IQ: straight from camera or after Post Processing?

Although M43 is ¼ the size of FF, the OP makes a valid point. Although FF starts off with less grain and can use the same apps for raw file optimization, the gains made with M43 are significant in reducing grain and adding sharpening to compete with FF. This is just one of the reasons M43 is the Sweet Spot.

When you throw in the fact that by far, most images are only ever viewed on-screen and M43 looks great printed up to 16x20”, the FF advantage only really exists for pixel peepers and rare times ISOs above 6400 are used or prints larger than 16x20”.
Or for when you want the best DR and/or like to blur backgrounds of course.....
Even for blur backgrounds the Adobe Lightroom Lens Blur feature does a great job and allows me to vary the depth and amount of blur in post processing. Helps negate the need for faster/larger lenses or larger sensors.
Yes, but that's not photography and will never look right.
Well, if it's not photography, then once you open any raw file in Photoshop and alter any settings....... neither is that.
I couldn't disagree more.

If we follow this argument , we may as well use auto-compose when it is available or use a phone.
OK, where do you draw your line where taking the image stops and making the image begins?
 
the web site is Magnifying the images at Greater than 100%, so if You choose to use the Maximum Magnification the images Look aweful.

Try looking at the images at Just 100% magnification.
I downloaded duck#2 and at 100% view it's a 2050x1600 jpeg so I'm either looking at a resized image or am I looking at a crop of the full sized image?

Whatever the case it does look like an aggressively sharpened image to me. DxO needs a firm hand to dampen its enthusiasm to sharpen.
PhotoLab offers you the option to start with modest settings as your default. I used to use the standard defaults. I found comparing neutral settings in C1 with standard settings in PhotoLab helpful. I got tired of fiddling with PhotoLab, so now I have modest settings as my default for a consistent approach between the two processors.

A
 
Although M43 is ¼ the size of FF, the OP makes a valid point. Although FF starts off with less grain and can use the same apps for raw file optimization, the gains made with M43 are significant in reducing grain and adding sharpening to compete with FF. This is just one of the reasons M43 is the Sweet Spot.

When you throw in the fact that by far, most images are only ever viewed on-screen and M43 looks great printed up to 16x20”, the FF advantage only really exists for pixel peepers and rare times ISOs above 6400 are used or prints larger than 16x20”.
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The process controls let you determine the amount of foreground and background that is in blur or not. I was responsible for the rodeo foreground being in focus. Of course, the tricky part is when we have something like a horse and rider that in not flat to the focus plane, but there are brush control to adjust as needed. Every iteration of this feature gets better and more controllable, but it still takes a fair bit of trial and error as part of the learning curve........ and I'm still on that curve.
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The process controls let you determine the amount of foreground and background that is in blur or not. I was responsible for the rodeo foreground being in focus. Of course, the tricky part is when we have something like a horse and rider that in not flat to the focus plane, but there are brush control to adjust as needed. Every iteration of this feature gets better and more controllable, but it still takes a fair bit of trial and error as part of the learning curve........ and I'm still on that curve.
Managing the point of critical focus and focus fall-off is essential in large high-resolution landscapes. I'm still on the learning curve too - boxed in between visual impact, diffraction and depth of field. That's not to mention managing lens aberrations... sorry "character".

View attachment 2a70c24c673d4b68a01b087285b5f48e.jpg
One for the benches thread

Where was the focus - guess the trick
Where was the focus - guess the trick

View attachment 864ab0a18d594d118a5961eca5c6b634.jpg
Know your subject

View attachment 4467c37ef0fd46e4b314fdcf4e4f6089.jpg
Processed or what?

:)

A

PS You need to look at these at least as large as a laptop 4k screen

--
Infinite are the arguments of mages. Truth is a jewel with many facets. Ursula K LeGuin
Please feel free to edit any images that I post
 
Last edited:
Even for blur backgrounds the Adobe Lightroom Lens Blur feature does a great job and allows me to vary the depth and amount of blur in post processing. Helps negate the need for faster/larger lenses or larger sensors.
Getting back to the original topic...

