Luminous Landscape Forum
Raw & Post Processing, Printing => Printing: Printers, Papers and Inks => Topic started by: BCRider on January 29, 2007, 02:17:29 am
-
Are there any DJ 130 users using Printfix Pro to build their profiles? If so, how do you find it compared to the stock profiles for the Premium Plus Satin?
I have a new DJ 130 and find the stock Satin profile to be pretty good but not perfect as it has a slight yellow and a very tiny bit of green cast. I found updated profiles (from mid last year) on the HP site for Satin (and Glossy) but sadly they weren't obviously better as they solved the green and yellow cast but introduced excess red (sigh).
So all this leads to the thought of buying Printfix Pro with the hope of improving on these profiles. Of course, having done so, I'd also like to make good profiles for the other papers. I'd buy the Eye One but it is pricey for an amater and perhaps not really any better.
Any help along these lines greatly appreciated!
Thanks,
Byron
-
Byron,
Your timing is perfect. A few months ago I wrote Fancy Graphics Galore: Some observations on Gamutvision and PrintFIX Pro which was published on this site. You may want to read that first, you can find it here:
http://luminous-landscape.com/reviews/acce..._graphics.shtml (http://luminous-landscape.com/reviews/accessories/fancy_graphics.shtml)
In the very, very near future Michael will publish an update to that article which will answer all your questions, and it will be a thrilling read as well. So please return to The Luminous Landscape in the upcoming few days, check the What's New page and be surprised! You may want to hold your buying decisions until you've read the article - it will be valuable information for you.
Gerard Kingma
www.kingma.nu (http://www.kingma.nu)
-
Yes you should probably wait until the update is put on the site.
-
Thanks for the reference to your article Gerard. Interesting reading. Seems you like the DJ 130 but weren't as pleased with Printfix Pro.
Given that I don't think I need to wait for further articles! I want something that will produce better than the stock Satin profiles and this doesn't seem to be it. Too bad, the price was almost affordable.
If others are using this combination and have different views please speak up! All input welcome!
-
Find an i1 Pro UV spectro on ebay and download and learn to use Argyll CMS.
For that matter, you could get an old Xrite Digital Swatchbook and do the same.
-
Byron,
Really, wait for the update. It contains a new perspective on PrintFIX Pro.
Greg, it won't be long - your patience will be rewarded as well...
Gerard Kingma
www.kingma.nu (http://www.kingma.nu)
-
Thanks for the reference to your article Gerard. Interesting reading. Seems you like the DJ 130 but weren't as pleased with Printfix Pro.
Given that I don't think I need to wait for further articles! I want something that will produce better than the stock Satin profiles and this doesn't seem to be it. Too bad, the price was almost affordable.
If others are using this combination and have different views please speak up! All input welcome!
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=98178\")
Byron,
Recognizing the need for a good profile indicates wisdom, so as a suggestion while you are sorting out software and hardware for that purpose, you might look into have a custom profile made for you. This may also give you a benchmark, so to speak, of what a good custom profile should be. There are a number of people who can make one for you, however, from personal experience I can only recommend one, [a href=\"http://www.digitaldog.net]Andrew Rodney[/url]. Remember too, it is not the type of thing that you buy on price alone, but on quality.
Good Luck,
Ed
-
Byron,
Due to an unfortunate turn of events the update will after all not be published within the very near future. Please contact me privately through my e-mail so we can discuss your options re. PrintFIX Pro.
Gerard Kingma
www.kingma.nu (http://www.kingma.nu)
-
To do printer profiles right, a spectrophotometer-based solution is by far the best. Scanner-based solutions like PrintFix really don't work well.
-
Scanner-based solutions like PrintFix really don't work well.
[a href=\"index.php?act=findpost&pid=98529\"][{POST_SNAPBACK}][/a]
Yes, I've played with Profile Prism enough to agree with that.
However we are talking about Printfix Pro, not Printfix. They are completely different products (despite the similar sounding names).
Byron
-
Printfix Pro's measuring device is still RGB-based; not a spectrophotometer.
-
Printfix Pro's measuring device is still RGB-based; not a spectrophotometer.
[a href=\"index.php?act=findpost&pid=98553\"][{POST_SNAPBACK}][/a]
Yes it is not a spectro...neither is it a RGB device...but I'm not overly concerned with that as there are many ways to solve problems...each with their own trade-offs.
People can get tangled up in theory all too easily and forget that real world results trump theory any day. Especially theory at the level a layman understands it.
The hardware design has enough potential that I think the question about how well it stacks up is worthwhile. But we need actual users to answer that, or better yet experts like Andrew Rodney to really wring it through its paces and tell us the real world results. (As far as I know he hasn't done that, which is surprising to me given he specializes in this field).
-
I see your updated article has finally appeared Gerard! A very nice read as usual!
As a fellow DJ130 user I found the same results regarding the HP generic profiles. They have tempermental color and very poor shadow detail. A custom profile really makes this printer sing!
I was surprised and disappointed at the poor showing of the EyeOne profile (given they've discontinued the Xrite line) and intrigued by the good showing from Printfix Pro. However the PFP seems quite variable as your previous print performed worse than even the stock HP profile. I've found this same kind of variability with products like Profile Prism (scanner based profiler) which I suspect is linked to metamerism issues due to the limited spectral bands of the device. It would be interesting to hear if you see the same variability from the Xrite profile.
It would have been nice to see less time spend in Gamutvision and more emphasis on selecting/printing real world images. Programs like Gamutvision/Colorthink are fun to play with but aren't easily translated into real world printing results.
I've also found the satin or glossy papers to be the easiest to profile. Although I am curious how much drying time elapsed before the profiles were actually made. In any case, it would be interesting to see how the combatants do on fine art papers!
All in all, a lot of hard work. I can appreciate the effort involved in doing these tests including the write-ups! So kudos to you and your profiling cohorts!
-
Well, the papers that were sent to me had however long it takes to cross the Atlantic to dry. It had to be at least 5 days in transit, plus a couple before shipping. The Xrite profile was made from the same number of patches as the PFP profile (729) the larger 1728 target might have shown slightly different results. Overall Gerard's printer showed pretty decent linearity, which you would expect from a printer that can be calibrated like the DJ130. That linearity is why I say that the larger target might produce different results.
-
A printer like th DJ130 which can re-calibrate is a perfect candidate for custom calibration, because you'll be able to return it to its initial state whenever you want.
I really don't understand why people who jst use one paper type don't go straight for a custom profile as soon as they get the printer, rather than think about buying a cheap color management solution.
I've had one client do this cleverly - she got an HP with a money-back warranty, and asked me to profile it - intending to keep or return according to how the results looked. She kept it.
Edmund
-
With all of the comparison testing I have been doing on the z3100 it was decided to use two measuring systems, the internal Eye one Spectrophotometer and the PrintFix Pro. Looking at the measurement data, I would say that the Eye one is better but only marginally as the readings from both devices are very close.
-
Byron, thanks for your kind words! It was a lot of work indeed.
To be on the safe side, I let all targets dry for at least twelve hours before packaging and shipping them to Jan and Greg, or before I measured them myself with PrintFIX Pro.
Gerard Kingma
www.kingma.nu (http://www.kingma.nu)
-
With all of the comparison testing I have been doing on the z3100 it was decided to use two measuring systems, the internal Eye one Spectrophotometer and the PrintFix Pro. Looking at the measurement data, I would say that the Eye one is better but only marginally as the readings from both devices are very close.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=106730\")
The Z3100 gamut/profiling discussions on LL intrigued me for several reasons. First of all that people were comparing the Z3100 prints with Epson prints based on different profiling solutions for the last. The Z3100 I have for two weeks now got its firmware upgrade right away so I can not comment on the old firmware. It works for me because I did read all the discussions as well. I have to thank you and Julian Mussi for that. Using the Litho-realistic media choice for HM Photorag (+ Bright White) and a tweak to the profile I made in Kodak's Custom Tools profile editor to get sky color correct to my monitor, they are very, very close to the Epson 4000 prints on EAM they have to match this time.
While reading the discussions I noticed you use the PrintFix Pro profiler as the reference. I had seen Gerard's review of PFP and Gamutvision before (good review Gerard !) and wondered whether there was relation between that observation and the Z3100 issue. There's more going on in theory though with all the profile creation packages and hardware used in the discussions but I will keep it limited in this thread. Between the PrintFix Pro Colori/Spectro hardware and the Z3100 spectrometer hardware: the Z3100 uses one white light LED as the lightsource that doesn't go beyond 380 Nm so in practice a UV filter is on the light source (there are also Spectrometers with a UV filter on the sensor, there are ones without any UV filtering like my SpectroCam). The PrintFix Pro most likely isn't adding UV light either with its 6 colored LEDs. What I'm more concerned about is whether the 6 color LED readings it performs cope well with N-Color printers like the Z3100, like the iPF9000. Apart from the wider gamuts it has to cover in some cases, there's also the three extra hues/pigments in the mix that may not be measured like the CcMmYK mixes the PFP solution probably was still based on when in development. Then there's the drying time of targets, HP is quite correct that the color is stable very fast after printing (all the printers and inks I have used in the past didn't show that behaviour) but I will test for any shifts that drying time could add. There's another thing, the HP solution measures with the black tablet behind the paper, that's theoretically correct for many uses but for art reproduction and photography the use of two or three extra layers of the same paper behind the target to be measured is now common practice. The Z3100 uses a patch reading system instead of rows and the black/dark patches reading is taken longer than the lighter patches. The patch reading happens with the Datacolor/PFP hardware too but I guess no extra time is taken for the darkest patches. Then there's the classification of the Eye One in the HP Z3100, it is closer to another GM model with its white LED light (can't recall the name) than to the normal Eye One spectrometer so even there some differences in measuring internally and externally could occur.
