Page 3 of 4 FirstFirst 1234 LastLast
Results 41 to 60 of 61

Thread: Need comments: What I understand about inkjet printer resolution...

  1. #41
    Senior Member
    Join Date
    Jan 2002
    Location
    Singapore, Singapore, Singapor
    Posts
    6,405

    Default

    Originally posted by roygoh


    The gentlemen who did the comparison and posted his conclusions is called John Mills. He is the boss of Weink I think. Weink is a supplier of 3rd party inks for inkjet printers. Go to the printers and printing forum in dpreview and do a search for "john mills" and you should be able to find his post.

    What he said in his post is of course subjective, and he has already been criticised by many other forum members. By the way, he used a print of his wife, thus comparing skin tones and not just Disney colours.

    I am quite interested to find out for myself how much impact there is by adding the 2 photo colours. Is it just a marketing gimick by the printer companies to sell more ink? If the PC and PM colours are half the intensity of the regular C and M ink, then can the same effect be acheived with the regular C and M inks by making the ink drop size even smaller?

    I supposed this is related to your question (1). How does a laser printer print shades of grey with only black toner? Half tone. By controlling the density of black dots on the white printing paper, the black dots blend with the background white when viewed at normal viewing distance to appear grey. The smaller the dots can be, the more continuous the apparent tone variation can be achieved.

    By the way, by adding the photo colours, does it really extend the colour gamut, or simply increase the colour resolution acheivable within the same gamut? Since the photo colours are a lighter version of the original C and M inks, it cannot really extend the gamut of CMYK printing, am I right?
    You have a valid point there. But still, using a PC and PM inks will result in better tonal graduations in the printout right? It's still better to have a lighter ink at a smaller droplet size than just a smaller droplet alone.

    For the same reason, until the new Epson 2100 with UltraChrome inks featuring "light black" ink, it's quite impossible to print good B&W on inkjet printers. Use black alone, and you don't get enough tonal variations. Use black + colour, you get colour casts.

    Regards
    CK

  2. #42
    Senior Member
    Join Date
    Jan 2002
    Location
    Northwest
    Posts
    5,011

    Default

    Originally posted by ckiang


    You have a valid point there. But still, using a PC and PM inks will result in better tonal graduations in the printout right? It's still better to have a lighter ink at a smaller droplet size than just a smaller droplet alone.
    Yes, I agree, because the current dpi is not high enough to generate fine enough tonal variations at higher ppi. However, since we are talking about practical printing which is mostly around 300ppi or less, and rarely makes any sense to go beyond 600ppi, then would the purpose of the PC and PM inks become not so important as the drop size reduce further?

    Imagine a printer cableble of 9600 dpi, and a paper capable of holding each drop of ink at its place, for discussion sake. Such a printer can lay 16X16 = 256 dots in a 600ppi pixel. If this is achievable, do we still need the PC and PM colours? I know this is not available yet. Do you think this will someday be possible?

    Originally posted by ckiang
    For the same reason, until the new Epson 2100 with UltraChrome inks featuring "light black" ink, it's quite impossible to print good B&W on inkjet printers. Use black alone, and you don't get enough tonal variations. Use black + colour, you get colour casts.

    Regards
    CK
    If my understanding is correct, with current printers, you can use quad tone inks with the correct print driver to achieve much higher tonal variations in B&W print. Quite a hassle when you want to switch between colour and B&W printing, but it is not impossible.
    As complexity rises, precise statements lose meaning and meaningful statements lose precision.

  3. #43

  4. #44
    Senior Member
    Join Date
    Jan 2002
    Location
    Northwest
    Posts
    5,011

    Default

    Originally posted by erwinx
    Gamut

    http://www.popphoto.com/pdfs/colormanagp2.pdf

    http://www.dp-now.com/Features/Print...mut/gamut.html
    for the latter look at the 4 colour HP vs the Epson/Canon
    It's my misunderstanding of the term Gamut. Thanks for the link!
    As complexity rises, precise statements lose meaning and meaningful statements lose precision.

