EXR DR (dynamic range) mode question.

[John Carson]

John Carson wrote:

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.

The only tangible difference being that Fuji's two exposures are 'simultaneous', where as the conventional 'multi-shot' method's exposures are 'consecutive' (albeit barely seconds in between, if fast 'bracketing' is used).

My reluctance to discuss the issue is because of limited patience, not because I am avoiding your killer argument. Just for the record, here are some brief words.

With multi-shot HDR one of the images will have a faster shutter speed and it will be selectively brightened in post processing. The combined effect of in-camera faster shutter speed and brightening in post processing is similar to, though not exactly the same as, using higher ISO in camera. The difference is that the brightening is selective. Just as with higher ISO, the reduced shutter speed will worsen the signal to noise ratio.

See what a big difference it made to discuss multi-shot HDR?

--
john carson

 

[Mark H]

John Carson wrote:

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.

The only tangible difference being that Fuji's two exposures are 'simultaneous', where as the conventional 'multi-shot' method's exposures are 'consecutive' (albeit barely seconds in between, if fast 'bracketing' is used).

My reluctance to discuss the issue is because of limited patience, not because I am avoiding your killer argument. Just for the record, here are some brief words.

With multi-shot HDR one of the images will have a faster shutter speed and it will be selectively brightened in post processing.

The combined effect of in-camera faster shutter speed and brightening in post processing is similar to, though not exactly the same as, using higher ISO in camera. The difference is that the brightening is selective. Just as with higher ISO, the reduced shutter speed will worsen the signal to noise ratio.

You keep repeating the same thing over and again - but to no avail as far as I'm concerned.

See what a big difference it made to discuss multi-shot HDR?

You are missing, or dismissing, the pertinent point - that in the 'multi-shot HDR' the camera's ISO setting would not be changed between exposures, and that is exactly the same as what is occurring with Fuji's EXR HD's dual exposure(s).

'Selectively brightening' part of an image in post processing is not increasing the captured image's ISO - no more so than playing with curves, levels or contrast would be deemed to.

Even though S/N is increased slightly (remember we are only brightening 'highlights') that does not equate to increasing the images' ISO.

It may be a bit 'like' ISO, but it is very far from being the same thing.

 

[John Carson]

'Selectively brightening' part of an image in post processing is not increasing the captured image's ISO - no more so than playing with curves, levels or contrast would be deemed to.

Your position is apparently that playing with curves in one way is raising ISO but playing with them in a somewhat different way is not. It is a rather subtle distinction to be so dogmatic about.

Think of the DR enhancement in the s100fs. The ISO value as recorded in the camera's settings must be increased when you increase the DR setting. And what is the effect of all this? Fuji raises the shutter speed and selectively brightens the image. So Fuji is LYING about having raised ISO. I think you should send them a letter of complaint.

--
john carson

 

[min32]

You explanation is confuse. I just made an analog, let said EXR sensor had two set of sensors equivalent to two camera. One camera set to ISO100 with normal 0EV exposure and the other camera set to ISO100 with -2EV exposure for DR400. When I press both camera at same time, one will be use faster shutter speed and the other will be slower shutter speed. Are you mean the the --2EV exposure came using higher ISO as camera setting. If you brighter up the under exposure photo and process it, still had higher noise?. If this what you mean, so all camera if under expose should consider higher ISO!!!

 

[John Carson]

You explanation is confuse. I just made an analog, let said EXR sensor had two set of sensors equivalent to two camera. One camera set to ISO100 with normal 0EV exposure and the other camera set to ISO100 with -2EV exposure for DR400. When I press both camera at same time, one will be use faster shutter speed and the other will be slower shutter speed. Are you mean the the --2EV exposure came using higher ISO as camera setting.

I'm saying it doesn't matter much what the camera tells you the setting is. What is important to note is that choosing two stops higher ISO has the same effect on exposure as choosing -2 EV.

If you brighter up the under exposure photo and process it, still had higher noise?.

Yes it still has higher noise, as I showed with an example.

If this what you mean, so all camera if under expose should consider higher ISO!!!

Choosing two stops higher ISO has the same effect on exposure as choosing -2 EV, so neither approach has any obvious advantage over the other.

--
john carson

 

[Steen Bay]

I'm saying it doesn't matter much what the camera tells you the setting is. What is important to note is that choosing two stops higher ISO has the same effect on exposure as choosing -2 EV.

Not quite, with -2 EV we'll get 2 stops more highlight headroom, with 2 stops higher ISO we won't.

 

[John Carson]

I'm saying it doesn't matter much what the camera tells you the setting is. What is important to note is that choosing two stops higher ISO has the same effect on exposure as choosing -2 EV.

Not quite, with -2 EV we'll get 2 stops more highlight headroom, with 2 stops higher ISO we won't.

By "exposure" I meant the shutter speed.

Of course, with 2 stops higher ISO, there is in-camera processing that doesn't occur with -2 EV, with the effect you state.

--
john carson

 

[min32]

For compact camera, how you capture the detail of highlight and at the same time the detail of low ligth area? Are you just under exposure it to get the highlight data and low ligth data by using higher ISO (then pp the photo) or just take two photo with different EV exposure at same iso setting? Or the camera had the feature can take two photos at once time follow by the other time then process it? Which appoal you prefered?

 

[John Carson]

For compact camera, how you capture the detail of highlight and at the same time the detail of low ligth area? Are you just under exposure it to get the highlight data and low ligth data by using higher ISO (then pp the photo) or just take two photo with different EV exposure at same iso setting? Or the camera had the feature can take two photos at once time follow by the other time then process it? Which appoal you prefered?

The best method for a compact camera is that used by the Fuji EXR cameras. They take two exposures simultaneously. Half the photosites use a longer exposure and half use a shorter exposure.

--
john carson

 

[Mark H]

'Selectively brightening' part of an image in post processing is not increasing the captured image's ISO - no more so than playing with curves, levels or contrast would be deemed to.

Your position is apparently that playing with curves in one way is raising ISO...

No, that is not my position - you've misread, or misunderstood my statement.

I stated that 'selectively brightening' part of an image is not changing the image's ISO - it is comparable to 'playing with curves, levels or contrast', which are also not ISO changes.

Think of the DR enhancement in the s100fs. The ISO value as recorded in the camera's settings must be increased when you increase the DR setting. And what is the effect of all this? Fuji raises the shutter speed and selectively brightens the image.

Comparison of the EXR 'HD' with the S100fs 'D-Rng' is invalid...

• The S100fs is under-exposing the 'whole frame', and very nearly the whole of the frame is 'brightened' The used/reported ISO may be higher because the 'whole frame' exposure is taken with just the one shutter speed, so the 'metering' is effectively different. And, there is still a crucial unanswered anomaly - how can it 'truly' be higher ISO, when 'true ISO' would have lost/discarded the extra highlight signal? The actual mechanism is actually more accurately described as 'under-exposure' at normal (not raised) ISO - and Fuji's choice to report raised ISO is questionable, even for the S100fs's method used here.

• The EXR 'HD' is under-exposing only 'one out of two (simultaneous) frames' - the other frame, with the majority of the final images' dynamic range is exposed normally. Only a 'part' of one frame is 'partly brightened' - and for what it's worth, the mutually exclusive 'part' of other frame is inevitably 'partly darkened' (there is no part darkening in the S100fs's 'D-Rng' method).

So Fuji is LYING about having raised ISO. I think you should send them a letter of complaint.

No - but it still is even questionable that Fuji chose to 'report' a higher ISO for the S100Fs 'D-Rng' (as raised above), because the exact ISO mechanism cannot be the same as it would normally function, but since the 'overall exposure settings' and the 'noise' would be similar to higher ISO, so Fuji 'chose' to treat it as so.

As regards EXR 'HD', the situation is very significantly different - and therefore there is essentially no need, nor any justification, for Fuji to report a higher ISO for the EXR's 'HD' mechanism - and sure enough, Fuji don't - the reported ISO can and is the same regular ISO as for that of the 'non-HD' exposure, and no extra lowest available ISO limitation restriction either (as there is on the S100fs).

 

[Mark H]

You explanation is confuse. I just made an analog, let said EXR sensor had two set of sensors equivalent to two camera. One camera set to ISO100 with normal 0EV exposure and the other camera set to ISO100 with -2EV exposure for DR400. When I press both camera at same time, one will be use faster shutter speed and the other will be slower shutter speed. Are you mean the the --2EV exposure came using higher ISO as camera setting. If you brighter up the under exposure photo and process it, still had higher noise?. If this what you mean, so all camera if under expose should consider higher ISO!!!

A key difference between -EV under-exposing, versus using higher ISO is...

With ' higher ISO ' - this would normally apply a gain/amplification of the sensor's signal captured, and so generally would immediately lose most of any potential extra highlights that may have been captured at the sensor.

With ' under-exposure ' - there is the option to only selectively amplify/boost the shadows and mid tones back to near normal, but to preserve the extra highlight range captured by not boosting the highlights (at least nothing like as much).

If you weren't adjusting the 'under-exposure' to preserve extra highlight/dynamic range, but just adjusted it for correct 'linear +EV exposure' - then the image noise would be near enough the same as the 'higher ISO'.

 

[Mark H]

You explanation is confuse. I just made an analog, let said EXR sensor had two set of sensors equivalent to two camera. One camera set to ISO100 with normal 0EV exposure and the other camera set to ISO100 with -2EV exposure for DR400. When I press both camera at same time, one will be use faster shutter speed and the other will be slower shutter speed. Are you mean the the --2EV exposure came using higher ISO as camera setting.

I'm saying it doesn't matter much what the camera tells you the setting is. What is important to note is that choosing two stops higher ISO has the same effect on exposure as choosing -2 EV.

Same aperture/shutter balance yes, but...

The higher ISO mechanism generally applies a signal/data gain that chops off the sensor's highlight range.

The -EV exposure - no (or less) extra gain is applied, and the sensor's highlight range is preserved.

If you brighter up the under exposure photo and process it, still had higher noise?.

Yes it still has higher noise, as I showed with an example.

If this what you mean, so all camera if under expose should consider higher ISO!!!

Depends what you mean by "if under expose" - I think you/John might not be talking about the same issue here?

If there is a problem with 'metering' that results in 'under-exposure' then the solution should be to correct the under-exposure with +EV compensation.

Raising to ISO will not correct under-exposure at all.

Choosing two stops higher ISO has the same effect on exposure as choosing -2 EV, so neither approach has any obvious advantage over the other.

No - the 'obvious advantage' is as I describe above (preservation of extra highlights).

 

[Mark H]

For compact camera, how you capture the detail of highlight and at the same time the detail of low ligth area? Are you just under exposure it to get the highlight data and low ligth data by using higher ISO (then pp the photo) or just take two photo with different EV exposure at same iso setting? Or the camera had the feature can take two photos at once time follow by the other time then process it? Which appoal you prefered?

The best method for a compact camera is that used by the Fuji EXR cameras. They take two exposures simultaneously. Half the photosites use a longer exposure and half use a shorter exposure.

That's a bit simplistic, to say it's 'the best'...

If time between exposures isn't a constraint (i.e. subject movement) then shooting two full resolution shots using + -EV would be better, because you get a 'full resolution' result.

Of course, the above requires a tripod, and some post-processing software work.

The Fuji EXR 'HD' advantage is that it will work with even moving subjects, and no tripod or PP work required - but its disadvantage is your image is only 50% of full resolution.