While we would like to get matching prints at the end of the day there's a lot that has to be compensated or matched in software and user's routines to get to that goal. In the end there still is a difference in color mixing, the Z3100's mixes are cleaner in most mixes compared to any Epson target patches I have gathered here. A good observer will see that difference while the spectrometer has to take an average of the area measured and the different gamut 3D renderers add their different routines to shape the data: there's no web equivalent for comparing two prints on a table. Matching the Z3100 prints to the Epson 4000 under different Kelvin light sources is a revelation too. If they are equal in 5500 K light I call it a day, the Z3100 prints behave far better in the rest of the light sources. I don't think the discussions will end soon and I am very happy with what I get right now from the Z3100.
Ernst Dinkla
try: [a href=\"http://groups.yahoo.com/group/Wide_Inkjet_Printers/]http://groups.yahoo.com/group/Wide_Inkjet_Printers/[/url]
-
Looking at the measurement data, I would say that the Eye one is better but only marginally as the readings from both devices are very close.
[a href=\"index.php?act=findpost&pid=106730\"][{POST_SNAPBACK}][/a]
It would be very useful (I think) to actually compare the readings for each patch and compute the differences. Supposedly the big "weakness" of printfix pro is the limited hardware...but how limited is it really? This would tell us. The rest would then be software (which can always be improved).
-
The Datacolor 1005 colorimeter is limited in that it can not measure spectral information, so improvements in the software are limited to what PFP can do with the measurements.
-
It would be very useful (I think) to actually compare the readings for each patch and compute the differences. Supposedly the big "weakness" of printfix pro is the limited hardware...but how limited is it really? This would tell us. The rest would then be software (which can always be improved).
[a href=\"index.php?act=findpost&pid=106898\"][{POST_SNAPBACK}][/a]
The i1 Pro tungsten light source has a faily large latency if you like. As the lamp heats up it influences readings. If you take a chart and read it cold, read it again after you've had your hand on it for a long time, you'll have delta e differences of over 0.5 usually only the L values show significant differences. I noticed this before but always thought it was the light scratching of the surface before the new ruler came out.
Graeme Gil brought it to the attention of the Apple Colorsync user forum that he was seeing these shift due to the light source.
Not much information is around on the LED side of operating condition expectancies variation.
From what reading I've done there are LED that for the reading of the charts that don't emit light in 380-410 nm yet certain spectros have LED's that do justly to make the inter-instrument agreement inline with past instruments.
So it is possible to have a non uv source have LED's for rebuilding the spectrum to make redundancy between instruments. It gets a bit confusing to know what the new LED's will do and what they won't, and or if you need to know.
-
Not much information is around on the LED side of operating condition expectancies variation.
From what reading I've done there are LED that for the reading of the charts that don't emit light in 380-410 nm yet certain spectros have LED's that do justly to make the inter-instrument agreement inline with past instruments.
So it is possible to have a non uv source have LED's for rebuilding the spectrum to make redundancy between instruments. It gets a bit confusing to know what the new LED's will do and what they won't, and or if you need to know.
[a href=\"index.php?act=findpost&pid=107154\"][{POST_SNAPBACK}][/a]
Yeah, iSis has the ability to make With-UV or UV-Cut measurements while using LED illuminants with an i1 spectro head.
I would speculate the same technology could find its way into an EyeOne Pro revision in the future.
Edmund
-
Where any of these tests done with the new PrintFIX PRO 2.0 software? Or do you not think it will make any difference? Looks like a nice upgrade.
-
Where any of these tests done with the new PrintFIX PRO 2.0 software? Or do you not think it will make any difference? Looks like a nice upgrade.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=107794\")
The testing for both articles was done with version 1.1.1 of the PrintFIX Pro software. The testing for Part II was done in November 2006, before version 2 was released, but due to many delays, unforeseen circumstances, mishaps and malfortunes, the article wasn't published until last week.
I downloaded 2.0 and built a new profile but haven't made any comparisons yet.
Gerard Kingma
[a href=\"http://www.kingma.nu]www.kingma.nu[/url]
-
The testing for both articles was done with version 1.1.1 of the PrintFIX Pro software. The testing for Part II was done in November 2006, before version 2 was released, but due to many delays, unforeseen circumstances, mishaps and malfortunes, the article wasn't published until last week.
I downloaded 2.0 and built a new profile but haven't made any comparisons yet.
Gerard Kingma
www.kingma.nu (http://www.kingma.nu)
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=107820\")
Part II is quite a read but worth it. Thank you and the contributors for the work done. That Graeme Gill of Argyll knows what has to be done in CM was clear from the moment I saw his first Colorsync list contributions.
