A900 DRO ???

[Dr. Harout]

I am not an expert in this, but I see some differences in the ACR shots you submitted Don. I mean in the JPEG shots the temperature and tint are set to '0' as for the RAW files they are set to different values. Even in the lower part there are different values (look at brightness, contrast, black...). Shouldn't you put the values to the same size for both shots?

[Dr. Harout]

Any other questions, Doc?

-Sonolta

Not for now.

[David Kilpatrick]

Don, you did not set -0.6 for the normal high ISO tests - all your shots show 1/10th at f9 as the exposure. That means the Hi200 shot received 0.6 stops less than the normal 200 (I may test this properly to see whether it is actually 0.6 or whether it is 1 stop - it looks like 0.6 on the ACR histogram). Secondly, you shot the test in a non-native WB condition, which means that the channels are not balanced, and your 'highlight' is green, which means that highlight clipping can not be assessed. To do this properly you need 5600K conditions, and a white highlight zone (like a wedding dress, which is what Hi200 was originally intended to deal with), and a short bracket series at 0.3 intervals with absolutely no shift in conditions or composition. Even then because Hi/Lo access an ISO setting which can't be set manually comparisons are difficult.

The Hi200 setting should - as the range curves you have posted elsewhere show - give identical shadow-midtone density detail to a normal exposure, maybe just a hint darker, in JPEG, but with much improved highlight density. In RAW. the shadow detail will be significantly darker, as it indeed is in your samples. I would point out that ACR's default transfer curve is much straighter and uses less of an S than the in-camera JPEG, so processing any raw file using ACR will tend to be much closer to either the Hi or Lo setting results, especially if you use Linear or Medium Contrast. To come close to an in-camera Standard look, 0 contrast adjustment result the Strong contrast curve should be used in ACR with a blackpoint adjustment.

David

[01af]

This is either something you understand at a glance [...] or you need a thousand word primer, and I really have no time to write that, nor any need as most people reading this will I am sure be able to understand it perfectly.

I am afraid even a thousand words won't do. In fact, you did write more than a thousand words on this topic after Sep 30th, 21:31 h---and he still doesn't get it. Obviously he doesn't understand how exposure and film speed (or sensor speed) interact. He also doesn't understand the difference between changing the meter's speed setting and changing the the film's (or sensor's) actual speed.

You are wasting my time, and everyone else's ...

Not the time of those who are able to make proper use of this forum's ignore list feature

-- Olaf

[Greg Beetham]

Hmmmm..a rug? I thought for a moment it might have been a cat.....persian 98 maybe.
Greg

[Greg Beetham]

David,
being semi-serious for a second, what you're saying is that Hi200 is approximate too taking an average exposure of a given scene/composition and then under exposing 2/3rd stop in normal ISO200, preferably with correct WB setting?
Greg

[Greg Beetham]

Don,
what would be the point of reducing the exposure in ACR on the normal ISO200 shot? the horse has bolted, the highlights are already overexposed by comparison to the Hi200 shot, you'd have to underexpose the normal ISO200 shot with the camera too preserve the highlights.
Greg

[David Kilpatrick]

Don,
what would be the point of reducing the exposure in ACR on the normal ISO200 shot? the horse has bolted, the highlights are already overexposed by comparison to the Hi200 shot, you'd have to underexpose the normal ISO200 shot with the camera too preserve the highlights.
Greg

That's what the Hi200 setting does. It leaves the camera metering set to ISO 200, but the sensor is set to ISO 125 (you never see 125 reported, it says 200, but the actual sensitivity is lowered). The result is a raw file which is underexposed by 0.6 stops. A special raw to JPEG conversion curve - rather like using highlight recovery in ACR plus some Brightness increase - is used to produce an in-camera JPEG with minimal shadow detail loss if any, and much improved highlight detail.

A similar process happens for Lo80, with the sensor again set to 125, but an exposure given which is correct for 80. This results in a raw file where there is less highlight headroom (the usual leeway has been used by the Lo80 setting) but has considerable better shadow and three-quarter tone detail, which a custom raw-to-JPEG conversion curve is again able to render optimally.

