• Welcome to Photrio!
    Registration is fast and free. Join today to unlock search, see fewer ads, and access all forum features.
    Click here to sign up

Epson 3800 Digital Neg Success

Recent Classifieds

Forum statistics

Threads
201,079
Messages
2,818,717
Members
100,521
Latest member
julia kan
Recent bookmarks
0

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
I've had a 3800 for a couple weeks now and have been experimenting with colorized negs using OHP and Clay's basic approach as well as RGB "black and white" negs. I was getting pretty good results with colored negs, but I was getting a little big of mid-tone grain using color R210G30 and a curve I derived using an iterative approach and a UV densitometer. The amount of grain would be invisible in an image with more texture. Tonight I got a very smooth pure palladium print using an RGB B&W negative. To make the neg, I apply my curve, invert the image, then print with +5% ink density. I got a DR of a little over 1.9 and very smooth tones. I'm not done yet, but I'm close and the results are very encouraging. I'd be happy to share my curve with anyone who has a 3800 and wants to use it as a starting point. Attached is the image I'm using for testing. It started as a Canon 5D RAW file. This is not a scan of the print - it's not quite dry yet.
 

Donsta

Member
Allowing Ads
Joined
Jun 11, 2004
Messages
191
Format
Multi Format
Kerik

I too have been spending some time experimenting and testing inkjet negatives on my 3800. Basically, following PDN, everytime I put green ink onto a negative, there was some "grittiness" - not visible in every image, but especially visible in images with very smooth tones (skin tones and open sky). Pure red ink produces prints which are exceptionally smoothly, but a regular dose of red ink only gets to about 1.6 UV density. However, if you increase the quantity of ink a little, the UV density increases dramatically. With an extra 3% ink, I was able to derive a good curve for printing with a No3S solution (6F,6Pd, 1 10%Na2). The results are very good - no trace of grain, dither or anything else. I have also used the RGB B&W negatives - they are better than two color mixes with green, but not as smooth as pure red... Give it a try.
 
Last edited by a moderator:
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
Donsta,

The tones with my RGB B&W negs are perfectly smooth with absolutely no grittiness. I chose the image I used for this specific reason. I also tried pure red and got a UV density up to about 1.8, but it completey shouldered off from about 70% on up. I could see changes in optical density on my step tablet, but the UV density didn't budge. I'll give it a try with greater ink density. But, your formula has way more Na2 than I want to standardize on. I prefer pure palladium, or, when necessary, a tiny bit of Na2. Would you be willing to share your curve?
 

Donsta

Member
Allowing Ads
Joined
Jun 11, 2004
Messages
191
Format
Multi Format
Sure Kerik - drop me an email and I will send it on (or anyone else interested). I use the 3s mix because I prefer slightly more neutral tones... However, I am pretty certain that you could get very close to pure Pd required density by just stepping up the ink density (perhaps around 10% - you'd have to test). You may also have to slow down the print head by pausing it between runs to allow the extra ink a bit of extra time to dry.

I also have been trying to use the ultrafine material from photowarehouse. However, a very interesting observation - the red ink on that substrate has far less UV density! - like 0.17. Seems really bizarre, but I presume that the reaction of the ink on the surface of Pictorico OHP and on the surface of that film produce very different chemical reactions.
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
I haven't tried ultrafine yet. That is pretty weird... Will definitely give your method a try.
 

Platonumb

Member
Allowing Ads
Joined
Mar 6, 2007
Messages
10
3800 over 2400?

Hello Kerik,
and thank-you for sharing your experience regarding the 3800,
and is it enough of an improvement over the 2400 other than the 17" width?

I felt the jump from 2200 to 2400 was a bit underwhelming for digi-neg
output.
Truly Platonumb
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
TP - I've not used a 2400, so I can't say. My previous printer was an Epson 1280. From what I've read, the dithering pattern on the 3800 is "new and improved" compared to any of the other existing Epson printers, including its bigger brothers.

A bit of an update... This weekend I taught a pt/pd workshop in my home with 5 students. We made 2 or 3 negs for each of them and they all printed extremely well. I think I probably sold a few printers for Epson this weekend.

I was really pleased to see this because up until then, all my testing had been on the one image I posted above which I chose as a torture test for tonal smoothness and because it would readily display any issues with banding and venetian blinds that until now has been a real frustration with using digital negs for me.

The curve I derived using my patent-pending Easy Digital Negative (EDN) system :smile: is giving me very nearly perfect linear response and I'm getting a UV density range of a little over 2.0. It prints pure palladium beautifully without any added contrast agent. Again, I'm printing B&W RBG negatives. The only colorizing I'm doing is adding a screen layer of G004 based on a recommendation from Clay. This bumps the UV density just a bit but doesn't cause any grittiness in the image. The tonal smoothness I'm getting is equivalent to in-camera negatives.

On top of all this happy news, the 3800 makes jaw-dropping prints. Until I save my pennies for a good RIP, I'm using the Epson Advanced B&W printing tool. I convert my images to 16-bit grayscale, which seems to work much better than RGB for this approach. After a little trial-and-error on small prints, I dial in the color that I like and save the settings. The prints look great and don't have any of the weird color transitions I was getting printing "toned" RGB images using the standard output options. It's also pretty fast. A 16x20 print takes about 6 or 7 minutes.

For my money, the 3800 is an absolute bargain considering it's relatively low price and top shelf output.
 

Platonumb

Member
Allowing Ads
Joined
Mar 6, 2007
Messages
10
Kerik, your system of using the BW-rgb printing system, gives you superior results to colorized negatives?
and how does one aquire a cd of your system?

Platonumb
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
Platonumb - Apologies... My "EDN system" was really a joke. I used a UV densitometer, a spreadsheet and some trial and error to dial in my curve. I started with colorized negs using a combination of R190 and G32. While these negs gave me good DR, the results had a slight grittiness that I didn't like. I tried to decrease the amount of Green which gave smoother tones, but not the same DR. I've also tried pure Red which gives very smooth tones with the 3800, but not a good tonal distribution in the highlights (for me - I understand pure red is working for others). So, yes, the B&W RGB negs are giving me the best digital negs I've ever used, including old-school imagesetter negs.

I'd be happy to share my curve with anyone who wants to try it on a 3800. It's not likely going to work as well on other printers...
 

Jeremy

Member
Allowing Ads
Joined
Oct 26, 2002
Messages
2,761
Location
Denton, TX
Format
Multi Format
Kerik,

I actually use B&W RGB negs for pt/pd using my 2200 and I'm loving the results (also using the EDN system--though mine involved shooting the outputted charts w/ my digital camera since I don't have a UV densitometer). Glad to hear the 3800 is kicking out such great negs for you. How do the wet plate inkjets look?
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
Glad to hear the 3800 is kicking out such great negs for you. How do the wet plate inkjets look?
Jeremy,

To die for. And I think you know my standards, I don't say that lightly.
 

Jeremy

Member
Allowing Ads
Joined
Oct 26, 2002
Messages
2,761
Location
Denton, TX
Format
Multi Format
Wow, I'd love to see some, will you be sending any down to the Watermark in Houston?
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
Yes, I'm going to send a few for them to check out with my next shipment to them. I'll let you know...
 

Davec101

Member
Allowing Ads
Joined
Jul 29, 2005
Messages
1,216
Location
Cambridge, U
Format
Large Format
This is great news Kerick, thanks for the research, i have a 3800 and will be going over to that printer from my 1290 pretty soon. From my initial observations it prints quite different to the 1290 with its dye based inks for my cyanotypes, but it seems with all these things if you put in the work you should get some results in the end.

Just a couple of questions, What type of OHP are you using at the moment, also how do your digital files from the 5D hold up when you apply the curve, is there less noise than your film based scans?

thanks
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
Davec,

I'm using the original OHP. I just bought a 17" wide roll and was surprised to find out that it's thinner than the packaged cut sheets I've been using for years.