I do the same now Gary, the Lens Blur in LR is really a nice tool.

I just can't shoot jpegs anymore because with all the software now I might as well shot RAW and get all the benefits

My workflow typically is

RAW
Import to Light Room, do any perspective corrections, lens corrections, basic adjustments, use the Lens Blur feature if applicable etc..

Import to Topaz Photo Ai for sharpening and noise reduction, upscale if going to be printed.

These tools really help with M43's short-comings.
 
Although M43 is ¼ the size of FF, the OP makes a valid point. Although FF starts off with less grain and can use the same apps for raw file optimization, the gains made with M43 are significant in reducing grain and adding sharpening to compete with FF. This is just one of the reasons M43 is the Sweet Spot.

When you throw in the fact that by far, most images are only ever viewed on-screen and M43 looks great printed up to 16x20”, the FF advantage only really exists for pixel peepers and rare times ISOs above 6400 are used or prints larger than 16x20”.
Or for when you want the best DR and/or like to blur backgrounds of course.....
Even for blur backgrounds the Adobe Lightroom Lens Blur feature does a great job and allows me to vary the depth and amount of blur in post processing. Helps negate the need for faster/larger lenses or larger sensors.
Yes, but that's not photography and will never look right.
Well, if it's not photography, then once you open any raw file in Photoshop and alter any settings....... neither is that.
I couldn't disagree more.

If we follow this argument , we may as well use auto-compose when it is available or use a phone.
OK, where do you draw your line where taking the image stops and making the image begins?
Personally, I only adjust the following if required-

Exposure

Highlight clipping

Colour balance for interiors

Raise shadows if required

Denoise if required

This is just my personal preference and I would never consider using blurring backgrounds or focus-stacking etc.

It's just my opinion and others will disagree I am sure.
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
 
Even for blur backgrounds the Adobe Lightroom Lens Blur feature does a great job and allows me to vary the depth and amount of blur in post processing. Helps negate the need for faster/larger lenses or larger sensors.
Getting back to the original topic...

I do the same now Gary, the Lens Blur in LR is really a nice tool.

I just can't shoot jpegs anymore because with all the software now I might as well shot RAW and get all the benefits

My workflow typically is

RAW
Import to Light Room, do any perspective corrections, lens corrections, basic adjustments, use the Lens Blur feature if applicable etc..

Import to Topaz Photo Ai for sharpening and noise reduction, upscale if going to be printed.

These tools really help with M43's short-comings.
Pretty similar here except I run the raw through Workspace first for the raw conversion before I import the tif into LR.
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
I find the need to make videos dramatic and visually exciting gets in the way of conveying information. It even happens with videos on how to replace parts but most of those are by people who want to share information and know how to do it.

Watching through five minutes of the life story and dining preferences of the presenter is the worst, but I've learned they tend to have limited useful information anyway.

That video looked like a how to video, which was promising, but the rapid cutting between images to show how clever the presenter was turned me off almost immediately. I can wade through pages of diagrams and detail because you can skim to find the useful bits.

That's just me.

A
 
But, man, those birds are so overprocessed they look posterized.

Back the NR down a bit and let the file info do its thing, and they will look a lot better.

-J
That is the Standard NR, Not dialed up or down. Shots were taken on a dark overcast day with light showers.

You probably only go out on sunny days LOL
The standard (default) NR in DXO is very excessive with sharpening. I never use it without dialing it down. However, it's all a matter of personal taste. Some like it, others do not. All that matters is that we're happy with our own photos.
AFAIK DxO NR on itsself doesn't apply (much) sharpening, but the standard settings contain a lot of processing among which sharpening. Many of those settings are way too strong for my taste.

