Olympus-OM
[Top] [All Lists]

Re: [OM] topaz sharpen AI

Subject: Re: [OM] topaz sharpen AI
From: Moose <olymoose@xxxxxxxxx>
Date: Mon, 18 Mar 2019 21:44:43 -0700
On 3/18/2019 4:01 PM, Mike Gordon via olympus wrote:
The Large even-toed ungulate, sometimes referred to as Eurasian Elk  in some 
locales, states:

<<<I mention it because, in my apparently unending test/experiments with Raw 
converters, NR and sharpening, it's running neck and neck with a combination of 
NeatImage and AI Sharpen for first place.

Will stayed tune to this channel for more updates on this race.  One would 
strongly predict that deconvolution would perform better on an unprocessed raw 
when linked to a full wave solution  for the lens including diffraction as used 
in DLO for Canyon.

Should do.

It is extremely likely DXO uses blind deconvolution  (R-L like FM) though 
applied in a weighted fashion where the lens is softer.

You don't think they create a full wave solution when they test each lens? I don't really know what's involved in that. And I know it wouldn't be the same as one created from the lens design, rather than one specimen.

Still, you've got the lens on the test bench . . .

Totally unclear if previous manipulation would deleteriously affect AI sharpen 
though it
appears to be trained on less manipulated images.

The only thing I've tried like that is running it on top of images already treated with NR, DxO Prime and ACR+NeatImage. Works fine, compared to running it on conversion w/o NR.

My Hooded Merganser test subject has all sorts of complex fine detail, and seeing the differences in different parts is time consuming. Samples may see the light of a screen near you - if I stop finding other options to try out. :-)

Some of the pixels are created out of whole cloth.

Yes, it does an amazing job of that, in some areas, while overdoing it a bit in 
others.

Current front runner, with this image, overall, is still DxO Prime+DxO Lens Sharpen. The way DxO varies application of both processes by area based on noise, detail, sharpness is quite impressive

Will have to await an opinion using empiric determination of optimal work flows 
from our highly esteemed ungulate.

Moose in Overalls

--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz