On Thu, December 1, 2011 03:43, Karl Shah-Jenner wrote: > I thought I was alone with my take on colour management but it seems > others > have 'discovered' the same thing as I. > > I'd also state straight up that I taught photography as a science course, > so > the emphasis was not on following what I say, but in having students TEST > what I say - compare it to other methods, compare it to 'what they know' > and > evaluate the results as objectively as possible. That is what science > is.. > degrees don't make a scientist - science is a philosophy of analysis. > > > So firstly... if you're a graphic designer working with colourmetrics, > you've a set of colours that are limited to your printer, and graphics > printers are generally 4 colour. > > This means: photographs will never reproduce well > Your calibration is based on inksets used in specific printers. > You will pay dearly to buy profiles. > You will be outputting to a very tiny selection of printers. > Photographers will hate your work and say you do not understand colour. > > > We are photographers. We work with many more inks than this, we don't > like > spending money, and we DO understand colour, just differently than > graphics > guys understand it. > > I was thinking about this one day while I was running a control strip > through the RA4 machine. For those who are unaware of the process, a > batch > of 'control strips' are bought from Kodak, Fuji or elsewhere that are > simply > lengths of RA4 paper, frozen, that have been exposed to give a selection > of > colour swatches. Instructions on the box include (specific to each batch) > standard colour correction variables - ie, each batch of paper is > different > so even though the exposure of the paper was done under a specific, > carefully monitored light source, the paper will not represent these > colours > 100% accurate from batch to batch. I never ran a color line; but I know what control strips are. And that proper use of them was one of the big differences between a pro lab and a consumer lab (the pro lab will be more consistent AND closer to "right"). > Running these through the RA4 machine each day you'd then pop the control > strip under a densitometer and check how close the colours were to what > they > were supposed to be (allowing for the included variation) then you'd head > back to the RA4 machine and tweak the chemistry.. > > This was a number of days after Fuji Australia had failed to reorder E6 > control strips and we'd run out.. 4 weeks wait and we neeeeeed these > things > to get the chemistry right! Kodak had none either (unbefreakingleavable) > > I got around this pretty easily by grabbing a batch of frozen film which > had always produced reliable colours, then I shot a few rolls of a Gretag > card under controlled lights (flash with voltage monitoring) at a > specified > distance at a specified exposure with a specific camera. - all carefully > noted in case I needed to go through the process again at a later time. > It > is critical you use exactly the same setup, any variations *here* will > affect stuff *there* - how you going to know which was which? > > After this, I snipped off a few frames and ran these with the last control > strip through the chemistry. Popping them under the densitometer, I > calibrated the control strip, then adjusted the values against the stock I > shot.. I calculated what the correct values for my control strips were, > adjusted the chemistry based on the last Fuji control strip then checked > mine again. Pretty close to perfect. > > Noting the various values for My Control Strips, the rest of the rolls > shot > were snipped up and frozen to be used daily.. When the Fuji control > strips > finally came in I checked the chemistry as we'd been using it against what > it was supposed to be and it was as though we'd been using the Fuji's all > along. > > Back to the RA4 saga.. > > I was tossing all this around in my head along with what I knew was being > taught about colour management to students and I had a bit of an epiphany > > Firstly, aside from the control strips, NO ONE sticks their prints under > the > densitometer unless they're testing papers! No final print is ever > checked > this way - they're viewed under a controlled 'daylight' light source by > eye, > but never a densitometer! We only calibrate the 'printer' - not the film, > not the lighting conditions, nothing but the printer (RA4) and then as we > print we eyeball the results and tweak the filtration until we get the > print > looking right. Once your light is right, for that entire roll of film (if > the subjects were shot under the same light) no further tweaking is > necessary. > > Sure we may filter light when shooting, and we may select specific film > stocks, but these are always to get 'best looking' results, not '100% > accurate' results (impossible anyway). Who filtered to eliminate the sky > colour in the evening? No one. Unless you were shooting for colour > accuracy, in which case you'd have been better off shooting in a studio.. > even then, your film stock would let you down. Remember how we picked > Kodak > for reds, Fuji for greens etc? What did accuracy have to do with it?? > The > closest film to accurate was Konica 100 and no one used that (that's an > objective analysis done by a compulsive film tester who blew loads of cash > testing every film available on the market talking ;) > > Ultimately, we went for what LOOKED right. > > Anyone who took a leaf and compared the leaf it's self to the print under > a > densitometer deserved to be dragged off to Trembling Willows and placed in > a > room with rubber wallpaper. We photographers unlike graphics folk, go for > what *looks* right. The managing director of advertising at Coke > doesn't > care if the logo looks green under fluro lights, what he cares about is > that > the logo looks exactly the same - exactly - as the pantone swatch value > for > his logo colour under the same lightsource, so all coke logos, coke labels > and coke party hats . > > We came to digital imaging as photographers long after the graphics guys > did > and we adopted their tools - which was all there was - and we basically > copied their techniques without ever asking if it was the right way.. > after > all there was a huge learning curve and the graphics guys seemed to have > it > nailed. but the fundamental philosophic difference was overlooked.. > This set me to work trying to find a more appropriate method of colour > management for photographers than that we were using. I was mixing with > graphics and printer guys not long after this at a Canon printer > repair/setup training course and I mentioned I was on this path - I was > soundly congratulated by all and sundry as (in their words) they were > heartily sick of photographers and how little they understood of colour. > (/hangs head in shame) The individuals who created all on Ilford > Australia's colour profiles were there, the biggest distributor of wide > format printers were there.. they all said the same thing. > > The thing was, once I nailed it I didn't actually have anything saleable > for anyone, it's actually very easy. > > > First thing to understand is printers, as these are often the intended > final output device for photos. More on screen colour later.. > > dpi means nothing, ppi means everything.. and printer manufacturers > confuse > the two to confuse you. DPI is how many splats of ink per inch the > printer > may use to make up a colour. Tamper with this value by restricting the > number of colours the printer can select and you'll lose your colour > range. > Canon doesn't let you near this value (nor should they) Epson does. > Great > for Graphics folks to limit their colour range, a disaster for > photographers. leave this alone. > > Media. Basically there's gloss (needs more ink squirted per drop), matte > (needs lots more ink squirted per drop) and plain (doesn't need much). > All > changing the media should do is change the amount squirted out per drop. > Canon and almost everyone else abides by this simple rule, Epson does not. > Don't believe me? Change the media setting and run three prints through > on > the same media, the colours will look pretty much the same out of a canon, > just more or less (paler or ink sloshing across the paper). Do it with an > Epson and watch the colours actually change. Epson like photographers. > Epson sells many, many profiles to photographers. Photographers love > Epson. Epson is a wealthy company. (Epson is Seiko) Okay, the alarm bells are starting to ring here. Epson doesn't sell profiles to photographers; they give profiles for Epson media and standard Epson printers. If you want a profile unique to your own printer, or for a combination of ink, printer, and paper that crosses company boundaries, you can either make it yourself, or use one of a number of commercial outfits that make custom profiles. But Epson isn't selling profiles anywhere I've ever heard of. And thus there is no benefit to Epson from making this "hard"; what Epson benefits from is satisfied customers. > > Quality. Draft (skip some lines (low dpi), make tiny drops, go fast). > Standard (don't be too fussy, put a reasonable amount of ink > down, use maybe a few more colours and quirts) > Graphics ( let the rip go stupid on saturating the more > perceptually 'vivid' colours, use lots of ink, avoid subtle tones please > RIP) > Photographic (ok guys, fire up the 'good' rip, be careful > with > colours, lay down reasonably heavy drops (check the 'media type') and use > all the dots you need to get the colours goodest) > > > Photographers should have their printers set to photographic quality, > gloss > or matte and nothing else. > > Set the PPI on the image before sending it to print and get it to bang on > 300ppi for the intended image size. If using an Epson and you go looking > at > dpi, set it to maximum, but restrict the print size to whatever it is you > intend, ie an 8x10 image should be sent through as 2400x3000 pixels, image > size 8x10, dpi= max. do not send 4000ppi images to print or your printer > will hate you.. While for extreme enlargements you can probably do better in your own software than in the printer, I find sending original-size images to my printer for ordinary print sizes to be the best practice. There are very modest gains in sharpness from excess data, and skipping the resizing step in the workflow saves me from a range of possible mistakes (saving the wrong thing, possibly the small version over my master version). > Now the colour side of things.. this bit is actually short. > > I made a swatch of colours, RGBYMC and a greyscale, including a headshot > of > a person under 'natural daylight' on 'normal' camera settings and bunged > these together as a 'Shirley' and sent it to print, setting the media to > gloss generic, photo quality. > > Look at the result under a decent light. Is it too red? go back into the > printer settings and change the 'gloss generic' setting and dial back the > red. Was it too pale (not enough ink)?, increase the saturation. Tweak > away and save it. > > Note: Some may view this a laborious and costly exercise, but potholes in > the road cannot be filled in with by the Transport Minister simply sending > an email - someone still needs to pick up a shovel. Same same as thinking > you can calibrate without using any ink.. sorry. > > Once you're happy with the result, save it. Repeat with the matte paper. > > (Now I know that with Epsons if I do this, it works, irrespective of the > paper brand - the images will all look the same. However if I tell the > Epson I'm using Iford paper for one print and Fuji for the next, the > colours > will be wildly different. Why do you do that Epson - it's crazy! If we > go > popping in the brand and style for each paper type on an Epson, we'd have > to > calibrate each one.. total madness.. but a good way for Epson to sell > profiles! Epson really got to annoying me..) Again, Epson doesn't sell profiles. And I don't even know what you mean by saying "I tell the Epson I'm using Ilford paper". The only meaning I can see is that you're using an Ilford profile? But that should be in Photoshop, not in the printer (at least that's the preferred workflow for everybody I've talked to about this). > Next, grab one of your unaltered digicam images at random , using these > print settings above, send it to print. Sit back and be amazed at how > good > the image looks. > > I've had people who have every calibration device known to man, every > software calibration tool loaded and follow meticulously all the chants > and > arm gestures. They've risked everything and tried this method only to > discover all their past image adjustment was simply fighting both ends of > the snake. Consider: you calibrate a profile for importing pics, you > alter > the way it looks by having a monitor profile 'adjust' the image to look > good, you have another calibration profile adjusting the printer.. now, > should one go wrong - which is it? Could it be they're all adjusting each > other to give you what you had in the first place? (and that's not even > going anywhere near the 'hidden' profiling done by the OS, browsers and > graphics cards) Not if you're doing it right. The opposite, in fact -- with proper profiling, what's going on is each device is coming as close as it can to properly representing the colors recorded in a file. If you start nudging each one individually, *that's* when you end up chasing your tail. Your approach also makes your files unique to you, not usable by anybody else. Because they encode all the adjustments you've made to your personal environment. > Next step is to shut down ALL image tampering your PC/Unix/Mac/Other does > and calibrate the screen - use software or the monitor hardware, no one > cares.. but make the screen match the print as closely as possible. > > Looks pretty OK, hey? <glares balefully at the Eye-One> > > Don't fret that your monitor is now 'off'. Odds are that 90% of the folks > viewing your images on the web do not have their monitor calibrated and > even > if they do it'll still not look anything like yours. Different phosphors, > video cards, Shell enhancements, browsers - they all impact colour. If it > looks OK to your eyes, it's OK. > > That's it. > > If you plan on using Joe Bloggs Printing Services to have Pegasys prints > made, no bother - make an 8x10 filled with 4x5 rows of those shirleys on > the > same page, but tweek each to create a colour ring (0 correction, +5 Red, > +5Red+5Blue, etc) Flatten it and save as a jpg 100%. When you first > send > these guys a digital file, make it this one - check the print when it > comes > back - if say the +10Green+5Blue looks best, before you send these guys > your > images to print batch apply these colour adjustments to your files and > send > them off - you've just applied a profile to their printer :) > > > There is no dark art to this, no hardware required, just a set of eyes and > a > change of thinking. Try it and see.. This is where we were before color management became available. I have to say that color management works monumentally better than this for me. -- David Dyer-Bennet, dd-b@xxxxxxxx; http://dd-b.net/ Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/ Photos: http://dd-b.net/photography/gallery/ Dragaera: http://dragaera.info
