DevilN Details

Man I can’t believe that the Lakers so totally blew it tonight, I’m bummed and am afraid that they are TOAST!

Now back to the topic at hand. Scaling this image (the one in my post on June 1st) turned out to be a good test for a re-evaluation of several scaling techniques. This time though I only did a 3 way comparison:

1. Digital Outback Photo’s (DOP) upsizing plugin
2. Photoshop’s bicubic smoother
3. Qimage

For this test I scaled my image up to 24×36″ (which is quite a bit larger than I can actually print).  There was very little difference between bicubic and the DOP plugin. They both looked ok, with no noticeable jaggies or objectionable artifacts. However, the output from Qimage did actually look better, it was sharper, a little better defined, and offered more contrast. I suspect that it was a combination of Qimage’s scaling and sharpening algorithms that made the difference. I haven’t gone ahead and purchased Qimage yet, but I’m thinking that I probably will now that I can actually see that it is doing better than the other techniques I’ve been using. What’s kinda holding me back from just buying it straight away is that I don’t particularly care for its interface. Every time that I go to use this program I find myself struggling with simple things like rotating the image or deciding on which scaling algorithm to use. To me the UI is confusing and difficult to use. I also am not so thrilled about having to go to another program outside of Photoshop to get my image printed. If and when I get around to buying the program I’m sure I’ll have more to say about it here at my DevilN Details blog.

I finally made  a successful print of my second portfolio image using Qimage on Tuesday. Everything worked out well and I didn’t find any problems in this second print. Next up will be the third portfolio image which you can see in the middle of my post on May 3rd. I’m hoping that once the three of these prints are finished that they will look well in a tryptich arrangement.

Finally, on another note I ordered a new lens yesterday, the Nikon 14-24 f/2.8 which is supposed to be the most awesome wide angle lens currently made. I’ll be doing some tests and comparing it to my 12-24 f/4 to see if it is worth the $600 premium. I also am expecting my first set of piezography inks to arrive next week. I might have a thing or two to say about them after I’ve had a chance to work with these new toys for a while.

Yesterday I finished print number two, which was a long time in coming. I actually made what I thought was the final version of print number two about a week ago. But after letting it sit out to dry for a day I looked at the image closely before I boxed it up. On close inspection I could see some digital artifacts in some of the rock detail that just weren’t visible to me in the smaller prints. These image flaws made some of the very abstract detail in the rocks look harsh and brittle (very digital-ly looking, now there’s a word for you). I didn’t like this at all and felt that this was something that I wouldn’t really be happy giving to someone else, so I went back to the computer to see what went wrong.

It turns out that the problem was twofold. While working on the image in Photoshop I tend to go no larger than 50% magnification (unless something requires very small scale editing). I had applied at least 2 rounds of sharpening to the image as I worked on it. This alone wasn’t a problem, as the image looked fine as 6×10″ and I’m pretty sure it still would have been fine up to 11×17″. The images for this portfolio are being scaled up to 14×21″ and on top of that the scaling algorithm that I was using does another final round of output sharpening as part of this enlargement process. It was this final round of sharpening after scaling the image up that did this print in.

Digitally enlarging an image is a pretty interesting process that can be done in any number of ways. Choosing the best way to scale an image has become a good deal easier in recent years. Back in the “old days” pre 2004 we used either Photoshop’s bicubic method or spent some money and purchased a program like Genuine Fractals (GF), which was a very popular program for doing these kinds of things. Before Photoshop CS/CS2, Genuine Fractals was noticeably better than Photoshop’s bicubic method. PS CS3’s bicubic smoother and sharper scaling algorithms are much better nowadays especially for things that are likely only going to be scaled from a 10+ megapixel camera up to a 17×22″ print size. There are also tools like AlienSkin Blow Up, and Qimage. A couple of years ago I looked closely at all of these products (and several techniques for scaling using Photoshop tools) and was never really able to produce demonstrably better enlargements (of my images, to my desired size) with these other programs than I could with what was already in Photoshop. So I just stuck with Photoshop’s bicubic smoother, or the upsizing routine from the folks at Digital Outback Photo.

I think that in my case, I just didn’t have a need to up rez my photos to the point that these more expensive (and hopefully more advanced) scaling algorithms had a chance to show their stuff. In this particular instance it was the Digital Outback Photo upsizing method that caused print number two to be botched. I’m pretty sure that had it not applied that final bit of sharpening the image would have looked ok. With that discovery I resized the image simply using Photoshop’s bicubic smoother and it looked fine, no longer harsh and brittle. I also decided that I’d take another look at Qimage now that I had an image that was actually kinda difficult to scale to the size that I wanted.

Qimage is a program that has a very avid following out there in the inkjet printer community. Qimage not only provides several advanced scaling algorithms, but also employs its own “intelligent” form of output sharpening, and also a page layout facility (which automates the most efficient placement of multiple images on a single page). Oh I forgot to mention that Qimage is also a fully color managed image viewer and print spooler too! Over the years I’ve wondered what it is about this program that causes so many of its users to scream its virtues at the top of their lungs, where I couldn’t see any real improvement over the scaling that I performed in Photoshop.

Its getting late and I’m pretty tired, so I’ll have to continue this post tomorrow…zzzz

I wonder if this feeling of self doubt is common to those who are trying to put something together like I am with this portfolio. Frequently when I get back from a shoot, I download my images and run through them quickly. As I am shooting I take note of the ones that I think are going to be especially interesting look at them first when reviewing the shots on the computer for the first time. Sometimes I’m right about what was a good capture and other times I’m pretty wrong. There are other images that take quite some time to grow on me, for me to notice something especially interesting about them, or for the potential of the image to be worked into something more than the shapes and light that were present when I tripped the shutter.

The very nature of many of the images that I’ve selected for this portfolio are such that I only had some general idea of what they would look like as I opened the shutter for any where from 0.75 to 5 seconds while the waves were crashing over the rocks. The three photos that will make up the triptych I am working on are examples of this delayed type of realization. It was only after spending a fair amount of time reviewing those images that I had shot several months prior to beginning this portfolio project when it dawned on me that some of the many photos of this rock formation could possibly fit together forming a single piece. I most always have a tendency to tackle the toughest problems first when working, so I’ve put of the photos I am more sure about until later in the project.

With the first photo printed I began editing the second image today (it is at the top of this post). In the preceding months I had already done some quick (maybe an hour total over the 5 months since I first shot it) and dirty tweaks of the image just to get myself an idea of what it might look like in final form. Trying to bring it into final form can be a fairly slow process for me. Today alone I’ve probably spent about 5 hours working on this photo. This involves doing a fresh conversion from the RAW color photo. Then doing a second conversion from color into black and white, followed by the tweaking of luminance curves, dodging and burning the rocks and clouds, and making around 10 test prints today. I don’t really know if I’m pitifully slow when working on an image, but there are so many things to consider: how much contrast is needed, where to set the black point, is there enough shadow/highlight detail? I spend a lot of time considering so many aspects of each photo, and then when I begin to make the print there is still even more time that goes into things like ink tone and paper choice. In this instance I have to keep things consistent across the portfolio so those two things are mostly decided already. But with this being the only the second image to be printed there is still time to change my mind I guess.

Here is the problem for me.  After spending so many hours staring at the same image, concentrating on all the areas that I see as potential problems I begin to doubt the image’s worth. I guess I kinda get tired of looking at it. So for a while there this evening it seemed as if every test print that I made was coming out too flat, too dull, too tired. In reality I guess I’m the one who is tired, but it is at this point that I really begin to wonder if I’m just wasting my time. Is anyone ever going to even begin to appreciate what I see in these images? Making art is a risky proposition, actually producing something and then putting it out there for the world to see is taking a chance. Maybe there is nothing at all there in what I’m doing. I certainly don’t expect any other person on the planet to sit there looking at what I produce for 5+ hours in a single day. Do the master’s of the art world go through the same thing when they’re producing? I certainly know that the creation of a painting or sculpture can take even longer than what I’m doing right now. What drives them, what is driving me to do this with no guarantee of reward or even recognition? The question isn’t what is art, it’s why is art?

Ten, twelve maybe 14 more to go. Today I finally buckled down and made some concrete decisions about how the portfolio would be printed. This involved sticking with my chosen paper (VFA) and deciding on how I would tone the inks for this series of prints. I had decided some time ago that I would start with the most difficult image in the group of candidates for the portfolio. The image that I’ve included with this post is pretty hard to appreciate on the web, or for that matter at any size less than 6×10″. It is the image that I’m probably the least sure of in this group of photographs. The reason that I kept in the portfolio along with everything else is because I think that it will work when combined with two others in order to make a triptych. When printed on a textured matte paper like VFA this photo has a certain type of subtle transition between paper white and the lightest shades of gray that make up the clouded waves.

The clouds in the waves that make up the images in this portfolio have inspired me to name this body of work Aqua-Valent which is a sort of play on the work of a rather famous early 20th centurty photographer.

In confirming my paper choice I printed several small copies of the image also on Hahnemuhle Photo Rag 308 (HPR), H. Museum Etching and H. German Etching. Even though I really love the HGE it didn’t work as well with this image because the highlights suffered from a lack of separation within the upper 5% of image density. Museum Etching simply came across as a little flat when compared to the others, and its shadows seemed to suffer in a similar way as the highlights did on HGE. That left it to the pair of VFA and HPR, which are without a doubt two of the most highly regarded matte papers out there. And for good reason, the image really did look best on these two, VFA is able to generate a superior black to HPR, however as I’m sure anyone would notice, there just isn’t much black in this image at all. The biggest differences between HPR and VFA are the paper color and the texture. VFA is a bright white paper, while HPR has a creamier color close to that of a natural paper wihtout OBAs. HPR still has OBAs nonetheless and that is the reason why it looked superior to the Museum Etching paper.

In the end VFA won out because of the fact that I really like the way that this image (and I expect the others also) look on textured paper. I had envisioned that this image would be rendered on the matte paper in such a way as to look as if it were almost painted on. That effect is more readily achieved with the textured surface of VFA or even HGE. This image when printed large on a sheet of VFA came with the look I was after. I moved the white point of the image down so that small parts of the clouds would be clipped to paper white. The lightest gray shades then are allowed to more obviously stand out from the paper white, giving those paper white areas a painterly feel, while the texture of the paper kinda sorta mimics the texture that a brush would impart upon oil based paints applied to canvas.

Choosing a color tone for the b/w images was one of the easier decisions to make, well at least it was that way since I had to put aside building my own ink mixtures in QTR. For some reason I never seem to want to make my prints as a straight up neutral b/w. In previous posts I had described my quest for a certain type of sepia/selenium split effect. I also like my images printed up as slightly warm with the Epson ABW driver (H:5, V:5), to a deeper (H:10, V:10). While fooling around with toning the ABW driver I discovered an interesting combination (H:19, V:-33) that works quite well on the creamy or natural papers without OBAs. However this last combination didn’t seem to fit my image when printed on VFA. I finally wound up trying an ABW setting suggested by Clayton Jones in one of his paper chase articles. Using H:2, V:8 gives a slightly warm rendition that seems to work in a pleasing fashion with the brightened white of VFA. My test print using this setting came out looking pretty nice, not neutral gray, but also not too warm either. After examining the test print’s tone, shadows and highlight separation with this setting I was satisfied. With that last decision made I was able to prep and print the first image that will make up my first portfolio Aqua-Valent.

The past week has been pretty busy, and not much has happened with the portfolio work in some time. The time that I’ve had to spend on photography has been centered around getting other prints made for sale, and struggling with paper and ink issues. I went to make a final print on a paper that I had done extensive testing with (Ilford Galerie Gold Fibre Silk(GFS)) on 8.5 x 11″ paper. I ordered a box of 13 x 19″ paper just for this particular print, this was the first time I had tried the paper in anything other than letter size.

I recently began reading the book Real World Color Management, along with several other web sites on the topic. All of this reading had reminded me that I never did a thorough analysis of the various media settings available when creating a paper profile and printing an image. So I spent a couple of days making test prints and waiting for them to dry before selecting the best paper type to use when printing on GFS. It turns out that the media setting I used when I first created a profile for this paper wasn’t really the best one (though it did offer the deepest blacks). So with my newly found paper setting I created a new profile for GFS and that is where the problems started. After printing the test sheets and allowing them to dry for 6 hours or so, I took a look at them under a strong light and with my glasses on. And I found something I had never seen before on this paper… the dreaded pizza wheel marks from my Epson printer. Pizza wheel marks are created by the feed mechanism Epson uses in their lower end (if you can call the 3800 that) printers without a vacuum suction transport. What pulls the paper through the printer are two sets of opposed rollers, rubber on the bottom and spiked wheels on the top. These spiked wheels sometimes dig into the ink and paper and leave columns of tracks that are affectionately known as pizza wheel marks.

