Is this the hill I will die on? - How to get better image quality from your Olympus...

[Brownie]

I don't understand why people continue to feed into his ego.

 

[994]

Moved to the backroom

 

[oldracer]

... in the practical sense digital gain will render more noise on the image. ...

Well, I'm not even sure what you mean by "digital gain" but any digital processing of the pixel values by definition changes them. Normalization, gamma adjustments, whatever. But no noise (in the engineering sense) is added. If you say "more noise on the image" that is a value judgment, not an engineering question. The values displayed, whether subjectively "noise" or your little sister's nose, are the ones that came from the ADC, adjusted by whatever calculations the camera software may have made.

Your original statement that noise is "generated" in the digital domain remains incorrect. Good try to dodge, though

(Really, your whole post #21 was largely an attack rather than an attempt to understand the OP. I could have commented on other statements in the post, but this one about digital noise just struck me as particularly prominent nonsense.)

 

[Robstar1963]

(Really, your whole post #21 was largely an attack rather than an attempt to understand the OP. I could have commented on other statements in the post, but this one about digital noise just struck me as particularly prominent nonsense.)

Off subject but I didn’t realise that there was a numerical reference to each post - I’ve learned something useful here today after all

 

[pdk42]

Well, I'm not even sure what you mean by "digital gain" but any digital processing of the pixel values by definition changes them. Normalization, gamma adjustments, whatever. But no noise (in the engineering sense) is added. If you say "more noise on the image" that is a value judgment, not an engineering question. The values displayed, whether subjectively "noise" or your little sister's nose, are the ones that came from the ADC, adjusted by whatever calculations the camera software may have made.

Your original statement that noise is "generated" in the digital domain remains incorrect. Good try to dodge, though

(Really, your whole post #21 was largely an attack rather than an attempt to understand the OP. I could have commented on other statements in the post, but this one about digital noise just struck me as particularly prominent nonsense.)

Well, I'm happy to accept your point about digital gain not introducing noise in the engineering sense. And by "digital gain" I mean brightening the image by increasing the pixel values numerically - is that not a reasonable phrase to use? But, my contention is that brightening an image post the ADC stage, esp in the darker regions, will clearly deteriorate image quality in the sense that the noise will be more apparent.

All that matters in the final analysis is what the resulting image looks like. If we consider two images side by side, one at an ISO before digital gain is added (variously quoted at around ISO 800 or 1600 but let's not argue about that since I'm not sure anyone really knows), and another at a higher ISO where digital gain is used (so no additional analogue amplification before the ADC), the one at the higher ISO will appear more "noisy". The same argument applies to PP treatments which raise brightness (all the things you mentioned and more). Clearly, even a perfect amplifier will amplify noise as well as signal and if the signal is weak (underexposed sensor) then the SNR will be worse.

Finally, was my post an attack? I've tried to understand Siegfried's posts many times before but I've failed. He constantly posts outrageous claims about FF conspiracies and then tries to support his position with muddled and complicated arguments about camera image processing. Maybe you got me on the point about digital gain, but are my other comments off the mark? I'm very happy to have a discussion about each and every one of them and I'll gladly accept where I'm wrong.

 

[pdk42]

?


Sorry Paul but I just could not ignore this one...

View attachment 848663

Sounded like the Gossip asked the OM1.... how many wild statements did this Paul guy make...? But I could have been wrong...

Does the light meter see the effect of the lens on the camera?, or its interaction with the sensor's micro-lenses?, does the light meter have the same calibration as the sensor?, does it have the same linear/non-linear response to changing light levels?; is the light meter capable of seeing very bright areas that the photographer would like to preserve? The answer to all these is "no".

By contrast, on-sensor exposure metering and then rendering the results in the live view (over/under indicators, histogram) is as close to the real exposure as you can get (although I accept it's still not perfect). It just has to be more accurate and informative than pointing a Sekonic at the scene!

 

[John King]

@pdk42 Paul, I keep on saying that (every) digital camera is the world's best light meter. It even gives you a photo of what it's 'thinking' if you press a button ... .

 

[pdk42]

@pdk42 Paul, I keep on saying that (every) digital camera is the world's best light meter. It even gives you a photo of what it's 'thinking' if you press a button ... .

Yep - totally true. The only compelling reason for a separate meter I can think of is for studio flash. Of course, you can do trial and error with that, but a flash meter makes things easier.

 

[John King]

Yep - totally true. The only compelling reason for a separate meter I can think of is for studio flash. Of course, you can do trial and error with that, but a flash meter makes things easier.

Yeah, I've got one of each type.

I really cannot stress how important it is for people to learn how to use the different metering modes in their camera/s.

 

[BushmanOrig]

Well, I'm happy to accept your point about digital gain not introducing noise in the engineering sense. And by "digital gain" I mean brightening the image by increasing the pixel values numerically - is that not a reasonable phrase to use? But, my contention is that brightening an image post the ADC stage, esp in the darker regions, will clearly deteriorate image quality in the sense that the noise will be more apparent.

