Is there a digital equivalent to a contact print

[Bob McCarthy]

I just don't know where these numbers are coming from. 300 or even 600 has nothing to do with any Epson I've printed on for over 10 years. 300 is a number that's been tossed around for no good reason for a long time, I think a holdover from prepress days as a recommended res. It's not native to any Epson, and Epsons can conditionally resolve more detail than 300 or 600. I think there may be some Canon printers with 600, not sure. Also it's not just the pixel math, lens performance of the enlarged 2&1/4 vs the 1:1 8x10 will be a factor as well as grain. 8x10 will definitely be an advantage for a variety of reasons even with a lowly inkjet.

My highest apparent resolution prints were from 10x12 negs drum scanned at 1000ppi (now we get into film resolution and aperture issues), resized in Photoshop to 720 ppi with no resample (therefore slight physical image size enlargement), and in my digitally addled brain- they are contact prints... digital.
Both input and output ppi were native to the devices, there is a direct path between the optically captured pixel to the pixel from which the ink dot dither is created.

That's it, I've bored even myself, send in the orderlies.
Tyler

The 300 and 360 dpi numbers commonly thrown about are a part of printing lore.

It was believed that 360 dpi was the magic print setting for Epsons and 300 dpi for Canon and HP printers.

There was the belief that the drivers operated at 720 dpi (Epson) and 600 dpi(Canon/HP). A whole digit divisor did no damage so 360 and 300 came into the venacular.

Actually I have followed this guideline myself. It makes a certain amount of sense as they are also whole divisors of the 2880 (Epson) and 2400 (HP/Canon) max dpi's of the printheads.

How did you derive your settings in the last paragraph. This is in line with my experiments, can you clarify your logic for me.

I have a Cezanne which will zoom the scanning lens to a specific area and focus quite nicely. Matching up native optical resolution to printer maximum was where this thread started.

bob

[Tyler Boley]

360 and 720 are well known amongst the printing community to be native Epson driver resolutions depending on model and settings (1440, 2880, finest detail, etc.) I don't know the particulars since I don't use the Epson driver. StudioPrint reports back to me 720... I can't remember where in the interface... in a blind test some detected a difference sending 1440 info as well. But other complexities come into play, perhaps 720 is native, so the RIP scales down from 1440, in some fashion that creates apparent sharpness. The entire issue is not as straightforward as it first seems.
A great app for printing, Qimage, unfortunately for PC only, also reports back the resolution of whatever printer is selected, including the effect of the driver settings. In fact it has all kinds of great features, various advanced scaling methods and sophisticated sharpening options on the fly for print size. Some great printers I know swear by it, and it's another very affordable shareware app. The author has a very informative page on the subject of scaling for print, some of the best info I've seen.
Again, native res through the process is not viable the vast majority of the time, nor necessarily worth it. But the manner in which you get from input res to final printer res can and should be a consideration, there are many methods.
I just thought it was conceptually relevant to the thread.
Tyler

[BradS]

This is because any time you change resolutions from the device's native resolution, unless you're downsampling by an even factor (like going from 1000ppi to 500ppi) you're going to be interpolating. Interpolation means using mathematical algorithms to invent entirely new pixels. There's no way to do this without at least some quality loss.

In practice, photoshop and the better rips and print drivers do a really good job of interpolation, especially when downsampling. But any time you can avoid it it's a good idea.

So, in a nutshell....anything other than the native resolution of the device degrades the image because a mathematical algorithm was used to process the raw data...is that correct?

[Brian Ellis]

I still am baffled...why would one bother to shoot 8x10 film only to scan and print with an inkjet? (or is that in essence, the OP's question too?)

I keep thinking that you give up so much when you scan and inkjet print that probably a 6x7 medium format original is good enough....or am I missing something?

You give some things up and gain others when you scan and make an ink jet print. On balance, for me and some others, it's actually a net gain rather than a loss.

[BradS]

You give some things up and gain others when you scan and make an ink jet print. On balance, for me and some others, it's actually a net gain rather than a loss.

Yes, I had neglected to consider the "I can fix it in Photoshop" aspect. I guess, like so many things...it is a subjective thing.

[Peter De Smidt]

So, in a nutshell....anything other than the native resolution of the device degrades the image because a mathematical algorithm was used to process the raw data...is that correct?

Down-sampling can increase quality. "Raw" is ambiguous. If it means "unprocessed", which it looks like your intending in this case, then no file from a digital camera or scanner is raw. Obviously there are other meanings.

[joncone@cone-editions.com]

The K7 systems come the closest.

Piezography K7 gets to exceed the particular vision of the printer manufacturer because I do not have to use the printer as intended. For arguments sake (this is a list group!), lets discard what happens in the shadows and very dark grays because the human eye has trouble discerning detail - and focus into the lighter shades where the human eye can resolve more detail.

The OEM driver uses either one or three blacks to produce a grayscale from dMin to dMax and therefore has to use some dithering in order to break up the three black shades. You can easily imagine it as a halftone with one black...where there are tiny dots of black ink with tons of paper space between them to give the brain a perception of light grays. Even with three blacks, this dithering is still necessary although the white space is reduced.

But with seven dilutions of black - which is what my Piezography K7 is - we present two additional lighter shades and two additional darker shades than the ABW 3 black system. (K7 shade 4 is equal to Epson LK and K7 shade 5 is equal to Epson LLK). As a result, I can create a "profile" or curve set for QTR that is printing the inks always in the higher frequency dithers, where the dots of ink are adjacent rather than spaced apart.