Normally these marks are only visible when looking very closely at the print under direct lighting at a specific angle. These marks only show up on glossy type papers. In this instance I had not seen them in my initial evaluations of the paper, and was fairly annoyed that they’ve only showed up just as I’m preparing a print for a client. So I went back and made a few changes to the paper thickness settings hoping that would alleviate the problem. I wound up picking a paper thickness of 0.5mm and a platen gap of “wider”. This seemed to get rid of the marks. With my new best possible profile for GFS in hand I made what I thought was the final print of the ForestFlora5 image. My initial inspection under the light showed nothing wrong with the print, so I boxed it up to let it sit for 24 hours before its final spray.

A couple of days later I removed the print and put it up on the line to shoot it with a dose of Premiere Art Print Shield which is a spray designed to add a layer of scuff protection and lightfastness to inkjet images. Once on the line I could see two small white spots on the print. Hoping that they were just dust or something on the surface I backtapped, then blew, and finally brushed the print with my hand. Nope, the ink had not adhered to the paper in two pinprick sized spots. Before I printed on this sheet, I cleaned it as I normally do with an anti-static brush (with orthogonal brush strokes). Then I blasted the paper with a can of compressed air as a final cleaning measure before putting it into the printer. This sequence of prepping the paper has worked quite well for me over the years, and I rarely ever get flakes on my cotton rag papers, let alone leave dust on a glossy type paper before printing. Assuming that somehow I just had some bad luck and some piece of detritus had found its way on to the paper just before printing I grumbled about wasted time and went back to make a new print.

On my second “final” print I followed the same paper cleaning procedure. Then after the printer was finished I carefully examined the print under my light (with my glasses on). And sure enough there were two other small defects in the image, these weren’t missing inks but defects in the paper (as I now suspect the first problems were). These defects are quite small less than a square millimeter in area and practically impossible to see on the paper before making the print. By now I was getting kinda mad and kicking myself for trying to do a job on a new box of paper in a size I had never tried before. I sat down and made a third “final” print and hopped that there would be no issues with this one. The third sheet was good, and suffered from no paper defects. It is now dry, sprayed and bagged, ready for delivery.

When using new products, you never know what is going to happen, or trip you up. I sincerely hope that these problems were just isolated incidents since I really like this paper’s look and cost. However, the issue with the pizza wheel marks remain, and my search for the perfect photo black type of paper for me might not be over after all.

I’ve been out of town for the past few days camping near Lake Cachuma with the family and my son’s scout troop. But just before I left my computer gremlins left another present for me. My new hard drive to replace the failed mirror drive started reporting errors on Thursday night. Of course this angered me to no end. So my time on Thursday was spent trying to figure out what was wrong, or if I just have bad luck (more on the luck thing later). I was all set to just rip the drive out and wait for a replacement from Western Digital (this is an RE2 drive which is supposed to be very reliable). When I thought that maybe there could be a hardware problem in my drive subsystem. So I ran SpinRite 6 on this drive all night long in mode 4 which is the deep cleaning and analysis mode. In the morning it reported that there were no errors on the drive.

When we returned from the camping trip yesterday I was pretty perplexed about what to do about the drive system. For the time being I’ve just replaced the SATA cable, and re-added it to the mirror. The drive has been functioning properly yesterday and today with no errors. I’m keeping my fingers crossed, but if the system once again starts logging errors on that drive while SpinRite claims that it is ok, I may have to look into getting a standalone RAID controller (which would totally suck).

Today when I got home my Epson 1400 printer had arrived. I had ordered one that was supposedly a “like new” printer that just was in an open box. Well that was being a bit generous. The printer has obviously been used, didn’t arrive in its original Epson box, was missing the install CD, and one CD printing accessory. I suspect that this may have been a floor model at the store I bought it from since there are signs of use in the form of light scratches on the cover and buttons. To make matters worse the printer didn’t work when I took it out of the box. It reported that the black ink cartridge was empty (even though the cartridge was new and in a sealed wrapper). After talking with Epson they said that the only way to tell if the problem was in the printer or a defective cartridge, was to put in a new cartridge. They shipped me one for free but it won’t get here ’till Friday. I couldn’t wait that long to find out if the printer really has a problem so I went to Frys Electronics and bought a replacement cart. Luckily the printer immediately fired up with the new cartridge, so maybe I just had a mid sized amount of bad luck going for me today. Luckily Epson does a pretty darned good job of standing by the warranty on their printers and is always ready to ship a replacement part (or whole printer if necessary) when something goes wrong.

