I’ve been on the fence about hardware UV filters until a recent experiment with Absolute Colormetric rendering. Using the an inkjet RIP I calibrated and profiled our Epson7800 with matte black on Epson Enhanced Matte stock. I built a paper profile with Monaco/DTP70 with the UV filter engaged. The resulting proof with an AbCol rendering was miles away from the CREO Spectrum proof. I remeasured the profile target without the UV filter and the resulting proof was a significant improvement to the color match with AbCol.
Lately I have found using a hardware UV filter on inkjet media with high quantities of optical brighteners is not yielding color accurate proofs. I always endup editing the destination profile to add more yellow. Examing the two profiles in CTPro shows the gamut volume is different too:
With UV filter = 342,525
Without UV filter = 353,536
So using a UV filter cuts the gamut down by 3% not huge loss but the visual results show where that loss counts.
The hardware UV filter seems to compromise an ICC profile. As a result, AbCol rendering is not simulating a match to a yellow substrate like Kodak Commercial base. This would also affect the highlights, but some of the flesh tones are lacking yellow too.
Even if the UV color is not visible to my eyes, I can see the benefit of turning off the filter when building a paper profile. Why use a hardware UV filter at all?