  5. #45
    Member
    Join Date
    Jul 2002
    Location
    Singapore
    Posts
    340

    Default

    Originally posted by roygoh
    If the PC and PM colours are half the intensity of the regular C and M ink, then can the same effect be acheived with the regular C and M inks by making the ink drop size even smaller?
    Yes, it can be achieved, at the cost of pixel resolution. A 6-color printer need to dither less than the 4-color printer if printing the light color because it has the ink. But if the dot resolution (dpi) of the 4-color printer is high enough to allow the dithering of light color so its pixel resolution can come to 300 ppi, then that printer can print as good as the 6-color competitor. (for normal viewing, not under the loupe).

    By the way, by adding the photo colours, does it really extend the colour gamut, or simply increase the colour resolution acheivable within the same gamut? Since the photo colours are a lighter version of the original C and M inks, it cannot really extend the gamut of CMYK printing, am I right?
    Theoritically you're right. The PC & PM ink will increase the color resolution within the same gamut. But on inkjet printer, color resolution decreases as the color become lighter, less pixel can be printed. If the ppi drop lower than 300, then we start to observe the grain. Because of this reason, some printer limit the gamut, especially on very light colors. With PM&PC inks, the printer can print light colors with the same pixel resolution as the normal one, thus it may increase the gamut to certain extend.

  6. #46
    Member
    Join Date
    Apr 2002
    Location
    Jurong
    Posts
    160

    Default

    Originally posted by roygoh
    When you claim that 300ppi is a bit soft, are you comparing it to the same picture printed at 600ppi? Also, even if you are comparing with 600 ppi, did yout print from an image that has that amount of resolution to start with, or did you interpolate to higher resolution for 600ppi print?

    Your next post is even more confusing. Now are you saying that you compared a 300 ppi print to desktop calendars and made the deduction that the reason for your print being comparatively soft is due to the ppi?

    Regarding the PC and PM colours, someone in dpreview posted that he actually prefered the output from Canon S750 (4-colour) better than the S9000 (6-colour).

    I wonder if there is a way to switch my S9000 into 4-colour mode so that I can make a comparison myself.
    Yes, I find that 300ppi is a bit soft either I do or do not compare to calendar prints. I mentioned calendar prints was to give an example of what I consider sharp prints.

    I compared 300ppi to 600ppi but not the same picture. I know you'll say that that's not valid. But I guess you can distinguish which CD has the better sound quality even if you compare different songs from each CD when the difference is easily noticeable, right?

    And the 600ppi picture I test printed was of course not interpolated from lower resolution. I think we all know software interpolation will NOT add any extra details to the picture. You'll get exactly same quality when print at same size. It only helps to give a smoother and less pixelation print at larger scale. Otherwise we wouldn't have keep pushing for larger CCD. We could just interpolate our 1/2MB up to 6/8MB and print. Well the only interpolation that does some magic is the SuperCCD of Fujifilm, because it's hardware interpolation.

    Regarding 4 colors vs 6 colors, I'd say that higher resolution plus smaller droplet size, for example maybe 5760dpi @ 1pl (the highest currently is 2880dpi @ 2pl) will compensate the lack of light cyan and light magenta and produce the same tone as 6 colors can.

    Why manufacturer introduced 6 colors printing was because it's much easier to add two more colors than technically breakthrough in printer technology. Another great reason is... ink cartridges of course. $$$$$$!

    When one day we achieve very high printer resolution and very small droplet size, CMYK is what we need, in theory and in fact.

    And the light version of C and M has disadvantages. They are much less stable. I compared the same prints printed by 4 colors and 6 colors, the latter exhibited faster fading. And also 6 colors needs tougher color management (calibration and profiling) before you get your colors right while 4 colors usually gives out-of-the-box color matching. I shifted from Epson's Stylus Photo printer to Stylus Color printer.

  7. #47
    Member
    Join Date
    Apr 2002
    Location
    Jurong
    Posts
    160

    Default

    Originally posted by ckiang

    ... But still, using a PC and PM inks will result in better tonal graduations in the printout right? It's still better to have a lighter ink at a smaller droplet size than just a smaller droplet alone.
    You're right, but not when one day we achieve droplet size that's beyond human eyes' visual capabilities, maybe 1pl or half pl. By that time, light cyan and light magenta will render useless, or pointless.
    Last edited by KamWeng; 29th July 2002 at 11:11 PM.

  8. #48
    Member
    Join Date
    Apr 2002
    Location
    Jurong
    Posts
    160

    Default

    And why we don't have light yellow??

