Inkwell: Authors and Artists

    

Welcome to Inkwell.vue, Ctein!
  

    

Ctein ( http://ctein.com) is a professional photographer best known for his
photographs of eclipses, aurora, natural and unnatural scenics, and space
launches, and his hand-printed fine-art books.

He is also the author of "Digital Restoration from Start to Finish: How to
Repair Old and Damaged Photographs," as well as "Post Exposure: Advanced
Techniques for the Photographic Printer," and hundreds of articles and
columns published in numerous photography and technology magazines,
including PHOTO Techniques, Camera & Darkroom, and InfoWorld.

Ctein's been an industrial consultant on computer displays, a technical
writer of computer manuals, has degrees in English and Physics from Caltech,
and has engaged in pollution research, astronomy, photocopy research, world
designing for CONTACT, and radical feminist queer activism.

He says that if he grows up, he wants to be a dilettante.

Inkwell.vue host David Adam Adelstein leads the conversation with Ctein.

David says he's been hopelessly addicted to photography ever since his 
parents bought him a Kodak instamatic for his eighth birthday. His parents 
bought their first computer a year later, so his father could design 
weaving patterns on it. This meant that in David's mind, the computer has 
always been first and foremost an artistic tool.  
  

    
Thanks, Cynthia, and Hi, Ctein -- it's good to get to meet you
virtually after so many years of following your work in those hundreds
of articles.

I first saw your name in the context of dye transfer prints, which I
understand is literally a dying art, since the materials are no longer
made.  Besides that of course you've been the source of a steady stream
of great technical articles on photographic issues.  

All of this extreme technique -- not to mention a keen eye -- is of
course key in doing something like photo restoration, but I'm curious
how you made the leap from doing your own work to restoring other
people's work.  What's the attraction for you?
  

    
Well, David, there is actually a connection between dye transfer and
print restoration, but before I go there, let me comment, as an aside,
the dye transfer isn't dying off as quickly as I expected.  When Kodak
killed off all the supplies 13 years ago, I don't think any of us
expected that we would still be doing dye transfer printing today. 
But, it appears that there are it least a couple of dozen others around
the world still doing it.  There's a whole cottage industry that has
grown up around people trying to make their own dyes and paper (with
varying degrees of success).  Stores of supplies that people have
squirreled away turn up periodically.  

While I would say that dye transfer is a dying craft, it appears that
the death is going to be extremely prolonged and the process may very
well outlive many of us.  

My interest in restoration came directly out of dye transfer.  Back
around 1990, Jim Wallace, who was running the photography department at
the Smithsonian at the time, contacted me about making some dye
transfer prints from a collection of old color-separation glass plates
that had been done for Life magazine in the 1940's.  The plates had
been poorly enough processed and were in bad enough shape that they
couldn't print them directly onto modern color materials.  

Dye transfer affords enough kinds of control the that I could  pull
extremely good prints from those plates.  At that time, there was no
other way to do that at a reasonable price.  It got me interested in
the possibilities for resurrecting seemingly-lost old photographs.  

Several years after that, I noticed a very badly faded portrait on a
friend's wall (http://photo-repair.com/bigsample_myrna.htm). I offered
to do a restoration using dye transfer printing methods to correct for
the severe color fading.  

What really happened was that the photograph sat around my house for
several years. Meanwhile, computing capabilities and software got
better and better, and in the late 90's I decided to tackle the job
digitally.  The results were good enough that I decided doing this kind
of restoration in the darkroom was completely obsolete.  

What's the attraction for me? I always been interested photo
conservation.  Personal value aside, photographs are the most important
cultural record we have of the 20th century.  It bugs the hell out of
me that so many of them were made on crappy materials that wouldn't
survive more than a few decades.  

On top of that, it's a real kick to magically make a photograph
reappear out of what seems to be a blank sheet of paper.  I really do
think it's a huge amount of fun, and people are really happy to get
back their valued photographs that they were sure were lost for all
time.  


                                 pax / Ctein 
[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
I can definitely see the excitement of bringing something back that
seemed otherwise lost.

One idea you touch on several times in the book is the need to bring
an aesthetic vision to restoration work, that it's not simply a
mechanical task.  What really interested me about this idea is that
there might be a significant difference between, say, your personal
aesthetic vision for your own work, and a vision for the restoration
work you do.  How do you make that distinction?  Or is there even a
distinction in your mind?
  

