LR RAW default sharpening

[spacecadet]

May I ask what you mean by 25-50-50? Sharpness-luminance-detail?

[MDM]

May I ask what you mean by 25-50-50? Sharpness-luminance-detail?

 

The 3 sliders for color noise in that case.

[spacecadet]

I only see 2, colour and detail. LR4.

[Armstrong]

Also where did you see/hear about the order of the sliders being important in LR/ACR? I've never heard this before. I'm genuinely interested.

George Jardine released a series of Podcasts in 2006 where he talked to the LR Engineering and UI team about the process that led to the creation of Lightroom. This was one of the aspects discussed. The idea behind the slider order was to help guide the user through the best sequence of steps for the RAW conversion.

 

Michael

[Armstrong]

Noise Reduction: Avoid unless essential. Never more than 10. If more than this is needed I consider the file to be unusable for Alamy.

 

I agree with most of this except for Noise Reduction and I'm not sure exactly what you mean by 10 there

 

That's just the 10 mark on the LR Slider. It seems to be about the setting that generally is the max I can go to with my camera before it softens the image too much. That's for Alamy. For other uses I might go higher. I'm interested in what you said about the ISO's where you start to need NR. I'm using a D7000 and on a well exposed image that hasn't had a lot of shadow recovery ISO 800 is where I start to see Noise that needs attention. Almost always luminance and hardly ever colour noise.

 

Michael

[spacecadet]

Helpful stuff, thanks all.

I seem to be able to apply quite a lot of luminance NR , 40 or 50, at high ISO, without too much loss of detail, or do I just need new glasses?

[Armstrong]

 

If previously you were shooting jpg and then importing and adjusting in LR then effectively you were 'baking the cake twice'. Effectively the JPG is an in camera RAW conversion.

If it helps

Michael

 

It does. Thankyou.

I have got used to a bit of saturation to make the thumbs 'pop' a bit but I will have a rethink as I haven't used vibrance before.

Where does a camera profile come from?

Have a look at this link:

https://fstoppers.com/originals/simple-lightroom-trick-restore-back-your-camera-look-38204

It's Adobe's way of getting LR to mimic the colour settings that your camera manufacturer uses to create the in camera JPG.

 

Michael

[Armstrong]

Helpful stuff, thanks all.

I seem to be able to apply quite a lot of luminance NR , 40 or 50, at high ISO, without too much loss of detail, or do I just need new glasses?

I would say that is high but we're probably on different camera kit. Perhaps post another 100% crop?

[MDM]

 

Also where did you see/hear about the order of the sliders being important in LR/ACR? I've never heard this before. I'm genuinely interested.

George Jardine released a series of Podcasts in 2006 where he talked to the LR Engineering and UI team about the process that led to the creation of Lightroom. This was one of the aspects discussed. The idea behind the slider order was to help guide the user through the best sequence of steps for the RAW conversion.

 

Michael

 

 

OK Michael. I couldn't envisage how the order could make any difference to the end result as the all the modifications are completely non-destructive and can be changed at any time.

 

I'm surprised you don't see color noise if you are not using any noise reduction. I've seen it in every digital camera I've owned, some a lot worse than others, so I always leave it at 25-50-50 as an import default. It doesn't have any softening effect on the lower ISO images but it's apparent in blue sky by ISO 400 on my D800 if I drop the color slider to 0. Any sharpening really shows it up.

[Armstrong]

 

 

Also where did you see/hear about the order of the sliders being important in LR/ACR? I've never heard this before. I'm genuinely interested.

 

George Jardine released a series of Podcasts in 2006 where he talked to the LR Engineering and UI team about the process that led to the creation of Lightroom. This was one of the aspects discussed. The idea behind the slider order was to help guide the user through the best sequence of steps for the RAW conversion.

Michael

OK Michael. I couldn't envisage how the order could make any difference to the end result as the all the modifications are completely non-destructive and can be changed at any time.

