How'd I do with this B&W conversion

cjmchch

I have tried to follow a few techniques from the tutorials here on how to make an image 'pop' and applied it to this image, with of course a conversion to black and white. I'd like some critique on the finished product please. ( I know I need to clone out the dead skin on his thumb.)

SloYerRoll

Hey CJM,

This isn't really a B&W conversion. There is still quite a bit of color in that image.
The black point isn't really set on this either, so it lacks that pop you were going after.
If you post an image in finishing school. I (and others) will be happy to help show you some best practices for converting to B&W.

Don't take this as rude. It's definitely not intended that way. Take it as constructive feedback

DavidTO

SloYerRoll wrote:

If you post an image in finishing school. I (and others) will be happy to help show you some best practices

I just went ahead and moved it, since it really is a Finishing School post. Technique is intended more for shooting technique (in-camera).

cjmchch

SloYerRoll wrote:

Hey CJM,

This isn't really a B&W conversion. There is still quite a bit of color in that image.
The black point isn't really set on this either, so it lacks that pop you were going after.
If you post an image in finishing school. I (and others) will be happy to help show you some best practices for converting to B&W.

Don't take this as rude. It's definitely not intended that way. Take it as constructive feedback

I'd really apprecaite some help. I must admit, one thing I did not do was set the black point. I'll load the original image for you to see, from which hopefully you will be able to guide me. I am completely green when it comes to photography so everything is learning for me. The lens I was using on this shot was the 24-70L USM.

The link to the 2.49mb file: http://www.outclassed.co.nz/IMG_6619.JPG

And DavidTO, thanks for moving it.

pathfinder

As Jon, said, this image really still has a little red in it.

With a Pixel Reader, like Photoshop, or the Digital Color Meter that comes on a Mac, you can read each pixel's RGB values. Your highlights tend to be pretty neutral ( R=G=B ) but in your shadows the Red channel values tend to be higher than the Green and Blue values, hence my statement that your image has a little red in it.

I have added a black and white point to your image, and a slight curve. I then converted to grayscale to eliminate any residual color, and then back to sRGB for the web, and 'attached' the image to dgrin here.

I am, explicitly, not saying this is better than your original image, I rather like the warm brown tonality, but this image is a true neutral B&W for your comparison.


Don't feel badly - most of the "B&W" images that I post are actually duotones also. Greg Gorman helped popularize this technique. Was this your original intention, perhaps, and Jon and I just missed it?

cjmchch

pathfinder wrote:

As Jon, said, this image really still has a little red in it.

With a Pixel Reader, like Photoshop, or the Digital Color Meter that comes on a Mac, you can read each pixel's RGB values. Your highlights tend to be pretty neutral ( R=G=B ) but in your shadows the Red channel values tend to be higher than the Green and Blue values, hence my statement that your image has a little red in it.

I have added a black and white point to your image, and a slight curve. I then converted to grayscale to eliminate any residual color, and then back to sRGB for the web, and 'attached' the image to dgrin here.

I am, explicitly, not saying this is better than your original image, I rather like the warm brown tonality, but this image is a true neutral B&W for your comparison.


Don't feel badly - most of the "B&W" images that I post are actually duotones also. Greg Gorman helped popularize this technique. Was this your original intention, perhaps, and Jon and I just missed it?

My original intent was to get the image to as close to true B&W as possible and given my lack of knowledge re reaching that point in CS3 I decided to do most of my adjustments to the original color image then in the final steps turned it into B&W using the Image/Adjustments/Black and White tool.

At the time I thought it looked good, until I saw your version, which is ultimately, the look I would like.

Would I be correct by saying I should create an action that has the following steps included:

Set black and white points -Image/Adjustments/Threshold
Curves
Levels
Unsharp mask using the Lab color/lightness routine
Convert to gray scale / eliminate any residual
plus any other tweaks I want to do like adding vignettes, layers etc
Save in RGB format

?

