Deconvolution, Smart Sharpen,... for web output sh

[lbenac]

Hello,

As indicated I am a beginner shooting RAW with Pentax gear. I use a very simplified work-flow based on ACR and CS3 and do not pretend to fully understand all the posts in this forum. So I am definitely not part of the "old boys" club and as such it might be the wrong forum to post.
I do very limited printing (not good enough yet) so output most of my photos for monitor or web (Zenfolio, Pentax Gallery).
I usually do capture sharpening in ACR and then create a Smart Object for all localized adjustments (masks, layers, filters, PT lens, Noiseware,..) and final sharpening in CS3. I store the DNG as my archives, keep PSD as my working copies and finally output a JPG with frame for viewing on my monitor or post on the web.
Recently thanks to this forum, I have tried PKS. I have found that either because of my limited knowledge or because I am not outputting for print, I preferred the results I got using Smart Sharpen on various Smart Layers.
Being curious by nature I have read some of the posts on deconvolution software and gave it a try. I could not try Focus Magic as I am using a Vista 64 bit machine but tried Focus Fixer plugin. On very limited test I really liked the result compared to Smart Sharpen. Now the draw-back is that it looks like the plugin will not work on Smart Objects. I really like to work with Smart Object and make changes of some of the adjustments at posteriori.
Would anybody know of a similar plugin that would work on Smart Objects (and do not cost too much)?
Should I reconsider my workflow using Smart Objects and start using normal PSD files and normal layers using adjustements tools in CS3 as opposed to ACR (outside of the initial conversion, toning,....)?

I am aware that everybody have their own preferences and work-flow but weighting in arguments from learned individuals  is usually conductive to faster progress.

Thank you.

Cheers,

[marcmccalmont]

I to have found focus fixer the best sharpening algorithm out there
They advise to use it early in the process so if for my 5D I set the sharpening in DxO optics to .25 and for my P30 I do no sharpening in Capture 1.
When I first open the file in Photoshop I sharpen the background layer using focusfixer (usually .9 sometimes up to 1.3) I then rename the background layer "Sharpened" and that's it. Looks good on prints from 8x10 to 17x24 (iPF5000).
Marc

[01af]

Working with Smart Objects makes sense when putting images into other images or into page layouts. It does not make sense when processing just a single photograph. For non-destructive edits, use layers. And before flattening and downsizing, always save a layered copy at full size.

-- Olaf

[Tim Gray]

Working with Smart Objects makes sense when putting images into other images or into page layouts. It does not make sense when processing just a single photograph. For non-destructive edits, use layers. And before flattening and downsizing, always save a layered copy at full size.

-- Olaf
[a href=\"index.php?act=findpost&pid=176621\"][{POST_SNAPBACK}][/a]

I think this is not the best way to look at smart objects.  There are some edits, eg shadows and highlights that commit the pixel changes regardless of the layering technique.  Also any of the filters really should be applied to a smart object if there's any possibility and wanting to go back and change eg: the amount of blur.

I personally do not open the file from the Raw converter as a smart object, but I do understand the reasons why someone would want to.

[lbenac]

I think this is not the best way to look at smart objects.  There are some edits, eg shadows and highlights that commit the pixel changes regardless of the layering technique.  Also any of the filters really should be applied to a smart object if there's any possibility and wanting to go back and change eg: the amount of blur.

I personally do not open the file from the Raw converter as a smart object, but I do understand the reasons why someone would want to.
[a href=\"index.php?act=findpost&pid=176635\"][{POST_SNAPBACK}][/a]

Hello Tim,

Yes this is exactly why I do it. I am sometime victim of the "what the hell was I thinking a week ago" syndrom and I go back and change local tone, noise reduction, sharpening.... on certain masks and Smart Objects make this really easy. On top I am dealing with the same set of controls in ACr which is a bonus for a beginner.
I found that when working with layer in a normal PSD your only choice is often carefull naming of the layer and deleting the layer to start over.
I found Smart Objects more intuitive in their use as describe in Real World Camera Raw for CS3 on page 192~197 dealing with local tone correction.
For a beginner I found this very usefull.

Cheers,

Luc.

[lbenac]

I to have found focus fixer the best sharpening algorithm out there
They advise to use it early in the process so if for my 5D I set the sharpening in DxO optics to .25 and for my P30 I do no sharpening in Capture 1.
When I first open the file in Photoshop I sharpen the background layer using focusfixer (usually .9 sometimes up to 1.3) I then rename the background layer "Sharpened" and that's it. Looks good on prints from 8x10 to 17x24 (iPF5000).
Marc
[a href=\"index.php?act=findpost&pid=176545\"][{POST_SNAPBACK}][/a]

Hello Marc,

By the look of it, no deconvolution plug-in will work as a Smart Filter.
After some trial and errors, I have changed my work-flow to come close to your comments.
1)I made all my normal adjustments in ACR.
2)Open the processed raw DNG in CS3 (not as a Smart Object)
3)Use Focus Fixer on the background layer (so far 0.8 to 1.9)
4)Save the file as a TIFF that I can then open via ACR as a Smart Object if local adjustments are needed (by that time what is really needed is a drink...!). Of course the file as a Smart Object become huge (10MB>35MB>150MB)

Regarding using FocusFixer early in the workflow, do you still do some capture sharpening in ACR or not? My limited understanding of the deconvolution would say that it would not be needed but as pointed out I am a beginner.