  9. #49
    Senior Member
    Join Date
    Jan 2002
    Location
    Singapore, Singapore, Singapor
    Posts
    6,405

    Default

    Originally posted by roygoh


    Yes, I agree, because the current dpi is not high enough to generate fine enough tonal variations at higher ppi. However, since we are talking about practical printing which is mostly around 300ppi or less, and rarely makes any sense to go beyond 600ppi, then would the purpose of the PC and PM inks become not so important as the drop size reduce further?

    Imagine a printer cableble of 9600 dpi, and a paper capable of holding each drop of ink at its place, for discussion sake. Such a printer can lay 16X16 = 256 dots in a 600ppi pixel. If this is achievable, do we still need the PC and PM colours? I know this is not available yet. Do you think this will someday be possible?
    I don't see why not? But given our inherent lust in specs and all, we would eventually want a 12 colour (or more), 1 pl printer.


    If my understanding is correct, with current printers, you can use quad tone inks with the correct print driver to achieve much higher tonal variations in B&W print. Quite a hassle when you want to switch between colour and B&W printing, but it is not impossible.
    Yes, that's possible, but like you said, quite a hassle. And you'll need to flush all remaining ink out of the reservoir.....

    Regards
    CK

  10. #50

    Default

    I compared 300ppi to 600ppi but not the same picture. I know you'll say that that's not valid. But I guess you can distinguish which CD has the better sound quality even if you compare different songs from each CD when the difference is easily noticeable, right?


    blues and greens are generally more important. for individual ink tanks, cyan runs out first. photo yellow would be rather underused. By the way, the Japanese Epson 950 uses photo yellow but not other versions. The A3 equivalent (dunno model number), uses light gray as the 7th colour.

    if you don't compare 300 dpi versus 600 dpi of the same picture, your comparison is pointless.

    Similarly, your CD example is flawed.
    If you listen to 2 different CDs you can tell which one sounds better, but can you determine whether the improvement is due to say the fact that its recorded in DSD versus 16/44.1 PCM?

    Similarly, if you compare 2 different prints, you can say that one is better, but how can you say that the difference is due to the 600dpi setting?

    As for light-fastness, if you are using an older generation of printers, that was of course a problem but things have moved on.

  11. #51
    Member
    Join Date
    Jul 2002
    Location
    Singapore
    Posts
    340

    Default

    Originally posted by KamWeng
    And why we don't have light yellow??
    Because human eyes perceive yellow as "light" color, yellow looks as lighter than cyan or magenta, even if printed at the same density side by side. So it is more difficult for human eyes to notice yellow grain than cyan/magenta grain.

  12. #52
    Member
    Join Date
    Jul 2002
    Location
    Singapore
    Posts
    340

    Default

    Originally posted by KamWeng

    You're right, but not when one day we achieve droplet size that's beyond human eyes' visual capabilities, maybe 1pl or half pl. By that time, light cyan and light magenta will render useless, or pointless.
    That's true, but not so fast. Beside the printer must be able to produce an extremely small droplet, the paper also has to be able to hold that super tiny droplet with minimal 'dot-gain'.

    Try to see a print-out of a 300 ppi dye-sub printer, which is only CMYK (without PC/PM). And compare it with the best print-out of ink-jet printer you can get. Then you will know that the ink-jet printer still need a long time to catch-up, despite of its exotic technology such as variable-size droplet, picoliter dot, 6-ink, 2880 dpi, etc.

  13. #53
    Member
    Join Date
    Apr 2002
    Location
    Jurong
    Posts
    160

    Default

    Originally posted by erwinx
    if you don't compare 300 dpi versus 600 dpi of the same picture, your comparison is pointless.

    Similarly, your CD example is flawed.
    If you listen to 2 different CDs you can tell which one sounds better, but can you determine whether the improvement is due to say the fact that its recorded in DSD versus 16/44.1 PCM?


    I knew you'd put it this way, I'd say that I already knew that which CD is recorded in which form (not guessing) and that further proven by listening to them, just as I already knew that one print was 300ppi and the other was 600ppi.

    Similarly, if you compare 2 different prints, you can say that one is better, but how can you say that the difference is due to the 600dpi setting?