 

I'm surprised you don't see color noise if you are not using any noise reduction. I've seen it in every digital camera I've owned, some a lot worse than others, so I always leave it at 25-50-50 as an import default. It doesn't have any softening effect on the lower ISO images but it's apparent in blue sky by ISO 400 on my D800 if I drop the color slider to 0. Any sharpening really shows it up.

It is non destructive but the order does affect the result. For example the contrast slider is above highlight recovery. That makes sense as boosting the contrast after highlight recovery would increase the highlights again. The same with contrast and saturation. Increasing contrast can boost saturation so you would want to make sure that contrast is done first.

 

I've only used a full frame camera (dF) for a brief period but I'm surprised that the D800 is that noisy. I thought one of the plus points of full frame was cleaner high ISO performance. The only time I've seen colour ISO noise at lower ISO was photographing a military aircraft that was uniform mid grey. At ISO 400 I only see problematic noise if I've had to correct the exposure too much or have done large amounts of lifting the shadows.

 

Michael

[MDM]

It is non destructive but the order does affect the result. For example the contrast slider is above highlight recovery. That makes sense as boosting the contrast after highlight recovery would increase the highlights again. The same with contrast and saturation. Increasing contrast can boost saturation so you would want to make sure that contrast is done first.

 

I've only used a full frame camera (dF) for a brief period but I'm surprised that the D800 is that noisy. I thought one of the plus points of full frame was cleaner high ISO performance. The only time I've seen colour ISO noise at lower ISO was photographing a military aircraft that was uniform mid grey. At ISO 400 I only see problematic noise if I've had to correct the exposure too much or have done large amounts of lifting the shadows.

 

Michael

 

 

Not wishing to drag this out too much but I think that what you are talking about is perceptual and may effect the result in how one makes decisions about slider strength etc but I don't believe it makes any difference to the end result whether one say sets the Exposure to +0.5,  the Vibrance to +20 and then the Clarity to +10 as against the the Vibrance to +20, the Clarity to +10 and Exposure to +0.5 in that order. This is just an example, it could be much more complex than this but would make no difference to the end result. The fact is that every process in the Lightroom Develop Module is reversible and has no impact on the underlying image as there is no pixel editing involved. This reversiblity means that the order is irrelevant to my mind. I'm quite happy to eat my words if I'm wrong.

 

I wasn't saying that the D800 is noisy - it's actually incredibly good at noise control as is the D700 - just that noise starts to be perceptible moving up the ISOs. I don't think it would be a problem with Alamy QC until it gets well into the thousands as it can be easily controlled in the raw conversion. But some colour noise is perceptible (not problematic) at relatively low ISO if not controlled.

[Allan Bell]

Another point continuing from the comment on the masking slider under luminance noise.

 

I have flung that if the masking slider is taken too high it can introduce what look like sharpening artifacts in flesh tones.

 

Just my experience.

 

Allan

[spacecadet]

I have just had my first sub from RAWs passed and whilst it predates the advice here I had done some of those things anyway- so thanks again to all.

[Armstrong]

Not wishing to drag this out too much but I think that what you are talking about is perceptual and may effect the result in how one makes decisions about slider strength etc but I don't believe it makes any difference to the end result

 

I wasn't saying that the D800 is noisy - it's actually incredibly good at noise control as is the D700 - just that noise starts to be perceptible moving up the ISOs. I don't think it would be a problem with Alamy QC until it gets well into the thousands as it can be easily controlled in the raw conversion. But some colour noise is perceptible (not problematic) at relatively low ISO if not controlled.

First of all apologies for selectively quoting! Yes, I agree that a given set of values produces the same result. You are right in what I was describing is perceptual. In that the order you use the sliders influences your decisions on the next set of sliders ie you think you've pulled back highlights enough then a contrast boost increases them again.

 

Glad to hear you have no problems with the D800 noise. I think we maybe got our lines crossed between perceptible noise and problematic for QC noise.