    
Dear David, 

Well, this question also relates back to my regular photographic work.
 A good part of my business is doing custom printing, where I print
other people's art.  That involves me applying my printing skills to
achieve their artistic goals, not mine.  

Of course, I can usually ask the artist what they have in mind for a
photograph, but sometimes they're not very articulate about it, and it
still works better if I can adopt their artistic frame of mind and
learn to think about photographs the way they do.  

Sometimes there's a big difference between the way I would print a
photograph for myself and the way they want it printed.  Not that one
version is better than the other;  Ansel Adams once said that the
negative is like a musical score and the print is like the performance.
 As a custom printer, part of my job is to figure out how to perform
the way the client would, if it were them doing the printing.  

Actually it's a little more abstract than that.  I am usually a much
better printer then my clients could ever be.  So what I really have to
do is perform the score the way they would perform it if they were as
skilled a performer as I am.  

When it comes to doing restorations, this doesn't come up as
explicitly.  I usually don't have any idea what the original
photographer had in mind.  All I can do is imagine myself in their
shoes and say to myself, "OK, if this were my photograph how would I
print it to look the very best." Unless I have a large body of the
photographers work to look at, there's no way for me to get inside
their head.  

There's one notable exception in my book, which is the self portrait
of Tee Corinne and Honey Lee Cottrell in Chapter 9, page 337  (
http://photo-repair.com/bigsample_tee.htm ).  In this case I could work
directly with Tee, just as if it were regular custom printing. In
fact, once I had a technically-correct restoration of the photograph, I
looked at it and thought to myself that this really could be better. 
So I made some further corrections to the photograph, above and beyond
what were in the original, and Tee agreed that it made it an even
better photograph.  That's the final version we went with.  So this was
kind of a mix of traditional custom printing and restoration.  


                                 pax / Ctein 
[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
Having the original photographer there seems like it would definitely
be a help for the restoration work!

That brings up another general point I wanted to talk about.  Aside
from all of the technique, and all of the repair work -- which we'll
talk about -- it may be more important to know *what* to do, than how
to do it.   

That seems like a difficult detective exercise at best.  How do you
approach figuring out how to start?  Is there some set of experiments
you try first?
  

    

Hi Ctein, I was delighted to read your veiled reference on p 3 of your
introduction to Adams's comment that the "negative is the score and the
print is the performance," and then to see it again in you post above.

Not to distract from David's question, I want to simply say that your focus
on workflow gives a context within which performance can become more
consistent.  When you say that you would try to achieve for others the
performance with their images that they would make if they knoew what you do
is a subtlty not found in the darkroom of every photo lab.  As a former
production printer in a San Francisco color lab during the 70s and 80s I can
testify to that.

Finally, mining the book for techniques may be the way many people will
approach it at first, and they will not be disappointed. Your treatment of
some of the more confusing tools in Photoshop, such as curves,
shadow/highlight and techniques such as using adjustment layers and masks
gives a detailed practical immersion in how these tools and techniques are
actually used.  For many who are just getting into digital photography and
post production this book is a very dense and useful reference.
  

    
That was definitely my reaction as well -- I don't expect to spend
that much time doing restoration, but a lot of the techniques mentioned
(hello, shadows and highlights tool I've never paid attention to
before) have immediate application to my personal work.
  

    

(Note: offsite readers with questions or comments may send them to
<inkwell@well.com> to have them added to this conversational thread)
  

    
Coming from a different angle entirely, on the question of artistic frames
of reference, it seems to me that in photo restoration there is often a
third set of eyes and aesthetics involved, that of the owner of the
photograph.  Since many of the photographs people want restored are of
people, and generally people with whom the owners have an emotional
connection, they are likely to view the quality of the work not just in 
terms of the technical quality of the restoration, but also in terms of 
how much it "looks like Grandma Fan", or even "how much it looks like my 
emotion-blurred memory of Grandma Fan".

On the two family photographs you have restored for me, this aspect was
very right-feeling, but we're good friends and you certainly have a good
picture of my familial emotional nuances.  Does this apply in a more 
abstract situation, and if so, how do you deal with it?
  

