L DR100 vs M DR400 (RAW) Dynamic range test [pics]

[Orion12]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

Those Fuji engineers are an Odd Bunch but they do manage to pull of some decent capability so I think their jobs are still intact. . . lol

Many Thanks for Your Inputs & Understanding with Cheers from Orion

 

[Orion12]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

You and he are both dead wrong then, because SN mode is the pixel binning mode. That means that two identical exposures are combined in pixel pairs to double the physical size of the pixel, which we know happens to improve the noise characteristics in shadows.

When two identical exposures are used, you are in DR100 mode by definition. That is, SN mode is always DR100 by definition.

So ... when selecting EXR Auto, what the camera is actually doing is selecting between HR, SN (both DR100, the latter with binned pixels) and DR modes ... and when it selects DR mode it can select DR200 or DR400.

And note that just selecting DR100 does not happen, because that would require that either SN or HR mode be selected, since you have to choose whether to demosaic after binning, or without binning. DR200 and DR400 are effectively blending modes, where the brighter and the darker exposures are essentially blended on a curve to give you a high dynamic range image. This is essentially an HDR mode with two exposures separated by 1 or 2 stops.

If you contemplate what the camera is trying to accomplish and what it must do to accomplish that, then it all becomes pretty clear ...

--
http://kimletkeman.blogspot.com

Looky here, you are dead wrong. The camera selected both SN mode and DR400. You probably never tried EXR Auto, otherwise you would have seen it happen. In EXR auto SN mode camera has the privilege of using both pixel binning and tone mapping. This is the equivalent to using M size whenever ISO is higher then 400, the camera is pairing the pixels and using software DR enhancement. But the point of Orion (and I was trying to clarify) was that it is possible to have SN mode and DR higher then 100.

This Image Info Display Says it All. . . Excellent Work DS21

F200EXR

I was talking about how the hardware modes work. If you are shooting EXR AUTO and the camera selects SN mode and DR200 or DR400, then can you guess what is happening?

SN mode is DR100 in hardware.

So if it selects DR200 or DR400 along with SN then it is binning the pixels in hardware (which is a DR100 operation) and then applying software based DR.

Now Kim one has to be "Careful" with the "Binning" word as when this is done in Software verses PHYSICALLY Combining 2-Separate Pixel-Sites together the RESULTS are Quite Different. . .

The following article does a good job at delineating the differences . . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

As I've said many times ... EXR AUTO is unpredictable and is to be avoided. At least by enthusiasts who have the ability to judge what they think is the most important goal in any one exposure.

I myself have had considerable Negative Reaction to this Strained-EXR-Implementation but My Reactions don't Change the Way those Fuji Engineers Implemented the EXR : SN + DR Functionality! So I've Simply Adopted MY Shooting Methods to be Able to Take Advantage of What I Personally Consider - The MOST Powerful Sensor Modes this Camera Possesses. . .

That's Not to Say I LIKE this Implementation and It's Certainly NOT how I would have done it BUT those Fuji Guys NEVER Asked Me - Maybe Next Time They WILL. . . lol

Thanks Again Kim for Your Inputs & Considerations

Many Cheers from Orion

 

[Kim Letkeman]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

You and he are both dead wrong then, because SN mode is the pixel binning mode. That means that two identical exposures are combined in pixel pairs to double the physical size of the pixel, which we know happens to improve the noise characteristics in shadows.

When two identical exposures are used, you are in DR100 mode by definition. That is, SN mode is always DR100 by definition.

So ... when selecting EXR Auto, what the camera is actually doing is selecting between HR, SN (both DR100, the latter with binned pixels) and DR modes ... and when it selects DR mode it can select DR200 or DR400.

And note that just selecting DR100 does not happen, because that would require that either SN or HR mode be selected, since you have to choose whether to demosaic after binning, or without binning. DR200 and DR400 are effectively blending modes, where the brighter and the darker exposures are essentially blended on a curve to give you a high dynamic range image. This is essentially an HDR mode with two exposures separated by 1 or 2 stops.

If you contemplate what the camera is trying to accomplish and what it must do to accomplish that, then it all becomes pretty clear ...

--
http://kimletkeman.blogspot.com

Looky here, you are dead wrong. The camera selected both SN mode and DR400. You probably never tried EXR Auto, otherwise you would have seen it happen. In EXR auto SN mode camera has the privilege of using both pixel binning and tone mapping. This is the equivalent to using M size whenever ISO is higher then 400, the camera is pairing the pixels and using software DR enhancement. But the point of Orion (and I was trying to clarify) was that it is possible to have SN mode and DR higher then 100.

This Image Info Display Says it All. . . Excellent Work DS21

F200EXR

I was talking about how the hardware modes work. If you are shooting EXR AUTO and the camera selects SN mode and DR200 or DR400, then can you guess what is happening?

SN mode is DR100 in hardware.

So if it selects DR200 or DR400 along with SN then it is binning the pixels in hardware (which is a DR100 operation) and then applying software based DR.

Now Kim one has to be "Careful" with the "Binning" word as when this is done in Software verses PHYSICALLY Combining 2-Separate Pixel-Sites together the RESULTS are Quite Different. . .

No, one does not have to be careful with the binning word because that has been the whole point of the EXR technology from day one.

3 modes: SN is bin in hardware by combining two side by side pixels (much, much better than standard bayer binning), DR is blend in hardware by combining adjacent pixels at different exposures on a curve, and HR is use all pixels as individuals.

The following article does a good job at delineating the differences . . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

As I said ... old school software binning is irrelevant since 2009 when EXR came out.

As I've said many times ... EXR AUTO is unpredictable and is to be avoided. At least by enthusiasts who have the ability to judge what they think is the most important goal in any one exposure.

I myself have had considerable Negative Reaction to this Strained-EXR-Implementation but My Reactions don't Change the Way those Fuji Engineers Implemented the EXR : SN + DR Functionality! So I've Simply Adopted MY Shooting Methods to be Able to Take Advantage of What I Personally Consider - The MOST Powerful Sensor Modes this Camera Possesses. . .

If you say so. I have simplified shooting this sensor down to settings that are very popular and used by many to get great captures. If you like SN mode plus software DR that is fine. I don't.

That's Not to Say I LIKE this Implementation and It's Certainly NOT how I would have done it BUT those Fuji Guys NEVER Asked Me - Maybe Next Time They WILL. . . lol

Thanks Again Kim for Your Inputs & Considerations

Many Cheers from Orion

--

Nothing Special

This blog is my personal space. Do not confuse this with statements or opinions on behalf of any employer I have ever had, have now, or will ever have in the future.

kimletkeman.blogspot.com

 

[Kim Letkeman]

Looking at some tests I'm running this morning, DR400 seems to get automatically switched off as the ISO goes up. With the camera set to DR400 (6MP), I get double sized raws at iso 100-400 but they go to half size as soon as ISO gets to 800. This seems to happen automatically and without any indication to tell you it's happened (which is annoying!).

Is this well understood amongst the experts?

I'm not an X10 expert but have ran into a Similar Phenomenon on the HS50EXR. . .

At this point in time I usually don't pay much concern to Captured DR Levels as I routinely edit these relationships in RAW and use My Own Personal Visual Preferences for My Photographic Ruler. . .

Well, the DR setting also controls how the Fuji cameras expose to some extent. Highlights are protected much better in DR400 mode than in SN mode because the camera underexposes more in the former mode. I need to spend some time experimenting with this again to nail down the amount.

Occasionally I shoot in EXR : Auto Mode as this the ONLY Method of getting an SN Mode + DR200 or DR400 RAW capture with an HS50EXR. . . ( There is an EXR Mode Indicator when 1/2 Pressing )

I don't understand that comment.
You get SN (DR100) or DR200 or DR400 in any one shot.

