Digital Spot Meter

[bob carnie]

Please excuse my ignorance if I ask a silly question about camera spot meter.

Is there a medium format back*digital* that can take spot readings in L channel readings?
I am thinking a hand held device is not the ticket, but rather something that spot meters through the lens with very small apeture settings to look for significant highlight and significant shadow L channel numbers.

[williamtheis]

is this for film or do you intend to get a digital back for a medium format camera? i am confused.... for film, a hand held device like a pentax digital spotmeter is a wonderful tool and you just measure the significant highlight/shadow and everything else in between... reads out in "EV" same as the scale on the Hasselblad and other camers which can then be used with a dial to pick shutter speed/aperature combinations. nothing could be simpler. with black & white film, put the shadow EV you read on zone III and measure the highlight EV to see where it falls... if on Zone VI, use normal development; if on Zone VII use N-1; if on Zone V use N.... etc

exactly why do you need to read out in "L" channel and what would you do with the information if it did?

[tlitody]

You could look at the following to see if it serves your purpose:

http://www.konicaminolta.com/instruments/products/light/luminance-meter/ls100-ls110/index.html

The normal minolta spot meter and spot meter F both give luminance readings but you need to use a lookup table to convert to cd/m². They don't have all the features of the above link. Minolta meters have a tripod mount hole in their base.

But as I understand it, the L,a,b colour space is three dimensional and an L reading on its own is meaningless without the a and b components, so if you are thinking of the L from L,a,b then I don't think any of the above meters will do what you want but I could be wrong.

Don't know if any digital backs are able to use part of the sensor as a spot meter but since its digital all you need to do is take the picture and then look at a single point of the image to see its L component.

The following may do it:

http://www.konicaminolta.com/instruments/products/color-measurement/spectrophotometer/cm700d-600d/specifications.html

[bob carnie]

Ok I knew it would be a bit confusing.I am sure the following will not be any easier.

I use the L numbers for my density settings in PS to set my aim points for various print media. Actually I have set L aim points for all the different material I use and keep very defined notes on what papers, processes hold tonality.

I am more of a printer than a photographer, and I am hoping that someday I will have a medium format digital back , or Betterlight scanning back to capture images. I am hoping to transport L readings from lets say clouds, and deep shadows and make sure that I keep these number in capture and then apply them to PS and my prints.
My thought process is to do this through the camera/lens system rather than through an hand held digital spot meter. I also want to have LAB readings and I am not sure if there is such a device.


A few years back I had a discussion with Sandy King and he stated something like ** give me a file that has 100 shades, from white to black and I can make a digital file that works with any material.**
This has always stuck in the back of my head , much like the 10 zones of the system to place critical info in the highlights and shadows, I am curious of a metering system that works in a 0-100 density system much like the L channel in LAB.

When capturing info , placing critical info at specific L numbers in my mind would be the ultimate in Zone System capture with current equipment and my many printing processes.
For example my silver prints off a lambda have highlight with detail and shadow with detail L numbers of 4 for shadow and 96 for highlight, anything beyond this is specular or deep black.
But for inkjet photo rag these numbers change to 7 for shadow and 93 for highlight.

You are right a digital meter handheld pretty well does what I need, but I am thinking about applying exact density numbers for specific output at time of capture rather than a 10 zone system. This ability would allow me to set the points I require to capture for each process at finer increments much like expansion and contraction of film.

Am i over thinking this?? or should I just consider keeping a flat file and bumping contrast in PS much like what is done in scanning.

I am sure this is pretty confusing but any help would be appreciated.



is this for film or do you intend to get a digital back for a medium format camera? i am confused.... for film, a hand held device like a pentax digital spotmeter is a wonderful tool and you just measure the significant highlight/shadow and everything else in between... reads out in "EV" same as the scale on the Hasselblad and other camers which can then be used with a dial to pick shutter speed/aperature combinations. nothing could be simpler. with black & white film, put the shadow EV you read on zone III and measure the highlight EV to see where it falls... if on Zone VI, use normal development; if on Zone VII use N-1; if on Zone V use N.... etc

exactly why do you need to read out in "L" channel and what would you do with the information if it did?

[bob carnie]

I could not find the link

If I could take lab number readings at time of capture what a bonus. Knowing the three channels and biases would be awesome.