 

Michael

[MDM]

 

Not wishing to drag this out too much but I think that what you are talking about is perceptual and may effect the result in how one makes decisions about slider strength etc but I don't believe it makes any difference to the end result

 

I wasn't saying that the D800 is noisy - it's actually incredibly good at noise control as is the D700 - just that noise starts to be perceptible moving up the ISOs. I don't think it would be a problem with Alamy QC until it gets well into the thousands as it can be easily controlled in the raw conversion. But some colour noise is perceptible (not problematic) at relatively low ISO if not controlled.

First of all apologies for selectively quoting! Yes, I agree that a given set of values produces the same result. You are right in what I was describing is perceptual. In that the order you use the sliders influences your decisions on the next set of sliders ie you think you've pulled back highlights enough then a contrast boost increases them again.

 

Glad to hear you have no problems with the D800 noise. I think we maybe got our lines crossed between perceptible noise and problematic for QC noise.

 

Michael

 

No need for aplogies Michael. There were some crossed wires there but I figured it was worth bashing it out. Nothing like a good pedantic discussion to clarify the mind.

 

It inspired me to have a detailed look at the effects of ISO and different types of noise reduction on some of my images. There is definitely noise coming in by ISO 400 but only noticeable at 100% view and easily dealt with. For my practical purposes, I find that noise only becomes problematic at 3200 or thereabouts on the D800 but I rarely go above 800. When I have a bit of time I would like to do some detailed testing on the effects of noise reduction on sharpness at different ISOs.

[Trevor Wintle]

I thought one of the criteria that has to be ticked for uploading images to Alamy was that "No sharpening has been applied to the image". Are you all saying that you ignore this and sharpen your images to Alamy? 

[NYCat]

I think the difference is whether you have applied OUTPUT sharpening. That is a no-no but people are using discretion about the default raw sharpening.

 

Paulette

[Ed Rooney]

For a long time Alamy said "no sharpening." In recent years they have said "no excessive sharpening." 

 

I live in a rent-stabilized apartment on Mulberry Street in NYC, and I have my stock at Alamy. There are three things I never say in print: what I pay for rent, the names of any of the Goodfellas on my block, and what I do about sharpening on the images I send to Alamy. 

[MDM]

Alamy have never formally declared (to my knowledge) that default LR input sharpening or non-excessive sharpening (however that would be defined) are acceptable. However, there was a thread on the old forum a few years back where somebody from Alamy did post that default LR input sharpening was acceptable, but that doesn’t appear anywhere in the Alamy guidance. I have never seen Alamy make any further statement on that since, despite several threads on the subject.

My guess is that it would be impossible to control if they came out and said that a little sharpening is ok as long as you use discretion because there is no way of measuring that - not everybody uses LR or ACR for one thing and image files are all different in terms of sharpness and what they can tolerate - so many factors here including camera, lens, technique, ISO, noise etc etc . It is simpler to keep the rule of no sharpening, which is what it still says in the guidance and in the tick boxes at upload. But there does seem to be an acceptance that many people are going to apply some input sharpening (or even in-camera sharpening of JPEGs) as nobody from Alamy has ever explicitly said in any thread that I have seen that it is not ok and they do read these threads as we know.

I liken this to speed limits. Many people break the speed limit - frequently it will be perfectly ok from a safety point of view to do so but other times it will be dangerous to drive at anything even approaching the speed limit. But if you get caught driving over the speed limit there are no excuses - “everybody does it” or “I didn’t see the signs” don’t cut it.

[Guest]

Alamy have never formally declared (to my knowledge) that default LR input sharpening or non-excessive sharpening (however that would be defined) are acceptable. However, there was a thread on the old forum a few years back where somebody from Alamy did post that default LR input sharpening was acceptable, but that doesn’t appear anywhere in the Alamy guidance. I have never seen Alamy make any further statement on that since, despite several threads on the subject.

 

 

 

Since QC policy is about excessive sharpening, the logic is that non-excessive sharpening in not a QC fail........

 

And perhaps a long time ago...... but James IIRC (maybe Alan) did answer a question about this at one of the London meetings and it was ok, but not too much ...a bit muddled like the current non-policy/policy. It was on video and on the site for many a moon..... perhaps can still be tracked down.