    I can say that one is sharper, much sharper, not the word 'better'. If both picture go thru the same software process, print by the same printer on same media at same resolution, the only difference is the ppi (again, not dpi).

    As for light-fastness, if you are using an older generation of printers, that was of course a problem but things have moved on.

    Moved on, but not too far. Even the once famous Epson 6 color 870/1270 which claimed 10 years of light fastness received complaints on fading. It was once the hottest topic at dpreview.com. 4 color was well known more stable. Now Epson don't dare to anyhow claim already. For the new 890, they just claim 'achieve high light fastness IF printed on matte paper AND framed under glass'.

    I've changed 3 printers during the past 3 years, from 4 color 640 to 6 color 710 back to 4 color 860 and have done over hundred of test prints (and kept for stability study) trying to find out how to print better and sharper when I was new and so enthusiastic to digital photography. I'm pretty sure about:

    1. which one between 300ppi and 600ppi is sharper,

    2. which one between 4 color and 6 color is more stable.

  14. #54
    Member
    Join Date
    Apr 2002
    Location
    Jurong
    Posts
    160

    Default

    Maybe I was too harsh on my previous statements. An addition to make is that, for you who find that the difference between 300ppi and 600ppi is not noticeable, you may not be wrong. Maybe you're not comparing with scenery, landscape or group photos which have enough fine details to exhibit the difference. For normal/large object photos, difference may not be seen at all.

  15. #55
    Senior Member
    Join Date
    Jan 2002
    Location
    Northwest
    Posts
    5,011

    Default

    Wow, you have revived this thread. I do not find your statements harsh at all.

    Here's one of my theories again. Feel free to help me refine it or criticise it:

    I feel that one factor that is often neglected or not scientifically associated in the discussion of ppi is the viewing distance.

    From JC physics class I remember that the resolving power of the human eye (or any optical device) can be expressed as the smallest angle between 2 point light sources than can be barely distinguished as separate light sources. This value most likely varies from person to person. (For me, I know this angle differs a lot between the times when I am wearing my glasses and when I am not! ). Anyway, I do not know what ange of this value applies to an avverage person with good eyesight.

    Now for a 300ppi print, when viewed at 50cm, the angular separation between 2 adjacent pixel centres is 0.0097 degrees (ATAN (pixel distance / viewing distance)). Would an average person have significantly better resolving power than this angle? If yes, then he will more likely be able to tell the difference between a 300ppi and a 600 ppi print. If the resolving power is weaker (larger angle than 0.0097 degrees), then he will not be able to tell the difference between 300 ppi and 600 ppi - at the viewing distance of 50cm.

    As the viewing distance decreases, the angular separation between pixels increases, thus making it easier for one to tell the difference. At the extreme, you can look at the print with a loupe and be able to distinguish the individual dots used to form the pixels.

    What do you think?

    So KamWeng, from your experience, what is the critical distance for you beyond which you will not be able to tell the difference between 300 and 600 ppi?

    Also, related to my initial question, do you notice any degradation in colour resolution when you print at 600ppi?

    By the way, regarding your analogy to CD, regardless of the type of audio processing, the final CD format is still 44.1kHz 16-bit, which correspond better to the print resolution. So even if you can tell that one CD sounds better than the other (even a difference in clarity and detail heard), it is not a good analogy to compare 300 and 600 ppi, which is a difference in the resolution of the fianl media. A better analogy would be to compare 44.1kHz/16-bit against 96kHz/20 bit.

    Thanks!

    Roy
    Last edited by roygoh; 10th August 2002 at 12:59 AM.
    As complexity rises, precise statements lose meaning and meaningful statements lose precision.

  16. #56
    Senior Member
    Join Date
    Jan 2002
    Location
    UK
    Posts
    3,911

    Default

    Here we flaming go again. I missed the first round of stuff because my fresh posts was reset during the Games. Now I've seen the lot, and here we go again.