SloYerRoll

cjmchch wrote:

My original intent was to get the image to as close to true B&W as possible

THe only thing B&W does is remove color. If you get close (as pathfinder said) you get a duo-tone. With try B&W you have 256 shades of grey (from blak to white).

cjmchch wrote:

Would I be correct by saying I should create an action that has the following steps included:

Set black and white points -Image/Adjustments/Threshold
Curves
Levels
Unsharp mask using the Lab color/lightness routine
Convert to gray scale / eliminate any residual
plus any other tweaks I want to do like adding vignettes, layers etc
Save in RGB format

With CS3 and real time adjustment sliders. I would do this differently.
DISCALIMER:
This is not the definitive way to do this. This is just my way.

• Levels (shortcut + ctrl + L) *hold down the alt key when adjusting in the levels tool. It gives you a preview of shadows and highlights.
• Convert to B&W (Image>adjustments>Black & White) (shortcut = alt + shift + ctrl + . Then adjust the sliders from top to bottom and see what effect it gives.
• Curves (if necessary) *You can do everything in Levels through the Curves dialog once you play around and learn this tool.
• Sharpen to taste.
• Save and admire how good you are at Photoshop.

arodney

pathfinder wrote:

With a Pixel Reader, like Photoshop, or the Digital Color Meter that comes on a Mac, you can read each pixel's RGB values.

Are you getting identical RGB values for the same image in Photoshop and DCM?

pathfinder

Good question, Andrew, it never really occurred to me to compare them, so I just did.

When I read pixels in the true neutral B&W ( passed through grayscale ) image I attached above, the tones higher than 128,128,128 seem to read within 1 digit in Photoshop and DCM. The lower quartertones vary significantly more, sometimes reading 12,12,12 in Photoshop and 20,20,20 in DCM. They do all read as neutral with either tool, since it is a true B&W.


My working space in PSCS3 is sRGB ( and I am reconsidering whether to change it back to aRGB for 2008:D ), but my photos are routinely edited in ProPhoto 16 bit until saving for smugmug in sRGB. As I am anticipating more printing on 8 color inkjets, I am reconsidering aRGB again.

I assume you can explain the difference in reading between PSCS3's Info tool and DCMs. I admit my ignorance here completely.

Merry Christmas, and Happy New Year to all.

arodney

pathfinder wrote:

I assume you can explain the difference in reading between PSCS3's Info tool and DCMs

First, the color space of the Photoshop doc plays a huge role.

Next, the numbers that one is reading (DCM, from the display) has nothing inherently to do with the numbers in the document.

pathfinder

The topic of conversion of color images to B&W in Photoshop is a long, and complex one, that I do not feel is well handled by a simple Action. You can quickly type ctrl-1, or ctrl-2, or ctrl-3 ( apple-1 on the Mac) {commands are different in CS4 } and see the individual red, green or blue channels, and this will give you a start at looking at the individual color channels and their contribution to contrast in the image. Or you can switch to LAB color space, and look at the Lightness channel with ctrl-1.

B&W is ultimately about contrast and shape, since color disappears with conversion.

The reason that an Action doesn't work well, is that color images vary so widely. When you convert from color to B&W, the colors that were present in the original image need to be handled properly in the conversion, or you will get the very same gray tones from markedly different colors. The different colors in color images need to be seen in differing gray tones or the image will lose a lot of its punch.

I Googled -- "B&W conversion technique : dgrin.com"

And found these threads for a start.

Why you need to pay atention to colors when converting to B&W from color by John Ruttenberg - read this!! It is well worth your time.
http://dgrin.smugmug.com/gallery/1134301

A list of the Sticky URLs at the top of Finishing School above
http://www.dgrin.com/showthread.php?t=56678

A tute on B&W conversion by Andy Williams the head mod and all around good guy
http://dgrin.smugmug.com/gallery/1126557

Another by Andy
http://www.dgrin.com/showthread.php?p=217438

A simple way is to shoot B&W and save as a RAW+jpg - not everyone favors this technique, but I wrote a little thread about it here
http://www.dgrin.com/showthread.php?t=3605

Some motorcycle engines in B&W ( sorry - sometimes I can't help myself )
http://www.dgrin.com/showthread.php?t=20909

My review of Chapter 7 - Keeping the Color in B&W from D Margulis "Professional Photoshop"
http://dgrin.com/showthread.php?t=52724

In all previous versions of Photoshop, most folks agreed that Channel Mixer was the mode most folks who were serious about the B&W conversion tended to use. Since the advent of Photoshop CS3, the new B&W conversion tool found via Image>Adjustments>B&W has really taken over.