All that matters in the final analysis is what the resulting image looks like. If we consider two images side by side, one at an ISO before digital gain is added (variously quoted at around ISO 800 or 1600 but let's not argue about that since I'm not sure anyone really knows), and another at a higher ISO where digital gain is used (so no additional analogue amplification before the ADC), the one at the higher ISO will appear more "noisy". The same argument applies to PP treatments which raise brightness (all the things you mentioned and more). Clearly, even a perfect amplifier will amplify noise as well as signal and if the signal is weak (underexposed sensor) then the SNR will be worse.

Finally, was my post an attack? I've tried to understand Siegfried's posts many times before but I've failed. He constantly posts outrageous claims about FF conspiracies and then tries to support his position with muddled and complicated arguments about camera image processing. Maybe you got me on the point about digital gain, but are my other comments off the mark? I'm very happy to have a discussion about each and every one of them and I'll gladly accept where I'm wrong.

Thank you Oldtimer for your kind input. Must say, I haven't experienced this level of open attack in many years... Here a few thoughts:-

- It's maybe a good idea to think before you attack again Paul, try a simple Google search
- You made so many wild statements - I saw no benefit answering PLUS way off-topic!!!
- I was amazed when you made statements about gain and then asked "like what"... weird
- Then the flip-flopping trying to weasel you out from more wild statements - amusing...
- Most amusing was when you thought you accused of trolling - losing all confidence??
- All your back and forth on light meters - go search youtube videos on "Sekonic"...
- As before you always seem to attract an interesting mob-like reaction - really sad
- Then all the personal attacks on me - why that or is that the way you talk to people?
- Worst of all - so many "young" M43 users would have benefitted from a tame intellectual discussion and not the wild show you put up - this is really disappointing and so sad...

All the best...

 

[pdk42]

Thank you Oldtimer for your kind input. Must say, I haven't experienced this level of open attack in many years... Here a few thoughts:-

- It's maybe a good idea to think before you attack again Paul, try a simple Google search
- You made so many wild statements - I saw no benefit answering PLUS way off-topic!!!
- I was amazed when you made statements about gain and then asked "like what"... weird
- Then the flip-flopping trying to weasel you out from more wild statements - amusing...
- Most amusing was when you thought you accused of trolling - losing all confidence??
- All your back and forth on light meters - go search youtube videos on "Sekonic"...
- As before you always seem to attract an interesting mob-like reaction - really sad
- Then all the personal attacks on me - why that or is that the way you talk to people?
- Worst of all - so many "young" M43 users would have benefitted from a tame intellectual discussion and not the wild show you put up - this is really disappointing and so sad...

All the best...

Whatever Siegfried - I'm tired of this. Think whatever you want to think.

 

[BushmanOrig]

Whatever Siegfried - I'm tired of this. Think whatever you want to think.

My guess Paul... this is what canceled any reasoning and made you see RED in my article...

"The ongoing sensor size debate turned subjects like image noise and image quality into unnecessary confusing subjects. The reason is these debates never propose real solutions or techniques that will help you master your "current" camera. The standard answer is, full-frame cameras are better while avoiding any behavioral problems you might transfer to your next camera. It's a shame in this day and age that one needs to spend this much effort to present real information to photographers."

You can see my thoughts on what we saw happening in this thread here...

 

[pdk42]

My guess Paul... this is what canceled any reasoning and made you see RED in my article...

"The ongoing sensor size debate turned subjects like image noise and image quality into unnecessary confusing subjects. The reason is these debates never propose real solutions or techniques that will help you master your "current" camera. The standard answer is, full-frame cameras are better while avoiding any behavioral problems you might transfer to your next camera. It's a shame in this day and age that one needs to spend this much effort to present real information to photographers."

You can see my thoughts on what we saw happening in this thread here...

Siegfried - for the record I’m not an advocate of FF in the sense that I see it as superior to m43. For sure there are some things it does better, but there are many areas where m43 has it beat. It’s horses for courses. And whatever camera you have, it’s no substitute for the eye of the photographer. Technically brilliant snapshots are still snapshots, not images with any inherent value beyond perhaps a personal one to the photographer or his/her friends and family.

 

[BushmanOrig]

Siegfried - for the record I’m not an advocate of FF in the sense that I see it as superior to m43. For sure there are some things it does better, but there are many areas where m43 has it beat. It’s horses for courses. And whatever camera you have, it’s no substitute for the eye of the photographer. Technically brilliant snapshots are still snapshots, not images with any inherent value beyond perhaps a personal one to the photographer or his/her friends and family.

I like to believe you Paul but you seem to always be in arguments with anybody who does not fall in line with your views, especially those linked to full-frame cameras... It's an interesting read to go back in history and read your replies...

 

[Robstar1963]

I like to believe you Paul but you seem to always be in arguments with anybody who does not fall in line with your views, especially those linked to full-frame cameras... It's an interesting read to go back in history and read your replies...

I thought that the post above from @pdk42 would have been a good point to let this one rest

 

[pdk42]

I like to believe you Paul but you seem to always be in arguments with anybody who does not fall in line with your views, especially those linked to full-frame cameras... It's an interesting read to go back in history and read your replies...

Siegfried - the only person l disagree with on this forum is you

 

[Phocal]

Subscribe to see EXIF info for this image (if available)

Do Not Feed the Trolls by Phocal Art, on Flickr