    [1] KamWeng, I assume from your postings throughout that in referring to 300dpi and 600dpi you are referring to detail resolution input into the printer. Now forget all this "people can tell the difference between two CDs even if they are not playing the same music" thing. Comparing two different things, whether CDs or prints, is just shoddy work ethic. Didn't you get in taught round about in Primary 4 or possibly earlier that in order to conduct a good experiment you must have a good control setup? Secondly, it doesn't matter if you had compared similar prints. The human eye is only able to resolve approximately 300dpi of detail, varying from sample to sample (as in person to person). So if your eye can resolve 600dpi of detail, all power to you. Thirdly, it doesn't matter if you had compared similar prints and you had superpower vision that could distinguish 600dpi of detail, commercial inkjet prints only print at 300dpi of detail resolution anyway, or just maybe up to 360 (I have heard conflicting reports for some printers between 240/360). As in, because I'd better spell this out for you, if you took a 600dpi image and fed it through your printer, it'd throw away the extra resolution and only print at 300dpi or whatever it prints at.

    [2] Pardon me, but this is in no way looking down at your equipment or expertise, but from using a sharpening setting of 9 in Camedia, that suggests to my an Olympus P&S digital with very basic and uniform unsharpmasking applied. Now compared that to calendars is not exactly fair play. I'd certainly consider examining [1] the camera's limitations [2] the software's limitations, especially in the face of what people have already been telling you. And no, maximum unsharp masking is not a bright idea, especially not with basic software and a P&S digital.

    [3] Midnight, the problem with laser printers is that they have funny resolution terms. You can't get anywhere near to a decent photo quality print with a 300dpi laser printer, or even a 600dpi laser printer for that matter. Furthermore they are printing with only one shade and cannot bleed.

    [4] Tsdh has already explained the difference between 6 and 4 colour inksets in inkjets. Roy as we were discussing, the current state of inkjet technology is already just about grasping the required quality, and any further sacrifice in colour resolution can be done without. Furthermore, bear in mind that there is a limit to how small your dots can go for the lighter shades as well. It is far preferable to print, say, 20 light dots in a grid of 100 dots (6 empty), as opposed to 10 darker, smaller dots in a grid of 100 small dots (covering the same area).

    [5] Well the only interpolation that does some magic is the SuperCCD of Fujifilm, because it's hardware interpolation. You know, for the rest of the paragraph until that point, you were making sense. Sorry, but interpolation is interpolation. A Fuji 3.3mp Super CCD still sees only 3.3mp of detail. No amount of marketing smooze, which obviously you've bought, will make it 6mp.

    [6] Regarding 4 colors vs 6 colors, I'd say that higher resolution plus smaller droplet size, for example maybe 5760dpi @ 1pl (the highest currently is 2880dpi @ 2pl) will compensate the lack of light cyan and light magenta and produce the same tone as 6 colors can. See above (pt. 4) on why that's not necessarily true. And more importantly, assuming you are right, why cut from 6 to 4 colours when you can have 6 colours @ 5760dpi @ 1pl. Thereby increasing your colour resolution.

    [7] Yes, you are right about print stability, light cyan and light magenta inks are more noticeable when they fade.

    [8] Do you understand colour spaces, and CMYK specifically? There has already been a discussion on gamut.

    [9] Why don't we have light yellow? Because people like you are already campaigning against light magenta and light cyan, what for you want light yellow?

    [10] At the risk of repeating myself, with 5760dpi @ 1pl, light magenta and light cyan will not be rendered useless simply because of what was discussed in the early part of this thread.

    [11] Dyesub v Inkjet. Tsdh has made his opinon clear, fair play to him, I'm stating mine. In terms of visual output, inkjet is there. I don't care what technology they use, inkjet is there. There are still major problems centering around other facets of the print, in particular light fastness, gas fastness, water fastness, and print speed. Again, my opinion.

    [12] Even the once famous Epson 6 color 870/1270 which claimed 10 years of light fastness received complaints on fading. It was once the hottest topic at dpreview.com. 4 color was well known more stable. Now Epson don't dare to anyhow claim already. For the new 890, they just claim 'achieve high light fastness IF printed on matte paper AND framed under glass'.

    This is (was) not a greatly known fact but I was fairly involved in this issue at the first hand. I was in contact with Epson UK at several stages of the situation and I have a good idea of what was going on. You sound like this is something you read about off hand. So listen to me. The Epson xx70 series, as the later ones such as the xx90 series, do not at all suffer from lightfastness problems, at least not more than was originally claimed and confirmed by Wilheim. The problem is gas fading. Which is why Epson ask for the prints to be framed under glass or in sleeves or laminates, to protect the print from exposure to the atmosphere. As far as dye based printers go, the Epson is still as good as it gets.