With all the disk/printer problems in the past few days I haven’t done any work on my portfolio. I just wonder why the hardware problems are around every corner of my life when I’m trying to work on this portfolio, its like the gods are frowning on me for even attempting to do this…

My first commissioned art piece was given the go ahead today. It is a print of the image above, that was originally part of a set I did for a gallery show last year. None of the pieces in that show sold, but now finally someone has said hey, I like that, I want that, I’ll give you money for that! Maybe there will be more printed for her in the future, maybe things will start to happen and I’ll finally have this whole photography thing start bringing in a little money to offset all my expenses that I’ve racked up over the past few years. Its funny that the work I’m doing for my portfolio moved away from these flora type images. Looking at the copy of this image in the browser doesn’t quite do it justice (it may be that the greens don’t properly fit in the sRGB space), but at least it’ll give you an idea of what I’m talking about.

Actually making a real effort to market my photographs has been on my todo list for quite sometime. I guess I’m lucky that something is selling considering I’ve done nothing other than that gallery show last year in order to get my images out for people to see. At some point I am going to get a professional web site designer to build something for my artwork; but I keep dragging my feet on that issue because it is something I should be able to do myself. The problem is making the time to actually get it done.

Before I get to Piezography, a little bit about this blog. I don’t really know if the issues I’m having with this blog are related to my hosting provider, problems with WordPress plugins, or simply my own misunderstandings about how they are all supposed to work together. None of the statistics for my website or my blog show any activity at all. Now it could be that almost no one has found this yet, but I’ve had a friend already thank me for listing him on my blogroll, and the only way he would have known that I had done so was by visiting by blog. If any of you WordPress experts out there have a clue what might be wrong please let me know.

I’ve received two sets of sample prints made with the Piezography system and they look pretty fantastic! Good enough that I’m going to pick up a new printer so I can try them out myself. Its kinda hard to say why they are better (maybe they aren’t, just sufficiently different in a pleasing way) than the b/w prints I’m making with the Epson ABW driver on my 3800. I think one thing about them that I can put my finger on is that they do seem to offer greater resolution. The Piezography website has a demonstration of this using 1pt text from Alice in Wonderland. When I first saw it I was pretty skeptical but after holding a few of their prints in hand I guess they must have a point. Its kinda hard to be sure without printing the same image with both systems, but I’m intrigued enough to plunk down some money in order to give it a try. I’ll be getting the Epson 1400 I mentioned a few days ago, but probably won’t get the Piezography inks right away. Currently the Piezography system for the 1400 only works with a continuous inkflow system (CIS). Which means that there are no refillable cartridges for this printer yet (nor are there any for my 3800). There is a possibility that new carts will be coming soon, so I’ll wait a little while in the hopes I can get some. If not then I’ll wind up paying more for the inks and CIS than for the printer itself. I’m pretty excited about getting to work with the Piezography system, I’ve been reading about it for years, hoping to see why this is such a big deal.

Back to my own work. I’ve put my whole selenium QTR curve project aside for now. Tonight I just concentrated on printing a couple of b/w images exploring what I can do with the Epson inks and to see if I can tell where the difference lies between them (Epson) and the Piezography inks. I didn’t really come to any conclusions, and I think that to the untrained eye there probably wouldn’t be very much difference at all. I guess one other thing that really surprised me about the Piezography prints is how subtle their color tones are. Their selenium tone ink is just barely perceptible to me, while the sepia is certainly more noticeable, it’s not at all like what most of us think of what an old sepia toned photograph would look like. I printed the image at the top of this posting several times tonight with several variations in color toning applied, or with different printing profiles. After quizzing my daughter about the differences between them I’m left with an observation no more concrete than this: photographers must be nuts to obsess over the minutiae of toning, metamerism, neutrality, or what ever else may be the topic of the day.