What happens is that detail that would normally end up as white space in the Epson driver (just simply not printed) is now actually represented in ink. We are able to drive the printer at a much higher "apparent" resolution.

We (anyone using the system) can actually produce a very high resolution print on a very low resolution printer when we choose (such as the older Epson 3000s, or the mid-90s 7000, 9000, etc...). Those we drove with 4 dilutions at a higher perceived resolution than did Epson. On the new printers with their tinier droplet sizes, we are getting superior results. Piezography on an Epson 1400 with six shades is still superior to ABW on an Epson 7880. And Piezography K7 on an Epson 7880 is a beautiful thing to behold.

As Tyler points out however - it does take some skill sets in order to produce fine prints - no matte the what. But, K7 gives you some extra finesse if your files are smooth. I actually shoot with both a 5D and a Blackberry. I happen to love low res Blackberry photos - I just want them to be printed with Piezography because tons more ink equates into significantly greater depth. I print probably about 20-40% more ink than does EPSON in the 1/4 tones to the highlights and you can see this as "depth" in the surface of the final K7 print.

One final thought here in terms of pixels is that the printer does not print pixels... it prints dots of ink - many of them to make up a pixel. Its interpretive algorithms for resampling are quite amazing because rather than have to recreate pixels, it utilizes new dots of ink. So, throw a ton of resolution at it through a QTR K7 curve and the result is levels of detail that can not be obtained by the Epson RGB driver because it still has to think in terms of dithering the inks. Even with a light cyan and a light magenta - or LK and LLK - it can not create it in a continuous tone.

Piezography has K, DDK, DK, LK, LLK, LLLK, LLLLK. The ability to render both continuous tone and much higher resolutions of detail is made easy by the absence of low frequency dithering. We do not need the low dithers because we have lighter shades of ink that we can print more full than can Epson.

Here is an example of a K7 curve...



The shades in the above curve go from lightest black shade 7 in the yellow position on the left to full black in the black position on the right. Each of the color inks is replaced with a progressively darker shade of black, even though in the "curve set" they are represented in the color position of the printer. The set of curves produce a continuous gray ramp that moves from 0% to 100%. The gray ramp is made by the seven overlapping shades of ink. The ink limit for each ink is at about the 40 level which is similar to that of the Epson driver. At this % - the driver is at full dithering printing a solid ink. And notice how quickly each ink ramps up and the amount of overlaps...and its easy to imagine that the only place K7 is not at a full dither (without dots of ink spaced far apart) is from 0 - 5% density which the eye would have trouble discerning considering how light our LLLLK is.

None of this is very applicable to the OEM driver - but it does throw in something - the printer driver and its ability to render dots of ink which make up a pixel. K7 can utilize more dots of ink per pixel than can ABW.

This also, is the trick behind our ability to render more detail than ABW in tiny type. It just illustrates the capability. If you're still reading to this point you can click on this link to see a super high resolution scan of Epson ABW side by side with Piezography K7 in trying to render a high resolution file which has 1pt type of the entire text of Alice in Wonderland. And if this high resolution type file was instead a highly detailed 8x10 negative - then you can see how K7 renders more resolution than does the OEM driver - even though it is the same printer.

http://www.piezography.com/PiezoPres...-to-epson-abw/

regards,

Jon Cone
Piezography

[Bob McCarthy]

Mr. Cone

This was from a previous post, can you enlighten me.

bob



Regarding Cone K7/MPS, I do have another question........

I am using a 7 ink printer (7600)which means the GLOP position is missing.

Is using a spray (I'm using Moab) which I have experience with, an equivalent to the GLOP?

[joncone@cone-editions.com]

It can do the same thing as GLOP but its not the same. The GO is printed so is very even and perfect and also does not put fine droplets of the spray in the air. So those are two pluses to using GO. But the spray can equalize the gloss and get rid of bronzing. I have used the Lumijet spray. I think that they are all the same.

Incidentally here at the studio we use the 8 ink Epson to be a matte/photo black and six shades so we have K7 matte and glossy on one printer. We use a spare 9600 with seven channels of GLoss Optimizer and print with a special curve. If you had any old 7000, 7500 or 7600 with one or more missing heads that were bad - we could make a GO curve for you and then run K7 in the good printer. Make sense?

People give away bad printers that have at least one or more good channels...

Jon

[Bob McCarthy]

It can do the same thing as GLOP but its not the same. The GO is printed so is very even and perfect and also does not put fine droplets of the spray in the air. So those are two pluses to using GO. But the spray can equalize the gloss and get rid of bronzing. I have used the Lumijet spray. I think that they are all the same.

Incidentally here at the studio we use the 8 ink Epson to be a matte/photo black and six shades so we have K7 matte and glossy on one printer. We use a spare 9600 with seven channels of GLoss Optimizer and print with a special curve. If you had any old 7000, 7500 or 7600 with one or more missing heads that were bad - we could make a GO curve for you and then run K7 in the good printer. Make sense?

People give away bad printers that have at least one or more good channels...

Jon

This printer is the one we wanted to dedicate to B&W only.

That is a clever way to apply glop in a volume shop.

Let me look for a file that I can send and have 2 prints made. One with glop and one without that I can spray to compare.

I can see the rational for your technique.

bob