Yahoo Groups archive
Thread
2005-08-19 by Ernst Dinkla
There was a problem reported in the lists with the 51 linearisation some time ago. Steve Kale ? Is that solved ? Will QTR work with a custom range of 41 steps if the 51 range is too subtle for measurements-computing ? So far I have not seen linearisation ranges that go beyond 40 patches. Ernst
2005-08-19 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > There was a problem reported in the lists with the 51 linearisation some > time ago. Steve Kale ? > Is that solved ? > Will QTR work with a custom range of 41 steps if the 51 range is too > subtle for measurements-computing ? > So far I have not seen linearisation ranges that go beyond 40 patches. > > Ernst The Linearization does work with 51 steps but since the step values are closer its a little more sensitive to lack of separation betweem the steps. For the most part I think 21 is plenty. Compared to color charts that are commonly used this is way more data points per dimension -- 1000 patches for color in 3 dimension is only 10 point per dimension. Roy
2005-08-20 by Ernst Dinkla
Roy Harrington wrote: >--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > > >>There was a problem reported in the lists with the 51 linearisation some >>time ago. Steve Kale ? >>Is that solved ? >>Will QTR work with a custom range of 41 steps if the 51 range is too >>subtle for measurements-computing ? >>So far I have not seen linearisation ranges that go beyond 40 patches. >> >>Ernst >> >> > >The Linearization does work with 51 steps but since the step values are >closer its a little more sensitive to lack of separation betweem the steps. > > That's what I expected to be the cause. >For the most part I think 21 is plenty. Compared to color charts that are >commonly used this is way more data points per dimension -- 1000 patches >for color in 3 dimension is only 10 point per dimension. > > > The targets you mention are the ICC profiling targets that usually build on linearised inkchannels. Linearisation targets exist in many shapes CMYK-CcMmYK-Ncolor and they have ranges of 11 to 41 patches per inkchannel. the lowest number can be used for printsystems that actually address the heads separately but on a CMYK driver that is linearised as CMYK but drives a CcMmYK printer it is better to use the 21 patches so there are 10 steps measured per head. With the 4800 that has 3Ks you could go up to 31 steps and that has been used for Colorspan models that have 3 ink dilutions of CMYK each = 12 heads. 21 should be good for B&W systems K2-K3-(K4). But for a K7 set it means 3 measure points per ink channel (head) which is too low in quantity. The CMYK linearisation of the Wasatch SoftRip has 14 patches per color, on the Cc and Mm inkchannels of the 9000 it means 7 patches per C,c,M,m head and I think that's about the lowest you can go. 51 for K7 is in that category. A linearisation target that falls in between the two would be nice for K4-K6 inksets. lt has more measurements per head than the 21 and more separation between the steps than the 51. While it wasn't an easy job to change the Wasatch SoftRip into quad printing it had the linearisation at inkchannel level. The CcMm inchannels had a K4 setup and together 28 steps for measurement. The toners each 14. On top of that the separation curves were build. I checked the linearisation of the total then with 21 or 44 steps to get the curves right and added a curve (very subjective) to get the perceptual linearisation. This is all much easier in QTR. However I think that the relation between the step number and the total of printer heads measured shouldn't be forgotten. The ICC profiling done with the 21 step target on top of the linearisation done with the same 21 step target may work as an iterative measurement as well and so average the result a bit (it produces another curve of course) but whether that is sufficient for B&W sets above K4 isn't clear. I think there's improvement possible. Ernst
2005-08-20 by Tyler Boley
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> ... snip... > The targets you mention are the ICC profiling targets that usually build > on linearised inkchannels. Linearisation targets exist in many shapes > CMYK-CcMmYK-Ncolor and they have ranges of 11 to 41 patches per > inkchannel. the lowest number can be used for printsystems that actually > address the heads separately but on a CMYK driver that is linearised as > CMYK but drives a CcMmYK printer it is better to use the 21 patches so > there are 10 steps measured per head. With the 4800 that has 3Ks you > could go up to 31 steps and that has been used for Colorspan models that > have 3 ink dilutions of CMYK each = 12 heads. My impression of the process is a bit different, correct me if I'm worng. Individual ink components of a color, C and c for example, or the 3Ks you mention, are generally not addressed individually in the linearization process. Some control is usually given with respect to how the transitions will happen, but then the color (all inclusive "C", or"K") is linearized as one. How exactly the total patch count may divide up between those indivicual components may not be very tidy. I'm sure this varies from RIP to RIP. I recall Roy asking if we thought there would be an advantage to individul ink linearization, and wonder the same thing myself. > 21 should be good for B&W systems K2-K3-(K4). I have on occassion seen an advantage to doing many more here with 4, and wound up at 80. I've seen a higher count like this fix problems between say 95% and 100%, more than once. > But for a K7 set it means > 3 measure points per ink channel (head) which is too low in quantity. The above may apply here again. It seems, under the curcumstances, we are somewhat at the mercy of the partitioning scheme. If it's very well behaved and designed for the ink set and media, a good overall linearization (with an adequate patch count as you are discussing) should work well. In fact this seems to be what we are all doing for now anyway. The downside is that a less then stellar partition scheme can be bad enough that overall linearization will