You can easily turn those off (or use more subtle settings) and save it as your personal default.
I only use Prime, my PC isn't up to the more advanced NR modes in Photolab - but after a bit of experimentation, I realised that PL defaults to '40' on the NR slider and this is quite aggressive. My personal starting default in all my own presets is now 0. This isn't actually nothing either - if you set it to 0, then turn NR off, a lot more noise appears in the preview window. I only actually increase it where an individual image might need it - but as I usually downsize images for online sharing, that mitigates a lot of visual noise too, so I rarely need much more.
Yes, most DxO defaults are too agressive for my taste. But my point was that NR doesn't apply the heavy sharpening unlike a previous poster commented. It's a different setting. You can apply heavy NR without any sharpening.
Indeed - NR and sharpening are separate processes. I just happened to reply to you as the last post in this side conversation, but in respect of what had been said in several posts - and reiterating what you said about DxO defaults being too robust for most purposes.
As for sharpening, I use a conservative lens softness correction at raw development stage and actually apply output sharpening at the final size. In reality what I mostly do is create a high pass sharpening layer at full size, downsize the image with the layer in place and make adjustments at that size so that it looks right at the actual viewing size - you can get different effects by changing layer blend modes for example. If I were going to print, I'd make a different version with appropriate sharpening for that process etc.
Seems a good way to keep controll, but also quite cumbersome. I rarely apply much sharpening myself, and for what I do the details slider in DxO is enough for a little enhancement without introducing artifacts. I find their lens corrections alone already improve sharpness of mediocre lenses. I get it some people want it even sharper than perfectly focussed with a sharp lens, but maybe they just need more MP.
It doesn't feel cumbersome - I suppose it might be if handling a lot of images for numerous different purposes. But for my personal workflow - which is what it is - I find it works well. I'm currently using Affinity and rarely 'publish' any photo without cloning out a bit of fluff, stray hair or tweaking something (I find it better than trying to do small repairs in DxO) or adding a border, so clicking my pre-set script for adding a layer and sharpening to my preferences takes a mere second.

For many purposes, it doesn't require much more than a quick tweak of the opacity slider to finish to taste, when viewing it at the size it will be shown. I prefer to finish my sharpening at the output size - I don't find a 'one master file fits all purposes' works for me.
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
I find the need to make videos dramatic and visually exciting gets in the way of conveying information. It even happens with videos on how to replace parts but most of those are by people who want to share information and know how to do it.

Watching through five minutes of the life story and dining preferences of the presenter is the worst, but I've learned they tend to have limited useful information anyway.

That video looked like a how to video, which was promising, but the rapid cutting between images to show how clever the presenter was turned me off almost immediately. I can wade through pages of diagrams and detail because you can skim to find the useful bits.

That's just me.

A
I’m more of a bounce-off-the-trees kind of guy when learning a new software. This blur thing is no different, but it gets easier.
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
I find the need to make videos dramatic and visually exciting gets in the way of conveying information. It even happens with videos on how to replace parts but most of those are by people who want to share information and know how to do it.

Watching through five minutes of the life story and dining preferences of the presenter is the worst, but I've learned they tend to have limited useful information anyway.

That video looked like a how to video, which was promising, but the rapid cutting between images to show how clever the presenter was turned me off almost immediately. I can wade through pages of diagrams and detail because you can skim to find the useful bits.

That's just me.

A
I’m more of a bounce-off-the-trees kind of guy when learning a new software. This blur thing is no different, but it gets easier.
The full C1 manual is over 600pp. That's a lot of trees!

:)
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
I find the need to make videos dramatic and visually exciting gets in the way of conveying information. It even happens with videos on how to replace parts but most of those are by people who want to share information and know how to do it.

Watching through five minutes of the life story and dining preferences of the presenter is the worst, but I've learned they tend to have limited useful information anyway.

That video looked like a how to video, which was promising, but the rapid cutting between images to show how clever the presenter was turned me off almost immediately. I can wade through pages of diagrams and detail because you can skim to find the useful bits.

That's just me.

