The site being DIY, I should think that reproduction is the major interest. That being said I have no trouble with discussion of either end of the chain. Just so long as we understand that the two things are completely separate - but equal.
On the production side everything is right; distortion, coloration, ... whatever the artist or producer wants is right, and only his or her opinion of the final art matters. Of course public opinion may determine financial success, but no art should ever be judged on its commercial success.
Quite the opposite, reproduction should never be judged by a personal opinion and accuracy can be measured and quantified. The perception of the masses is what counts so long as nuisance variables have been accounted for. That means double blind testing in controlled settings.
Unless we learn to keep these two areas of interest clearly defined and demarcated we can have meaningful discussions. Mix them together in a discussion and it will soon cease to make sense or have any validity.
I always view it that if someone states 'I prefer' then fine as long as they don't try to claim that their way is the one truth with zero evidence to back it up.
Interestingly three people on here have claimed my speakers are rubbish and that is why I can't hear magic interconnects etc. In each case they haven't bothered to find out what I am using before trashing it.
Interestingly three people on here have claimed my speakers are rubbish and that is why I can't hear magic interconnects etc. In each case they haven't bothered to find out what I am using before trashing it.
+1 I just wish I could put things as succinctly as you do Earl.
While I am sympathetic to this argument in principle and in spirit, it mostly doesn't work that way in the real world. I know, because I've spent a lot of time doing that. It's an imortant method to an end, but the end often isn't where you'd like it to be. No camera or printer is completely accurate, so we choose what parts of the accuracy we think are important.
Yes, it is. I learned those skills in Paris and honed them in the US.
Firstly; most film, and now digital, isn't accurate and isn't meant to be. It's meant to be pleasing to look at and sell photos. There are ways around that, but I won't go into those here. No photography or printer is fully accurate, you have to choose what you can live with and what you can't.
We could routinely make proofs that the artist could not distinguish from the original. Fairly easy with watercolors, very difficult with oil, acrylics are somewhere in between. But even when we fooled the artitst, it's rarely what they wanted. No, they wanted it to "pop" more. More contrast, more saturation, sharper. They wanted it better than the original. That bothered me for some time because I worked so hard for accuracy, but there are good reasons for it - some have to do with the faults of the medium, some have to do with ego. No matter how you measure and analyze it, the copy is never perfect and often lacks a certain "feel" of the original. That could be texture, reflectivity or other things. You have to make that up in subtle tweaks to the reproduction to achieve the same feel, to convey the same emotion.
So in making a fine art print, accuracy is important and needed - you can't be too far off or your work will be rejected. But the goal is rarely 100% accuracy. It isn't what the client wants and they won't pay for it. Some of that is because you simply can't be 100% accurate, so you fudge a little here, enhance a little there, to make up for the fact that the mimic isn't completely true to the original. Interestingly the closer the mediums, the more often a very accurate print will be accepted. Oil paint is an extremely rich medium and hard to match for "feel" even if you do match the color. It takes a lot of time and tweaking.
Accuracy or truth in images is important, but it's only someone's definition of accuracy, a choice of what to measure and what to match. It's still a kludge, though a highly skilled one. But then it has to sell, and that's a another matter.
It may be too late, consumers are already modifying playback to taste: DSP/Effect plugins for Winamp Media Player - WinampHeritage.com
I have a friend who listens to all music through a vintage spring reverb and he likes classical as well as 60's - 80's rock.
Never saw that comment, the pigments are the same but the vehicles are very different any idea why there would be an extreme difference? I'm a fan of Han blue (BaCuSi4O10) but the manufacture is long and sensitive to the temperature profile. Peacock Blue as a pure pigment is essentially lost but the difference to the eye on all of them is dramatic. I do know the particle size makes a dramatic difference in the perception of the paint color. lately I've been buying pure chemical pigments or ground mineral pigments from around the world and mixing my own paint.
Last edited:
I don't really know Scott, except in somewhat general terms. Pigments are very complex, be they bound in water, oil or acrylic.
When I was printing with the old Iris printer it was simple 4 color dye ink. The results were uncanny with water color and even pencil. The imitation of oil or acrylic was spectacular, but not as convincing as the watercolor. Watercolor on Arches paper could be almost impossible to tell from the original. Sometimes I had to use a loupe to look for the dot pattern of the print.
Moving to the Epson pigment CcMmYKk printers just wasn't the same. Beautiful printers and reliable, but much harder to get results as convincing, even with the extra color gamut. Oil is so difficult to get right. Maybe it is the oil itself, or some other carrier or additive that give it its luminosity and depth, it's richness. Some of what is so difficult to imitate is that richness and luminosity and I think that's one of the reasons we had to "pump it up" to get the same feeling out of the pigment inks as oil.
That's what I was hoping to explain in my previous post - reproduction isn't exact, but it can get amazingly close. Science is a fast and consistent way to get very close. It's those last few percent that are so devilishly difficult. Not unlike audio.
When I was printing with the old Iris printer it was simple 4 color dye ink. The results were uncanny with water color and even pencil. The imitation of oil or acrylic was spectacular, but not as convincing as the watercolor. Watercolor on Arches paper could be almost impossible to tell from the original. Sometimes I had to use a loupe to look for the dot pattern of the print.
Moving to the Epson pigment CcMmYKk printers just wasn't the same. Beautiful printers and reliable, but much harder to get results as convincing, even with the extra color gamut. Oil is so difficult to get right. Maybe it is the oil itself, or some other carrier or additive that give it its luminosity and depth, it's richness. Some of what is so difficult to imitate is that richness and luminosity and I think that's one of the reasons we had to "pump it up" to get the same feeling out of the pigment inks as oil.