 

There is a way of measuring capture sharpening, at least GI were able to measure it.... they posted maximum guidelines back in the day for one method.

 

There are many creative techniques that use forms of sharpening, frequency separation for skin retouching is one well known beauty retouching one - it uses the high pass...would that be outlawed???

[John Mitchell]

My impression has always been that Alamy's QC inspectors are trained to "look through" unsharpened images to check for correct focus. No?

 

That said, all the RAW processors that I've tried appear to add a bit of capture sharpening. RAW files can look very soft without it.

[MDM]

 

Alamy have never formally declared (to my knowledge) that default LR input sharpening or non-excessive sharpening (however that would be defined) are acceptable. However, there was a thread on the old forum a few years back where somebody from Alamy did post that default LR input sharpening was acceptable, but that doesn’t appear anywhere in the Alamy guidance. I have never seen Alamy make any further statement on that since, despite several threads on the subject.

 

 

 

Since QC policy is about excessive sharpening, the logic is that non-excessive sharpening in not a QC fail........

 

And perhaps a long time ago...... but James IIRC (maybe Alan) did answer a question about this at one of the London meetings and it was ok, but not too much ...a bit muddled like the current non-policy/policy. It was on video and on the site for many a moon..... perhaps can still be tracked down.

 

There is a way of measuring capture sharpening, at least GI were able to measure it.... they posted maximum guidelines back in the day for one method.

 

There are many creative techniques that use forms of sharpening, frequency separation for skin retouching is one well known beauty retouching one - it uses the high pass...would that be outlawed???

 

Yes I've heard that before about the video now that you mention it but I've not seen it. The raw conversion parameters are held in the metadata from ACR into PS but not necessarily from LR depending on how the JPEG is created. Basically I think, as with many things in life, common sense works best and a small amount of capture sharpening is unlikely to be harmful for subsequent users of the image but, by the letter of the Alamy law as I read it, sharpening is still explicitly forbidden.

[MDM]

My impression has always been that Alamy's QC inspectors are trained to "look through" unsharpened images to check for correct focus. No?

 

That said, all the RAW processors that I've tried appear to add a bit of capture sharpening. RAW files can look very soft without it.

 

That is what I've always assumed. That certainly was the case in the days of having to upsize my 12MP D700 files to 48MB.  They never looked pinsharp but would jump into focus if sharpening was applied and the assumption was that QC would see through that as you say. Indeed I was looking back recently through some of my 12MP images taken in 2008-11 that I never uploaded because they didn't look sharp enough when upsized but they look fine at their native size. I'll start uploading some of these now in fact. For all the talk about QC being more difficult now, I think it's probably actually easier because of the fact that upsizing is no longer required but maybe they are less tolerant in terms of sharpness because of that.

[dustydingo]

I look at sharpening the same as I look at HDR . . . if you can tell it's been applied, bin it.

 

dd

[John Mitchell]

 

My impression has always been that Alamy's QC inspectors are trained to "look through" unsharpened images to check for correct focus. No?

 

That said, all the RAW processors that I've tried appear to add a bit of capture sharpening. RAW files can look very soft without it.

 

That is what I've always assumed. That certainly was the case in the days of having to upsize my 12MP D700 files to 48MB.  They never looked pinsharp but would jump into focus if sharpening was applied and the assumption was that QC would see through that as you say. Indeed I was looking back recently through some of my 12MP images taken in 2008-11 that I never uploaded because they didn't look sharp enough when upsized but they look fine at their native size. I'll start uploading some of these now in fact. For all the talk about QC being more difficult now, I think it's probably actually easier because of the fact that upsizing is no longer required but maybe they are less tolerant in terms of sharpness because of that.

 

 

Same with scans. They certainly don't look all that sharp, even when in perfect focus. If an image "jumps" into sharpness, as you say, when I experiment with adding a touch of sharpening, I too assume it's OK as is.