    Also, what Epson has done is to bring their pigment based range into reach of serious amateur photographers. The 2100 is arguably a much better printer than its predecessor (and one that I'd own as a former owner of the 2000p just for its speed alone), yet comes in at a pricepoint about 30% less than what its ancestor retailed at. And offhand, only about 25% more expensive than the 1290. Now assuming we avoid problems with atmospheric fading, these prints outlast RA4 prints, dyesub prints, Ilfochrome prints, and quite possibly even conventional black and white prints.

    Now if you think I'm pro-Epson, figure this one. Of the entire UK, I was the person worst hit by the 870/1270 situation. Which is why I was in contact with them. But at least as a person directly involved in this situation I can comfortably throw in my $0.02. But frankly what I've heard from you just typifies the problem of the Internet. People hearing things third hand and then speaking as an authority on the situation at hand.

  17. #57
    Member
    Join Date
    Apr 2002
    Location
    Jurong
    Posts
    160

    Default

    Originally posted by roygoh
    I feel that one factor that is often neglected or not scientifically associated in the discussion of ppi is the viewing distance.

    Agree absolutely. That's what lead us into the debate.

    From JC physics class I remember that the resolving power of the human eye (or any optical device) can be expressed as the smallest angle between 2 point light sources than can be barely distinguished as separate light sources. This value most likely varies from person to person. (For me, I know this angle differs a lot between the times when I am wearing my glasses and when I am not!). Anyway, I do not know what ange of this value applies to an avverage person with good eyesight.

    Oh, why I never learn that 20 years back in JC? I guess we have generation gap.

    Now for a 300ppi print, when viewed at 50cm, the angular separation between 2 adjacent pixel centres is 0.0097 degrees (ATAN (pixel distance / viewing distance)). Would an average person have significantly better resolving power than this angle? If yes, then he will more likely be able to tell the difference between a 300ppi and a 600 ppi print. If the resolving power is weaker (larger angle than 0.0097 degrees), then he will not be able to tell the difference between 300 ppi and 600 ppi - at the viewing distance of 50cm.

    As the viewing distance decreases, the angular separation between pixels increases, thus making it easier for one to tell the difference. At the extreme, you can look at the print with a loupe and be able to distinguish the individual dots used to form the pixels.

    What do you think?


    I can't agree more.

    So KamWeng, from your experience, what is the critical distance for you beyond which you will not be able to tell the difference between 300 and 600 ppi?

    Maybe beyond 30cm or more. I have a strange habit when examining a print which I think normal person won't look at that way : first I look from a distance of around 30cm (not 50cm for me) and then at a distance of ~10cm without my glasses, moving the print back and forth until I can see it clearly (manual focus I call it). In this way, I can see the details which most of my colleagues I asked could not tell.

    I'm a perfectionist. So maybe it's myself who're trying to find fault. As I've pointed out in my previous post, maybe I was asking for too much. I know people normally won't look at a print that careful. I'm just trying to find out the ppi that possibly make prints sharper (in order to impress people), but since most of you don't see that there's difference, there's no point for me to print 'sharper'.

    Also, related to my initial question, do you notice any degradation in colour resolution when you print at 600ppi?

    No. I can only notice the increment in details on fine objects i.e. the titles of the books on a book shelve () , tree leaves etc.

    By the way, regarding your analogy to CD, regardless of the type of audio processing, the final CD format is still 44.1kHz 16-bit, which correspond better to the print resolution. So even if you can tell that one CD sounds better than the other (even a difference in clarity and detail heard), it is not a good analogy to compare 300 and 600 ppi, which is a difference in the resolution of the fianl media. A better analogy would be to compare 44.1kHz/16-bit against 96kHz/20 bit.

    No comment. I was just replying the guy with his own example. I didn't analyzed further.

    Thanks!

    Your open and friendly discussion is most welcomed.

  18. #58
    Member
    Join Date
    Apr 2002
    Location
    Jurong
    Posts
    160

    Default

    Originally posted by Jed
    Here we flaming go again. I missed the first round of stuff because my fresh posts was reset during the Games. Now I've seen the lot, and here we go again.