A
There is always a balance videos between too much detail and keeping the viewer watching. I have often thought Julieanne does a great job doing that to INTRODUCE the viewer into a new subject and get them to try it in real life, which is where you really learn it…which, unfortunately, one cannot do without having Lightroom.

I understand you statement on Wading through a very detailed write up….which I often do with user manuals. Unfortunately,often with processing software, we get an introduction to features and then need to try and find what works for our tastes and images.

here is a short write up which you may find interesting.
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
I find the need to make videos dramatic and visually exciting gets in the way of conveying information. It even happens with videos on how to replace parts but most of those are by people who want to share information and know how to do it.

Watching through five minutes of the life story and dining preferences of the presenter is the worst, but I've learned they tend to have limited useful information anyway.

That video looked like a how to video, which was promising, but the rapid cutting between images to show how clever the presenter was turned me off almost immediately. I can wade through pages of diagrams and detail because you can skim to find the useful bits.

That's just me.

A
I’m more of a bounce-off-the-trees kind of guy when learning a new software. This blur thing is no different, but it gets easier.
The full C1 manual is over 600pp. That's a lot of trees!

:)
Lightroom Queen’s LR Classic book is almost 800…..but PDF’s save trees 😀
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
I find the need to make videos dramatic and visually exciting gets in the way of conveying information. It even happens with videos on how to replace parts but most of those are by people who want to share information and know how to do it.

Watching through five minutes of the life story and dining preferences of the presenter is the worst, but I've learned they tend to have limited useful information anyway.

That video looked like a how to video, which was promising, but the rapid cutting between images to show how clever the presenter was turned me off almost immediately. I can wade through pages of diagrams and detail because you can skim to find the useful bits.

That's just me.

A
There is always a balance videos between too much detail and keeping the viewer watching. I have often thought Julieanne does a great job doing that to INTRODUCE the viewer into a new subject and get them to try it in real life, which is where you really learn it…which, unfortunately, one cannot do without having Lightroom.

I understand you statement on Wading through a very detailed write up….which I often do with user manuals. Unfortunately,often with processing software, we get an introduction to features and then need to try and find what works for our tastes and images.

here is a short write up which you may find interesting.
https://blog.adobe.com/en/publish/2024/06/26/inside-lens-blur
I appreciate your thoughtfulness.

A
 
Thank you Gary, that's really helpful. Like the shots a lot.

Those look odd to me because the background is blurred in the rodeo one but the foreground DoF is very large, so it looks like two images Photoshopped together. The beautiful robe also looks odd because the transition is so abrupt and the robe has such a hard edge.

However, these are so much better than anything I've seen previously that I can imagine a day when the technique works for me too.

I've seen images taken with inappropriate use of shallow DoF that also look very odd, so I guess it depends on expectations.

Andrew
The important thing to remember is that you have lots of control, via the depth map and brushes, to control what is blurred and what isn’t….even the transition between them….plus have foreground and background blur. Also, the type of blur can be modified. It is 5he reason I linked the YT for your reference.
I find the need to make videos dramatic and visually exciting gets in the way of conveying information. It even happens with videos on how to replace parts but most of those are by people who want to share information and know how to do it.

Watching through five minutes of the life story and dining preferences of the presenter is the worst, but I've learned they tend to have limited useful information anyway.

That video looked like a how to video, which was promising, but the rapid cutting between images to show how clever the presenter was turned me off almost immediately. I can wade through pages of diagrams and detail because you can skim to find the useful bits.

That's just me.

A
There is always a balance videos between too much detail and keeping the viewer watching. I have often thought Julieanne does a great job doing that to INTRODUCE the viewer into a new subject and get them to try it in real life, which is where you really learn it…which, unfortunately, one cannot do without having Lightroom.