One thing seems to be missing in this overview, a Spectrometer that measures into UV as well. The 3 hardware tools mentioned are all UV filtered one way or another. The HP Premium Plus Photo Satin may not be loaded with OBAs, it does have optical brighteners. HP's canned profile doesn't look like one that is based on unfiltered readings either. This can be expected in HPs conservative approach as sketched in the review. BTW, I have seen another approach in the first Epson 9000 (dye ink) canned profiles 8 years ago, in that case they made sure that any Epson 9000 delivered would fit within the gamut of that inflated profile. No customer complained about the printed gamut size but they all got something different from their printers :-) At that time you needed a custom profile for precisely the opposite: get actual color control.
I wonder what a normal (non UV filtered) SpectroCam (Xenon flash light source) + ACMS would deliver. In view of Xrite/GM dominance on the market another software competitor + a second hardware source is more than welcome. PFP II isn't filling that gap alone in my opinion.
Ernst Dinkla
try: [a href=\"http://groups.yahoo.com/group/Wide_Inkjet_Printers/]http://groups.yahoo.com/group/Wide_Inkjet_Printers/[/url]
-
I agree that it would be useful to test how each product handles FWA's although they should be avoided like the plague.
-
Speaking of OB's, this went up on the ColorSync list yesterday and it's worth its weight in gold (many thanks to Robin)
For those interested in the OBA/FWA (Optical Brightening Agent/Fluorescent Whitening Agent) issue, I have just finished putting together a page on the issue of brightened papers (www.rmimaging.com/information/fine_art_paper.html). The paper includes data from paper samples obtained at the PMA show. Unfortunately I did not get the Innova paper mentioned by Andrew Rodney, but I did get papers from Moab, Hahnemuehle, Epson, BF Inkjet and Crane. I measured all of these papers with a Spectrolino with and without a UV blocking filter, so I could detect even small amounts of brightener. The page does not have the spectra, but it does discuss the issue of brighteners for fine art print papers and it has a chart showing which papers have brighteners and which do not.
If anyone would like to have a particular paper added to the list, let me know. If anyone needs actual spectral data, also let me know.
Robin Myers
robin@rmimaging.com
-
I agree that it would be useful to test how each product handles FWA's although they should be avoided like the plague.
[a href=\"index.php?act=findpost&pid=107860\"][{POST_SNAPBACK}][/a]
There are not many papers without OBA's. The resellers believe we want them. With education finally users will have to turn this around.
As much as I lov the Innova Ultra, they told me it was free of OBA. Curious, as the white is very bright, I put it under blacklight. It is loaded.
A little will not show significant change over displayed years, but a lot will.
I'd prefer to see a lot less use of OBA.
-
There are not many papers without OBA's. The resellers believe we want them. With education finally users will have to turn this around.
As much as I lov the Innova Ultra, they told me it was free of OBA. Curious, as the white is very bright, I put it under blacklight. It is loaded.
A little will not show significant change over displayed years, but a lot will.
I'd prefer to see a lot less use of OBA.
[a href=\"index.php?act=findpost&pid=107863\"][{POST_SNAPBACK}][/a]
Less is better for sure. There's a bit in Premium Luster. When I ran a black light over the Canon paper I was sent with my iPF 5000 (premium photo satin) I was almost blinded. Put it next to Luster and luster looks yellow. But I really hate the Canon paper anyway. Under a GTI booth, it looks awful (for obvious reasons). Way, way too 'white'.
-
To do printer profiles right, a spectrophotometer-based solution is by far the best. Scanner-based solutions like PrintFix really don't work well.
[a href=\"index.php?act=findpost&pid=98529\"][{POST_SNAPBACK}][/a]
Except that PrintFIX PRO -isn't- a scanner based solution. It's a spectrocolorimeter. And according to Gerard's latest article, it holds its own very well against the other solutions which use a spectrophotometer instead.
What you're seeing here is that both spectrocolorimeters (the PFP spectro) and spectrophotometers (the other, more expensive instruments) all provide the same kind of raw material. The more expensive devices do it more quickly, and, if someone needs spectral data for some particular reason, that's the way to go. But for inkjet profiling, the differences come down to the software, not the hardware.
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
-
Printfix Pro's measuring device is still RGB-based; not a spectrophotometer.
[a href=\"index.php?act=findpost&pid=98553\"][{POST_SNAPBACK}][/a]
No, you're wrong again. It's not RGB based. It's a spectrocolorimeter that supplies Lab values to the PrintFIX PRO software.
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
-
The Datacolor 1005 colorimeter is limited in that it can not measure spectral information, so improvements in the software are limited to what PFP can do with the measurements.
[a href=\"index.php?act=findpost&pid=107016\"][{POST_SNAPBACK}][/a]
The usefulness of spectral measurements is being able to change the illuminant and to recalculate Lab values from the same base set of spectral data. That's a high end feature that wouldn't be used by people who are looking to spend $500 or so for a printer profiling package.
Improvements in the PFP software aren't limited by the measurement values. The most important new feature in the 2.0 version was the use of the extended grays target to provide very accurate neutrals and near-neutrals using the same profile that would be used for doing color prints, and this improvement wasn't limited by the fact that the PFP spectro doesn't provide spectral values...:-)
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
-
What you're seeing here is that both spectrocolorimeters (the PFP spectro) and spectrophotometers (the other, more expensive instruments) all provide the same kind of raw material. The more expensive devices do it more quickly, and, if someone needs spectral data for some particular reason, that's the way to go. But for inkjet profiling, the differences come down to the software, not the hardware.
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=108215\")
Which is the thing I am interested in. Do the 6 color LEDs illuminate the N-color printer patches build with 6 hues CMYRGB as good as a continuous spectral light source does ? There's a difference in the secondary mixes (RGB) of a CMY inkset compared to the primary RGB patches a CMYRGB inkset makes. Not to mention the UV filtering in the lightsource that is practical outcome of the Datacolor spectrocolorimeter design. The RGB mixed patches for a given CMY inkset are quite predictably related to the CMY ink specs, the RGB hues of a CMYRGB inkset can be selected more or less independtly of the CMY hues. Copying the spectral behaviour of RGB mixes made with CMY inks isn't the thing one would expect of the N-color printer designers, the RGB inks are there to enhance the inkset. Both with small and larger targets the measurements could end in different results if the same software has to handle the different measurements and the resulting RGB profile may not fit the CMYRGB inkset in relation to that. Theory from my side as I have no PFP II profile creator.
Ernst Dinkla
try: [a href=\"http://groups.yahoo.com/group/Wide_Inkjet_Printers/]http://groups.yahoo.com/group/Wide_Inkjet_Printers/[/url]
-
The usefulness of spectral measurements is being able to change the illuminant and to recalculate Lab values from the same base set of spectral data. That's a high end feature that wouldn't be used by people who are looking to spend $500 or so for a printer profiling package.
Sure it would if someone had a device at that price point. Its a very useful feature and provides important options when we run into FWA's (optical brighteners). And if you're printing under anything but D50, its also useful to use that in the profile white point definition.
How can you say it wouldn't be used by these people, that's silly.
What you're seeing here is that both spectrocolorimeters (the PFP spectro) and spectrophotometers (the other, more expensive instruments) all provide the same kind of raw material.
Raw material? What the heck is that supposed to mean and be? Spectral data? First you make up a name for an instrument that has spectral in it, and then say all the instruments provide the same 'raw material'. Would you clarify exactly what you're capturing and what they are capturing as raw isn't a term anyone here should except in this context. Or are you referring to the data file a digital camera prodcues?
-
Here we go again
Andrew, I'll leave this fight for you.
-
Here we go again
Andrew, I'll leave this fight for you.
[a href=\"index.php?act=findpost&pid=108252\"][{POST_SNAPBACK}][/a]
Please, don't. Between the two color management vendors on this site who post questionable marketing speak, I can't call Karl Lang that often <g>.
I'm going off to Home Depot for a few more measuring tapes and some high rubber boots.
-
Please, don't. Between the two color management vendors on this site who post questionable marketing speak, I can't call Karl Lang that often <g>.
I'm going off to Home Depot for a few more measuring tapes and some high rubber boots.
[a href=\"index.php?act=findpost&pid=108254\"][{POST_SNAPBACK}][/a]
The chest waders are better. Make sure to get the ones with the good strong suspenders so that you can't be pulled out of the waders. Even if it doesn't get that high, nothing can spill over into your boots.
Having worked in floods, nothing is worse than filling your knee high boot with water. Even worse with waist high boots.
-
The chest waders are better. Make sure to get the ones with the good strong suspenders so that you can't be pulled out of the waders. Even if it doesn't get that high, nothing can spill over into your boots.
Having worked in floods, nothing is worse than filling your knee high boot with water. Even worse with waist high boots.
[a href=\"index.php?act=findpost&pid=108256\"][{POST_SNAPBACK}][/a]
A bit OT but I actually DO need to look into these for Michael’s trip to the Amazon. He said we'd need em. Never used them before and NO I'm not going in the water.
But they could do double duty here (and on DP Review where it really gets thick).
-
But they could do double duty here (and on DP Review where it really gets thick).
[a href=\"index.php?act=findpost&pid=108257\"][{POST_SNAPBACK}][/a]
You are a braver man than I, Andrew. I left the infighting a long time ago over at DPR. It is amusing sometimes to watch from the sidelines, though.
-
Here we go again
Andrew, I'll leave this fight for you.
[a href=\"index.php?act=findpost&pid=108252\"][{POST_SNAPBACK}][/a]
I don't know what to say about this.
I'm not here to fight with anyone. I saw a few posts, said a few simple things, and I'm not the one who's using words or making statements like:
(Andrew:)
"silly"
"questionable marketing specs"
"Raw material? What the heck is that supposed to mean and be?"
"First you make up a name for an instrument that has spectral in it"
etc etc.
Why would anyone like me (who develops color management software) want to come into a forum like this and be greeted by this kind of response?
Andrew, I worked for several years at Integrated Color Solutions as one of the primary developers on all of their basICColor software product line and also on portions of their Remote Director product, all of which (as you probably know) is based on spectral data when the measurements are taken with a sensor that supports it... including Display3 for both Integrated Color and Color Solutions, when used with an EyeOne. I think I know a little more about this than you're giving me credit for.
If anyone would like to have a reasonable discourse about any of this, then I'd be willing to continue, but with all due respect, this is a bunch of nonsense and I'm not going to get into a war of words with you over it.
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
-
I don't know what to say about this.
I'm not here to fight with anyone. I saw a few posts, said a few simple things, and I'm not the one who's using words or making statements like:
Why would anyone like me (who develops color management software) want to come into a forum like this and be greeted by this kind of response?
Andrew, I worked for several years at Integrated Color Solutions as one of the primary developers on all of their basICColor software product line and also on portions of their Remote Director ...[a href=\"index.php?act=findpost&pid=108263\"][{POST_SNAPBACK}][/a]
Great then maybe you can answer the questions addressed to Jack in the series of posts "Losing sleep over monitor calibration" as well as those I posted above.
I'm not here to fight but I am here to ask for clarity on what appears to be mainly marketing speak. If its not and you have technical answers to dismiss this, I'm all ears.
If the questions are nonsense, you may want to explain why to those lurking here (and me). I'm not asking you to explain the origins of life or debating religion. The questions IMHO are valid and remain unanswered.
What's raw data mean in the context of what these devices are producing? Are you capturing spectral data and if not, what's the rational behind the naming of your device? What matrix did you use to express the idea that users don't need the capabilities of captured spectral data. What's an accurate profile mean etc.
What about the question Enrst asked?
-
If anyone would like to have a reasonable discourse about any of this, then I'd be willing to continue, but with all due respect, this is a bunch of nonsense and I'm not going to get into a war of words with you over it.
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=108263\")
Well the issues I raised (earlier and later in this thread) were phrased reasonably but have not been answered and you seemed the one to have the answers on that. However typing: - Datacolor narrow band LEDs - in Google brought me to a thread on DPreview:
[a href=\"http://forums.dpreview.com/forums/read.asp?forum=1003&message=21501325]http://forums.dpreview.com/forums/read.asp...essage=21501325[/url]
that gave me my answers and they didn't come from you there either.
Ernst Dinkla
try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/ (http://groups.yahoo.com/group/Wide_Inkjet_Printers/)
-
...If anyone would like to have a reasonable discourse about any of this, then I'd be willing to continue...
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
Well I'm not interested in name calling...just the facts honestly reported...please continue in that vein.
It seems to me people are most concerned about your hardware device being able to provide accurate values to the profiling software. You assert it is as good as a real spectro...I can find accuracy specs for the spectros but none for your device. Do you have an accuracy spec? If so, what is it?
-
However typing: - Datacolor narrow band LEDs - in Google brought me to a thread on DPreview:
http://forums.dpreview.com/forums/read.asp...essage=21501325 (http://forums.dpreview.com/forums/read.asp?forum=1003&message=21501325)
that gave me my answers and they didn't come from you there either.
Yup, an interesting series of posts where once again, we hear claims of 'accuracy' which reared its ugly head on the recent posts about display calibration. That term is really overused and never properly defined.
In the DP post we see this:
> The comparison charts for the products show that the values coming
> from the spectrocolorimeter are just as applicable for building
> high quality profiles as the values coming from the
> spectrophotometer.
No, it doesn't. I've already explained the issue of accuracy to you.
http://forums.dpreview.com/forums/read.asp...essage=21426454 (http://forums.dpreview.com/forums/read.asp?forum=1000&message=21426454)
So its nice to see Joe make the call (again) to a color management vendor to define what they are talking about (and once again, to see this go unanswered).
Is this an example of 'cut and run' I read about?
-
http://forums.dpreview.com/forums/read.asp...essage=21426454 (http://forums.dpreview.com/forums/read.asp?forum=1000&message=21426454)
Ernst, maybe this helps too?
-
http://www.aviantechnologies.com/products/...eflect.html#FTS (http://www.aviantechnologies.com/products/standards/reflect.html#FTS)
Almost getting down into the realm where mere mortals might tread.
