I know the right answer is to work the raw so that it becomes the jpg, but ...

Sometimes I'll work on a jpg (smaller file size) as a draft (CS6) on my laptop and wind up doing some iterative processing (may include forced flattening as I move across 32bit, LAB, etc.) that takes me down a road that I wasn't intending to go ... winding up with something I like from what I thought was going to be a bust. My brain tells me that I need to repeat it on the RAW for optimal quality ... but my heart says "oh crap" not that again ... I don't know if I could repeat it as well again.

So, I've wondered about the prospects of uprezzing the jpgs (3MB) when the draft turns into something more ... or how should I be best proceeding with a workflow on a small file size (yeah, RAW on a bigger computer, I know) for drafts that might turn into more. Is there a way to apply all that was done to the jpg to the raw ?

Thanks.

While it will use more memory, one thing you could do is to do most of your corrections by using Adjustment Layers. Then, when you're finished (if you like the results), you can save each adjustment and reapply them to the raw file.

I gave up LAB converisons a long time ago. As Bob mentioned, you can use adjustment layers (and smart objects) and with the various blending options, petty much everything you could have done in LAB mode. And Smart Sharpen performs a lot of its magic addressing luminosity -- so again no need to go there. There are ways of controlling USM artifacts as well.

As for 32 bit work, Play with it. See what works. And then back off an go back to the raw file...

You can always downsample your raw file, build your layers, and then introduce the full resolution file when done / close to what you want.

Thanks ... two good thoughts that could work in various situations ... and in concert with each other.

BobCollette wrote:
While it will use more memory, one thing you could do is to do most of your corrections by using Adjustment Layers. Then, when you're finished (if you like the results), you can save each adjustment and reapply them to the raw file.

Yes, I use layers ... how to save each adjustment / reapply

Bernie wrote:

You can always downsample your raw file, build your layers, and then introduce the full resolution file when done / close to what you want.

This too might have some merit ... could you expand on your introduce method

I figure that everything will need to be re-tweaked after the larger file is in play, but at least the "steps will be intact" .. sans the bulk of the iterative process. It sounds like I could (if I find myself needing to) also reintroduce the full file prior to "flattening" if need be, then again downsample for 'next phase' of iteratives.

What would you recommend for downsampling to ... although I guess it isn't uber-critical, since I would likely have to repeat anything that is actual pixels anyway (i.e. cloning, spotting, etc.),

I would suggest that if you just want to see if something works in terms of cropping, color, and composition, don't be concerned about cloning and spotting. That goes for other pixel level stuff like downsampling. Just use CS6 bicubic automatic.

When you've built your background layer or smart object with the other layers, you can just set your canvas back to the original size, edit your object. If you're using smart objects, you could have done spotting in ACR. Do the rest of your pixel level editing and then save your work to the smart object or cut & paste to the background layer.

I always save the PSD file so I know what I did. I have too many different ways of editing -- and they change through time. Altough I may finally be settling on some "styles". The web version is completely flattened. Print versions may have layers which adjust specifically for the print process (signature, contrast, tonality, final sharpening, saturation).

RustyBug wrote:
Yes, I use layers ... how to save each adjustment / reapply

Most image adjustment tools in PS have a Save & Load button, allowing you to save an adjustment and then load it to use it again.

Hell, just drag the Adjustment Layers from one document to the other. If there are masks in that Adjustment Layer, they won't automatically scale to the larger document, but anything that's a straight adjustment it won't matter anyway. I do this ALL the time. Drag and Drop. No Save and Load needed.

Peter ... thanks, it seems to be working fairly well @ Drag and Drop.

Not perfect @ color profile variance, sharpness etc., but certainly in the ballpark to skip all the iterations as if I had recorded an action all along the way. A handy piece of the puzzle to be sure.

As long as I remind myself to downsample the raw, instead of using the jpg, the color profile variance won't be an issue ... and remember that if I do any pixel editing (i.e. spotting, sharpening, blur, etc.) it will have to be repeated. Those things are pretty straight forward to repeat. It's those tonal values that can be a pain to replicate.

Hmmm, at mask size differential.

Bob,

I still haven't figured out the save load @ layers ... a little help

Bernie,

+1 @ multiple workflows. I'm doing a fair amount of mask painting in conjunction with the layers currently. The thing I like about that is that I can not only dial in the adjustment and the opacity, but it can be further refined via mask opacity and feathering, etc., particularly as the iterative process progresses ... easy to backtrack and change the mask opacity selectively.

I haven't ventured into "smart objects" yet.

RustyBug wrote:
Bob,

I still haven't figured out the save load @ layers ... a little help

Sorry, I should have been a little clearer. You don't save the adjustment layers per se. The adjustment layers just "keep track" of the adjustments you've made (Levels, Hue/Sat, Selective Color, etc.). When you're finished (the low res image looks the way you want it to), open each adjustment layer and save the adjustment value. Once you've saved all the adjustments, you can then open the full res image and load each adjustment value that you saved.

Yes, you *can* save the adjustment values, but what's the point if you're going to keep a layered lo-res anyway, where it's so much faster to just drag from one file to another. You can even drag multiple layers at one time. I guess because I do this all day long, I'm always looking for the fastest, most efficient and most effective way to get things done.

Peter, apparently you're a more experienced Photoshop user than I am. I wasn't aware that you could just drag a set of adjustment layers from one image to another. Your way is much more efficient.

There are times when you might want to save and load, but I can't remember the last time I actually did that - maybe to send a specific adjustment to someone for them to load.

For the Drag and Drop, even if you have a mask, it will come over, but it won't be the right size going to the larger file. If you hold the Shift key when dragging, the mask will be centered and you can then Option-Click on it to show it, and then scale it manually using Free Transform and the Ctr + Shift and it will scale to fit. If it's a simple mask that may be enough, or you may have to do some fine tuning as well.

Bob ... oic @ save each, good to know.

Peter ... you read my mind @ mask scaling to resize ... that should work fine for my mask painting style ... it typically is feathered anyway, so not uber-critical, plus the mask is very "editable".

Thanks ... I think these ideas will be helpful as part of a "plan B" workflow workaround when needed.

Gotta love FM'ers ... thanks guys.