    [1] KamWeng, I assume from your postings throughout that in referring to 300dpi and 600dpi you are referring to detail resolution input into the printer. Now forget all this "people can tell the difference between two CDs even if they are not playing the same music" thing. Comparing two different things, whether CDs or prints, is just shoddy work ethic. Didn't you get in taught round about in Primary 4 or possibly earlier that in order to conduct a good experiment you must have a good control setup? Secondly, it doesn't matter if you had compared similar prints. The human eye is only able to resolve approximately 300dpi of detail, varying from sample to sample (as in person to person). So if your eye can resolve 600dpi of detail, all power to you. Thirdly, it doesn't matter if you had compared similar prints and you had superpower vision that could distinguish 600dpi of detail, commercial inkjet prints only print at 300dpi of detail resolution anyway, or just maybe up to 360 (I have heard conflicting reports for some printers between 240/360). As in, because I'd better spell this out for you, if you took a 600dpi image and fed it through your printer, it'd throw away the extra resolution and only print at 300dpi or whatever it prints at.

    [2] Pardon me, but this is in no way looking down at your equipment or expertise, but from using a sharpening setting of 9 in Camedia, that suggests to my an Olympus P&S digital with very basic and uniform unsharpmasking applied. Now compared that to calendars is not exactly fair play. I'd certainly consider examining [1] the camera's limitations [2] the software's limitations, especially in the face of what people have already been telling you. And no, maximum unsharp masking is not a bright idea, especially not with basic software and a P&S digital.

    [3] Midnight, the problem with laser printers is that they have funny resolution terms. You can't get anywhere near to a decent photo quality print with a 300dpi laser printer, or even a 600dpi laser printer for that matter. Furthermore they are printing with only one shade and cannot bleed.

    [4] Tsdh has already explained the difference between 6 and 4 colour inksets in inkjets. Roy as we were discussing, the current state of inkjet technology is already just about grasping the required quality, and any further sacrifice in colour resolution can be done without. Furthermore, bear in mind that there is a limit to how small your dots can go for the lighter shades as well. It is far preferable to print, say, 20 light dots in a grid of 100 dots (6 empty), as opposed to 10 darker, smaller dots in a grid of 100 small dots (covering the same area).

    [5] Well the only interpolation that does some magic is the SuperCCD of Fujifilm, because it's hardware interpolation. You know, for the rest of the paragraph until that point, you were making sense. Sorry, but interpolation is interpolation. A Fuji 3.3mp Super CCD still sees only 3.3mp of detail. No amount of marketing smooze, which obviously you've bought, will make it 6mp.

    [6] Regarding 4 colors vs 6 colors, I'd say that higher resolution plus smaller droplet size, for example maybe 5760dpi @ 1pl (the highest currently is 2880dpi @ 2pl) will compensate the lack of light cyan and light magenta and produce the same tone as 6 colors can. See above (pt. 4) on why that's not necessarily true. And more importantly, assuming you are right, why cut from 6 to 4 colours when you can have 6 colours @ 5760dpi @ 1pl. Thereby increasing your colour resolution.

    [7] Yes, you are right about print stability, light cyan and light magenta inks are more noticeable when they fade.

    [8] Do you understand colour spaces, and CMYK specifically? There has already been a discussion on gamut.

    [9] Why don't we have light yellow? Because people like you are already campaigning against light magenta and light cyan, what for you want light yellow?

    [10] At the risk of repeating myself, with 5760dpi @ 1pl, light magenta and light cyan will not be rendered useless simply because of what was discussed in the early part of this thread.

    [11] Dyesub v Inkjet. Tsdh has made his opinon clear, fair play to him, I'm stating mine. In terms of visual output, inkjet is there. I don't care what technology they use, inkjet is there. There are still major problems centering around other facets of the print, in particular light fastness, gas fastness, water fastness, and print speed. Again, my opinion.

    [12] Even the once famous Epson 6 color 870/1270 which claimed 10 years of light fastness received complaints on fading. It was once the hottest topic at dpreview.com. 4 color was well known more stable. Now Epson don't dare to anyhow claim already. For the new 890, they just claim 'achieve high light fastness IF printed on matte paper AND framed under glass'.

