IM Vivid Magenta in Epson Pro 3800 = Epson Pro 3880 performance ?

[pharmacist]

I'm wondering if I could use a particular aftermarket Epson K3 Vivid Magenta compatible ink in my Epson Pro 3800. According to the specs Epson altered the printhead to deal with this more viscous ink, but not sure how the vivid magenta equivalent from Image Specialists behaves in a Epson Pro 3800 or similar Epson printer like the Epson R2400 or R2200 ? Maybe Mikling can give me some advise about this.

 

[Bobo2006]

Cone Color (Inkjet Mall) provides a set of K3 inks for the 3800. From their web page:

"ConeColor K3 Vivid inks are compatible with the Epson 3880 but are being offered as a color space upgrade to the Epson 3800 printer. You will need to make ICC profiles for your media to accommodate the larger color space of K3 Vivid. As of this writing we do not know if the Epson 3880 ICC profiles are compatible with the Epson 3800 driver. We are suggesting the ColorMunki as a low-cost, high-quality ICC profiling product - if you buy this version of inks for your Epson 3800."

I have purchased a set of cartridges and ink but have not yet installed them as I still have ink in my OEM cartridges.

 

[mikling]

I will be testing this on my list of projects. My plan is to see whether the new magentas can be used on Claria based machines. While not on the 3800, it should be indicative of potential improvements. Not that I will bias the result but I suspect the differences might indeed be slight and possibly the accuracy and care of creating the profile will have more influence.

I generally do not recommend going away from stock balanced inksets because I have found that 1. it creates havoc when you need to use your printer and the application is not color managed and so many extra intermediary steps need to be done. 2. It could be me, but unbalanced inksets just seem so much more sensitive and extra care is required when profiling.

 

[Grandad35]

.....snip.....
2. It could be me, but unbalanced inksets just seem so much more sensitive and extra care is required when profiling.

If you think about it, each printer's internal RIP is designed around the printer's OEM ink colors to get optimum color accuracy and still allow small tweaks in the profile to compensate for different papers, etc. If an ink's color varies too far from the OEM colors, the profile must try to compensate for large color differences that are best handled in the RIP (like the color interactions between inks, the transitions as more ink is applied, the transitions from PM&PC to M&C and the transitions from 1pl to 5pl nozzles).

 

[mikling]

Yes, makes perfect sense. That is principally the reason why I often wonder when some make claims that certain inksets can increase gamut but when printed with standard drivers are too intense and not like the OEM tones. There must be a compromise somewhere in there.

At the same time, I've been able to get profiles with pigment inks on the Epson Claria family which are very very good. These same inks when printed with standard drivers will output unusable photos. The process of getting the profile required very careful testing of the correct media setting and using a high number of sample colors. On balanced inksets whose colors are very close to the OEM much fewer patches yield excellent results but not with this inkset. I would think that there are a lot of adjustments going on. However, in consideration of the end results of using a very low cost printer such as the Epson Artisan 50 ( 6 color inkset) ($40 on sale) using a pigment inkset and yielding very very good archival photos is so compelling. If I can further improve it another notch, then it would be so much more exciting.

So I would say then that your prognosis is correct that the closer the match to OEM the better the chances of getting a good profile with minimal tweaking. While all we see is a file, the level of adjustment contained within the file might be far simpler. It just makes sense.

Another thought crossed my mind when you mentioned the two sized droplets. When there are variable sized droplets, obviously the printer knows how to control the size of the droplets within the dither pattern to create the color desired. On the newer Epson heads, I believe that there are 5 different droplet sizes that can be generated from the same nozzle. When a custom profile is used, how does the interpolation and even extrapolation of the colors desired affect the way the printer outputs the colors? In a more detailed view, the decision to use what sized droplets is programmed into the internal RIP which is not modified by the profile. Could it be that when a profile is applied, the internal RIP now becomes suboptimal in its choice of what size and color droplets to use in producing a certain color? Only an engineer who has programmed the internal RIP software could probably answer this question? I suppose?

 

[mikling]

Pharmacist, can you point me to where you found the information that Epson altered the head to deal with this more viscous ink?

It could very well be true. What bothers me about this is that when you think about it, from a manufacturing and overall cost point of view it would not make sense for Epson. Why? most of what constitutes the makeup of the ink is the carrier. I'm just thinking that to modify the carrier to retain the properties ( if it is viscosity) of the inks as in previous shades of the family ( Only the Magentas were apprently tweaked) would be far simpler and much less costly than creating new tooling for just two channels of the printhead. Additonally you would want the identical carriers within the inkset to prevent differntial gloss issues as well. So on these two points my curiosity grows.