    
(replying to #6) 

Well, David, it is a bit of a tough problem.  It's also the part of
the book I'm least happy with.  Originally I planned to give people a
kind of overall road map for restoration, something of a "quick start"
guide.  I simply couldn't figure out a way to do it.  Too many choices,
too many decision branches, it just never jelled in my head.  But I
have some ideas for how to handle it better in the next edition of the
book.  

99% of the time, the first thing to do is to make a good scan, as
described in Chapter 4.  Basically, that's a scan where each of the
red, green, and blue channels has information spanning 90% of the tonal
range for that channel.  

Just about every time, that will not only get you a lot closer to a
final restoration, but it will let you see what's actually in the
photograph so you know what you have to work with.  Sometimes there are
real surprises, like in the "faded baby" photo (
http://photo-repair.com/bigsample_fadedbaby.htm ). Honestly, the
original was even more faded and less clear than it looks here; the
piece of paper looked almost blank.  I was truly amazed what came out
after I did the scan.  

Beyond getting a good scan, repair work mostly falls into three broad
categories: fixing fading and overall color and tone changes, fixing
stains and discoloration (like tarnish), and fixing physical damage
like cracks and tears.  The first is pretty easy; mostly that's just
playing with things like curves and the shadow/highlight tool. You may
need to do some masking to apply corrections more selectively, but
fundamentally this is not rocket science.  

Fixing stains and discoloration is a lot easier once you catch on to
the masking tricks I talk about in Chapter 7 that let you isolate
damaged areas by color and tone differences. Otherwise it's pretty
painful.  

Fixing the physical damage is the real killer.  The brute-force way is
to do it pixel by pixel, and I really try to avoid that because it's
incredibly time-consuming and boring (and expensive for the clients). 
That's why so much of Chapters 7 and 8 are devoted to physical
problems, because there really is no simple, elegant way to handle them
in the general case. Even with a pretty clean photo, I'd probably end
up spending half my time bit-twiddling.  If the photo's really messed
up, almost all the time is spent trying to clean up the garbage 


                                 pax / Ctein 
[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
(replying to #7)

Thanks for the nice words, Jeff. I still feel the workflow is the
weakest part of the book. As David observed, figuring out how to get
where you want to go is an awfully big part of restoration, and I wish
I had been able to provide more specifics and less arm-waving.  So it's
nice to hear that you found it helpful.  

Regarding putting oneself in the head of the client, remember that the
kind of custom printing I do for other clients is very high end.  With
rare exceptions, it's strictly dye transfer printing, and so people
are paying $1,200 and up for the first print.  I figure for that kind
of money they're really entitled to get exactly what THEY want.  Not
that I'm at all reluctant to tell them how I think something should be
printed, and more often than not I see potentials in their photographs
that they didn't.  Even so, it still has to happen within their
aesthetic, not mine. 

I spend so much time talking about curves in the book, because curves
really are absolutely necessary to doing good digital work of any sort
with any image processing program, not just Photoshop. At the same
time, most photographers don't naturally think that way. Especially
when it comes to thinking in terms of separations, which is really what
channels are all about. 

It's second nature for me, partly because of the way my head works,
but more so because of decades of experience making dye transfer
prints, but I think for most photographers it isn't even fifth nature. 
So laying out a bunch of "basic dance moves" seemed like a really,
really good idea.  

Photoshop is an insanely difficult program to use.  It wasn't designed
by photographers to begin with, and it still doesn't think the way a
photographer does.  I don't like the program at all; it just happens to
be the 500 lb. gorilla on the block.  But it works about as much like
a photographer as Microsoft Word works like an author writes. 


                                 pax / Ctein 
[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
(replying to # 10) 

Now that you bring it up, Sarah, that really is the way it should
work.  I hadn't thought about that before. 

As I think a bit more, I'm realizing, though, that the issue doesn't
seem to come up.  Which makes me wonder how much of a visual memory
people really do have.  Do people really remember what someone looked
like vs what they see in the photograph? Assuming that it's not a
really crappy photo, of course; we've all seen photos it didn't look
remotely like the person did in real life.  But I'm hypothesizing that
unless it's an egregiously bad photo, with the passage of time people
may more and more forget what their memory tells them and just rely on
what they see in the photo.  

