to four colour profiles embedded within!! I processed the same two files (one a psd & the other a tiff) no less than seven times with totally random results. I could not return a pattern. Sometimes sRGB & sometimes Adobe! A quick way to see it if you do not want to go the hex editor way is to download Qimage http://www.ddisoftware.com/qimage/ and wave the mouse over the file. The colour space among other details show in the status bar at the bottom. It may only appear after saving as a jpg file. I have not proved this conclusively as yet.

The only way I can get around it is to : Edit/Convert to profile/AdobeRGB 1998 and then do a SAVE AS on the original file. The jpeg that I then save off this to the print output folder is fine. This seems to ensure me the result I need to print accurately.
Reference to this problem at:
http://feedback.photoshop.com/photoshop_family/topics/corrupt_jpg_from_12_0_4
http://ddisoftware.com/tech/qimage/off-color-problem/30/

Note to ADOBE - NOT HAPPY!!

[Matt. K]

Colour printing is half science and half art. Try running 4 Epson photo quality printers in a classroom and watch the problems multiply. First step is....of course....calibrate the monitor. Then try and get some kind of reference to work with and go through some logical comparisons to identify the weak link. It is often the printer and something as simple as a cleaning cycle might solve your problem. Sometimes 15 to 20 cycles might be required to unblock a faulty jet. If you can borrow a printer and compare results this can be useful. One thing I tend to believe....a well exposed image from a modern digital camera is likely to have fairly accurate colours, straight out of the camera. If you find you are making endless colour corrections then the problem is likely a colour profile mismatch with paper type or some other problem. Be warned that trying to solve the problem may possibly drive you mad. I've been to that place.