I still examine each of my images quickly with ctrl-1, ctrl-2, and ctrl-3 to see just where I have contrast or where I need contrast.

Hang around here awhile, and I am sure there will be another thread on B&W conversion.

B&W conversion is great fun, and as really come into its own since the digital age.

cjmchch

pathfinder wrote:

CJMCHCH,

The topic of conversion of color images to B&W in Photoshop is a long, and complex one, that I do not feel is well handled by a simple Action. You can quickly type ctrl-1, or ctrl-2, or ctrl-3 ( apple-1 on the Mac) and see the individual red, green or blue channels, and this will give you a start at looking at the individual color channels and their contribution to contrast in the image. Or you can switch to LAB color space, and look at the Lightness channel with ctrl-1.

B&W is ultimately about contrast and shape, since color disappears with conversion.

The reason that an Action doesn't work well, is that color images vary so widely. When you convert from color to B&W, the colors that were present in the original image need to be handled properly in the conversion, or you will get the very same gray tones from markedly different colors. The different colors in color images need to be seen in differing gray tones or the image will lose a lot of its punch.

I Googled -- "B&W conversion technique : dgrin.com"

And found these threads for a start.

Why you need to pay atention to colors when converting to B&W from color by John Ruttenberg - read this!! It is well worth your time.
http://dgrin.smugmug.com/gallery/1134301

A list of the Sticky URLs at the top of Finishing School above
http://www.dgrin.com/showthread.php?t=56678

A tute on B&W conversion by Andy Williams the head mod and all around good guy
http://dgrin.smugmug.com/gallery/1126557

Another by Andy
http://www.dgrin.com/showthread.php?p=217438

A simple way is to shoot B&W and save as a RAW+jpg - not everyone favors this technique, but I wrote a little thread about it here
http://www.dgrin.com/showthread.php?t=3605

Some motorcycle engines in B&W ( sorry - sometimes I can't help myself )
http://www.dgrin.com/showthread.php?t=20909

My review of Chapter 7 - Keeping the Color in B&W from D Margulis "Professional Photoshop"
http://dgrin.com/showthread.php?t=52724

In all previous versions of Photoshop, most folks agreed that Channel Mixer was the mode most folks who were serious about the B&W conversion tended to use. Since the advent of Photoshop CS3, the new B&W conversion tool found via Image>Adjustments>B&W has really taken over.

I still examine each of my images quickly with ctrl-1, ctrl-2, and ctrl-3 to see just where I have contrast or where I need contrast.

Hang around here awhile, and I am sure there will be another thread on B&W conversion, it is about time for one again. B&W conversion is great fun, and as really come into its own since the digital age.

Thanks heaps for the detailed info above and also to the others for thier posts. I'll read through this a stuff a few times then will try applying it to the image I first posted and will see how it pans out.

Once again thanks.

davev

Another way of converting to B&W is to use the Gradient Map.

It's normally what use.
A quick tute is here: link

jdryan3

pathfinder wrote:

In all previous versions of Photoshop, most folks agreed that Channel Mixer was the mode most folks who were serious about the B&W conversion tended to use. Since the advent of Photoshop CS3, the new B&W conversion tool found via Image>Adjustments>B&W has really taken over.

As with the time to truly learn ACR 4.x & LR, I have been a little stymied in the CS3 B+W learning curve. I have used the 2 Hue/Sat layer method, channel mixer, calculations, (and others), with the last 2 being my primary methods. I also use Andy's fill color enhancement.

Is CS3 B+W truly as good as those? Why is it better or worse?

SloYerRoll

jdryan3 wrote:

Is CS3 B+W truly as good as those? Why is it better or worse?

