Editing for B & W

[Shadowman]

This is a crop of an image that I used to test the B & W capability of my printer (hp 7960). Images looked fine on my monitor but dark gray values were coming out pitch black. I tried different editing modes and Elements Smart Tone feature offered the best output, however I think the image still needs some tweaking.

1/1000s, ISO 160, f/5.3

[Polar01]

John when I go to print an image and if it is darker than my on screen image the first thing I look at is my monitor too bright. If it is a LCD type maybe lowering the output to 100 than the 120 that is recommended maybe a place to start.

Cheers: Allan

[Shadowman]

John when I go to print an image and if it is darker than my on screen image the first thing I look at is my monitor too bright. If it is a LCD type maybe lowering the output to 100 than the 120 that is recommended maybe a place to start.

Cheers: Allan

Hi Allan,

I usually start with the captured image onscreen and if I can see details in the darks or good variations within highlights/shadows I feel I'm on the right track. Those good indicators are usually seen in textures; I'm working my way up to skin tones. I have a ColorMunki display device on my monitor so I think the brightness is fine, however I did just read about ColorMunki Printer device and wonder if that would help, the device costs about $500 so I'm looking for alternatives for now. The ColorMunki Printer device supposedly helps with B & W printing output. .

[Manfred M]

John - +1 to what Allan has written.

I have found exactly the same as he has and I have determined (through making test prints; for both colour and B&W) that I have to lighten the image on my screen a fair bit to get a print that looks right. The good news is that once you determine the amount you can just remember the value and keep using the same adjustment over and over. The only time you will have to repeat the test prints is if you make changes to the brightness setting of your computer screen.

As for the ColorMunki print, it is for profiling papers and inks (instead of using the paper manufacturer's ICC profile) in addition to ensuring that the colours on your computer screen are accurate. I don't know if it will calibrate the two operations. On the other hand, doing test prints is so much less expensive.

I guess coming from the colour and B&W darkroom, I don't have any issues with doing test prints, as they were always the way we worked in those days.

[Shadowman]

John - +1 to what Allan has written.

I have found exactly the same as he has and I have determined (through making test prints; for both colour and B&W) that I have to lighten the image on my screen a fair bit to get a print that looks right. The good news is that once you determine the amount you can just remember the value and keep using the same adjustment over and over. The only time you will have to repeat the test prints is if you make changes to the brightness setting of your computer screen.

As for the ColorMunki print, it is for profiling papers and inks (instead of using the paper manufacturer's ICC profile) in addition to ensuring that the colours on your computer screen are accurate. I don't know if it will calibrate the two operations. On the other hand, doing test prints is so much less expensive.

I guess coming from the colour and B&W darkroom, I don't have any issues with doing test prints, as they were always the way we worked in those days.

Hi Manfred,

Doesn't ColorMunki make the brightness adjustment based on the profiling scheme? I do notice a brightness adjustment based on the time of day so the profiling must be working, also I know that the profile is based on my selections of the test patterns. And yes the test prints do come in handy.

[Dave Humphries]

I rarely print, like once a year - or less, so I have no experience (that I can remember well).

I do a test print and adjust 'Levels' as seems necessary to improve dark tone blocking.

When you talk about 'remembering the number', are you (effectively) talking about simply increasing the Black point number? (over what looks good on screen)
Or are you using the mid-point, to stretch out the darker end of the range.

TIA, Dave

[Polar01]

John I would say that the ColorMunko Photo would not really help as the difference with the Photo is I can scan test print colour patterns part into the computer and the program can then correct the colour printed to be correct to what it should be.
Now here is where it may help after all that. A suggestion that was given to me from a X-Rite video was to print out a grey scale of the 255 shades of grey but outputted as 50 patches, then scan they back into the system as it helps the overall system better figure out tones. So I did this, and it seem that the printer now does a better job printing say going from a dark blue sky to a lighter blue. Now another thing I did as suggested from an article in Luminous-Landscape was I did what I call a output sample sheet, this lets me see where I block up in darks depending on the stock used. On some printers you may not see any difference until you get to output 23, for me with my Epson 4900 printer 0 to and including 5 the printer can not print the difference in tones whoever I can see the printed different between the output 5 and 6 on Epson Luster. The thing is with a 256 toned grey scale on my monitor I can see the difference between 0 and 1 it is hard, but there is no way that the printer can output that tonal difference until 6, at the top end my printer can only tell the difference between 247 and lower, to it a 247 output is the same as a 255 all look the same, again I can see the difference on my monitor.
There are two Luminous-Landscape articles they are Beyond Calibration and Beyond Calibration 2 both a good read however 2 is more on the finding out how well you printer will output.