when you’re colorblind. I spent way too much time fooling around in QTR making my own ink curve definitions. I’ve picked up where I left off over a week ago in trying to craft my own ink blend that will give me deep purple shadows so I can try and emulate the most awesome sepia/selenium split that I’ve ever seen. The problem is that while I can usually see differences between colors, I’m unable to recognize why they are different. This problem is especially acute for me when trying to isolate colors like I’m doing while building these curve descriptors for QTR. The process works as follows:

• Start off with an ink descriptor file that someone has already built for a paper and ink combo which is similar to what you’re trying to develop an ink descriptor for. Unfortunately for me, I’ve been unable to find anyone whose done this already on any type of paper at all.
• Determine the ink limits for your black and gray inks, and then partition them such that the inks overlap, but the lighter ones fall off rapidly as the darker shades of gray build up their densities.
• Apply your colored inks Cyan, Magenta, Yellow, Light Cyan and Light Magenta in the appropriate percentages such that they together produce the color and tone you are after.

Just by kinda guessing and having a slight familiarity with ink curves I was able to make a purple shadow tone with out too much trouble, but the 3/4 and half tones were all over the place ranging from gray to blue, to pink, to purple, to warm. Its been a pretty rough day trying to figure out what percentages of CMYK (cyan, magenta, yellow and black) inks make the shade of purple I was after. And then have that purple gracefully evaporate into paper white. My brain hurts and I’m tired now. Several times I was ready to throw in the towel because I just couldn’t figure out how to control some particular peak or valley of warmth, or cool.

Eventually I hit upon an idea for figuring out the rough percentages of the inks to build my particular cherry/plum flavor of selenium. I went in to Photoshop and brought up the color picker, then selected a close approximation of this purple on the coarse vertical color ramp. Then in the fine grained color box I found the particular shade of shadow I was after. At this point the Photoshop color picker displays the numerical values for the selected color in RGB, HSL, Lab and yes CMYK. Next I filled a canvas with this deep purple over solid black and applied a linear gradient allowing for a graceful and steep ramp up from deep purple to max black. Finally I applied a layer of white above this, and then using another linear gradient mask applied over the white I now had a completely linear progression from pure white through deep purple and finally to max black. Saving this file I can now just open it in Photoshop and point to any place in the file, where the info tab will show me the CMYK percentages that make up that particular tone of my cherry/plum selenium.

Well that was all well and good, it allowed me to get my selenium curve to about 80% of where I wanted it to be. Smoothing out the bumps on my tone ramp and taming the warmth in the mid tones has been exasperating today. Knowing the CMYK values in Photoshop is only a rough approximation, that combined with the Light Cyan and Light Magenta inks (throwing off the percentages) has only got me close. I feel like my color perception problems are really making this difficult for me, and that I may be wasting my time with this (agggh, but I’m so close). And I’m not printing my portfolio, I’m printing pages and pages of test strips

So I’m attending this lecture series at the Julia Dean photo workshops that are hosted by Aline Smithson. On Thursday night I got to meet another extremely cool photographer Melanie Pullen. She’s achieved a good deal of notice in the photography world with her High Fashion Crime Scenes photographs. I had seen her work here and there before, but never really paid much attention to it until last night. Well it was just great to hear her tell the stories of how she came up on the idea for the series and then all the crazy, brazen things that she did in order to get her shoots done.

I’m sure that a lot of you other photographers out there have run into problems with people trying to prevent you from photographing whatever it is you might be interested in (I’ve certainly been hassled a lot). Well Melanie has on so many occasions just managed to talk her way through whatever obstacles were placed in front of her, whether it was policemen, highly agitated subway train conductors or just obstinant people with a chip on their shoulders. She’s done things like convince someone that her tripod, wasn’t really a tripod, but a “unipod” or some such thing that wasn’t explicitly disallowed by whatever set of made up rules this person was trying to enforce. The point of all this is that maybe we shouldn’t be so ready to pack up and go home just because some worry wart has nothing better to do than bother you for doing something that causes no one any harm at all.

Getting to hear from other photographers about what has worked for them (and also what hasn’t) is a worthwhile experience. There were several others in this lecture series that were also great to hear from:

• Ann Cutting
• Gil Garcetti (Way cooler and more interesting than I expected)
• Stephen Berkmen