David

[David Kilpatrick]

LOL...Now you are dogging AWB? Get a clue! Those are AWB balanced shots!

Don does not understand AWB either. He is apparently unaware that white balance adjustments have no effect on raw file data, only on in-camera conversions and the tags used by ACR or other conversion programs to assist with raw conversion. There is no such as an 'AWB balanced' shot - that would call for an adjusted channel gain on the output from the sensor. That does not happen, the raw file remains the same for a tungsten shot as for a daylight shot - which is useful, as you can correct WB errors in raw conversion.

If you want the very best tungsten light digital shots, use an 80B blue filter just the same way you used to with film, and daylight WB. You gain about one stop in highlight clipping headroom, and reduce blue channel chroma noise in the shadows.

David

[maratrs13]

If you download all four of the files and tweak them for highlight recovery and/or optimum IQ it will become quite clear to everyone that ZONE MATCHING is indeed working terrifically on both JPG and MRW files. You will find that both of the ISO200 samples (JPG and Raw) are blown to bits and they are not recoverable by any means. When you tweak the HI200 files, you will find out that you can recover almost all of the highlight details on both the MRW and the JPG file. In-fact, you will find the two zone-matched files to be very similar at the end of the day.

This test again very clearly shows that KM style Zone Matching can be a major benefit when shooting both raw and jpg files.

-Sonolta

Don, sorry for intervening into your rather hot discussion with DK but it seems to me that you are missing his point a bit - he tells that the ZM works only with JPEGs and that's true! If you'd be able to set the KM5 to ISO 125 and apply -2/3 compensation you would get exactly the same .MRW file as you got with HI200! The fact that the camera reports ISO 200 in exif data for HI200 file only means that the metering system used this value for EV calculation, but real sensor sensitivity was set to ISO 125 and more than that - the exposure was biased by -2/3 stop. These alterations are not written in the exif data but that is the way how it works in the camera. Other than that nothing special is done with .MRW file. And only because of the exposure bias you get enough headroom in the highlights area. After that a special curve is applied to .MRW file in order to lighten up dark and midtones but preserve highlights. Unfortunately, you cannot set the ISO to 125 on KM5 though it is possible to set -2/3 stop compensation (therefore you won't be able to get identical .MRW files for HI200 setting and normal ISO setting). In this sense I would agree with you that the ZM affects .MRW files but yet the ZM is designed primarily for JPEGs.

BTW, quite the opposite happens with LO80 setting - the sensor sensitivity is set to ISO 125 and +2/3 exposure compensation is applied, but the camera still reports ISO 80 without any compensation in exif data. The result is much brighter .MRW file which is supposed to bring out shadow details. Then a curve is applied to the raw file which darkens midtones and highlights a bit but preserves shadows at the same level. Again, all this is applied to JPEG file, not to .MRW! In both cases .MRW files undergo no changes but exposure compensations but JPEGs are processed in special ways compared to regular processing.

Marat.

[[SiC]]

it is IMPOSSIBLE to set ISO 125!

Yeah, manually - but you could do it in FW, right?

BR,
/Zeb!

[Greg Beetham]

As far as I know, the ISO settings correspond with amplifier gain on the sensor from low to high gain in distinct steps, (controlled by the user or by Auto in the camera), ie 100, 200, 400 etc. There is also exposure bias + or - in either 1/2 or 1/3 stops/steps. You could argue that these represent nothing more than fine graduations of amp gain as well.
Sooo it then becomes possible too set close-ish to ISO125 using these fractional stops, it would most likely be closer to ISO133.....but hey, good enough for the bush......or as in this current discussion, for a rough comparison between Hi200, and ISO200 biased with -2/3 stop.
Greg

[maratrs13]

You CAN NOT set ISO 125 it is IMPOSSIBLE! That is why Zone matching is so great!

I know that I cannot. But you can easily set the camera ISO to 100 and set -2/3 compensation and get almost the same .MRW file. But in this case (we are talking about regular ISO setting, no HI or LO) you'll get underexposed JPEG file out of the camera (if you shoot RAW+JPEG). It means that no curve is being applied to the final JPEG image while in either HI or LO setting it is (though different in each case). But if you open the MRW file in ACR and apply a curve which lightens up midtones and shadows but does little to the highlights you'll get practically the same result as if you were shooting HI200.