Cheers: Allan

[Shadowman]

John I would say that the ColorMunko Photo would not really help as the difference with the Photo is I can scan test print colour patterns part into the computer and the program can then correct the colour printed to be correct to what it should be.
Now here is where it may help after all that. A suggestion that was given to me from a X-Rite video was to print out a grey scale of the 255 shades of grey but outputted as 50 patches, then scan they back into the system as it helps the overall system better figure out tones. So I did this, and it seem that the printer now does a better job printing say going from a dark blue sky to a lighter blue. Now another thing I did as suggested from an article in Luminous-Landscape was I did what I call a output sample sheet, this lets me see where I block up in darks depending on the stock used. On some printers you may not see any difference until you get to output 23, for me with my Epson 4900 printer 0 to and including 5 the printer can not print the difference in tones whoever I can see the printed different between the output 5 and 6 on Epson Luster. The thing is with a 256 toned grey scale on my monitor I can see the difference between 0 and 1 it is hard, but there is no way that the printer can output that tonal difference until 6, at the top end my printer can only tell the difference between 247 and lower, to it a 247 output is the same as a 255 all look the same, again I can see the difference on my monitor.
There are two Luminous-Landscape articles they are Beyond Calibration and Beyond Calibration 2 both a good read however 2 is more on the finding out how well you printer will output.

Cheers: Allan

Hi Allan,

I read about the test image for determining your printer's black/white point. I'll download and do an evaluation.

[Polar01]

John sorry to say you can not download the output grid you have to make it yourself. Ya a really pain, but once you make it you never have to make it again, that is why Beyond Calibration 2 is more important as it explains it and how to produce one yourself.
Or know someone who has done it and is willing to you give let you have a copy of the seventy test patterns. Hint, hint.
PM me and we talk.

Cheers: Allan

[Shadowman]

John sorry to say you can not download the output grid you have to make it yourself. Ya a really pain, but once you make it you never have to make it again, that is why Beyond Calibration 2 is more important as it explains it and how to produce one yourself.
Or know someone who has done it and is willing to you give let you have a copy of the seventy test patterns. Hint, hint.
PM me and we talk.

Cheers: Allan

Thanks Allan,

I thought you were referring to a grayscale ramp.

[Polar01]

John I originally did a Grayscale ramp (could not thing of the last part of the name "ramp", I printed it out in 50 tonal patches as the ColorMunki Photo requires 50 patches to be scanned.
The other I printed out as per Beyond Calibration 2 is also a in some ways a grey scale it is so you can see where your printer blocks up in the blacks ( it is not able to print the different between one tone and another) for me on Luster everything black point 0 to 5 all print the same yet black point output 6 I can see the difference between it and output 5.
Knowing this has allowed me to get better prints in both my B&W's and colour images.

Cheers: Allan

[Shadowman]

John I originally did a Grayscale ramp (could not thing of the last part of the name "ramp", I printed it out in 50 tonal patches as the ColorMunki Photo requires 50 patches to be scanned.
The other I printed out as per Beyond Calibration 2 is also a in some ways a grey scale it is so you can see where your printer blocks up in the blacks ( it is not able to print the different between one tone and another) for me on Luster everything black point 0 to 5 all print the same yet black point output 6 I can see the difference between it and output 5.
Knowing this has allowed me to get better prints in both my B&W's and colour images.

Cheers: Allan

Allan,

I printed the test ramp and there were areas on the grayscale that looked exactly alike. I'll let the test image dry and evaluate again. For one analysis it is suggested that my monitor might be too bright, on another analysis it suggests that my printer is not suitable for fine art printing. However, it also suggests that I should do the evaluation for every type of paper I might use.

[Manfred M]

Hi Manfred,

Doesn't ColorMunki make the brightness adjustment based on the profiling scheme? I do notice a brightness adjustment based on the time of day so the profiling must be working, also I know that the profile is based on my selections of the test patterns. And yes the test prints do come in handy.

I do my photo editing work in a basement office with no natural light coming in. It has fairly low level of ambient light and gray walls, so I don't need to have the ColorMunki plugged in and monitoring the lighting conditions after the initial setup, as the light levels never change.

When I created the profiles, I set up the computer screen to the recommended 120 cd/m2.

From a printing standpoint I use a combination of what I learned in the classroom, one of the Kelby Training videos and stuff I've picked up reading here and there. I print 100% of my images using Photoshop, so the methods I use are not directly transferable to other software.

I suspect that some of the results are going to be paper specific, and as I did all of my testing on Epson Lustre paper on the Epson 3880 printer, the absolute numbers need to be treated with a bit of caution when looking at other papers and printers.

1. The brightest white you can get is the colour of the paper you are printing on. If your printer doesn't deposit some ink on the paper you get areas with a different finish that the parts of the paper with ink on it, so it is important to adjust your output to ensure some ink is deposited in areas that will be pure white; specular highlights, bright lights, the sun, etc.