Now checking the printhead part number will not yield definitive proof because in the 3880 I believe Epson added the teflon coated undersides. Thus the part numbers for the 3800 and 3880 will be different. While Epson does not state this, I suspect that the Teflon adds to better controlled behavior when the bubble detaches from the printhead yielding better performance in terms of less splatter, better more uniform shape on the paper and finally probably most important less ink buildup on the underside of the head which does indeed create clogs long term if not cleansed.

I have gotten multiple requests for support when users encounter clogging of their printers and most times, the clog is due to ink buildup on the underside of the head. All of the printers are older models which have been used over the years and an underside cleanup solves the problem. By having a smooth underside that is teflon coated, the wiper assembly is probably more effective in keeping this clean. The matte ceramic face was too rough and actually encourage the ink to hold on thus leading to buildup.

 

[Grandad35]

......Another thought crossed my mind when you mentioned the two sized droplets. When there are variable sized droplets, obviously the printer knows how to control the size of the droplets within the dither pattern to create the color desired. On the newer Epson heads, I believe that there are 5 different droplet sizes that can be generated from the same nozzle. When a custom profile is used, how does the interpolation and even extrapolation of the colors desired affect the way the printer outputs the colors? In a more detailed view, the decision to use what sized droplets is programmed into the internal RIP which is not modified by the profile. Could it be that when a profile is applied, the internal RIP now becomes suboptimal in its choice of what size and color droplets to use in producing a certain color? Only an engineer who has programmed the internal RIP software could probably answer this question? I suppose?

This is exactly the type of problem that I was referring to. From previous posts, we know that the Canon RIPs make a gradual transition from the "Photo" M/C inks to the full strength M/C inks to prevent a step in the printed color. There is every reason to believe that the same type of transition would also be necessary for variable droplet sizes, and that this transition is fully under the control of the internal printer RIP (and therefore inaccessible to the user).

If an ink is either lighter or darker that the OEM ink at all intensities, it should be an easy thing for a profile to correct. If the ink has the same color at full intensity, but is either lighter or darker in the mid-tones, the profile has to make non-linear corrections. When you throw in the possibility of non-linear cross-interactions with the other ink colors, it's easy to see why getting a good profile is sometimes a challenge. This is probably why it is recommended to stick with a full ink set with colors formulated to work together and to avoid changing a single color.

 

[mikling]

Hear Hear, well said. I guess sticking with inksets that mimic the OEM as best possible is probably the best way to true high performance printing as far as bulk inks are concerned for a whole host of reasons.

Now that still does not stop me from extracting all I can from that lowly Epson Claria photoprinter family. I have to print some wedding photos for my niece, I'll fire one up and test it. I'll wring every drop of performance from these inexpensive printers that can be tweaked to become wicked archival printers. Beyond the Magentas I have some tricks I'd like to try and see what the results are.

 

[pharmacist]

From what I know the Epson Pro 3800 already has this particular telfon coating on its printhead, but I read on www.photo-i.co.uk that the latest Pro 3880 has some modifications on its printhead. I really don't know what this improvement is. The LUT (Lookup Table) for the colour conversion is said to be improved.

Actually with your IM-K4 inkset I purchased from you and after creating custom profiles the results are very impressive, even beter than the original Epson inks and especially saturated colours are much better without the dreadful pooling which was common with the standard Epson K3 inkset when you want to achieve the same intensity. I have one type of cheap glossy paper which absorbs the original Epson K3 ink so poorly, that I had to reduce the intensity with 20 % to neutralize the pooling, but the gamut suffers dramatically by this action and the picture came out rather dull and flat. Now I can print with my Epson Pro 3800 and your IM-K4 inkset with much better quality and without the pooling on the paper.

 

[mikling]

Pharmacist, I tried the Vivid Magenta inks with a profile made with the standard magenta and I could see that it offers potentially rosier fleshtones or even brighter reds. However, after profiling I could not tell the difference. OK there was a slight difference but that difference I could easily have dialed in or out during raw conversion. So why bother?

My take is that the time is better spent trying different papers and seeing what improvements that might yield. If you're happy with the current color balance then concentrate on the image capture and getting the shots. It's more fun.

After a certain level is reached, it's like hifi. You can dial in what your preferences are by various methods and components. In the end, it is the music we want to hear and similarly it is the image that we want to convey.

Happy shooting!