And yet, you're the one that brings my name up out of nowhere after I haven't posted in months. You're a walking talking contradiction...but you'll blame everyone else for your bs, just like your fearless leader did.
I can see what both 'John Carson' & 'Steen Bay', and others, are trying to say - but all the talk of using different ISO's are completely wrong.
Yes, Fuji's 'trick' is simply to combine two images shot simultaneously with different shutter speed. I agree that's a good (and not too confusing) way to decribe what's going on. ;-)
Yep - I think you may well be right John (theory/explanation below).
Hi John - I think you may well be on to something there, although I don't think it would follow the ISO > DR rule exactly , but would appear to, more or less.
Indeed - I would support your 'theory'.
If you read my original post
and later
This last paragraph rather heavily qualifies the earlier one and confirms that the use of a faster shutter speed in EXR camera is a lot like higher ISO, though not exactly the same.
Yes - except on the one crucial definition (at end of this message)...
That is correct as regards 'photo-sensitivity' i.e. Signal/Area straight off the pixels before any amplification.
That is also correct, in 'digital' photography - the effective ISO is the product of the camera/processing system as a whole.
I'm going to disagree slightly here.
It's only the 'exposure' setting of the extra 'short shutter' exposure that is 'like' higher ISO - but it still is not higher ISO, it is only 'lower exposure' at same ISO.
You are overlooking the one crucial difference...
Please note:- I am aware that a large part of a 'software-only DR' image is 'boosted' by the application of a mid-tone curve lift - but that isn't quite the same thing as ISO which boosts the entire image capture uniformly.
No, it isn't quite the same thing, but it is very similar.
Re 'single exposure - software HDR' - I will concede it is 'similar' (to increased ISO).
No - that's just 'all one and the same thing' - the 'original noise' is not increasing, it is the 'boosting' of the low signal, that also 'boosts' the noise at the same time.
No, not at all - faster shutter speed alone can be** just -EV exposure at same ISO, as is the case with 'multi-exposure HDR'.
I don't clearly see the point behind your last statements - but no matter..
Yes, I know. Both the signal and the noise get boosted. The point is that the noise is perceived to be a problem because it is high relative to the signal and that high ratio arises from the short exposure time.
So we agree then
Again, you look at boost as the critical part of higher ISO. I don't. Further, there is a boost even in Fuji multiple exposures.
At the end of the day, there are three fundamentals: the light from the image subject, the lens aperture, and the exposure time. All else is smoke and mirrors — and processing.
No - noise itself is not increased by 'shorter exposure time' alone (in-fact, if anything, a shorter exposure time [alone] results in less noise).
No - the discussion/the point of debate, is not about 'normal single exposures'So we agree then
No I don't at all.
No, there is not any ISO boost. There is only tonal compression of highlights, but this is not ISO boost/shift, e.g. the highest highlight of the short exposure does not move.
Haven't you forgotten 'ISO' that you keep wanting to include?
What?
No - that is so completely wrong...
I'm beginning to wonder.
I don't agree.
Yes and no.. the shot/photon noise = squareroot of the signal, so if the signal is 100 photons the shot noise is 10 photons (SNR 10:1), and with a 4x faster shutterspeed the signal will be 25 photons, and the noise 5 photons (SNR 5:1). So, measured in photons the noise decreases from 10 to 5 photons with a 4x shorter exposure time, but the signal has decreased more than the noise, meaning that SNR has decreased, and the noise has increased from 10% to 20% of the signal.
'Steen Bay' I happily acknowledge/concede your correction of me on that technical point - although in the context of our 'HDR' discussion, we aren't actually dealing with low signal, but in-fact just the opposite.
Read what I wrote. I didn't say that noise itself is increased. I said that the ratio of noise to the signal is increased.
It is a completely valid association. Below is an experiment. I took two photos at ISO 800, but used EV of -2 on one of them. The shutter speed was 1/9 sec in the straight ISO 800 shot and 1/38 sec in the EV -2 shot. Aperture and focal length were constant.
Well, it is about ISO in the first instance.No - the discussion/the point of debate, is not about 'normal single exposures'So we agree then
There is still a boost, even though it isn't a uniform one.
No, I haven't forgotten ISO. ISO is not fundamental. It is "smoke and mirrors — and processing". I want to include it because people talk about it a lot and so it is necessary to understand it.
Given light source and aperture, the signal to noise ratio and hence perceived noise is fundamentally the product of exposure length. See above.
This second shot was uploaded by mistake. It is an ISO400 image, as the file name states. The correct image is below. This doesn't affect the crops, which were taken from the correct images.The full shots are below, first the normal exposure shot with shutter speed of 1/9 sec:
Next the short exposure shot with 1/38 sec shutter speed:
--Now some crops. In the first crop, it is clear that the noise is much more visible in the short exposure case on the right due to the worse signal to noise ratio.
In the next crop, the short exposure crop is so dark that it is difficult to see much.
Below, I brighten the short exposure crop (thus increasing both signal and noise). I couldn't match the colors without processing the image so heavily that it would be unclear what was really responsible for the outcome. Accordingly, I have just adjusted brightness and contrast to get a ballpark similarity in tone:
It is pretty clear that the noise is worse in the short exposure crop.
I accept that I overlooked that you stated 'ratio'.
It is not valid in the context of 'multi-exposure HDR', which is what this thread is about.
First off - double check - you appear to have either quoted the wrong numbers (ISO), and shutter speed 'typo', or you have posted the wrong images (or possibly both wrong numbers and images).
1/9 sec, ISO 800 I believe?
1/4 sec, ISO 400 is it not?
Next the short exposure shot with 1/38 sec shutter speed:
That is the point of dispute - but only re Fuji's 'multi-exposure EXR-HD' function.
Well, it is about ISO in the first instance.No - the discussion/the point of debate, is not about 'normal single exposures'So we agree then
Tone curve/compression is not ISO boost.
It is a fundamental part of the the total digital camera system - it is not "smoke and mirrors" .
Then perhaps that's better discussed in its own separate topic/thread - some of us have covered simple noise/exposure theory many years ago.
No, it's plainly not that simple when dealing with the selective combination of two different exposures (HDR) - although I don't see any difficulty in recognising the simple logic of 'two different exposures (different shutter speeds) at one ISO).
It doesn't do it 'selectively' (e.g. per pixel) it does it 'blanket 50% of pixels', and only when either the user, or the camera's 'Auto-EXR', thinks it beneficial.
Yes - because that would just produce under-expos and/or require ISO boosting.
The full shots are below, first the normal exposure shot with shutter speed of 1/9 sec:
Next the short exposure shot with 1/38 sec shutter speed:
Now some crops. In the first crop, it is clear that the noise is much more visible in the short exposure case on the right due to the worse signal to noise ratio.
I think this has gotten silly.
I wasn't asking for a discussion.
I was trying to illustrate to you that the 'multi-shot HDR' method is of exactly the same exposure and processing as Fuji's 'single shot/multi-exposure EXR HD' - and thus that neither involve ISO changes.
Yes I've heard your take on it so many times now, and I understand what you've said - but I still completely disagree with your interpretation of 'the two exposures' in Fuji EXR HD.
That's O.K. - I guess that's where I'll give up trying to persuade you otherwise.
No that's not entirely true - I only object to the reference to using 'higher ISO' in the 'multi-exposures', when in fact it is just 'different exposures - at same ISO'.
Well, you seem to be unwilling to even discuss what probably is the best definition, the best analogue, that I had just offered.
It's not the "preferred presentation" - it's the the correct definition, and the proper description.