I understand you statement on Wading through a very detailed write up….which I often do with user manuals. Unfortunately,often with processing software, we get an introduction to features and then need to try and find what works for our tastes and images.

here is a short write up which you may find interesting.
https://blog.adobe.com/en/publish/2024/06/26/inside-lens-blur
Thanks for the link to the write up. It cleared up some things for me.
 
the web site is Magnifying the images at Greater than 100%, so if You choose to use the Maximum Magnification the images Look aweful.

Try looking at the images at Just 100% magnification.
I downloaded duck#2 and at 100% view it's a 2050x1600 jpeg so I'm either looking at a resized image or am I looking at a crop of the full sized image?

Whatever the case it does look like an aggressively sharpened image to me. DxO needs a firm hand to dampen its enthusiasm to sharpen.
PhotoLab offers you the option to start with modest settings as your default. I used to use the standard defaults. I found comparing neutral settings in C1 with standard settings in PhotoLab helpful. I got tired of fiddling with PhotoLab, so now I have modest settings as my default for a consistent approach between the two processors.

A
With Photolab I seem to use the canned HDR-Realistic often and then adjust from there. But I do "collect" presets I have made for various of my older (they are all old!) cameras and apply them as needed, usually again with some quick fine tune adjusts for each different image. For example I have made a few Panasonic LX3 presets that really do make that 2008 vintage camera really jump up in image quality, makes it into a useful new camera.

When using DeepPRIME XD (in my current PL7) then playing with Force Details (-80 or so) often takes the sometimes edgy sharpness away.

Green seems to be a problem with Photolab so some careful detuning of the green enthusiasm is needed to get reality back and not to look like some ugly travel brochure shot.

But overall, even without much care then Photolab+Raw files easily does better than out of camera jpegs so it's my normal conversion tool. Sometimes I take a side trip to Silkypix to try their way of doing things. OM Workspace doesn't get used any more.
 
Duck #1 original image is 4268x 3201 = 13.66 mp

Duck #2 original image is 2050x1900 = 3.9 mp

Optical Corrections was selected

Deep Prime XD/XD noise reduction was applied.

DXO Photo Lab 8.9 build 39 was used.

Digital photo software appears to be double enlarging Duck #1 which distorts the image.

Both Duck images were downloaded from my FlickR page at Original resolution, which looks good on FlickR. No additional modifications were made to the FlickR downloads.
 
Last edited:
Although M43 is ¼ the size of FF, the OP makes a valid point. Although FF starts off with less grain and can use the same apps for raw file optimization, the gains made with M43 are significant in reducing grain and adding sharpening to compete with FF. This is just one of the reasons M43 is the Sweet Spot.

When you throw in the fact that by far, most images are only ever viewed on-screen and M43 looks great printed up to 16x20”, the FF advantage only really exists for pixel peepers and rare times ISOs above 6400 are used or prints larger than 16x20”.
Or for when you want the best DR and/or like to blur backgrounds of course.....
Even for blur backgrounds the Adobe Lightroom Lens Blur feature does a great job and allows me to vary the depth and amount of blur in post processing. Helps negate the need for faster/larger lenses or larger sensors.
I've never seen a good example. Could you share one? I don't have LR, so I can't try for myself.

thanks

Andrew
Thank you but you can't assess the impact from an image that's on the screen less than 0.5s. Is there an example which is not a video?

thanks

Andrew
I think like very other filter there is a tendency to over egg the pudding :-) Messing about with it I find it can be effective for tidying up a messy background . it has a lot of tweaks to work with . With a subtle touch it does a decent job , certainly will not turn a kit lens shot taken at F/8 into an F/1.2

https://www.dpreview.com/sample-gal...system-om-1-mark-ii-sample-gallery/1555503864



7881ea8fead24159b59afd833bf6c28e.jpg





--
Jim Stirling:
"Cogito, ergo sum" Descartes
Feel free to tinker with any photos I post
 

Attachments

  • 453326d8c0de46d69d8505bedf0eabc4.jpg
    453326d8c0de46d69d8505bedf0eabc4.jpg
    3.4 MB · Views: 0
Last edited:

Keyboard shortcuts

Back
Top