Or not.  I'm making this up as I go.  But you've got me definitely
wondering why this hasn't come up as an issue particularly. 


                                 pax / Ctein 
[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
Your last paragraph so perfectly captures my experience with
photoshop--a giant, awkward, bloated monstrosity, full of ways to screw
up your images without having any idea what you're doing, while making
it hard to do what you do want to do.

That said, I have come to terms with if sufficiently that I'm really
looking forward to the book (hoping my copy has made it home before me
tonight).  I've gotten past the first baby steps with it, and feel
ready to add some more advanced steps to my repetoire.  My sister (who
I hope will join us by e-mail) is the keeper of the family genealogy
and family photos, but I have my fair share, plus more than a few still
film-only images of my own I'd like to work with. 

And your word salad is pretty clean.  My typos are strictly my own.

--diane in st. louis
  

    
I just received the book yesterday, which I was interested in receiving
because aside from my lifelong interest in photography, I have a LOT of old
family photographs and I am absolutely dying to restore someday.

I have a small amount of photoshop knowledge, but am still using v. 7 so
suspect an upgrade will be in order.

I have no scanner - yet. At least one capable of doing what I'll need (have
an old 600dpi Canon but I think one may not do).

At least my computer is up to snuff!

I think my knowledge level is far below some of the other folks who will be
the active Well posters here, but will definitely follow along.
  

    
Heh! I claim the place at the bottom of the knowledge level tree. 
What little I know about this comes entirely from looking at
photographs with Ctein and him telling me what, exactly, I'm seeing. 
(It's been one hell of a good visual education, too.)

About visual memory, Ctein, I suspect you may be right.  I have always
been kind of iffy about taking and keeping many photos, because I want
things to look like my memories, and sometimes photos don't. (Often
current photographs don't look like my reality either, because the I
which I see in my mind isn't entirely congruent with the I which I see
in the mirror.  But if a photograph exists and is somewhere I see it a
fair amount, it tends to become the "official" memory image. 
  

    
(replying to #15)

Before you assume you need to upgrade everything, go read Example 3 in
Chapter 10 (page 362). It was done with much inferior hardware and
software than you have (Photoshop 4 and a *cheap* scanner of similar
vintage). Biggest disadvantage of Photoshop 7 over later Photoshop is
no 16-bit layer support. So, if you find you need layers, get the tone
and color approximately corrected (via a good scan and curves) in the
16-bit scan and then convert the image to 8-bit. Or spend more time
fiddling with the scan and run 8-bit from the start (read Chapter 4 for
more details and clarity).

pax / Ctein
  

    
I'm with you on visual memory.  After enough time, the picture is the
more current and more active memory.  Who really knows what color the
carpet was in the old house? It's dark green in the photo.

Your section on hardware requirements was definitely a refreshing
change from the usual "You must have the latest and greatest or you
should just return this book" attitude of photoshop books.

One question I had for you though, is your mention of monitor
calibration as not being in the must-have category.  That's counter to
my experience; can you talk about why you don't think it's required?
  

    
Just got the book, and for a free book, I can see that it's going to
be very expensive.  That is to say, the introduction and examples
inspired me to rush over to my photo albums and film binders.

I realize that there are quite a few that could use restoration, and
many others that would gain new life simply from a quality scan job
(slides that never became prints, for example, B&W negatives that were
never printed, and have not been kindly treated through numerous
moves.....).  

Having found a few examples ready to go, I then discovered that my
scanner is no longer supported with up to date software by HP, and I am
going to have to buy a new one to play with all of these techniques
properly, especially with the film, since my scanner/copier/printer
combo will only suffice for limited practice with prints.
  

    
 Hi Ctein, Welcome. Perhaps you can give me some advice about some
restoration.
  While I was in college, my father gave me several picture books - a
coffee table type book of Weston pictures two coffee table type picture
books by  Ansel Adams, one autographed by Ansel himself. My father had
met Ansel at Yosemite on a hike.  Family treasures.

  And several years ago after grad. school, I had all my worldly
belongings packed in boxes, and stored in the basement of a friends
house on the north side of Chicago. The rains came, the flood waters
rose, and my friend called me after he discovered that his basement
flooded. All the books sustained some water damage. Some of the pages
still have water marks.
 