Any comments or suggestions would be very welcome.

Cheers,

Luc.

[JeffKohn]

I could not try Focus Magic as I am using a Vista 64 bit machine but tried Focus Fixer plugin.

Focus Magic works just fine on Vista 64 for me, although it doesn't work with smart-filters (neither do most of my 3rd-party plugins). I'm not so sure that's an issue with sharpening though, my approach is to over-sharpen a bit on a duplicate layer and the adjust to taste with opacity slider.

I haven't compared it to FocusFixer though, so I don't know how it stacks up.

[marcmccalmont]

Hello Marc,

By the look of it, no deconvolution plug-in will work as a Smart Filter.
After some trial and errors, I have changed my work-flow to come close to your comments.
1)I made all my normal adjustments in ACR.
2)Open the processed raw DNG in CS3 (not as a Smart Object)
3)Use Focus Fixer on the background layer (so far 0.8 to 1.9)
4)Save the file as a TIFF that I can then open via ACR as a Smart Object if local adjustments are needed (by that time what is really needed is a drink...!). Of course the file as a Smart Object become huge (10MB>35MB>150MB)

Regarding using FocusFixer early in the workflow, do you still do some capture sharpening in ACR or not? My limited understanding of the deconvolution would say that it would not be needed but as pointed out I am a beginner.

Any comments or suggestions would be very welcome.

Cheers,

Luc.
[a href=\"index.php?act=findpost&pid=177166\"][{POST_SNAPBACK}][/a]

Luc
For my 5D I use DxO and sharpening at .25 (low) and for my P30 I use C1v4 and no sharpening. Focus fixer seems to work well as a one step solution. Experiment a little but I think you will find it will replace your usual capture sharpening.
Marc

[lbenac]

Luc
For my 5D I use DxO and sharpening at .25 (low) and for my P30 I use C1v4 and no sharpening. Focus fixer seems to work well as a one step solution. Experiment a little but I think you will find it will replace your usual capture sharpening.
Marc
[a href=\"index.php?act=findpost&pid=177244\"][{POST_SNAPBACK}][/a]

Hello Marc,

Yes that seems to be the case on the few photos I have been processing for testing purpose.

Cheers,

Luc.

[Philmar]

Maybe I am totally off base here but the intended medium is for web based viewing on a computer monitor, not prints. It is not my intention to denigrate your work, after all web viewing is my sole medium as well. I too am a photohobbyist that does not do critical work for clients. I too agonised over my workflow and tried to do whatever workflow it took to get the best possible results. But ultimately your work is going to be viewed on a computer screen which in my opinion means that if you were to compare the results you get using the many different methods and products available to you that you'd find you really have to strain your eyes and pixel peep before you really saw any big difference betwen the various tools and workflows. My decision was to try to do as much as possible in ACR and CS3 because I was finding my workflow way too cumbersome and time consuming (I'm not getting paid for this). So now I use a workflow that gets the best possible results within an easy and quick workflow. This would change of course, if I were a pro churning out critical work for clients.
Perhaps I am totally wrong but I don't know how much of an improvement you get adopting a complicated workflow using smart objects instead of adjustment layers and various third party plug ins that require going in and out of Photoshop. You might see a discernable difference with printed work but would it really make that much of a difference on a downrezzed web based image? I didn't see much of a difference with my web images that warranted a workflow that doubled my time to process a photo. That was my decision and may not be yours. I wasn't interested in the Zen of a perfect photo, just an excellent photo.
Others may feel free to set me straight. I always want to learn.

[lbenac]

Maybe I am totally off base here but the intended medium is for web based viewing on a computer monitor, not prints. It is not my intention to denigrate your work, after all web viewing is my sole medium as well. I too am a photohobbyist that does not do critical work for clients. I too agonised over my workflow and tried to do whatever workflow it took to get the best possible results. But ultimately your work is going to be viewed on a computer screen which in my opinion means that if you were to compare the results you get using the many different methods and products available to you that you'd find you really have to strain your eyes and pixel peep before you really saw any big difference betwen the various tools and workflows. My decision was to try to do as much as possible in ACR and CS3 because I was finding my workflow way too cumbersome and time consuming (I'm not getting paid for this). So now I use a workflow that gets the best possible results within an easy and quick workflow. This would change of course, if I were a pro churning out critical work for clients.
Perhaps I am totally wrong but I don't know how much of an improvement you get adopting a complicated workflow using smart objects instead of adjustment layers and various third party plug ins that require going in and out of Photoshop. You might see a discernable difference with printed work but would it really make that much of a difference on a downrezzed web based image? I didn't see much of a difference with my web images that warranted a workflow that doubled my time to process a photo. That was my decision and may not be yours. I wasn't interested in the Zen of a perfect photo, just an excellent photo.
Others may feel free to set me straight. I always want to learn.
[a href=\"index.php?act=findpost&pid=177555\"][{POST_SNAPBACK}][/a]

I will answer in two parts:

1) I do not pixel peep because I can't - I do not have this knowledge - but when I processed several images with Smart Sharpen on one hand and Focus Fixer on the other hand - I found on many (not all) of the images a very visible difference to my eyes. Hence my interest into the deconvolution approach.

2) Smart Objects
As stated I am a beginner, which means that what I do one evening I might undo the day after because I am still learning and looking at some PP after some time just slap me in the face to make a change. This is not the case on many pictures but on some and at that point I want to be able to make these changes without starting from scratch (aka . DNG file). I found that I learn from this process and hopefully improve.
Would I one day master all the possibilities of CS3 - I doubt it. But I certainly don't mind investing some time into processing some of my images into Smart Objects which will save me starting over if I want to make changes to the PP. I found it very gratifying to come back to an image that you really liked with fresh eyes and making a small change that make this image even more likable.
That said I agree with you that I would not process all my images like that.

And yes my wife complains that I spend too much time doing PP but on the other hand she would love me to take vacation or family and friends portraits.

Cheers,

Luc.

[Philmar]

Hey I'm not trying to discourage you from your current workflow. I just notice that many of the people in this forum are experts and pros and they spend far much more time on a single photo than many photohobbyists would. They have developed workflows that get them the best results possible with large prints, critical prints for clients and their workflow MAY not be the best one for someone who, like me, is a hobbyist creating webbased photos 1100 pixels long. I don't have 2 hours or more each day to devote to post-processing. My girlfriend sees to that! I just wanted to let you know that I too looked for the best sharpening plug-ins and the best downsizing method as well as the best RAW converter as well as the best noise reduction strategy. It created a long complicated workflow that was too long for me. It may not be for you. I decided to do everything in calibrated ACR/CS3 with adjustment layers, recorded actions and the results may not be optimum, but more than satisfactory. I just wanted to give you my perspective which may end up being different than yours. My concern was that the many pros and experts here would give you advice on workflows that more benefit someone creating critical prints rather than small web based photos. I may be totally offbase in my perspective and I was also hoping for a little feedback in that regard as well from the many knowledgable folks here. Do you not agree that a more simpler workflow is appropriate for photos that ultimately are for web viewing at say 1100 pixels long as opposed to a workflow for a magazine ad or 24x36 art print?
Best of luck in your journey!

[lbenac]

Hey I'm not trying to discourage you from your current workflow. I just notice that many of the people in this forum are experts and pros and they spend far much more time on a single photo than many photohobbyists would. They have developed workflows that get them the best results possible with large prints, critical prints for clients and their workflow MAY not be the best one for someone who, like me, is a hobbyist creating webbased photos 1100 pixels long. I don't have 2 hours or more each day to devote to post-processing. My girlfriend sees to that! I just wanted to let you know that I too looked for the best sharpening plug-ins and the best downsizing method as well as the best RAW converter as well as the best noise reduction strategy. It created a long complicated workflow that was too long for me. It may not be for you. I decided to do everything in calibrated ACR/CS3 with adjustment layers, recorded actions and the results may not be optimum, but more than satisfactory. I just wanted to give you my perspective which may end up being different than yours. My concern was that the many pros and experts here would give you advice on workflows that more benefit someone creating critical prints rather than small web based photos. I may be totally offbase in my perspective and I was also hoping for a little feedback in that regard as well from the many knowledgable folks here. Do you not agree that a more simpler workflow is appropriate for photos that ultimately are for web viewing at say 1100 pixels long as opposed to a workflow for a magazine ad or 24x36 art print?
Best of luck in your journey!
[a href=\"index.php?act=findpost&pid=177703\"][{POST_SNAPBACK}][/a]

Thank you for your advice, I hope I did not sound acrimonious in my previous post. You are right that I do not want to spend a lot  of time on certain images but yes I can spend a lot of time on a specific image I really like specially on a rainy week-end afternoon.
I usually print these few on 11*8 on top.
As far as viewing the image on the monitor I am of course limited to what a 21" LCD will give. One day it will be a 30" but not before several years.

All the best to you to.