That's what I was hoping to explain in my previous post - reproduction isn't exact, but it can get amazingly close. Science is a fast and consistent way to get very close. It's those last few percent that are so devilishly difficult. Not unlike audio.
My damascus moment on printing was over the work of Ansel Adams. I'd read all about his stuff, but when I got a book of his prints was underwhelmed. A few years back got a chance to see his original prints on display in UK so jumped at the chance. And realised the problem with the book was that they just couldn't match the dynamic range of his special printing process. The original prints were stunning.
A poor analogy to sound is the difference between the speakers disappearing and the room disappearing. The latter is rare in far field listening but when it occurs most enjoyable.
A poor analogy to sound is the difference between the speakers disappearing and the room disappearing. The latter is rare in far field listening but when it occurs most enjoyable.
Most DSLR are completely inaccurate.
It is possible to have a better color rendering at home, here is a extremely simplified prodcedure:
How to create a custom color profile for your DSLR - Pinkbike
Since the natural light is varying every second (clouds) this procedure should only work with artifical lights.
Yes. And of course color profiles were a large and tedious ongoing part of my job. Camera, printers, monitors. We used professional scanning 4x5 back cameras and started with as flat and neutral a scan as possible. That was never anything you'd want to print, it wasn't pretty. . But the flat and neutral scan gave the color correction crew the best pallet to work with.
There is another difficulty in shooting art for reproduction that most people don't know about - specular highlights. A huge problem. We had to cross polarize every shot - polarizing filter on the lens and on the lights. That ate about 4 stops. But you have to do it, or get tiny white specs all over the image. Those are specs you normally see, so the camera is accurate in picking them up, but they simply don't look right in a print. Why? Because they aren't dynamic, they don't change as you move, they way real highlights will. The camera shows you what's really there, but it does not look real.
There are equivalents to this in audio.
. But the flat and neutral scan gave the color correction crew the best pallet to work with. There is another difficulty in shooting art for reproduction that most people don't know about - specular highlights. A huge problem. We had to cross polarize every shot - polarizing filter on the lens and on the lights. That ate about 4 stops. But you have to do it, or get tiny white specs all over the image. Those are specs you normally see, so the camera is accurate in picking them up, but they simply don't look right in a print. Why? Because they aren't dynamic, they don't change as you move, they way real highlights will. The camera shows you what's really there, but it does not look real.
There are equivalents to this in audio.
Unfortunately the Chinese have taken advantage of this, on my last trip the amount of fake brush paintings had risen dramatically. As you might expect the print shops don't fool around the inks come in >gallon drums, they do not pay $100's for a set of those tiny cartridges. It was an eye opener to see 100's of foot rolls of paper on the Chinese equivalent of a giant Epson printer and giant bottles of ink on the floor with electric pumps running continuously printing classic paintings.
I think the luminosity is the key, from what I read that is one property controlled by the particle size and vehicle interface. One of my teachers has picked up on the totally not traditional technique of mounting and coating ink on "rice" paper (it's actually mulberry) paintings because it really pumps up the colors.
Last edited:
That is one of the major reason, yes. Professionals are not immune from fads, just like the rest of use.
24 bit recording and a 32 or 64 bit float workspace gives you huge room to work and tweak. There is so much more room than in 16 bit signed space. 192 is overkill for a final product IMO, and so is 24 bit. But for editing and effects? Amazing.
In color correction we had to work for years in 8 bits per color channel. The hardware and the software wouldn't handle 16 bit color, at least not for advanced features. In 8 bit, there are only 256 possible levels. Heavy image manipulation would result in the colors "breaking apart" as we called it. In other words, they would break up gradients and clump together like a bad jpeg. That meant huge amounts of work and masking just to keep the colors from forming clumps. In 16 bit color, there is sooooo much more room to work that colors clumping or breaking up is not much of a problem.
But for final output to a screen or a printer? 8 bit is fine. Even high quality Jpeg is OK. Just like Rebook audio or high bitrate MP3.
24/96 is probably more common than 24/192 for professional use. Basically, it sounds better and more detailed to many recording engineers, there is extra headroom for recording levels, and some effects processing sounds better.
There's also the equivalent that the digital era has never made it easier for amateurs/end users to get it "right".
The current era of inkjets have way, way wider gamut and blackpoint compared to color chemistries (can't remember on b/w, but thought dynamic range was pretty equivalent), and so many more fine art papers to work with. Metamerism is still a big problem and I still prefer my uncle's silver prints to my own work in b/w inkjet, but that could be for any number of reasons.
In the same vein, a fantastic photo is still a fantastic photo, and that's the same for music.
The current era of inkjets have way, way wider gamut and blackpoint compared to color chemistries (can't remember on b/w, but thought dynamic range was pretty equivalent), and so many more fine art papers to work with. Metamerism is still a big problem and I still prefer my uncle's silver prints to my own work in b/w inkjet, but that could be for any number of reasons.
In the same vein, a fantastic photo is still a fantastic photo, and that's the same for music.
I'll have to look into that. The cool thing about the Iris printer is that it would print on anything, it did not need to be a treated substrate. That's why we used real watercolor paper.
I printed miles and miles of canvas, but for the normal inkjet printers it needs a coating, as does paper. Think about printing on normal paper vs photo paper and that's the difference. The coated stuff is $$$ but you have to use it. I've been out of the biz for awhile, maybe the new printers do a better job on uncoated substrates.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.