I really love working LAB for defining colour balance and setting my end points but since I am always at the end of the system, darkroom I have not done much research on meters and how they work within camera systems.
I would imagine if you can open PS very quickly like with the phase system or probably Betterlight you would then be able to see the channels , but I am specifically thinking of a metering system onboard that reads in LAB as you work and not having to open the image in PS to see your info pallette.



You could look at the following to see if it serves your purpose:

http://www.konicaminolta.com/instruments/products/light/luminance-meter/ls100-ls110/index.html

The normal minolta spot meter and spot meter F both give luminance readings but you need to use a lookup table to convert to cd/m². They don't have all the features of the above link. Minolta meters have a tripod mount hole in their base.

But as I understand it, the L,a,b colour space is three dimensional and an L reading on its own is meaningless without the a and b components, so if you are thinking of the L from L,a,b then I don't think any of the above meters will do what you want but I could be wrong.

Don't know if any digital backs are able to use part of the sensor as a spot meter but since its digital all you need to do is take the picture and then look at a single point of the image to see its L component.

The following may do it:

http://www.konicaminolta.com/instruments/products/color-measurement/spectrophotometer/cm700d-600d/specifications.html

[tlitody]

this unit works in L,a,b

http://www.konicaminolta.com/instruments/products/color-measurement/spectrophotometer/cm700d-600d/specifications.html

but you have to put it against the subject. I'm not sure what you are metering. A print or something else? Are you thinking that you can tell your customers what L numbers to use so you can match them in printing or what?

[bob carnie]

LAB is not a colour space a lot of workers are comfortable in. Some of my clients get it and others would think I am talking in my mother tongue Moronica.
But I do tell them now to read the L channel in PS and make sure they are not exceeding the end points with significant shadow or highlight.
I am wondering if I can apply this logic at time of original capture.

I am referring to being able when photographing , metering significant highlight and significant shadow and applying L density numbers to them . This could mean highlights on skin, highlights on clouds, shadows in deep shade .
Kind of like tailor making exposure at time of capture to minimize PS manipulation to solve bad capture.

by putting the end points where I want them then all the information would fall between, which is easy to manipulate.

This is something I think about when not working at my job , which is kind of pitiful, but whenever I am outside I am trying to guess the L values of different objects.
I need a life.
thanks for the link btw




this unit works in L,a,b

http://www.konicaminolta.com/instruments/products/color-measurement/spectrophotometer/cm700d-600d/specifications.html

but you have to put it against the subject. I'm not sure what you are metering. A print or something else? Are you thinking that you can tell your customers what L numbers to use so you can match them in printing or what?

[tlitody]

If the photographer is using a tethered laptop then they see the image immediately and check and correct lighting/exposure accordingly. So I don't see that checking L,a,b values is a problem unless they don't have a tethered system. A lot of studios use tethered to get perfect exposure. A lot of location photographers use tethered too. So telling them to work to high and low points shouldn't be a problem except that what they see on a tethered system may not be what you can get out of the image with raw conversion.

[tlitody]

what you are suggesting is that the photographer should be able to adjust levels in camera. i.e. adjust the dynamic range of the camera. B+W film users do this via ISO and development controls. Colour film users do it to a much lesser degree. I don't know if digital backs can do it. But that is why people chimp at histograms, to check if everything is in range of the sensor.

[bob carnie]

Yes, but rather than looking at histograms, I would want a numbering system where I can place siginficant tones to a number that I would then know transfers to my end output.
There probably is such a system but I just haven't looked hard enough.
There is a phase here and my business partner uses it tethered to a laptop but I do not think he can quickly get L numbers while shooting.

I am going to go on a bit of a tangent here but here goes.

What do sky's look light , is there areas of really non information, is the bottom of clouds a different colour and monster difference in density, is there blue in deep shadows , How do we percieve nature and then transfer to specific looks on paper.

I recently have started taking photographs a bit more outdoors, and being a printer, I look at densitys and colours from a finished print perspective. I need to see the results.
Years ago I met and saw a grouping of photographs by Gordon Hutchings at Eight Elm, I believe Richard K from this forum brought him to Canada.