Setting the output white value (in the levels adjustment) to 240 will assure that some ink is deposited in those areas of the print. I suspect I could back that off a bit on matte papers, but suspect that I won't be able to drop below 245.

2. Photo printers (those with additional cartridges over and beyond CMYK) can print a wider gamut than AdobeRGB, BUT the number of discrete steps is far lower than what we see on our screen. I've seen estimates that suggest printers can produce between 350,000 and 650,000 distinct shades (after all in a print a combination of the colours in the cartridges and dot size is how the specific shades are produced).

What that effectively means that pure black is deposited once you get to very low colour values. I've seen some calculations that suggest any value lower than 12 will be output as pure black, so going a bit conservative, I set my output black point to 15 in order to preserve shadow details a bit better.

3. In printing, I want to preserve my black point and white point, yet push the brightness up the overall image so it comes out looking right as a print.

For printing I use the "stamp" command <Ctrl><Alt><Shft> E to create a new layer with all of the adjustments I've made applied to it. I then change the blending mode to "Screen", which preserves the black point and white point, but brightens everything up.

My test prints show that by setting the opacity of this layer to 20%, my print will come out looking the way I want it to.

[Polar01]

Yes for each paper type and supplier, rag paper will print different than a luster or gloss, (rag=matte Black, luster and glosses=Photo) two rags from two different will not print the same or look the same, every stock is different.
Adding the output grid to this thread.

Cheers: Allan

[Shadowman]

I do my photo editing work in a basement office with no natural light coming in. It has fairly low level of ambient light and gray walls, so I don't need to have the ColorMunki plugged in and monitoring the lighting conditions after the initial setup, as the light levels never change.

When I created the profiles, I set up the computer screen to the recommended 120 cd/m2.

From a printing standpoint I use a combination of what I learned in the classroom, one of the Kelby Training videos and stuff I've picked up reading here and there. I print 100% of my images using Photoshop, so the methods I use are not directly transferable to other software.

I suspect that some of the results are going to be paper specific, and as I did all of my testing on Epson Lustre paper on the Epson 3880 printer, the absolute numbers need to be treated with a bit of caution when looking at other papers and printers.

1. The brightest white you can get is the colour of the paper you are printing on. If your printer doesn't deposit some ink on the paper you get areas with a different finish that the parts of the paper with ink on it, so it is important to adjust your output to ensure some ink is deposited in areas that will be pure white; specular highlights, bright lights, the sun, etc.

Setting the output white value (in the levels adjustment) to 240 will assure that some ink is deposited in those areas of the print. I suspect I could back that off a bit on matte papers, but suspect that I won't be able to drop below 245.

2. Photo printers (those with additional cartridges over and beyond CMYK) can print a wider gamut than AdobeRGB, BUT the number of discrete steps is far lower than what we see on our screen. I've seen estimates that suggest printers can produce between 350,000 and 650,000 distinct shades (after all in a print a combination of the colours in the cartridges and dot size is how the specific shades are produced).

What that effectively means that pure black is deposited once you get to very low colour values. I've seen some calculations that suggest any value lower than 12 will be output as pure black, so going a bit conservative, I set my output black point to 15 in order to preserve shadow details a bit better.

3. In printing, I want to preserve my black point and white point, yet push the brightness up the overall image so it comes out looking right as a print.

For printing I use the "stamp" command <Ctrl><Alt><Shft> E to create a new layer with all of the adjustments I've made applied to it. I then change the blending mode to "Screen", which preserves the black point and white point, but brightens everything up.

My test prints show that by setting the opacity of this layer to 20%, my print will come out looking the way I want it to.

I set my white/black points within Elements to 10/240 as suggested by Kelby's tutorial, he probably chose different values for Photoshop. I'll have to do another monitor profile to see if my brightness needs to be adjusted, however I can see differences on the monitor; it's the output that doesn't match. Also, I have profiles for glossy paper but not for luster, the guide says one can be used for the other, however I also notice a big difference in the blacks between images printed on each type of paper.

[Manfred M]

I set my white/black points within Elements to 10/240 as suggested by Kelby's tutorial, he probably chose different values for Photoshop. I'll have to do another monitor profile to see if my brightness needs to be adjusted, however I can see differences on the monitor; it's the output that doesn't match. Also, I have profiles for glossy paper but not for luster, the guide says one can be used for the other, however I also notice a big difference in the blacks between images printed on each type of paper.

The values I gave do not come from Kelby, but from other sources including some testing by me.

The view on the screen with the adjustments I suggested will look too bright on the screen, but if you have set the level of the adjustment layer correctly (based on test prints), it should be a good match to what you see on your screen with the adjustment layer disabled.

[Shadowman]

Here is a link to some evaluation images for testing your printer, I used the one from Outback.

http://www.northlight-images.co.uk/a...st_images.html