Might be nice to have a set of these once in a while when this type of discussion comes up:
http://www.aviantechnologies.com/products/...lect.html#ceram (http://www.aviantechnologies.com/products/standards/reflect.html#ceram)
Anyone want to donate to the effort?
-
Yup, an interesting series of posts where once again, we hear claims of 'accuracy' which reared its ugly head on the recent posts about display calibration. That term is really overused and never properly defined.
In the DP post we see this:
So its nice to see Joe make the call (again) to a color management vendor to define what they are talking about (and once again, to see this go unanswered).
Is this an example of 'cut and run' I read about?
[a href=\"index.php?act=findpost&pid=108314\"][{POST_SNAPBACK}][/a]
"reared ugly head"? "Cut and run"?
Maybe it hasn't occurred to you that I'm actually doing development work on a product and that I didn't have a chance to get back in here to look for the continuing thread, until this morning.
If you really think this is the appropriate way to write or discuss color management, in what's supposed to be a friendly forum, then there's no place for me here, and I certainly don't feel welcome to comment on any of this, in the areas where it would be appropriate for me to comment.
Sorry to have intruded. I'm not cutting and running; but I am leaving.
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
-
Well the issues I raised (earlier and later in this thread) were phrased reasonably but have not been answered and you seemed the one to have the answers on that. However typing: - Datacolor narrow band LEDs - in Google brought me to a thread on DPreview:
http://forums.dpreview.com/forums/read.asp...essage=21501325 (http://forums.dpreview.com/forums/read.asp?forum=1003&message=21501325)
that gave me my answers and they didn't come from you there either.
Ernst Dinkla
try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/ (http://groups.yahoo.com/group/Wide_Inkjet_Printers/)
[a href=\"index.php?act=findpost&pid=108276\"][{POST_SNAPBACK}][/a]
Ernst,
I wouldn't have been the right person to answer hardware specific questions about the spectro. I work solely on the software/profiling end of things, not in any areas that are related to the internals of the spectro itself (this is the Datacolor 1005 spectrocolorimeter I'm talking about, of course).
Best regards,
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
-
Well I'm not interested in name calling...just the facts honestly reported...please continue in that vein.
It seems to me people are most concerned about your hardware device being able to provide accurate values to the profiling software. You assert it is as good as a real spectro...I can find accuracy specs for the spectros but none for your device. Do you have an accuracy spec? If so, what is it?
[a href=\"index.php?act=findpost&pid=108286\"][{POST_SNAPBACK}][/a]
Thanks, I'd like to, but there's such a negative sentiment here from other quarters that it makes it impossible to continue.
The only spec I'm aware of is the one that's been previously published and referred to, and which is at the bottom of the Northlight Images review and which is probably also somewhere on the ColorVision web site.
If the Datacolor 1005 spectrocolorimeter wasn't providing suitably accurate Lab values for building printer profiles, then the PrintFIX PRO package wouldn't have finished in the top 3 of the profilers, whether some people like that; or not. The results are there; if someone would like to do more extensive testing and comparisons with other test images, great!
If anyone has any additional questions about PrintFIX PRO, please direct them to me at davem@colorvision.com and I'll do my best to answer them privately via email.
Best regards,
David Miller
Senior Software Developer, Digital Color Solutions
ColorVision
-
Thanks, I'd like to, but there's such a negative sentiment here from other quarters that it makes it impossible to continue.
Impossible? Interesting. We didn't get the answers over on DP Review, I guess I shouldn't have expected it here.
The question Ernst asked was perfectly appropriate and polite but you're going to ignore it?
If the Datacolor 1005 spectrocolorimeter wasn't providing suitably accurate Lab values for building printer profiles, then the PrintFIX PRO package wouldn't have finished in the top 3 of the profilers, whether some people like that; or not. The results are there; if someone would like to do more extensive testing and comparisons with other test images, great!
I don't recall anyone claiming they were not 'accurate' although in the context here, it's meaningless since we can't get any real data from you. But what about the other questions from Ernst and me (do you capture spectral data? Why is Spectro used in the name if not?).
Sorry if you think we are (I am) personally picking on you but you came here to discuss the review and you're the representative from the company. Its not fair for us to ask you for specific, pertinent technical questions and expect answers or we should just let them pass by as we've seen elsewhere and just gloat about the article? Set the record straight and answer the questions. Now you really are cutting and running.
-
In the spirit of goodwill in these trying times, cut David some slack . He did state he's a software developer, you guys are asking mostly hardware related stuff. We can always fast-forward through the marketing speak, but we may get some other useful info too.
Also I tend to agree with him about the target audience not being that interested in raw data. I think a lot of people don't want to get to that level of sophistication, me included. I want to get a nice profile as painlessly as possible, and quite frankly I found the software well designed in this aspect. Remember, he doesn't define the product, the marketing people do. It's mostly out of his hands. Plus every additional level of functionality can add significantly to the development, testing and production costs, and Colorvision like most corporations is out there to make money for the stockholders, although hopefully providing a useful product to ensure this.
Also remember he's here as a representative of his company, which places constraints on what he says and how, and the fact he'd like to keep his job. Frankly I admired his courage for even participating in the discussion, especially after seeing the shall we say tough audience he's facing .
Just another perspective ...
Having said that, I'm having a hard time getting a proper profile using PFP 2.0 for my iPF5000 through the standard Windows driver (not through Photoshop) and maybe he can help me out . I guess I should start another appropriately named thread ...
-
In the spirit of goodwill in these trying times, cut David some slack . He did state he's a software developer, you guys are asking mostly hardware related stuff. We can always fast-forward through the marketing speak, but we may get some other useful info too.
OK maybe he can get some info about the hardware then although I suspect many of the questions asked (does it capture spectral data) would be known to someone who's handling the software.
Also I tend to agree with him about the target audience not being that interested in raw data.
I have no idea what he's speaking about when he says raw data. I asked for clarification but that's not happened.
Also remember he's here as a representative of his company, which places constraints on what he says and how, and the fact he'd like to keep his job.
If we're asking anything that is limited by an NDA or is proprietary info, he just has to say so. We know the device ends up providing LAB values, he's said so. What's it measuring to get that? Is it or isn't it a Spectrophotometer or to put it another way, is it gathering spectral data? The name, perhaps the marketing folks have used for the instrument kind of, sort of, gives that impression but no one at CV that I know of have said so one way or the other.
-
...He did state he's a software developer, you guys are asking mostly hardware related stuff.
I asked my question because David said the hardware provided input data to the software that was comparable to a real spectro in accuracy. I'm trying to understand what he means by that, although it seems it is really just marketing speak in the end.
The white paper from the original designer demonstrated an 8 band device and managed about 5 dE accuracy. The current Colorvision hardware is a 6 band device which increases the error. Is 5 dE accuracy (and it may be worse) good enough? Probably for most people it is OK...especially if the software is well written in terms of producing profiles that avoid "bad behaviour" whilst perhaps not being able to produce the "best behaviour". However these error levels probably aren't good enough for critical use.
The absence of accuracy specs (while including other specs) is telling. Understandable however since the hardware *is* more limited and can't compete in that area with a real spectro. But in that regard I think the concerns over its name and the fact it doesn't produce real spectral data is a bit overdone. It is clearly inbetween a scanner and a real spectro so having an inbetween name isn't so bad. And lacking real spectro data only matters if the competition is making better profiles with their spectral data...and that hasn't been shown one way or the other as far as I can see.
In the end, I think it is an interesting package and would like to try it. Who knows, it might be good enough for me. I also never liked the way EyeOne upsells their software just to get full functionality inkjet profiling. But alas, the local stores simply won't allow returns. So it remains a theoretical exercise.
PS. At the risk of writing a book...someone who has both PFP and a real spectro should scan some targets and post the dE differences in the resulting Lab values. That way we can actually add some further facts to this discussion rather than churning through the same old tired ones.
-
I asked my question because David said the hardware provided input data to the software that was comparable to a real spectro in accuracy. I'm trying to understand what he means by that, although it seems it is really just marketing speak in the end.
Exactly. He brought up the subject.
The absence of accuracy specs (while including other specs) is telling. Understandable however since the hardware *is* more limited and can't compete in that area with a real spectro. But in that regard I think the concerns over its name and the fact it doesn't produce real spectral data is a bit overdone. It is clearly inbetween a scanner and a real spectro so having an inbetween name isn't so bad. And lacking real spectro data only matters if the competition is making better profiles with their spectral data...and that hasn't been shown one way or the other as far as I can see.
I'd agree, its a bit overdone (on my part). But some clarity from the manufacturer would go a long way in putting this to rest.
As for the competition, they (GretagMacbeth) does use spectral data for among other functionally, handling optical brighteners and providing compensation as the profile is built along with custom illuminant support in building the profiles.
In the end, I think it is an interesting package and would like to try it. Who knows, it might be good enough for me. I also never liked the way EyeOne upsells their software just to get full functionality inkjet profiling.
Way, way too many flavors and options that's confusing to just about anyone looking at their solutions. There should be one flavor of Match and then you move up into ProfileMaker Pro. End users and potential customers need to hammer this home and maybe X-Rite will do something about it.
PS. At the risk of writing a book...someone who has both PFP and a real spectro should scan some targets and post the dE differences in the resulting Lab values. That way we can actually add some further facts to this discussion rather than churning through the same old tired ones.
[a href=\"index.php?act=findpost&pid=108475\"][{POST_SNAPBACK}][/a]
That would be useful. I'm afraid I can't assist, I don't have the hardware either.
-
If I had the datacolor 1005 a comparison between it and one of my spectros would have been the first thing that I did. But I don't, and probably never will.