At the time I thought his prints were way to bright, over the top in fact, but was grudgingly acceptance of his remarkable highlight detail.
Les Mclean from England makes very beautiful prints and one of his specialtys was holding highlight detail.
Many Many years ago I was a maskmaker for Cibachrome prints and we would not only make contrast reducing masks, but make highlight protection masks that then would be used with the transparancy to keep critical detail in the upper highlights when making the contrast reducing Mask.


All the above are using different methods to achieve somewhat the same thing, Pyro Hardening developer which limits highlight migration or blooming while the neg develops out.
Split contrast printing with using 0 and 5 as a team on highlights to create separation.
and Mask making to protect delicate details that either get blown out with exposure or dulled down with contrast reducing mask.

Now after about 5 years of intense PS learning curve , I am wondering is there not a way digitally to set points exactly where you want.
How does one squeeze out precision detail at both end points?
I am thinking an onboard densitomether on the camera that speaks the same language that PS and final printers understand.

If this sounds crazy, go out look at the sky, you will be amazed that there are points within the clouds that acutally record 0 information, and by printing this area down to detail may be a mistake.
same goes for deep shadows.
Our eyes adjust for these differences and a way to control them would be awesome.

I think Mr Adams and others were visionary with the Zone system and if there system worked in a 0-100 world, why would we want to work in 0-255 RGB but rather a density system much more like the ZS .

I know this may show my limited experience behind the camera , but enquiring minds need to know.





what you are suggesting is that the photographer should be able to adjust levels in camera. i.e. adjust the dynamic range of the camera. B+W film users do this via ISO and development controls. Colour film users do it to a much lesser degree. I don't know if digital backs can do it. But that is why people chimp at histograms, to check if everything is in range of the sensor.

[williamtheis]

here goes a rant: I use a Betterlight (BL) back and I use film with a spotmeter. When taking a picture, you need to set the exposure: shutter speed (line capture time on BL) and the f/stop on the lens. If you knew L, you still don't know how to set these values in the field. the usual number used is the "EV" or exposure value that is translated into these quantities and you usually measure such that you don't block up either the high or low values (true for film, true for BL) unless your'e on of them digital HDR guys.... the fact that the Minolta will display L does not mean it is useful in the field even with a tethered Laptop like I have with the BL since the time to open an image and convert to LAB (YES the CAPTURE is RGB, just like film... it does not natively capture in LAB--this is true of conventional dSLRs as well) takes so long that knowing the values is not truly useful. Theoretically, you could analyze the L values to see that something is not blocked up and adjust the shutter speed/capture time or the f/stop to refine it but you can do the same with the EV scale... just MHO

[bob carnie]

I am having my business partner look into this on the phase, You are absolutely right , he knows he can give me RGB values, and he will see how long it takes to covert to LAB to see the numbers... Probably the time it takes is not practical.
I am interested in the Betterlight as I think it has many attributes that are compelling to me.

I am moving into a whole new area of print making where each process will have aimpoint specifics , whether RGB, Greyscale or in my preferred LAB.
most of my output require RGB but I look at the L numbers for working in PS.

I would love to be able to purchase some type of high end back that allows me to see the numbers from beginning to end.

Please excuse my ignorance , but how exactly does the EVscale spot meter work , will it give you readings in a 0-100 scale??


here goes a rant: I use a Betterlight (BL) back and I use film with a spotmeter. When taking a picture, you need to set the exposure: shutter speed (line capture time on BL) and the f/stop on the lens. If you knew L, you still don't know how to set these values in the field. the usual number used is the "EV" or exposure value that is translated into these quantities and you usually measure such that you don't block up either the high or low values (true for film, true for BL) unless your'e on of them digital HDR guys.... the fact that the Minolta will display L does not mean it is useful in the field even with a tethered Laptop like I have with the BL since the time to open an image and convert to LAB (YES the CAPTURE is RGB, just like film... it does not natively capture in LAB--this is true of conventional dSLRs as well) takes so long that knowing the values is not truly useful. Theoretically, you could analyze the L values to see that something is not blocked up and adjust the shutter speed/capture time or the f/stop to refine it but you can do the same with the EV scale... just MHO

[tlitody]

I know you want to control the L value but apertures only go in one third stop and time in one third stop increments. When camera is in auto mode you may be able to control it finer but I'm not sure about digital backs. The lens is not always connected for control by camera back settings. What I'm saying is that whilst you have very fine aim points, the limitations of setting lens and shutter controls may not allow it. Therefore controls as they already exist have to be lived with. You can probably adjust exposure to be within limits but that is no more than you can already do if you know what you are doing with a standard spot meter.