 Fortunately, almost all the pictures have no visible damage. A few
photographs in one of the Adams books are damaged at the edges, and the
picture gone.  I think the biggest concern I have now, is to preserve
the book without having too much further damage.  It would be nice to
have the damaged pictures restored in some fashion.
   I'd be grateful for any suggestions.



   
  

    
(replying to #14) 

Personally, I much prefer Picture Window Pro, when it comes to design
and thinking like a photographer.  And it's expensive, compact, and the
technical support is a hell of a lot better.  Unfortunately, it just
doesn't have a some features, like layers, that I simply can't live
without.  Although in some ways it is much more capable than Photoshop,
I need that stuff.  But for someone who hasn't already bought into
Photoshop, it's really worth a look. 


                                 pax / Ctein 

[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
(replying to #18) 

 The reason I don't think monitor calibration is a must is because it
doesn't actually improve the quality of your prints ( unlike printer
calibration).  It produces a closer match between what you see on the
monitor and what should get out in a print, but it still won't be
exact.  It might save you one or two test prints, but it won't
eliminate the need for them.  I definitely think it's a nice thing to
have, but I don't think it's a necessity.  


                                 pax / Ctein 
[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
(reply to #20) 

I don't know a thing about repairing and restoring books, so I can't
help you on that aspect of it.  As for restoring the pictures, it would
depend on what the water damage looked like.  I presume most of what
you are looking at is some kind of staining, right? If the stains have
any sort of a color cast to them, you can use the tricks in Chapter 7
for making masks based on color to isolate those parts of the image. 
Then you can probably clean them up pretty easily with a curves
adjustment layer.  Look at the way I deal with tarnish to get the idea.
 The importance starting point is to make a scan that preserves or
even accentuates any color differences and then exaggerate them as much
as possible in Photoshop, to make it easy to grab on to them with some
kind of color selection tool.  

If there's no difference whatsoever in color between the stains and
the photographs,  I can't think of an easy way to deal with this off
the top of my head.  


                                 pax / Ctein 
[[ Please excuse any word-salad. ViaVoice in training! ]] 
  

    
That tarnish technique is outstanding.  There are so many interesting
uses of commands and techniques I thought I understood in this book;
it's really expanded my understanding of some of them.  Can you talk a
bit about where these techniques come from? Do you find a new problem
and think through how to solve it, or is it more of a case of fiddling
with a filter or tool and thinking "hey, this could fix problem x"?
  

    
Ok, I have a project!  Tonight I was at a meeting of my fish club,
held at a member's house, and she brought out some memorabilia from the
fish store she used to work at, including some old photos.  I asked if
there was one she'd like me to work on, and she selected this one:

http://www.flickr.com/photos/debunix/485920036/

I think here the intention is historical, as a record of this man, who
founded a regionally important aquarium business, so that the goal
will be maximizing recovery of detail, and minimizing the impact of the
nibbled bits without inventing information.

The first scan was done on my little all-in-one, an 8-bit 600 dpi scan
of an 8x10 photos, as something to play with until I can use the
higher-quality & adjustible scanner at work.  But already I can see
that there is more there to be recovered.

There is a rusty spot to be removed, cracks and tears, the obvious
critter nibbles, tone to be recovered and perhaps improved, details to
be sharpened, and the odd perhaps water damaged altered tone in the
lower left.

How would you prioritize the various steps on a project like this?

My general photoshop workflow for my new images is to open them with
camera raw to adjust curves & sliders; dodge and burn if needed; if I
am doing a B&W conversion, to do that next; then mask off areas for
unsharp mask & or blurring; and after sharpening, do any spot removal,
last thing before a crop if needed.  My idea is to leave the sharpening
until I see by the light adjustments better which areas I want to
emphasize or not.  Then after I sharpen, I can see all the spots I want
to work on, which may include some obvious from the get go, like spots
from the lens, and some subtler things, like minor dust from the
sensor, that might not become obvious until the unsharp mask makes them
obvious. 

But here I am unsure how to prioritize the tasks.  Is it better to
work on the rusty spot or the cracks & missing areas first?  The
altered tone in the lower left or the spots therein?  How does what I
first repair limit or alter what I am going to do next?