I was struck by how useful a tool AIG was and has been so far and bought it
Since then I have read of and seen some good examples of Sharpen AI, though have not trialled it for myself at this stage.
In my use of AIG I have done all my initial processing (exposure adjustments, capture sharpening, noise reduction etc) then opened the file in AIG to enlarge it >>>>> finally output sharpened as needed.
Just where in such a workflow should Sharpen AI come into play???
TIA for the insights.
NB at this stage I have mainly been using AIG to enlarge either the whole image or a modest crop (for compositional reasons) of up to 50%. In future I may crop much tighter and some images might exhibit a little motion blur or softness due to level of crop?
My personal thought is that a image, or portion thereof, that is to be enlarged should be in the cleanest possible condition before enlarging. This means correct tone, noise removed and sharpened … then after enlarging you may need to apply extra sharpening.
So, in my case, I would use Sharpen AI pre GigaPixel AI and then, if needs be, after enlarging.
I have always used Sharpening as a multi-tool. Both pre, post and pre publication…
You are echoing the general guidance about AIG i.e. if the stater file is already (noise) clean and ready for finalising then go Gigapixel…though and the reason I am maybe over-thinking it, is a note mentioned in the case of say a smartphone image the risk is PP artifacts creeping in if processed first, therefore enlarge ‘as is’ to avoid losing data.
So, Sharpen Ai can be used (as appropriate) in the second (creative) & the final (output) sharpening stages.
I will just have trial Sharpen Ai and see how I get on using it as needed in the PP workflow. (FWIW I use LR v6.12 for my DAM and initial prep…(and now) I use DxO PL2 for the noise control and other PP [used to be LR and then PS v6] and the final sharpening was PS Smart Sharpen but also for half tone or inkjet commercially printed I have Pixel Genius. This latter software is now old but was designed to sharpen as required for the specific printer technology.
Sure it can be used, but I have had better results with the AI Clear adjustment in Studio with smartphone images.
Again PS Smart Sharpen is OK but I tend to use Frequency Separation in Affinity Photo because it doesn’t affect solid color areas such as clouds and skin.
Ah! good insight as to the way you (others) do PP to get the same(?) end result.
In regard to your mention of AI Clear in Studio, I need to more fully learn DxO PL2 but it does great work in noise control and intermediate sharpening…so will stick with that.
As for your mention of Frequency Separation in AP, this I had not heard of. But as I got a free license for AP when I bought my X-Rite Passport I will need to try that…but have not even ‘got my toes wet’ let alone jumped in to AP.
The one thing I have learned over some years with PP and software choices is that there is no one tool that combines all the best of every sub process
Gavin Seim has a lengthy but good review of SAI and GAI on Youtube. He addresses using both and got much better results using SAI then GAI than in the reverse order.
I was looking through posts etc on Luminous Landscape (I am a subscriber there as well) and there (not surprisingly) have been discussions and users of Topaz posting.
Interestingly and maybe significantly…at least one user states that Sharpen AI is intended to be used only or at least primarily at the Capture Sharpening stage and to use it further into the PP workflow defeats its design purpose of usage.
He says something along the lines of ‘SAI needs to work on the purest of RAW data…once the image file has been processed in any way it is like asking SAI to unscramble the omelette…’
Has there been (I may have missed?) any stated guidance by the Topaz team about the “best practice workflow…” i.e. as per the LuLa user, where in the workflow did Topaz designers intend it to be used and as appropriate their designer insight as to its use in other points during the PP workflow ?
If not, as this might affect more than just a learner like me,what is the best way to draw this to the Topaz teams attention?
Remember that all of these comments are personal opinions on preferred workflow. What will work for you depends on the quality of the images, sensor and RAW converter you are using.
Personally I use sharpening where it is necessary, either as output sharpening or even as pre-sharpening where the images might need it after I convert to a TIFF in this case.
Fair point…but sometimes(?) when I read the history of specific contributors they appear to have a more authoritive opinion i.e. some greater insight???
I think everyone can claim their workflow is superlative and, for me anyway, the only type of sharpening that would be relevant as pre-sharpening would be to correct camera shake or focus before you go ahead with your processing.
In any case Sharpen AI has major drawbacks insofar as it doesn’t have selective sharpening options except when used as a plugin in a host such as PS where you can brush out the effect. Sharpening cannot be applied across 100% of an image as you need to take into account shadows, highlights, solid color areas, skin etc., etc.
Always remember that sharpening is an art, not a science and is also governed by your needs and perceptions.
Now I don’t have any deeper insight but I categorize my sharpening into 3 different parts of the image development process:
Pre-sharpening to correct camera shake, soft focus etc.
Image development with Sharpening in those areas that need it
Post-sharpening to sharpen for a specific display medium such as screen, matt paper, gloss paper etc.
It all comes down to what you want and how you apply it … art!