[bob carnie]

Thanks

I know my ignorance in Spot meters is showing, I am looking for 1point accuracy within a 0-100 scale.

bob

I know you want to control the L value but apertures only go in one third stop and time in one third stop increments. When camera is in auto mode you may be able to control it finer but I'm not sure about digital backs. The lens is not always connected for control by camera back settings. What I'm saying is that whilst you have very fine aim points, the limitations of setting lens and shutter controls may not allow it. Therefore controls as they already exist have to be lived with. You can probably adjust exposure to be within limits but that is no more than you can already do if you know what you are doing with a standard spot meter.

[williamtheis]

1/3 stop on f/stop, factors of two (1/25, 1/50, 1/100, ...) on the shutter for film, slightly better for the betterlight since scan times are 1/8 sec, 1/10, 1/15, ...

nothing approaches 1 part in 100 and it is not necessary to get the capture to this accuracy IMHO. you should do setting of final endpoints in CMYK or RGB but LAB is a little too brute force. (yes, I am VERY VERY VERY familiar with LAB having taking Dan Margulis's classes)

EVs are for 0 to about 20, each represent a doubling but spot meters conveniently read in 1/3 EV (so 2 1/3 is 1/3 stop more than 2) which matches lenses but not shutters

[bob carnie]

I know I am nitpicking here and asking for something that is probably not required.
I am just concerned about the top five shades of highlight detail and the bottom five shades of shadow detail.
I have never used a spot meter and do not own a camera with a meter , other than the phase that is available to me , which btw I have no idea how to work.
So taking my preference of LAB numbers out of the equation, my probable naive goal is to find a system that can give me exacting measurments that I can then transfer to print.
Thanks for your explanation on EVs.

1/3 stop on f/stop, factors of two (1/25, 1/50, 1/100, ...) on the shutter for film, slightly better for the betterlight since scan times are 1/8 sec, 1/10, 1/15, ...

nothing approaches 1 part in 100 and it is not necessary to get the capture to this accuracy IMHO. you should do setting of final endpoints in CMYK or RGB but LAB is a little too brute force. (yes, I am VERY VERY VERY familiar with LAB having taking Dan Margulis's classes)

EVs are for 0 to about 20, each represent a doubling but spot meters conveniently read in 1/3 EV (so 2 1/3 is 1/3 stop more than 2) which matches lenses but not shutters

[tlitody]

I know I am nitpicking here and asking for something that is probably not required.
I am just concerned about the top five shades of highlight detail and the bottom five shades of shadow detail.
I have never used a spot meter and do not own a camera with a meter , other than the phase that is available to me , which btw I have no idea how to work.
So taking my preference of LAB numbers out of the equation, my probable naive goal is to find a system that can give me exacting measurments that I can then transfer to print.
Thanks for your explanation on EVs.

Basically the zone system or variant will do it. What you do is find a flat evenly illuminated surface which is a mid grey. Doesn't have to be exactly 18% but just middleish grey. Could be another middling colour. Nothing to vibrant or dark.
Align camera with it and focus. 10ft away should be plenty. Then standing behind camera point spot meter at centre of surface and take a reading. If reading says for example 125th sec at f8, then set digital camera to 7 stops less exposure than that. So that would be 1000th at f32. make image at that exposure. Then open up 1/3rd stop and make image. Repeat another 7 times. Then set exposure to intial reading + 7 stops and make exposure. So that would be 1/8th at f2.8. Then close down 1/3 stop and make image. Repeat another 7 times.
Now you have 18 images. open in PS and check which of the dark ones is the first one with an L value higher than your low limit. You can then work out how many stops that is from the metered reading. So it may be 5 1/3rd stops for example. You now know that whatever you meter that you can't close down more than 5 1/3 stops without it going out of range. You then find from the light images which is the first image which is within your range. Say that is also 5 1/3rd stops more exposure than your metered reading for example. So now you know that whatever you meter, you can't open up more than 5 1/3rd stops without it going out of range.
You also now know that you have 10 2/3 stops useable dynamic range with detail if your end points in printing show detail at those high and low values.
You should then always be within a 1/3 stop of your high or low value. And given that 10 2/3 stops is 32/3 and each 1/3 is approx 3% of that, then you should always be between 0 and 3 points of hitting either your high or low target values which is pretty darn close for most practical purposes. And if you take the mean, then on average you will be within 1.5 points of hitting your target values. If that is not close enough for you then I think you will have to do your photography in Lab conditions (pun intended).

It then becomes easy to check if you are in range. You may have to change levels a tad in PS but thats OK. You can also make decisions where in the dynamic range to put the exposure if the subject is say only 6 stops in range. You can pick higher key or lower key in the knowledge that you are staying within your limits.

essentially all you have done is to check the dynamic range of your camera sensor and provided yourself with limits on playing with exposure. Basically thats zone system metering only in digital you don't control it through film speed and development like you do with film. Call it the zone system for digital.

Note I have assumed the dynamic range of your camera is at least 10 2/3rd stops. It may be less in which case you should do the repeats an additional 3 times at each end and that would give values for a sensor with only 8 2/3 stops dynamic range.
Also the lighting and camera position must constant through the test.
And also if you use daylight for the test then the test is only valid for daylight. If you use tungsten then its valid only for tungsten etc. So don't use strobes for the test if you are going to be shooting in daylight.

So then if you are photographing storm clouds you just meter the absolute lightest part of the clouds and open up 5 1/3 stops from that reading and expose. Of course check that deepest shadow is not more 10 2/3 stops less than metered reading of clouds and if it isn't then all will be in range and you can adjust shadows with levels to get to your low point of 4.
A digital spot meter can tell you this. The minolta spot meter or spotmeter F allows you to store the initial reading and then scan the subject telling you the number of stops difference from the original reading in 1/10th stop increments. i.e. it is very accurate.

[bob carnie]

Wow, I am impressed
thank you for that thats a lot of info to digest. I made ring arounds in College that were of the same scope, I am going to ask for a spot meter for christmas.

[tlitody]

Wow, I am impressed
thank you for that thats a lot of info to digest. I made ring arounds in College that were of the same scope, I am going to ask for a spot meter for christmas.

Yes but you only have to do this once for any lighting type and then you have the limits of you sensor. Not sure what you will find.

[D. Bryant]

A digital spot meter can tell you this. The minolta spot meter or spotmeter F allows you to store the initial reading and then scan the subject telling you the number of stops difference from the original reading in 1/10th stop increments. i.e. it is very accurate.

Minolta meters are no longer manufactured though Kenko purchased the rights to that product line and markets meters under their name.

I prefer Sekonic meters though I still use my Minolta Flash Meter II for B&W work; it is durable, very feature rich and relatively inexpensive on the used market.

As for connecting LAB values to exposure meter readings - I just don't see the point. To each their own. Once you can predict how and how well a sensor will record light, metering becomes straight forward.

Do digital backs use LAB color space? Perhaps I'm just ignorant. Since I primarily use ACR/Lightroom3 for RAW processing I have adopted xRite ColorChecker to calibrate my cameras and as a WB tool. I wish Capture One Pro would support DNG profiles for that reason. I think C1P is perhaps the best image processing utility around, particularly for RAW processing. I know from experience the output scaling works better than Adobe's for 10 to 12x enlargements (don't know about bigger personally but I've heard that it is awesome.)

LAB in post can be very powerful but there are certainly split camps out there about it's utility as a way of living. :)

Don Bryant

[bob carnie]

I know there are a lot of folks not liking Lab and consider it a sledgehammer when all that really is needed is a ball peen hammer.

I have been using it for a few years now , and FWIW feel having the info pallette set to LAB is the only way to go, as it allows me to see the underlying densitys of an image and as well the A B numbers are second nature to me and so easy to work with. When in RGB I am looking at the L numbers for my reference, and I do a lot of work in RGB and there lies some of the confusion. The L numbers for me are intuitive, the RGB and CMYK numbers , do not translate well in my brain.
I spent many hours using dichroic filters on many colour enlargers and working with A B is second nature.


There are simple corrections that just make sense on the A B curves that may seem extremely difficult , but I happen to find them sensible to my workflow.