Again, look at the results and understand the fact that ISO125 is IMPOSSIBLE to set. Please remember also that ZM can give -6 to 4.5EV which is ACR Max numbers.

The same can be done with any MRW file if it is underexposed/overexposed properly. There is no reason why you cannot achieve ZM results with other ISO setting - it's a matter of postprocessing only. ZM is designed for convenience purpose only so you don't have to bother with curves later. But it does not alter the MRW file in a special way, it just applies appropriate compensation and writes ISO used in the EV calculation in each case (HI or LO).


Look at the files and understand the fact that you CAN NOT set 125!

Once again - I know that. It seems that you are trying to tell me the thing that I know but for some strange reason you think that I don't understand it!


Again. Something special is done, it is minimum a 125 -EV exposure and this is not a setting that can be set any other way than by using Zone Matching!

Actually, you can achieve the same result if you take a shot at any ISO with -2/3 compensation and apply the curve described above in the postprocessing - nothing special. There is nothing magic in the ZM itself, it's there for your convenience only!

LOL...again you can not set 125 in camera, you need to use ZM.
If I need the same result but do not have ZM in my camera (though I have it) I could do it by applying 2/3 negative compensation and apply the curve in PS. I don't need to set the camera to ISO 125 in order to achieve the same result.

Show me how to set 125 on my camera! Better yet, if you have KM style zone matching try it out yourseld on some high key stuff!

I tried it and it works. But you can easily DIY if you need without the ZM - it's a matter of some postprocessing.


FINALLY! Reality sets in! You need to use ZM to get the benefit of 125!
What is so special about ISO 125 besides that it is set in ZM only?

FINALLY...and all you need to do is tweak the examples I have provided to see how much more highlight detail is preserved both raw and jpg!

Sure they are preserved because the shot is deliberately underexposed!

LOL...I know exactly what Lo80 does...geesh, I am the guy that actually uses the ZM settings!
Good for you! If you use it and is satisfied with the result - that's good!

Again, you don't need to explain ZM to me as I have used it in my real world shooting for the past three years. I am the one that posted the KM explanations, I am the one that posted the DR numbers, I am the one that has shown the additional highlight retention, and I am the one that has been telling everyone here that you can not set ISO 125 on the camera...it is impossible!
Again, what is so special in ISO 125? Does it have some magic in itself?


Thanks man, but the samples clearly show the hi200 raw has probably the most preserved detail, followed by the hi200 jpg, and both of those blow the ISO200 shot to the moon. You can not dial back -o.6 from the raw and recover the highlights and this situation is a very similar scenario as shooting flowers on a bright Sunny day.
Sure you cannot dial back -2/3 compensation and recover lost highlights, but you can do it *DURING" the shot and then recover midtones and shadows to the appropriate level!


Bottom line is ZM can give top DR at both ends of the spectrum and you can not set ISO125 in camera! All of the arguments saying it is the "same thing as" are moot since it is IMPOSSIBLE to set ISO 125!
For the third time - what is so special about ISO 125?

Marat.

[Greg Beetham]

Or, putting it another way,
when you select Hi200, the camera sets an exposure for ISO200, BUT reduces the gain on the sensor back to ISO125, and as Sonolta says is not accessible by the photographer as we only have 1/3rd steps available not 1/4 steps, but -2/3 bias on ISO200 is reasonably close enough to see if Hi200 is something "special" or not in post PP....surely.
Greg

[[SiC]]

Or, putting it another way,
when you select Hi200, the camera sets an exposure for ISO200, BUT reduces the gain on the sensor back to ISO125, and as Sonolta says is not accessible by the photographer as we only have 1/3rd steps available not 1/4 steps, but -2/3 bias on ISO200 is reasonably close enough to see if Hi200 is something "special" or not in post PP....surely.
Greg

So we're back to what DK is saying..
But according to Don this seems like magic and cannot be achieved even by the FW
Who knows for sure?

/Z!