If I understand your question. The Advantage in CS3 is the sliders provide real time results. So when I'm in B&W and I adjust the red slider. I see these changes happen right in front of me. So when I select, "OK" I already know exactly what the image will look like. This is a HUGE improvement over the preceding versions of CS3. It also gives a more intuitive user interface so you don't have to really understand the channel mixer to do a good conversion. You just have to have a good eye. (which is a skill in itself)

FYI: It's a good idea to mention the specific product in CS3. There are tons of applications that fall under the umbrella of "CS3".

arodney

jdryan3 wrote:

As with the time to truly learn ACR 4.x & LR, I have been a little stymied in the CS3 B+W learning curve. I have used the 2 Hue/Sat layer method, channel mixer, calculations, (and others), with the last 2 being my primary methods. I also use Andy's fill color enhancement.

Is CS3 B+W truly as good as those? Why is it better or worse?

The tools in Lightroom are truly amazing, plus you're working in a linear gamma space with unlimited undo's and iterations:

http://lightroom-news.com/2007/08/24/tips-for-better-black-and-white-conversions/

DavidTO

arodney wrote:

The tools in Lightroom are truly amazing, plus you're working in a linear gamma space with unlimited undo's and iterations:

http://lightroom-news.com/2007/08/24/tips-for-better-black-and-white-conversions/

:jawdrop

BinaryFx

Just to raise a quick point that has not been noted so far:

The instrument is an important part of the composition, so I personally would use some method to lighten and bring out detail (shadow/highlight, curves etc).

This would of course be done on the full colour image data before the data is made monotone - not simply lightening the mono pic in the shadows.


Hope this helps,

Stephen Marsh.
http://members.ozemail.com.au/~binaryfx/

jdryan3

SloYerRoll wrote:

If I understand your question. The Advantage in CS3 is the sliders provide real time results. So when I'm in B&W and I adjust the red slider. I see these changes happen right in front of me. So when I select, "OK" I already know exactly what the image will look like. This is a HUGE improvement over the preceding versions of CS3. It also gives a more intuitive user interface so you don't have to really understand the channel mixer to do a good conversion. You just have to have a good eye. (which is a skill in itself)

FYI: It's a good idea to mention the specific product in CS3. There are tons of applications that fall under the umbrella of "CS3".

Jon - Thanks for your feedback.
I guess the question is:
The CS3 version (of PS with it's B+W conversion tool as mentioned by pathfinder): Is it easier than the other methods I mentioned and with results as good or better?

Does it give you the flexibility and nuance need for good B+W conversions, especially compared to the other methods? Is it more like the "auto-grayscale" tool on steroids? (OK, I know the answer to that one )

arodney wrote:

The tools in Lightroom are truly amazing, plus you're working in a linear gamma space with unlimited undo's and iterations:

http://lightroom-news.com/2007/08/24/tips-for-better-black-and-white-conversions/

Andrew - I did get a copy of the Fraser/Schewe Camera Raw with Photoshop CS3 book to delve into ACR 4.x more. As the tools in LR and that are the same, you may have hit on the bigger issue: Use ACR / LR or PS CS3 for B+W?

I got the book since I use ACR now a lot and move to PS when needed, especially for B+W conversions. I believe LR is ultimately where I want to head, but haven't been able to get my head around it. I figured learning ACR 4.x was the best way to leverage my time and hedge my bets.

arodney

jdryan3 wrote:

Andrew - I did get a copy of the Fraser/Schewe Camera Raw with Photoshop CS3 book to delve into ACR 4.x more. As the tools in LR and that are the same, you may have hit on the bigger issue: Use ACR / LR or PS CS3 for B+W?

They share the same processing pipeline, use either (whatever your preference). The same things Martin did in LR you can do in CR.

jdryan3

arodney wrote:

They share the same processing pipeline, use either (whatever your preference). The same things Martin did in LR you can do in CR.

Agreed. My point exactly. The issue is to move to doing B+W conversions in LR versus Photoshop using the new tool (or the old methods).