Since I do no work currently to Off set Press , rather than my own RGB printers, I rarely delve into CMYK, but some of the new projects I am working with primaraly multiple colour gum I think I will. Since I will not need to worry about ink limits on press I doubt I will be working with UCR and GCR.

so for now I also sharpen on the L channel with DMargulis different suggestions. I have tested the RGB fade to luminosity, High Pass , but found for my work I like to sharpen in L.

I am really only using LAB mode for sharpening , contrast, and some colour enhancements. For my work I start and finish in RGB always.

In the future for the Colour Gum I may find myself only in CMYK , as I cannot imagine the colour gamut of tri colour gum on paper exceeding cmyk's capabilitys.

My quest for a metering system is a bit of pipe dreaming, and trying to get a bigger grasp on my own photography.

[D. Bryant]

[QUOTE]In the future for the Colour Gum I may find myself only in CMYK , as I cannot imagine the colour gamut of tri colour gum on paper exceeding cmyk's capabilitys.

Capture in RGB. Edit in LAB if you wish but don't edit in CMYK for tri-color gum. Convert to CMYK or CMY for seps.


My quest for a metering system is a bit of pipe dreaming, and trying to get a bigger grasp on my own photography.

There is really no need to do it that way. It just muddies the water.

[bob carnie]

Hi Don

Why not edit in CMYK?
I understand the colour space is smaller but I would imagine the Gum process could fit nicely.
GCR and UCR would not be an issue since I am not worring about ink limit.
The K channel has wonderful sharpening and secondary mask potential

other than the smaller colour spaces what would be your concerns.

Bob

[QUOTE=bob carnie;612515]



Capture in RGB. Edit in LAB if you wish but don't edit in CMYK for tri-color gum. Convert to CMYK or CMY for seps.



There is really no need to do it that way. It just muddies the water.

[D. Bryant]

Why not edit in CMYK?
I understand the colour space is smaller but I would imagine the Gum process could fit nicely.
GCR and UCR would not be an issue since I am not worring about ink limit.
The K channel has wonderful sharpening and secondary mask potential

other than the smaller colour spaces what would be your concerns.




Bob,

1) I'm no expert but I'm a believer in the late Bruce Fraser writings. He advised to edit in RGB as long as possible until you need to convert to CMY or CMYK. Once you convert you loose data from your original that makes editing color less precise and dicey.

2) Learn to print gum before you jump off the deep end of the pool. You are going to be surprised about what you don't know about gum printing until you have some practice under your belt. All that mumbo jumbo about GCR and UCR will be forgotten quickly until you gain skills with gum printing. We don't have the control of ink from press to paper with gum.

3) Get a good ICC profile from your gum print and see where you stand when you print an ICC test palette in gum from RGB converted to CMY or CMYK.

4) If you profile in CMYK then there is an automatic dot gain compensation going from RGB to CMYK. Handling out of gamut color in CMYK can be very tricky from what I understand. Profiles built for CMYK have a lot more limitations than RGB.

So pick your poison.:)

Don

[bob carnie]

Thanks Don

I am always overthinking , but I have some tricks up my sleeve that makes working in all Modes exciting.

btw I will be at Sandys for the last week Sept, give take a couple of days, then off to Louisville to see some gallerys.
I also signed up for the Cristina Anderson tri colour gum at Project Basho in Nov.
My team is put together now and we are putting a forum where I want you to be an advisor. Four other from my team are going to Project Basho with me so by Nov we will start the learning curve of laying gum on paper.

love to see you at Sandy's and see some of your prints live.

regards
Bob
Bob

Bob,

1) I'm no expert but I'm a believer in the late Bruce Fraser writings. He advised to edit in RGB as long as possible until you need to convert to CMY or CMYK. Once you convert you loose data from your original that makes editing color less precise and dicey.

2) Learn to print gum before you jump off the deep end of the pool. You are going to be surprised about what you don't know about gum printing until you have some practice under your belt. All that mumbo jumbo about GCR and UCR will be forgotten quickly until you gain skills with gum printing. We don't have the control of ink from press to paper with gum.

3) Get a good ICC profile from your gum print and see where you stand when you print an ICC test palette in gum from RGB converted to CMY or CMYK.

4) If you profile in CMYK then there is an automatic dot gain compensation going from RGB to CMYK. Handling out of gamut color in CMYK can be very tricky from what I understand. Profiles built for CMYK have a lot more limitations than RGB.

So pick your poison.:)

Don