    This is (was) not a greatly known fact but I was fairly involved in this issue at the first hand. I was in contact with Epson UK at several stages of the situation and I have a good idea of what was going on. You sound like this is something you read about off hand. So listen to me. The Epson xx70 series, as the later ones such as the xx90 series, do not at all suffer from lightfastness problems, at least not more than was originally claimed and confirmed by Wilheim. The problem is gas fading. Which is why Epson ask for the prints to be framed under glass or in sleeves or laminates, to protect the print from exposure to the atmosphere. As far as dye based printers go, the Epson is still as good as it gets.

    Also, what Epson has done is to bring their pigment based range into reach of serious amateur photographers. The 2100 is arguably a much better printer than its predecessor (and one that I'd own as a former owner of the 2000p just for its speed alone), yet comes in at a pricepoint about 30% less than what its ancestor retailed at. And offhand, only about 25% more expensive than the 1290. Now assuming we avoid problems with atmospheric fading, these prints outlast RA4 prints, dyesub prints, Ilfochrome prints, and quite possibly even conventional black and white prints.

    Now if you think I'm pro-Epson, figure this one. Of the entire UK, I was the person worst hit by the 870/1270 situation. Which is why I was in contact with them. But at least as a person directly involved in this situation I can comfortably throw in my $0.02. But frankly what I've heard from you just typifies the problem of the Internet. People hearing things third hand and then speaking as an authority on the situation at hand.
    Wah! This is probably The Longest Post Award winner.

    I admit defeated. Though I definitely find arguable points within your lecture but I refuse to answer/debate further because either I

    1. don't know how to, or
    2. don't have that much time even if I know how to, or
    3. too tired after work even if I have time to.

    Guess I should learn to feel satisfy with my 2mp camera and 1440dpi 3rd Epson printer and be happy from here on... Who needs that fake '6mp' (since I can get the same thing by software interpolation, oops, are you saying that Fuji's engineers are trying to bulls**t us?) and 2880dpi (which anyway prints at 300dpi)?

    You win.

    P/S: I just realized that my previous post was not harsh at all. Yeah, compare tells the difference.
    Last edited by KamWeng; 10th August 2002 at 10:48 PM.

  19. #59
    Senior Member
    Join Date
    Jan 2002
    Location
    UK
    Posts
    3,911

    Default

    Originally posted by KamWeng
    Wah! This is probably The Longest Post Award winner.
    You obviously haven't seen some of my stuff have you?

    Guess I should learn to feel satisfy with my 2mp camera and 1440dpi 3rd Epson printer and be happy from here on... Who needs that fake '6mp' (since I can get the same thing by software interpolation, oops, are you saying that Fuji's engineers are trying to bulls**t us?) and 2880dpi (which anyway prints at 300dpi)?
    Good textbook example of sarcasm. Playing the "you win" game yet trying to make your point. Very well done. You are [1] so sad, and [2] haven't a clue.

    And just for the record, I've already answered the Fuji issue, and 2880dpi printers don't print at 300dpi, they print at 240/360dpi visual resolution, go read my answer properly.

    And then go read my signature.

  20. #60

    Default

    Originally posted by KamWeng


    Guess I should learn to feel satisfy with my 2mp camera and 1440dpi 3rd Epson printer and be happy from here on... Who needs that fake '6mp' (since I can get the same thing by software interpolation, oops, are you saying that Fuji's engineers are trying to bulls**t us?) and 2880dpi (which anyway prints at 300dpi)?


    As Norman Koren would say:
    "The printer's dpi specification is meaningless ; it is merely the stepper motor pitch. Epson specifies the "resolution" of its new 1280 printer as 2880 x720 dpi. Horsefeathers! (For non-English speakers, "Horsefeathers" is the title of an old Groucho Marx film, implying the ridiculous or absurd.)"
    http://www.normankoren.com/Tutorials/MTF3.html

    Create a photoshop image 600x600 pixels. Draw 300 1 pixel vertical (or horizontal) lines of varying colours (include some very light pastel shades) with 1 pixel white space between them. Print the image at 1 inchx1 inch. Examine printout with magnifier. Do you have 300 distinct lines?

    Then compare it to printing the same image at 2 inch x 2 inch.

Page 3 of 4 FirstFirst 1234 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •