16 Bit images

[Simon Knight]

Hi,
At present I have Elements 3 plus the Hidden Power book and add ins. I have just attended a presentation on the subject of producing monochrome images and the speaker stated that for best results the image should remain at 16 bit per channel while it was edited. I have been attempting to discover what if any new 16 bit edits PSE 4 allows when compared with PSE 3. I'm also interested in PSE 5 and hope that Adobe might release a Mac universal binary sometime in the future.

Many thanks

Simon

[Richard_Lynch]

I have just attended a presentation on the subject of producing monochrome images and the speaker stated that for best results the image should remain at 16 bit per channel while it was edited. I have been attempting to discover what if any new 16 bit edits PSE 4 allows when compared with PSE 3. I'm also interested in PSE 5 and hope that Adobe might release a Mac universal binary sometime in the future.

There are some advantages for remaining in 16-bit, but you will not have those in Elements, even with Elements 5. My website (hiddenelements.com) lists the important additions to the program for each version, and Elements is still without 16-bit support for layers. As most solutions that I would use in Elements require layers, you won't be achieving these results in Elements is 16-bit. On the other hand, you also don't have any support for 16-bit images in print, so the additional information at this point is really somewhat superfluous.

While I agree that technically editing in 16-bit is a good idea, except for instances of drastic change you will likely get very little actual benefit from 16-bit.

If you tell me more about the techniques, I can let you know more about the effect.

[Simon Knight]

Dear Richard,
Thank you for your reply. The speaker mentioned in my original post was making various modifications to the tonal range of his images mostly by using layer copies with layer masks to extend the dynamic range. He was concerned that 8 bit images would become posterised (?spelling) possibly during the conversion from colour to mono. The conversion was achieved using two hue/saturation adjustment layers; the top with just the saturation set to fully desaturated, the one below used to 'mix' the colour channels with the layer mode set to 'color'. The speaker was keen to stress that 16 bit depth had some 4096 grey steps compared with 256 at 8 bits.

I have decided that I will print some test images at both bit depths to see if I am able to detect any differences especially as I suspect that my printer is unlikely to produce more that 20 odd grey steps.

Simon

[Richard_Lynch]

The only time you will see differences is if you out-and-out damage an image by trying to force it to do something the data can't handle. If you are just going to print images without alteration, the printer will have to convert a 16-bit image to 8-bit for printing anyway. I approve of your intelligent decision to test, but it is my considered opinion that you won't see a difference. If you do, it will not be a difference that makes/breaks the image.

Most of the features that have come out with Photoshop since version 4 have not done much to affect the way you approach image editing. That is not to say 16-bit is not helpful or desireable to use for archival purposes and initial adjustments to your image for exposure correction, but that perhaps a little much is made of what it can and can't do realistically. If you take a colorful RGB image and convert to GIF with a measley 256 colors and compare that side-by-side, you will see some change, but remarkably little considering that you are smashing 16 million colors into a paltry 256. The conversion from 35 Billion or so to 16 Million sounds large, but is far harder to detect in actual practice.

If the speaker is warning, yet showing no examples of the difference, it is likely because they either haven't seen any, or that they really don't even use the technique they are proferring except on the example they have selected for their presentation.

I am not sure who first developed the Hue/Saturation trick, but I've seen it in a few places in the past few weeks. John Nate, Raymond Shay, and others have tutorials on this on their websites, and Martin Evening even touts it, but attributes it to Russel Brown. It is neat, and quicker than fussing with Channel Mixer, but I am still of the opinion that you get the best conversion by knowing how to manipulate color component information, which offers the most options (and the most creditable science).

[Simon Knight]

Dear Richard,
Thanks again for such a comprehensive reply. I hope to do some B&W printing this weekend comparing 16 bit and 8 bit sources.

I have just received a copy of your book hppe4 to get an idea of what PSE4 adds over version 3. My first impressions are that 4 does not offer anything extra except for a new cut out / selection process and in fact Adobe have removed the ability to run saved PS action scripts - Thanks Adobe! I'm still searching for the reason to spend sixty pounds on the upgrade.

best wishes

Simon

[Richard_Lynch]

Originally Posted by Simon Knight

I have just received a copy of your book hppe4 to get an idea of what PSE4 adds over version 3. My first impressions are that 4 does not offer anything extra except for a new cut out / selection process and in fact Adobe have removed the ability to run saved PS action scripts - Thanks Adobe! I'm still searching for the reason to spend sixty pounds on the upgrade.

My books are never about the newest features. I have expanded my tool set from 60 to 100 tools between versions, partly to cover the gape Adobe left. They insist they want to make the software easier...If they had the market right, no one would want my tools.

They did NOT remove the ability to run actions, just certain things in actions -- and there are work-arounds for those (but you'll need access to Photoshop to edit the actions). I have a white paper on my hiddenelements.com website that tells how to install actions.

I agree, though, not a lot of reason to move from Elements 3 to 4.