Unfortunately the Dedicated EXR : SN Mode has a Fixed DR100 Attribute.

As I said ... this is what SN mode does by definition.

It appears you are making an "M" Size / EXR SN Mode Switching Assumption. . .
Remember that the EXR Sensors are Manufactured with 2 Equal Sized Image Planes. . .

I really have not forgotten. And I am not making an assumption. I am defining SN mode.

On the HS50EXR the 16-Mp Sensor "M" Size = 1/2 Size or a Single 8-Mp Image Plane. . .

I just defined it all above ...

IF You Shoot in RAW Only in PASM Modes you will Discover the Default Sensor Mode = 16-Mp

Yes. Raw only is HR mode. But as testing has shown, there is advantage in the raw files to setting DR400, which I always do. Not sure what Fuji does to get such latitude ... perhaps it is worth more testing to isolate that but I have seen the difference in tests.

IF You Shoot in RAW Only in EXR SN Mode you have a Default Sensor Mode Size = 8-Mp

Yes, binned in hardware is the default for SN, which is why the files are half sized.

When Shooting in PASM Fashion ( Which DOES Give You the DR100, DR200 & DR400 User Selectable Options ). . . The DEFAULT Sensor Mode - Appears to be the Full Frame Buffer Output made up of (2) 1/2 Resolution Image Planes and NOT EXR : SN Mode Twinned-Pixel Output. . .

PASM requires that the raw converter process the sensor output. But in JPEG it is for all practical purposes the same.

This imho is one of the Reasons we STILL get Arguments on PASM Mode "M" or "L" results as these are Actually Using the SAME Sensor Configuration albeit in an "M" you are getting one of the 1/2 Frame Outputs but also Have 1/2 of the Pixels to Process so the Camera has More Time to More Precisely Process them. . . But the Output is Still Not RADICALLY Different as the Same Pixel-Site Sizes are used in both. ( That's My Take )

I really don't think the camera tosses half the data in M size.

I Totally WISH Your Assumption was True and Was the Way the SN Sensor Mode is implemented but the Fuji EXIF Data ( Which Tags Specific EXR Sensor Modes ) doesn't indicate EXR : SN Sensor Mode usage in ANY Images - UNLESS they were shot in either Dialed EXR : SN mode or Dialed EXR : AUTO mode ( & Then Only when the Camera Selects SN DR100 default mode or SN mode + DR200 or DR400 ) Otherwise Fuji does Clearly EXIF tag this data stream.

Because it is not SN mode. SN mode is on the EXR dial. PASM mode simply shoots the same way without all the automation. At least, that's what the empirical results indicate.

I have compared the DR modes extensively and can see no difference in the output when shot in EXR DR or PASM DR400 ... I believe that SN modes look about the same too.

The raw file can be processed however you want to process it if you shoot L size, but the pixels are effectively shot at DR100 in L size.

In M size, you can choose to protect tones with DR400 or you can choose to try to maximize noise reduction with SN or PASM DR100,

I personally hold the EXR SN Mode in Highest Regards as you already know ( Which is a 8-Mp "M" Size Sensor Output - In Fact there is NO Possible Way to Shoot an SN Mode "L" size file PERIOD ) . . . ( I know YOU already Know this but some others may not )

I defined it above. And explained why I do not hold SN mode in high regard, as it requires shooting at DR100.

If you choose to shoot in EXR AUTO so you can allow the camera to make the choice of which software DR mode it throws on top then you are giving up far too much control for an enthusiast ... that's why I would never even test such a mode.

When I first got my HS50EXR I struggled to find the "BEST" Sensor Shooting Modes. . .
And AFTER More than 10,000 Shots - I've Landed on TWO Shooting Modalities that Serve ME the BEST - Those are. . .

1. Dialed EXR : SN Mode ( Semi-Automatic but with User Controlled Exposure, White Balance, Metering Method, Color, Sharpness ect ect ect )

2. Dialed EXR : AUTO Mode ( Full Auto Mode and the ONLY Method I know of to Get an SN Mode DR 200 or DR400 RAW File )

And I would never shoot those modes. After far more than 10,000 shots on just about every EXR camera that has graced the planet.

but anything else is mostly just dicking around to create the illusion of control.

I definitely agree with the above line. . .

The OTHER Method is getting SN Mode RAW captures is in EXR : SN Mode which has a FIXED DR100 Aspect.

Yes, it does. But what you describe above only serves to confuse the issues, since you are not separating what matters (dynamic range / noise) from the various ways in which the camera can be set up ...

I can only make this as Easy as It's come to me which Wasn't Totally Easy or Particularly Straight Forward and the Fuji Implementation of it's EXR Technology UNFORTUNATELY isn't a No-Brainer to Follow or Figure Out. . . I also have to Say It's Got Some Amazing Capability and I LOVE my HS50EXR - Quirks and All. . .

The HS50 is a fine camera ... but requires a lot of technique to shoot well at 1000mm ...

--
http://kimletkeman.blogspot.com

btw AFTER reading Your Comments --> I did some Additional Test Shooting on a Tuff DR Subject and Newly Discovered that I Could Get Some ADDITIONAL DR from my EXR : SN + DR400 RAW File. . . ( Beyond the DR Levels I WAS Previously Getting from my RAW Processing )

NOW this Even Further Pushes Me in the Direction of Switching to EXR : AUTO ( RAW ) when facing "Challenging" DR Subjects. . . As it's STILL the ONLY Method I KNOW of Getting an EXR : SN Mode DR200 or DR400 RAW Capture ! I really Wish their were OTHER Ways but I'm thankful anyway.

When shooting at high ISO, all modes are essentially done in software anyway ... I wonder if you would see any difference between fake AUTO SN/DR mode and any other mode at 1600 ISO once you run them through the raw converters ... and I certainly would not accept the loss of control by using an auto mode at high ISO ...

Ask Trevor G about that .... he's done the most testing on it I believe ...

These are from over 200ft away ...

1600 ISO ... RAW only ...

800 ISO ... RAW only ... cropped ...

--

Nothing Special

This blog is my personal space. Do not confuse this with statements or opinions on behalf of any employer I have ever had, have now, or will ever have in the future.

kimletkeman.blogspot.com

 

[DS21]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

You and he are both dead wrong then, because SN mode is the pixel binning mode. That means that two identical exposures are combined in pixel pairs to double the physical size of the pixel, which we know happens to improve the noise characteristics in shadows.

When two identical exposures are used, you are in DR100 mode by definition. That is, SN mode is always DR100 by definition.

So ... when selecting EXR Auto, what the camera is actually doing is selecting between HR, SN (both DR100, the latter with binned pixels) and DR modes ... and when it selects DR mode it can select DR200 or DR400.

And note that just selecting DR100 does not happen, because that would require that either SN or HR mode be selected, since you have to choose whether to demosaic after binning, or without binning. DR200 and DR400 are effectively blending modes, where the brighter and the darker exposures are essentially blended on a curve to give you a high dynamic range image. This is essentially an HDR mode with two exposures separated by 1 or 2 stops.

If you contemplate what the camera is trying to accomplish and what it must do to accomplish that, then it all becomes pretty clear ...

--
http://kimletkeman.blogspot.com

Looky here, you are dead wrong. The camera selected both SN mode and DR400. You probably never tried EXR Auto, otherwise you would have seen it happen. In EXR auto SN mode camera has the privilege of using both pixel binning and tone mapping. This is the equivalent to using M size whenever ISO is higher then 400, the camera is pairing the pixels and using software DR enhancement. But the point of Orion (and I was trying to clarify) was that it is possible to have SN mode and DR higher then 100.

This Image Info Display Says it All. . . Excellent Work DS21

F200EXR

I was talking about how the hardware modes work. If you are shooting EXR AUTO and the camera selects SN mode and DR200 or DR400, then can you guess what is happening?

SN mode is DR100 in hardware.

So if it selects DR200 or DR400 along with SN then it is binning the pixels in hardware (which is a DR100 operation) and then applying software based DR.

Now Kim one has to be "Careful" with the "Binning" word as when this is done in Software verses PHYSICALLY Combining 2-Separate Pixel-Sites together the RESULTS are Quite Different. . .

The following article does a good job at delineating the differences . . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

As I've said many times ... EXR AUTO is unpredictable and is to be avoided. At least by enthusiasts who have the ability to judge what they think is the most important goal in any one exposure.

I myself have had considerable Negative Reaction to this Strained-EXR-Implementation but My Reactions don't Change the Way those Fuji Engineers Implemented the EXR : SN + DR Functionality! So I've Simply Adopted MY Shooting Methods to be Able to Take Advantage of What I Personally Consider - The MOST Powerful Sensor Modes this Camera Possesses. . .

That's Not to Say I LIKE this Implementation and It's Certainly NOT how I would have done it BUT those Fuji Guys NEVER Asked Me - Maybe Next Time They WILL. . . lol

Thanks Again Kim for Your Inputs & Considerations

Many Cheers from Orion

The irony in all of this is that KIm's beloved settings work exactly the same as "dreaded" SN mode with added software DR, whenever the ISO is higher then 400. If you like EXR SN implementation in auto EXR mode, you can safely use the undocumented (by Fuji) feature of camera using it whenever in PASM in M size mode (in lower light, with ISO higher then 400), with DR set to Auto (then the camera is behaving almost exactly the same as in SN mode in Auto EXR}, or by setting DR to 400 (like KIm advocates) all the time to take the decision making process out of the camera's software hands. The only difference in EXR auto could be the slightly different amount of noise reduction the camera uses, but not concerning you if you shoot RAW.

 

[Orion12]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

You and he are both dead wrong then, because SN mode is the pixel binning mode. That means that two identical exposures are combined in pixel pairs to double the physical size of the pixel, which we know happens to improve the noise characteristics in shadows.

Now I'm Starting to SEE WHERE the Semantics are going South. . .

You my friend Kim are Incorrectly Using English Descriptors in your Above Statement of Pixel Binning !

In SN Mode there are NOT "two identical exposures... combined in pixel pairs. . ." their are (2) individual Pixel-Sites that are Electro-Mechanically or if you prefer Physically Tied together into ONE Functional Double Sized Pixel-Site. . . And FROM this Twinned-Pair you are Getting ONE Exposure.

When two identical exposures are used, you are in DR100 mode by definition. That is, SN mode is always DR100 by definition.

So ... when selecting EXR Auto, what the camera is actually doing is selecting between HR, SN (both DR100, the latter with binned pixels) and DR modes ... and when it selects DR mode it can select DR200 or DR400.

Actually the EXR : DR ( D-Range Priority ) Sensor Mode is a Totally Different Operation than the SOFTWARE DR Operations. . . Although Similar or Closely Related in Function.

In the Dedicated EXR : DR Mode the Sensor is Divided into it's TWO Image Frames and EACH of these TWO Frames Samples & Stores a DIFFERENT EXPOSURE. . . It's using the Small / Single Size or Non-Twinned Pixel-Sites. So on an HS50EXR you have (2) 8-Mp Image Frames EACH with a Different Exposure. . . Having these TWO Separate Image Frames allows Up to DR=1600 Range.

The SOFTWARE Only Version of this Operation is Limited to DR=400 Range.

And note that just selecting DR100 does not happen, because that would require that either SN or HR mode be selected, since you have to choose whether to demosaic after binning, or without binning. DR200 and DR400 are effectively blending modes, where the brighter and the darker exposures are essentially blended on a curve to give you a high dynamic range image. This is essentially an HDR mode with two exposures separated by 1 or 2 stops.

If you contemplate what the camera is trying to accomplish and what it must do to accomplish that, then it all becomes pretty clear ...

--
http://kimletkeman.blogspot.com

Looky here, you are dead wrong. The camera selected both SN mode and DR400. You probably never tried EXR Auto, otherwise you would have seen it happen. In EXR auto SN mode camera has the privilege of using both pixel binning and tone mapping. This is the equivalent to using M size whenever ISO is higher then 400, the camera is pairing the pixels and using software DR enhancement. But the point of Orion (and I was trying to clarify) was that it is possible to have SN mode and DR higher then 100.

This Image Info Display Says it All. . . Excellent Work DS21

F200EXR

I was talking about how the hardware modes work. If you are shooting EXR AUTO and the camera selects SN mode and DR200 or DR400, then can you guess what is happening?

SN mode is DR100 in hardware.

So if it selects DR200 or DR400 along with SN then it is binning the pixels in hardware (which is a DR100 operation) and then applying software based DR.

Now Kim one has to be "Careful" with the "Binning" word as when this is done in Software verses PHYSICALLY Combining 2-Separate Pixel-Sites together the RESULTS are Quite Different. . .

No, one does not have to be careful with the binning word because that has been the whole point of the EXR technology from day one.

3 modes: SN is bin in hardware by combining two side by side pixels (much, much better than standard bayer binning), DR is blend in hardware by combining adjacent pixels at different exposures on a curve, and HR is use all pixels as individuals.

The following article does a good job at delineating the differences . . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

As I said ... old school software binning is irrelevant since 2009 when EXR came out.

As I've said many times ... EXR AUTO is unpredictable and is to be avoided. At least by enthusiasts who have the ability to judge what they think is the most important goal in any one exposure.

I myself have had considerable Negative Reaction to this Strained-EXR-Implementation but My Reactions don't Change the Way those Fuji Engineers Implemented the EXR : SN + DR Functionality! So I've Simply Adopted MY Shooting Methods to be Able to Take Advantage of What I Personally Consider - The MOST Powerful Sensor Modes this Camera Possesses. . .

If you say so. I have simplified shooting this sensor down to settings that are very popular and used by many to get great captures. If you like SN mode plus software DR that is fine. I don't.

That's Not to Say I LIKE this Implementation and It's Certainly NOT how I would have done it BUT those Fuji Guys NEVER Asked Me - Maybe Next Time They WILL. . . lol

Thanks Again Kim for Your Inputs & Considerations

Many Cheers from Orion

--
http://kimletkeman.blogspot.com

More Cheers from Orion

 

[Orion12]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

You and he are both dead wrong then, because SN mode is the pixel binning mode. That means that two identical exposures are combined in pixel pairs to double the physical size of the pixel, which we know happens to improve the noise characteristics in shadows.

When two identical exposures are used, you are in DR100 mode by definition. That is, SN mode is always DR100 by definition.

So ... when selecting EXR Auto, what the camera is actually doing is selecting between HR, SN (both DR100, the latter with binned pixels) and DR modes ... and when it selects DR mode it can select DR200 or DR400.

And note that just selecting DR100 does not happen, because that would require that either SN or HR mode be selected, since you have to choose whether to demosaic after binning, or without binning. DR200 and DR400 are effectively blending modes, where the brighter and the darker exposures are essentially blended on a curve to give you a high dynamic range image. This is essentially an HDR mode with two exposures separated by 1 or 2 stops.

If you contemplate what the camera is trying to accomplish and what it must do to accomplish that, then it all becomes pretty clear ...

--
http://kimletkeman.blogspot.com

Looky here, you are dead wrong. The camera selected both SN mode and DR400. You probably never tried EXR Auto, otherwise you would have seen it happen. In EXR auto SN mode camera has the privilege of using both pixel binning and tone mapping. This is the equivalent to using M size whenever ISO is higher then 400, the camera is pairing the pixels and using software DR enhancement. But the point of Orion (and I was trying to clarify) was that it is possible to have SN mode and DR higher then 100.

This Image Info Display Says it All. . . Excellent Work DS21

F200EXR

I was talking about how the hardware modes work. If you are shooting EXR AUTO and the camera selects SN mode and DR200 or DR400, then can you guess what is happening?

SN mode is DR100 in hardware.

So if it selects DR200 or DR400 along with SN then it is binning the pixels in hardware (which is a DR100 operation) and then applying software based DR.

Now Kim one has to be "Careful" with the "Binning" word as when this is done in Software verses PHYSICALLY Combining 2-Separate Pixel-Sites together the RESULTS are Quite Different. . .

The following article does a good job at delineating the differences . . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

As I've said many times ... EXR AUTO is unpredictable and is to be avoided. At least by enthusiasts who have the ability to judge what they think is the most important goal in any one exposure.

I myself have had considerable Negative Reaction to this Strained-EXR-Implementation but My Reactions don't Change the Way those Fuji Engineers Implemented the EXR : SN + DR Functionality! So I've Simply Adopted MY Shooting Methods to be Able to Take Advantage of What I Personally Consider - The MOST Powerful Sensor Modes this Camera Possesses. . .

That's Not to Say I LIKE this Implementation and It's Certainly NOT how I would have done it BUT those Fuji Guys NEVER Asked Me - Maybe Next Time They WILL. . . lol

Thanks Again Kim for Your Inputs & Considerations

Many Cheers from Orion

The irony in all of this is that KIm's beloved settings work exactly the same as "dreaded" SN mode with added software DR, whenever the ISO is higher then 400. If you like EXR SN implementation in auto EXR mode, you can safely use the undocumented (by Fuji) feature of camera using it whenever in PASM in M size mode (in lower light, with ISO higher then 400), with DR set to Auto (then the camera is behaving almost exactly the same as in SN mode in Auto EXR}, or by setting DR to 400 (like KIm advocates) all the time to take the decision making process out of the camera's software hands. The only difference in EXR auto could be the slightly different amount of noise reduction the camera uses, but not concerning you if you shoot RAW.

Yes the RAW Aspect is of Critical Import in this escapade. . . I'm thinking a RAW ONLY Shoot Out Will Be a Deciding Factor. . . I happen to Already Know what Those Results would Look Like. . .
However We May Still Have to Figure In the "Influence" of "M" & "L" File Size Selection. . .

Here's a Fairly Definitive RAW TEST RESULT for HIGH-ISO ( 3200 ) SN 8-Mp RAW verses PASM 16-Mp RAW. . . Both Processed in Photo Ninja. . .

PASM 16-Mp RAW on Left BOTH ISO-3200 / SAME Manual Focus EXR : SN 8-Mp RAW on Right

One of the NEXT Rounds will be to TEST the PASM 12-Mb Raw Output AGAINST the SN 12-Mb Raw

IF the ISO-3200 PASM 12-Mb Raw STAND UP as EQUAL in NOISE with the EXR : SN Mode 12-Mb Raw THEN - Their Can Be ADDITIONAL Celebration in Fuji Land. . . BUT For NOW. . . The Jury is STILL OUT :-|

Great Cheers from Orion

 

[Kim Letkeman]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

You and he are both dead wrong then, because SN mode is the pixel binning mode. That means that two identical exposures are combined in pixel pairs to double the physical size of the pixel, which we know happens to improve the noise characteristics in shadows.

Now I'm Starting to SEE WHERE the Semantics are going South. . .

You my friend Kim are Incorrectly Using English Descriptors in your Above Statement of Pixel Binning !

In SN Mode there are NOT "two identical exposures... combined in pixel pairs. . ." their are (2) individual Pixel-Sites that are Electro-Mechanically or if you prefer Physically Tied together into ONE Functional Double Sized Pixel-Site. . . And FROM this Twinned-Pair you are Getting ONE Exposure.

1) That's not what I have read. They are effectively twinned, that's true, since the exposure is always identical in SN or HR modes, or any time DR100 is selected.

2) It would make no difference if it was true. Same end result whether they are electrically coupled with one read operation, or combined after the fact from two read operations and an addition.

When two identical exposures are used, you are in DR100 mode by definition. That is, SN mode is always DR100 by definition.

So ... when selecting EXR Auto, what the camera is actually doing is selecting between HR, SN (both DR100, the latter with binned pixels) and DR modes ... and when it selects DR mode it can select DR200 or DR400.

Actually the EXR : DR ( D-Range Priority ) Sensor Mode is a Totally Different Operation than the SOFTWARE DR Operations. . . Although Similar or Closely Related in Function.

In the Dedicated EXR : DR Mode the Sensor is Divided into it's TWO Image Frames and EACH of these TWO Frames Samples & Stores a DIFFERENT EXPOSURE. . . It's using the Small / Single Size or Non-Twinned Pixel-Sites. So on an HS50EXR you have (2) 8-Mp Image Frames EACH with a Different Exposure. . . Having these TWO Separate Image Frames allows Up to DR=1600 Range.

The SOFTWARE Only Version of this Operation is Limited to DR=400 Range.

That's a very confused explanation. This is what I understand:

• Hardware DR is used when it is possible to separate by shutter speed (obviously using electronic shutter independently on each half of the sensor by terminating the shorter exposure either 1 or 2 stops early.) These pixels are blended on a curve in order to get a smooth final result.
• Whenever DR100 is selected (HR mode, SN mode, and any other mode where DR100 is explicit), the two halves of the sensor are exposed identically.
• When SN is selected, the two halves are combined before any further processing takes place. This is hardware binning of adjacent pixels. It matters no whether there is one read operation or two, but the descriptions I have read say that there are two, as there are two physical photosites.
• When HR is selected, or whenever L size is shot, all photosites are treated independently.
• When ISO hits 400 and above, software DR is always used (although I feel that the jury is out on whether there are sensitivity tricks available to the two independent sensor halves.)
• Having two separate exposures allows DR200 (one stop difference) or DR400 (2 stops difference) only. DR800 and DR1600 are one and two stops of additional software DR and they are an ugly hack. Truly horrible smoothing of tones.

What this all amounts to is that you can set PASM with auto ISO 1600 or 3200 and shoot M size with DR400 and get about as good a result as possible under all conditions when shooting raw + jpg.

And note that just selecting DR100 does not happen, because that would require that either SN or HR mode be selected, since you have to choose whether to demosaic after binning, or without binning. DR200 and DR400 are effectively blending modes, where the brighter and the darker exposures are essentially blended on a curve to give you a high dynamic range image. This is essentially an HDR mode with two exposures separated by 1 or 2 stops.

If you contemplate what the camera is trying to accomplish and what it must do to accomplish that, then it all becomes pretty clear ...

--
http://kimletkeman.blogspot.com

Looky here, you are dead wrong. The camera selected both SN mode and DR400. You probably never tried EXR Auto, otherwise you would have seen it happen. In EXR auto SN mode camera has the privilege of using both pixel binning and tone mapping. This is the equivalent to using M size whenever ISO is higher then 400, the camera is pairing the pixels and using software DR enhancement. But the point of Orion (and I was trying to clarify) was that it is possible to have SN mode and DR higher then 100.

This Image Info Display Says it All. . . Excellent Work DS21

F200EXR

I was talking about how the hardware modes work. If you are shooting EXR AUTO and the camera selects SN mode and DR200 or DR400, then can you guess what is happening?

SN mode is DR100 in hardware.

So if it selects DR200 or DR400 along with SN then it is binning the pixels in hardware (which is a DR100 operation) and then applying software based DR.

Now Kim one has to be "Careful" with the "Binning" word as when this is done in Software verses PHYSICALLY Combining 2-Separate Pixel-Sites together the RESULTS are Quite Different. . .

No, one does not have to be careful with the binning word because that has been the whole point of the EXR technology from day one.

3 modes: SN is bin in hardware by combining two side by side pixels (much, much better than standard bayer binning), DR is blend in hardware by combining adjacent pixels at different exposures on a curve, and HR is use all pixels as individuals.

The following article does a good job at delineating the differences . . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

As I said ... old school software binning is irrelevant since 2009 when EXR came out.

As I've said many times ... EXR AUTO is unpredictable and is to be avoided. At least by enthusiasts who have the ability to judge what they think is the most important goal in any one exposure.

I myself have had considerable Negative Reaction to this Strained-EXR-Implementation but My Reactions don't Change the Way those Fuji Engineers Implemented the EXR : SN + DR Functionality! So I've Simply Adopted MY Shooting Methods to be Able to Take Advantage of What I Personally Consider - The MOST Powerful Sensor Modes this Camera Possesses. . .

If you say so. I have simplified shooting this sensor down to settings that are very popular and used by many to get great captures. If you like SN mode plus software DR that is fine. I don't.

That's Not to Say I LIKE this Implementation and It's Certainly NOT how I would have done it BUT those Fuji Guys NEVER Asked Me - Maybe Next Time They WILL. . . lol

Thanks Again Kim for Your Inputs & Considerations

Many Cheers from Orion

--
http://kimletkeman.blogspot.com

More Cheers from Orion

--

Nothing Special

This blog is my personal space. Do not confuse this with statements or opinions on behalf of any employer I have ever had, have now, or will ever have in the future.

kimletkeman.blogspot.com

 

[Kim Letkeman]

The irony in all of this is that KIm's beloved settings work exactly the same as "dreaded" SN mode with added software DR, whenever the ISO is higher then 400.

Not really. I keep control over the exposures whereas running EXR AUTO gives up a lot of control.

If you like EXR SN implementation in auto EXR mode, you can safely use the undocumented (by Fuji) feature of camera using it whenever in PASM in M size mode (in lower light, with ISO higher then 400), with DR set to Auto (then the camera is behaving almost exactly the same as in SN mode in Auto EXR}, or by setting DR to 400 (like KIm advocates) all the time to take the decision making process out of the camera's software hands. The only difference in EXR auto could be the slightly different amount of noise reduction the camera uses, but not concerning you if you shoot RAW.

EXR AUTO is not the same as AUTO EXR. The former is a mode where the camera makes all decisions and the latter is a mode where you allow only DR to float (which is silly, since the camera often makes bad DR decisions.)

If you actually test the F200EXR (for example) you will see that setting EXR SN fixes DR at 100. So only EXR AUTIO has the ability to mix the hardware binning with software DR and that is a full auto mode. I.e. pointless to an enthusiast that wants to control how the camera exposes every aspect.

 

[Kim Letkeman]

Yes the RAW Aspect is of Critical Import in this escapade. . . I'm thinking a RAW ONLY Shoot Out Will Be a Deciding Factor. . . I happen to Already Know what Those Results would Look Like. . .
However We May Still Have to Figure In the "Influence" of "M" & "L" File Size Selection. . .

Here's a Fairly Definitive RAW TEST RESULT for HIGH-ISO ( 3200 ) SN 8-Mp RAW verses PASM 16-Mp RAW. . . Both Processed in Photo Ninja. . .

PASM 16-Mp RAW on Left BOTH ISO-3200 / SAME Manual Focus EXR : SN 8-Mp RAW on Right

How are these being displayed? The left hand image is thoroughly distorted. Did you just let some application downsize them both?

I suggest that you create a pair of real 100% crops after properly downsizing to match with a decent algorithm. That way we can see what is really in these two.

One of the NEXT Rounds will be to TEST the PASM 12-Mb Raw Output AGAINST the SN 12-Mb Raw

What SN 12mp raw? SN bins by definition.

IF the ISO-3200 PASM 12-Mb Raw STAND UP as EQUAL in NOISE with the EXR : SN Mode 12-Mb Raw THEN - Their Can Be ADDITIONAL Celebration in Fuji Land. . . BUT For NOW. . . The Jury is STILL OUT :-|

Well, since there is no full sized SN mode, I don't quite see what would be celebrated. And further, there have been many tests of Sn versus DR versus PASM over the years ... there is really not much in the differences ...

--

Nothing Special

This blog is my personal space. Do not confuse this with statements or opinions on behalf of any employer I have ever had, have now, or will ever have in the future.

kimletkeman.blogspot.com

 

[Orion12]

Yes the RAW Aspect is of Critical Import in this escapade. . . I'm thinking a RAW ONLY Shoot Out Will Be a Deciding Factor. . . I happen to Already Know what Those Results would Look Like. . .
However We May Still Have to Figure In the "Influence" of "M" & "L" File Size Selection. . .

Here's a Fairly Definitive RAW TEST RESULT for HIGH-ISO ( 3200 ) SN 8-Mp RAW verses PASM 16-Mp RAW. . . Both Processed in Photo Ninja. . .

PASM 16-Mp RAW on Left BOTH ISO-3200 / SAME Manual Focus EXR : SN 8-Mp RAW on Right

How are these being displayed? The left hand image is thoroughly distorted. Did you just let some application downsize them both?

I suggest that you create a pair of real 100% crops after properly downsizing to match with a decent algorithm. That way we can see what is really in these two.

One of the NEXT Rounds will be to TEST the PASM 12-Mb Raw Output AGAINST the SN 12-Mb Raw

What SN 12mp raw?

NOT SN 12mp raw ---> SN 12Mb raw Size Files. . .

For the HS50EXR the Default and ONLY SN Mode Raw Files are 12Mb ( MegaByte ) / 8-Mp ( MegaPixel) Output
( I think I've seen a Very Small Number of 24Mb SN Raw Files which is an Oddity and it's possible I didn't track them correctly so this could be just an observational error )

SN bins by definition.

IF the ISO-3200 PASM 12-Mb Raw STAND UP as EQUAL in NOISE with the EXR : SN Mode 12-Mb Raw THEN - Their Can Be ADDITIONAL Celebration in Fuji Land. . . BUT For NOW. . . The Jury is STILL OUT :-|

Well, since there is no full sized SN mode, I don't quite see what would be celebrated. And further, there have been many tests of Sn versus DR versus PASM over the years ... there is really not much in the differences ...

MY RAW Shooting Tests as Above Certainly Have a Notable Difference in Clarity & Noise Levels. . .
And My Scaling Methods and NOT Treacherous to results One Way Or Another.

I'm starting to believe that BECAUSE you are so Adverse to the Horrible EXR SN Mode Shooting Implementation, that You've Maybe NEVER Seriously TESTED Actual HS50EXR -->

Dedicated EXR SN MODE 12-Mb / 8-Mp Output RAW Files
or
Dedicated EXR SN MODE 12-Mb / 8-Mp Output + DR200 or DR400 RAW Files

After all the ONLY Way I Know of to Get Them are from those (2) "Amateur" Shooting Modes. . .

Have to Admit it's a "Tricky Challenge" to sort out what that Little Sensor is Doing at times. . .

I'll be doing MORE Testing also as this Inter-Active Discourse has raised a few Questions and Possibilities. . . Thanks again of ALL of Your Inputs & Considerations. . .

Cheers from Orion

SOOC ISO-800 Fuji Embedded Jpeg for EXR : SN Mode RAW Capture

SAME Shot RAW Processed in Photo Ninja + Further Tweaked in FastStone Viewer 5.0

Auditorium Capture from last weekend's the Deschutes County Expo Center - Home & Garden Show

 

[Orion12]

He is saying that in EXR Auto mode, if camera selects EXR SN, it is possible to get DR200 or DR400 as well if the scene requires it (the camera will do it automatically). This is the only way to get DR other then 100 in SN mode. Got to love Fuji engineers.

You and he are both dead wrong then, because SN mode is the pixel binning mode. That means that two identical exposures are combined in pixel pairs to double the physical size of the pixel, which we know happens to improve the noise characteristics in shadows.

Now I'm Starting to SEE WHERE the Semantics are going South. . .

You my friend Kim are Incorrectly Using English Descriptors in your Above Statement of Pixel Binning !

In SN Mode there are NOT "two identical exposures... combined in pixel pairs. . ." their are (2) individual Pixel-Sites that are Electro-Mechanically or if you prefer Physically Tied together into ONE Functional Double Sized Pixel-Site. . . And FROM this Twinned-Pair you are Getting ONE Exposure.

1) That's not what I have read. They are effectively twinned, that's true, since the exposure is always identical in SN or HR modes, or any time DR100 is selected.

It's NOT just a MATH Solution as Each of the Twinned-Pixel Light Collecting sites are PHYSICALLY Double Sized AND Electrically on the SAME Light Collection Plane - You get ADDITIONAL Sensor Performance --> Above & Beyond the Minor Gains of Simple Software ( MATH Only ) Binning.

2) It would make no difference if it was true. Same end result whether they are electrically coupled with one read operation, or combined after the fact from two read operations and an addition.

I want you to Read or Re-Read this Article in a "New Unit of Time" and really get the "Difference" in Binning Methods. . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

When two identical exposures are used, you are in DR100 mode by definition. That is, SN mode is always DR100 by definition.

So ... when selecting EXR Auto, what the camera is actually doing is selecting between HR, SN (both DR100, the latter with binned pixels) and DR modes ... and when it selects DR mode it can select DR200 or DR400.

Actually the EXR : DR ( D-Range Priority ) Sensor Mode is a Totally Different Operation than the SOFTWARE DR Operations. . . Although Similar or Closely Related in Function.

In the Dedicated EXR : DR Mode the Sensor is Divided into it's TWO Image Frames and EACH of these TWO Frames Samples & Stores a DIFFERENT EXPOSURE. . . It's using the Small / Single Size or Non-Twinned Pixel-Sites. So on an HS50EXR you have (2) 8-Mp Image Frames EACH with a Different Exposure. . . Having these TWO Separate Image Frames allows Up to DR=1600 Range.

The SOFTWARE Only Version of this Operation is Limited to DR=400 Range.

That's a very confused explanation. This is what I understand:

You can Easily Check my Above Descriptions and Verify for YourSelf by going into the Dialed EXR : DR Mode ( D-Range Priority ) and You will Discover that this DEDICATED EXR DR Mode is the ONLY Method in the Entire Camera where You can SELECT and SHOOT Up to DR1600 Jpeg & Raw Files.

It's Another one of those "Amateur" shooting modes that many don't find "Technically Thrilling". . .

• Hardware DR is used when it is possible to separate by shutter speed (obviously using electronic shutter independently on each half of the sensor by terminating the shorter exposure either 1 or 2 stops early.) These pixels are blended on a curve in order to get a smooth final result.
• Whenever DR100 is selected (HR mode, SN mode, and any other mode where DR100 is explicit), the two halves of the sensor are exposed identically.
• When SN is selected, the two halves are combined before any further processing takes place. This is hardware binning of adjacent pixels. It matters no whether there is one read operation or two, but the descriptions I have read say that there are two, as there are two physical photosites.
• When HR is selected, or whenever L size is shot, all photosites are treated independently.
• When ISO hits 400 and above, software DR is always used (although I feel that the jury is out on whether there are sensitivity tricks available to the two independent sensor halves.)
• Having two separate exposures allows DR200 (one stop difference) or DR400 (2 stops difference) only. DR800 and DR1600 are one and two stops of additional software DR and they are an ugly hack. Truly horrible smoothing of tones.

I agree with your last Bullet - Those are Typically Nasty Transitions and have only found ONE Scene Type that Shot Well with the Dedicated EXR : DR Mode ( Which Definitely Locks in "L" Size Pixel Sites abeit using (2) 8-Mp Image Planes of them.

Cheers from Orion

What this all amounts to is that you can set PASM with auto ISO 1600 or 3200 and shoot M size with DR400 and get about as good a result as possible under all conditions when shooting raw + jpg.

And note that just selecting DR100 does not happen, because that would require that either SN or HR mode be selected, since you have to choose whether to demosaic after binning, or without binning. DR200 and DR400 are effectively blending modes, where the brighter and the darker exposures are essentially blended on a curve to give you a high dynamic range image. This is essentially an HDR mode with two exposures separated by 1 or 2 stops.

If you contemplate what the camera is trying to accomplish and what it must do to accomplish that, then it all becomes pretty clear ...

--
http://kimletkeman.blogspot.com

Looky here, you are dead wrong. The camera selected both SN mode and DR400. You probably never tried EXR Auto, otherwise you would have seen it happen. In EXR auto SN mode camera has the privilege of using both pixel binning and tone mapping. This is the equivalent to using M size whenever ISO is higher then 400, the camera is pairing the pixels and using software DR enhancement. But the point of Orion (and I was trying to clarify) was that it is possible to have SN mode and DR higher then 100.

This Image Info Display Says it All. . . Excellent Work DS21

F200EXR

I was talking about how the hardware modes work. If you are shooting EXR AUTO and the camera selects SN mode and DR200 or DR400, then can you guess what is happening?

SN mode is DR100 in hardware.

So if it selects DR200 or DR400 along with SN then it is binning the pixels in hardware (which is a DR100 operation) and then applying software based DR.

Now Kim one has to be "Careful" with the "Binning" word as when this is done in Software verses PHYSICALLY Combining 2-Separate Pixel-Sites together the RESULTS are Quite Different. . .

No, one does not have to be careful with the binning word because that has been the whole point of the EXR technology from day one.

3 modes: SN is bin in hardware by combining two side by side pixels (much, much better than standard bayer binning), DR is blend in hardware by combining adjacent pixels at different exposures on a curve, and HR is use all pixels as individuals.

The following article does a good job at delineating the differences . . . http://lifeinmegapixels.com/blog/2012/01/pixel-binning-does-it-work/

As I said ... old school software binning is irrelevant since 2009 when EXR came out.

As I've said many times ... EXR AUTO is unpredictable and is to be avoided. At least by enthusiasts who have the ability to judge what they think is the most important goal in any one exposure.

I myself have had considerable Negative Reaction to this Strained-EXR-Implementation but My Reactions don't Change the Way those Fuji Engineers Implemented the EXR : SN + DR Functionality! So I've Simply Adopted MY Shooting Methods to be Able to Take Advantage of What I Personally Consider - The MOST Powerful Sensor Modes this Camera Possesses. . .

If you say so. I have simplified shooting this sensor down to settings that are very popular and used by many to get great captures. If you like SN mode plus software DR that is fine. I don't.

That's Not to Say I LIKE this Implementation and It's Certainly NOT how I would have done it BUT those Fuji Guys NEVER Asked Me - Maybe Next Time They WILL. . . lol

Thanks Again Kim for Your Inputs & Considerations

Many Cheers from Orion

--
http://kimletkeman.blogspot.com

More Cheers from Orion

--
http://kimletkeman.blogspot.com

 

[Orion12]

Looking at some tests I'm running this morning, DR400 seems to get automatically switched off as the ISO goes up. With the camera set to DR400 (6MP), I get double sized raws at iso 100-400 but they go to half size as soon as ISO gets to 800. This seems to happen automatically and without any indication to tell you it's happened (which is annoying!).

Is this well understood amongst the experts?

I'm not an X10 expert but have ran into a Similar Phenomenon on the HS50EXR. . .

At this point in time I usually don't pay much concern to Captured DR Levels as I routinely edit these relationships in RAW and use My Own Personal Visual Preferences for My Photographic Ruler. . .

Well, the DR setting also controls how the Fuji cameras expose to some extent. Highlights are protected much better in DR400 mode than in SN mode because the camera underexposes more in the former mode. I need to spend some time experimenting with this again to nail down the amount.

Occasionally I shoot in EXR : Auto Mode as this the ONLY Method of getting an SN Mode + DR200 or DR400 RAW capture with an HS50EXR. . . ( There is an EXR Mode Indicator when 1/2 Pressing )

I don't understand that comment.
You get SN (DR100) or DR200 or DR400 in any one shot.

Unfortunately the Dedicated EXR : SN Mode has a Fixed DR100 Attribute.

As I said ... this is what SN mode does by definition.

It appears you are making an "M" Size / EXR SN Mode Switching Assumption. . .
Remember that the EXR Sensors are Manufactured with 2 Equal Sized Image Planes. . .

I really have not forgotten. And I am not making an assumption. I am defining SN mode.

On the HS50EXR the 16-Mp Sensor "M" Size = 1/2 Size or a Single 8-Mp Image Plane. . .

I just defined it all above ...

IF You Shoot in RAW Only in PASM Modes you will Discover the Default Sensor Mode = 16-Mp

Yes. Raw only is HR mode. But as testing has shown, there is advantage in the raw files to setting DR400, which I always do. Not sure what Fuji does to get such latitude ... perhaps it is worth more testing to isolate that but I have seen the difference in tests.

IF You Shoot in RAW Only in EXR SN Mode you have a Default Sensor Mode Size = 8-Mp

Yes, binned in hardware is the default for SN, which is why the files are half sized.

When Shooting in PASM Fashion ( Which DOES Give You the DR100, DR200 & DR400 User Selectable Options ). . . The DEFAULT Sensor Mode - Appears to be the Full Frame Buffer Output made up of (2) 1/2 Resolution Image Planes and NOT EXR : SN Mode Twinned-Pixel Output. . .

PASM requires that the raw converter process the sensor output. But in JPEG it is for all practical purposes the same.

This imho is one of the Reasons we STILL get Arguments on PASM Mode "M" or "L" results as these are Actually Using the SAME Sensor Configuration albeit in an "M" you are getting one of the 1/2 Frame Outputs but also Have 1/2 of the Pixels to Process so the Camera has More Time to More Precisely Process them. . . But the Output is Still Not RADICALLY Different as the Same Pixel-Site Sizes are used in both. ( That's My Take )

I really don't think the camera tosses half the data in M size.

I Totally WISH Your Assumption was True and Was the Way the SN Sensor Mode is implemented but the Fuji EXIF Data ( Which Tags Specific EXR Sensor Modes ) doesn't indicate EXR : SN Sensor Mode usage in ANY Images - UNLESS they were shot in either Dialed EXR : SN mode or Dialed EXR : AUTO mode ( & Then Only when the Camera Selects SN DR100 default mode or SN mode + DR200 or DR400 ) Otherwise Fuji does Clearly EXIF tag this data stream.

Because it is not SN mode. SN mode is on the EXR dial. PASM mode simply shoots the same way without all the automation. At least, that's what the empirical results indicate.

I have compared the DR modes extensively and can see no difference in the output when shot in EXR DR or PASM DR400 ... I believe that SN modes look about the same too.

The raw file can be processed however you want to process it if you shoot L size, but the pixels are effectively shot at DR100 in L size.

In M size, you can choose to protect tones with DR400 or you can choose to try to maximize noise reduction with SN or PASM DR100,

I personally hold the EXR SN Mode in Highest Regards as you already know ( Which is a 8-Mp "M" Size Sensor Output - In Fact there is NO Possible Way to Shoot an SN Mode "L" size file PERIOD ) . . . ( I know YOU already Know this but some others may not )

I defined it above. And explained why I do not hold SN mode in high regard, as it requires shooting at DR100.

If you choose to shoot in EXR AUTO so you can allow the camera to make the choice of which software DR mode it throws on top then you are giving up far too much control for an enthusiast ... that's why I would never even test such a mode.

When I first got my HS50EXR I struggled to find the "BEST" Sensor Shooting Modes. . .
And AFTER More than 10,000 Shots - I've Landed on TWO Shooting Modalities that Serve ME the BEST - Those are. . .

1. Dialed EXR : SN Mode ( Semi-Automatic but with User Controlled Exposure, White Balance, Metering Method, Color, Sharpness ect ect ect )

2. Dialed EXR : AUTO Mode ( Full Auto Mode and the ONLY Method I know of to Get an SN Mode DR 200 or DR400 RAW File )

And I would never shoot those modes. After far more than 10,000 shots on just about every EXR camera that has graced the planet.

but anything else is mostly just dicking around to create the illusion of control.

I definitely agree with the above line. . .

The OTHER Method is getting SN Mode RAW captures is in EXR : SN Mode which has a FIXED DR100 Aspect.

Yes, it does. But what you describe above only serves to confuse the issues, since you are not separating what matters (dynamic range / noise) from the various ways in which the camera can be set up ...

I can only make this as Easy as It's come to me which Wasn't Totally Easy or Particularly Straight Forward and the Fuji Implementation of it's EXR Technology UNFORTUNATELY isn't a No-Brainer to Follow or Figure Out. . . I also have to Say It's Got Some Amazing Capability and I LOVE my HS50EXR - Quirks and All. . .

The HS50 is a fine camera ... but requires a lot of technique to shoot well at 1000mm ...

--
http://kimletkeman.blogspot.com

btw AFTER reading Your Comments --> I did some Additional Test Shooting on a Tuff DR Subject and Newly Discovered that I Could Get Some ADDITIONAL DR from my EXR : SN + DR400 RAW File. . . ( Beyond the DR Levels I WAS Previously Getting from my RAW Processing )

NOW this Even Further Pushes Me in the Direction of Switching to EXR : AUTO ( RAW ) when facing "Challenging" DR Subjects. . . As it's STILL the ONLY Method I KNOW of Getting an EXR : SN Mode DR200 or DR400 RAW Capture ! I really Wish their were OTHER Ways but I'm thankful anyway.

When shooting at high ISO, all modes are essentially done in software anyway ... I wonder if you would see any difference between fake AUTO SN/DR mode and any other mode at 1600 ISO once you run them through the raw converters ... and I certainly would not accept the loss of control by using an auto mode at high ISO ...

Ask Trevor G about that .... he's done the most testing on it I believe ...

These are from over 200ft away ...

1600 ISO ... RAW only ...

800 ISO ... RAW only ... cropped ...

Totally Cool Shots of Elton John ( Well Done ! ) but the Image Definition is a bit Under what I've been able to get from EXR : SN Mode RAW Caputes @ ISO-3200. . . The Lighting in Not the Same so it's not a No-Brainer or Absolute comparison. . .


ISO-3200 EXR : SN Mode RAW Processed in Photo Ninja converter. . . View in Original Size

The Pixel Definition Appears more Crisp in my EXR : SN Mode ISO-3200 Capture. . . That's My Take

More Cheers from Orion

The Above is a Much CLOSER Shot then Your "Elton John Shots" so that's also a factor. . .
The Following is Shot from about 1-Mile from the Low Light Subject but @ ISO-400 EXR SN Raw


Photo Ninja RAW Processed / Dialed EXR : SN Mode DR100 / ISO400 ( View Original Size )

I know You've Seen this One Before :-(

 

[Lisetta]

Question:

Who'll be the first to have one post take up a whole page?

But I'm not complaining! It's heaps of fun to scroll (repeatedly) through long cut and pastes of repeated passages of the same text and photos over and over with a few new things added in here or there.

Fun and interesting reading.

So worthwhile to scroll through.

 

[DickyAus]

Question:

Who'll be the first to have one post take up a whole page?

But I'm not complaining! It's heaps of fun to scroll (repeatedly) through long cut and pastes of repeated passages of the same text and photos over and over with a few new things added in here or there.

Fun and interesting reading.

So worthwhile to scroll through.

Oscar Wilde — 'Sarcasm is the lowest form of wit.'

 

[Andy Hewitt]

Question:

Who'll be the first to have one post take up a whole page?

Blimey, How big is your screen? I have a 27" iMac here, and it's got past a full page long ago.

But I'm not complaining! It's heaps of fun to scroll (repeatedly) through long cut and pastes of repeated passages of the same text and photos over and over with a few new things added in here or there.

Fun and interesting reading.

So worthwhile to scroll through.

LOL. FWIW, I was also about to post a comment about the rubbish editing and threading that people manage with their replies.

Also FWIW, it's a subject matter in itself, and has also been argued since the beginnings of network conversations.

 

[Andy Hewitt]

Perhaps someone can confirm this:

on the X10 in raw+jpg mode and size=M (6MP) when you set DR200 or 400 we know you get the HDR blended exposure effect. But what happens at DR100?

The logical thing to happen would be that you get the SN pixel binning effect as the sensor is no longer doing the HDR blending.

It would be useful if it did work like this as you could then reduce noise when you don't need the dynamic range boost. In effect, DR100 would be like having a 6MP sensor with the pixel size of the 1 inch sensor used by Nikon and Sony.

If it doesn't then, there is no point at all to the 6MP DR100 setting.

Anyone know for sure?

Know what, I think I'm beyond caring.

The biggest thing that is being ignored, is what people are happy with in the final result, in what they see in their final output medium.

What I see with my own eyes in front of me from my screen here, and the occasional print I do, as well as the few I publish on the web, is all that is important, as far as I'm concerned.

As I've said many times, for my own needs, and preferences, it seems to me that an L size Raw image produces the most pleasing results overall, on most circumstances. However, I do agree that some of the other EXR modes are beneficial in the scenes that make the best use of them.

I'm not sure if any of the recent threads have even helped anyone understand anymore about it either. With talk of pixel 'binning' - which is not really a grammatically correct term for what happens - ISO boosting, and various combinations of guesswork, no-one actually seems to have any 'facts', other than from their own testing. Of course extensive testing by a competent person should give us some good data to work from, and I suspect somewhere in all this the truth does lay.

I have to say that my own arguments against some of the 'facts' represented here have been born from scepticism by some methods used. For example, I see an awful lot of images posted for viewing here, and my own interest is in results from Raw files only. However, to be able to view any, they of course need to be exported for web viewing, which usually means a highly compressed, and scaled down version of the original.

If we're supposed to actually make proper judgements from those, then it's no wonder we have so many arguments. Surely those versions are not necessarily going to represent anything seen by the originating poster on their own screens, in their software of choice.

To be honest, much of this arguing is moot, unless you can replicate all the conditions equally for every viewer. Every graphics card will work slightly differently, as will every display - not many here mention anything about calibration, so arguing about colours or tonal quality is pointless.

I also see mention of up/down scaling images so they match. That is also another factor that can affect the image, and the quality of the algorithms used may well cause an image to be represented falsely. No-one mentions what settings, or methods are used for the scaling operation.

Ultimately though, and as I say, it's what you see in your end results that matter. If you're happy with them, then why not stop trying to fix something that isn't broken?

If you're trying to improve things, then the best thing to do is to go out an learn the capabilities of your equipment, and it's limitations, and also try to improve on your own abilities, rather than simply take a set of numbers you found on the web, and assume they are the best.

 

[Kim Letkeman]

MY RAW Shooting Tests as Above Certainly Have a Notable Difference in Clarity & Noise Levels. . .
And My Scaling Methods and NOT Treacherous to results One Way Or Another.

I'm starting to believe that BECAUSE you are so Adverse to the Horrible EXR SN Mode Shooting Implementation, that You've Maybe NEVER Seriously TESTED Actual HS50EXR -->

Quite right. I stopped after writing 19 articles on it.

Dedicated EXR SN MODE 12-Mb / 8-Mp Output RAW Files

Yes, this is how the camera works normally.

or
Dedicated EXR SN MODE 12-Mb / 8-Mp Output + DR200 or DR400 RAW Files

And how does that work again? SN mode allows you to set DR200 or DR400 directly? That I did not know. You said earlier that you got there using AUTO EXR, which I was able to duplicate on my F200EXR. That is not a "dedicated" mode.

After all the ONLY Way I Know of to Get Them are from those (2) "Amateur" Shooting Modes. . .

Maybe your use of "dedicated" is confusing me. If you select EXR SN, you cannot set DR as it defaults to 100.

If you set EXR AUTO then the camera is free to mix hardware binning and software DR in a melange of I don't know what ...

So which is it?

Have to Admit it's a "Tricky Challenge" to sort out what that Little Sensor is Doing at times. . .

I don't see why. If it says SN then it is binning. If it says that it is also shooting DR200 or DR400 then it is applying software DR. That's not complex.

I'll be doing MORE Testing also as this Inter-Active Discourse has raised a few Questions and Possibilities. . . Thanks again of ALL of Your Inputs & Considerations. . .

Cheers from Orion

SOOC ISO-800 Fuji Embedded Jpeg for EXR : SN Mode RAW Capture

SAME Shot RAW Processed in Photo Ninja + Further Tweaked in FastStone Viewer 5.0

Auditorium Capture from last weekend's the Deschutes County Expo Center - Home & Garden Show

--

Nothing Special

This blog is my personal space. Do not confuse this with statements or opinions on behalf of any employer I have ever had, have now, or will ever have in the future.

kimletkeman.blogspot.com

 

[Kim Letkeman]

but the Image Definition is a bit Under what I've been able to get from EXR : SN Mode RAW Caputes @ ISO-3200. . . The Lighting in Not the Same so it's not a No-Brainer or Absolute comparison. . .

If you already know that the comparison is pointless, why even write the words down?

ISO-3200 EXR : SN Mode RAW Processed in Photo Ninja converter. . . View in Original Size

The Pixel Definition Appears more Crisp in my EXR : SN Mode ISO-3200 Capture. . . That's My Take

I shot from 200+ feet away and cropped ... so how exactly are you drawing any conclusion at all?

And since mine are rendered images, how can you use words like "pixel definition?"

More Cheers from Orion

The Above is a Much CLOSER Shot then Your "Elton John Shots" so that's also a factor. . .

No kidding.

The Following is Shot from about 1-Mile from the Low Light Subject but @ ISO-400 EXR SN Raw

No idea what that shot is meant to convey? There is no dynamic range in that shot at all from what I can see.

Photo Ninja RAW Processed / Dialed EXR : SN Mode DR100 / ISO400 ( View Original Size )

I know You've Seen this One Before :-(

--

Nothing Special

This blog is my personal space. Do not confuse this with statements or opinions on behalf of any employer I have ever had, have now, or will ever have in the future.

kimletkeman.blogspot.com

 

[DS21]

Question:

Who'll be the first to have one post take up a whole page?

Blimey, How big is your screen? I have a 27" iMac here, and it's got past a full page long ago.

But I'm not complaining! It's heaps of fun to scroll (repeatedly) through long cut and pastes of repeated passages of the same text and photos over and over with a few new things added in here or there.

Fun and interesting reading.

So worthwhile to scroll through.

LOL. FWIW, I was also about to post a comment about the rubbish editing and threading that people manage with their replies.

Also FWIW, it's a subject matter in itself, and has also been argued since the beginnings of network conversations.