So far the 5D files are holding up great. The image I posted at the beginning of this thread has since been output as a 16x20-ish negative and the palladium print from this neg looks fantastic. This image is about a 75% crop from the original. I upsized it a bit in Photoshop using bicubic smoother followed by a little bit of smart sharpening on the tree itself. I did some noise reduction on the foggy background using Neat Image. The foggy tones are baby-butt smooth. Only a true print-sniffing photonerd would know the difference between a print made from this neg and a print made from a film neg. I'm not about to abandon my MF and ULF cameras or film, but the Canon 5D with a good lens (24-105L) and the Epson 3800 are giving me the results from digital negs that have not previously been possible.

In general, I find film scans a bit noisier than my 5D files, but I have an older scanner and I'm not all that adapt at getting the best scans.
 

Anonymous

Member
Allowing Ads
Joined
May 14, 2003
Messages
59
Which 3800 is the one to buy? I see that there are two different printers.
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
Donald,

The only difference is one comes with the Colorburst RIP software and the other does not. Being cheap, I got the one without the RIP. So far, I haven't missed it. I know Clay uses Imageprint RIP with his 7800 and LOVES it. I've downloaded the QTR RIP, but haven't started fiddling with that yet.
 

clay

Member
Allowing Ads
Joined
Nov 21, 2002
Messages
1,335
Location
Asheville, N
Format
Multi Format
An aside about the RIPs for the X800 printers. I do love the Imageprint RIP, but my primary reason for getting it was the huge custom profile library that it gives you access to. IOW, if you plan on doing a lot color printing on just about any kind of paper, it will save you weeks of time and expense you might spend either having profiles made or making one yourself.

I previously had a very frustrating experience using the PrintFix 2 profiling hardware and software when I was using the 2200 for color printing. It literally took whole boxes of paper to get the profile built satisfactorily. The problem stemmed from the inconsistent readings that the PrintFix measurement tool was giving me. I found I had to read these 100+ square patches several times to zero in on a value. As you can imagine, the idea of doing this for more than one paper is just not something I want to spend time doing. So, on the recommendation of a fantastic color printer I know here locally, I swallowed hard and bought the Imageprint RIP. It offers a few advantages over the built in Epson driver:

1) you can tile images on your paper very efficiently and cut the amount of wasted paper you create
2) it is at least twice if not three times as fast as the Epson driver
3) it takes 16 bit files as input and (from what they tell me) uses the 16 bit values for printing - so in theory - smoooooth transitions
4) it uses its own dithering routine that to my eye is smoother than the Epson
5) the monochrome profiles that you can use are designed specifically for monochrome printing, and give you a wide range of tonality from cool to warm and the ability to create custom split and dual tones that look very much like some of split sepia/selenium toning I like to do in silver

I was able to create a very nice custom curve for the RIP for doing palladium negatives using an RGB file colorized as Kerik mentioned above in this thread (we've been talking)

That said, if the heartstopping price of the Imageprint solution gives you pause, then I heartily recommend the Quadtone RIP. It is not as slick and complete as the Imageprint RIP, but if you are willing to devote a little time, it can give excellent results. Thanks to some generous tips and help from Michael Mutmansky, I have been able to build some very nice palladium negatives using the Quadtone RIP.

The advantage this offers over the 'apply a curve' approach is that all of the curve building is done just once and embedded in your Quadtone RIP profile. So all you have to do to build a negative is just invert the file to a negative, flip it left-right, and print. Quick, easy, and no one gets hurt. By using QTR, you are able to skip some of the more perverse behavior of the Epson driver, which is designed for visual smoothness, not UV transmission density smoothness.

This, in my opinion, is the Achilles heel of the PDN approach when using the X800 printers. The PDN system is tied to the Epson driver, and it does things when laying down color that give strange UV density reversals at the high end of the color percentages. In fact, it makes it impossible to use the PDN system by the book. I managed to generate a PDN curve eventually by deviating from the by-the-book approach, but found the Quadtone RIP negatives and the Imageprint RIP negatives to be smoother and less grainy. So I punted on that approach, and am sticking to QTR for my negatives. My two cents.

Thanks Kerik, that helps.
 

Anonymous

Member
Allowing Ads
Joined
May 14, 2003
Messages
59
Clay,

Sofar as your experience making digital negs, does imageprint offer any advantage over QTR? Or vice versa? What density are you able to obtain? I know you have a larger printer than the 3800 but the inks are the same, or am I correct in this?
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
Clay sent me a 7800 QTR profile for diginegs to try out on my 3800. I printed a B&W RGB step wedge on OHP with it last night. If I did everything right, it showed me that there is plenty of density available. The 100% black gave me a UV density of around 2.7, which is significantly higher than I'm getting with the Epson driver - around 2.1. It's more than I need for palladium, but would be useful for longer scale processes. It also showed me that the 7800 and 3800 print differently, despite sharing the same inks. I didn't get anything close to linear response, but I think that will be easy to fix once I figure out how to turn the dials in QTR.

As a side note, I also used QTR to make "toned" B&W prints with the 3800 and they look really nice. I will definitely be sending Roy Harrington my 50 bucks for QTR.
 
OP
OP
Kerik

Kerik

Member
Allowing Ads
Joined
Nov 24, 2002
Messages
1,634
Location
California
Format
Large Format
I was able to create a very nice custom curve for the RIP for doing palladium negatives using an RGB file colorized as Kerik mentioned above in this thread.
Clay - Did you add the 4 points of green to your B&W RGB neg? I didn't do that on the step wedge I made and got boo-koo density.
 

clay

Member
Allowing Ads
Joined
Nov 21, 2002
Messages
1,335
Location
Asheville, N
Format
Multi Format
Depends on what you need to do with the printer. I would say that if you are pretty much doing all monochrome prints, diginegs and the occasional color print with some pretty common papers (e.g. Epson Premium Luster or Crane's Museo Silver Rag) you probably can get by just fine with the combination of QTR and the paper manufacturer's canned profiles. There is a difference in the prints between the Imageprint color prints and the canned profiles, but it is subtle. I would say if you are primarily going to be doing monochrome work, just send Roy your $50 and dive in. That said, the QTR interface is a little bit of a stone-knives-and-bearskins approach. It will require a little more user knowledge than the essentially drag-and-drop nature of Imageprint.

As far as density goes, you can get all you need with the any of the X800 printers as long as you don't have a problem using the black inks. Since there are three black inks in these printers, there is no problem getting very smooth results. The joker in the deck, in my opinion is the Yellow pigment, which Michael Mutmansky and I both think is the one that can bring the grain up. When you use the QTR ink evaluation feature and plot the UV densities of each individual ink, the yellow ink is the big-time UV blocker, much more than the black. I believe too much of it can contribute to graininess in the negative

Clay,

Sofar as your experience making digital negs, does imageprint offer any advantage over QTR? Or vice versa? What density are you able to obtain? I know you have a larger printer than the 3800 but the inks are the same, or am I correct in this?
 
Last edited by a moderator:

Anonymous

Member
Allowing Ads
Joined
May 14, 2003
Messages
59
Thanks Clay, What you say about the yellow ink makes perfect sense.
 

Anonymous

Member
Allowing Ads
Joined
May 14, 2003
Messages
59
As an aside, for those who may be interested, I just got off the phone to Epson a moment ago. I had a question about supply voltage to run the 3800 since I may operate part of the year here in the US and part of the year out of Italy.

Tech support at Epson indicated the printer will operate at 100-240 and 50/60 hz. so one printer will do double duty if needed.
 
Photrio.com contains affiliate links to products. We may receive a commission for purchases made through these links.
To read our full affiliate disclosure statement please click Here.

PHOTRIO PARTNERS EQUALLY FUNDING OUR COMMUNITY:



Ilford ADOX Freestyle Photographic Stearman Press Weldon Color Lab Blue Moon Camera & Machine
Top Bottom