not fix it. With what most people are using, that doesn't seem to happen much. > The CMYK linearisation of the Wasatch SoftRip has 14 patches per color, > on the Cc and Mm inkchannels of the 9000 it means 7 patches per C,c,M,m > head and I think that's about the lowest you can go. So it does allow you to linearize each color component individually? Studioprint does not, though it allows one to enter any per color patch count number they want, I'm sure there is an upper limit, I think 100? Set's with toners present a challenge as well, it's not a common color model SP is preset to work with. Sounds like your RIP has some options there you have bent to your will <G>. ...snip... > The ICC profiling done with the 21 step target on top of the > linearisation done with the same 21 step target may work as an iterative > measurement as well and so average the result a bit (it produces another > curve of course) but whether that is sufficient for B&W sets above K4 > isn't clear. I think there's improvement possible. no doubt. But the partitioning step needs to be considered as a major part of the overall performance picture. How many patches and how much work needs to be done at the end can have a lot to do with how well behaved the ramp is to begin with. That few have trouble with QTR speaks well of Roy's existing partitioning scheme. Setting it up to pull 7 inks in and out without major problems must have been fun, or not! Tyler
2005-08-20 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > Roy Harrington wrote: > > >--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > > > > > >>There was a problem reported in the lists with the 51 linearisation some > >>time ago. Steve Kale ? > >>Is that solved ? > >>Will QTR work with a custom range of 41 steps if the 51 range is too > >>subtle for measurements-computing ? > >>So far I have not seen linearisation ranges that go beyond 40 patches. > >> > >>Ernst > >> > >> > > > >The Linearization does work with 51 steps but since the step values are > >closer its a little more sensitive to lack of separation betweem the steps. > > > > > That's what I expected to be the cause. > > > >For the most part I think 21 is plenty. Compared to color charts that are > >commonly used this is way more data points per dimension -- 1000 patches > >for color in 3 dimension is only 10 point per dimension. > > > > > > > The targets you mention are the ICC profiling targets that usually build > on linearised inkchannels. Linearisation targets exist in many shapes > CMYK-CcMmYK-Ncolor and they have ranges of 11 to 41 patches per > inkchannel. the lowest number can be used for printsystems that actually > address the heads separately but on a CMYK driver that is linearised as > CMYK but drives a CcMmYK printer it is better to use the 21 patches so > there are 10 steps measured per head. With the 4800 that has 3Ks you > could go up to 31 steps and that has been used for Colorspan models that > have 3 ink dilutions of CMYK each = 12 heads. > > 21 should be good for B&W systems K2-K3-(K4). But for a K7 set it means > 3 measure points per ink channel (head) which is too low in quantity. > The CMYK linearisation of the Wasatch SoftRip has 14 patches per color, > on the Cc and Mm inkchannels of the 9000 it means 7 patches per C,c,M,m > head and I think that's about the lowest you can go. 51 for K7 is in > that category. A linearisation target that falls in between the two > would be nice for K4-K6 inksets. lt has more measurements per head than > the 21 and more separation between the steps than the 51. > > While it wasn't an easy job to change the Wasatch SoftRip into quad > printing it had the linearisation at inkchannel level. The CcMm > inchannels had a K4 setup and together 28 steps for measurement. The > toners each 14. On top of that the separation curves were build. I > checked the linearisation of the total then with 21 or 44 steps to get > the curves right and added a curve (very subjective) to get the > perceptual linearisation. This is all much easier in QTR. However I > think that the relation between the step number and the total of printer > heads measured shouldn't be forgotten. > > The ICC profiling done with the 21 step target on top of the > linearisation done with the same 21 step target may work as an iterative > measurement as well and so average the result a bit (it produces another > curve of course) but whether that is sufficient for B&W sets above K4 > isn't clear. I think there's improvement possible. > > Ernst Interesting comments. I hadn't thought to compare the number of steps with the number of ink partitions. It probably depends somewhat on the method used to partition the inks. QTR uses a special algorithm that takes into account the relative densities of the inks so the transitions don't have flat spots and all 256 values per ink are calculated rather than just smoothly connected. In any case I've made some K7 profiles and I haven't noticed any issues with it. Another factor that I'm beginning to think is also more important is the variable dotsize of the printers. My main printer has been the 7500 which has only one drop size and linearity has always been very smooth. But my smaller printers are all variable dot and I think that may be a large factor in their more bumpy linearity per ink. Once you add a bunch of inks the bumps usually even out. Roy
2005-08-20 by Tyler Boley
--- In QuadtoneRIP@yahoogroups.com, "Roy Harrington" <roy@h...> wrote: ... > Another factor that I'm beginning to think is also more important is the > variable dotsize of the printers. My main printer has been the 7500 which has > only one drop size and linearity has always been very smooth. But my smaller > printers are all variable dot and I think that may be a large factor in their more > bumpy linearity per ink. Yes! Seeing that here too... Basically, with the x600s and SP, variable dot can't be linearized with over two partitions. At least I haven't been able to do it. They're going to have to make it happen for the x800s though, somehow, with the K3s. What I'd really like... not vareiable, but to have different dots sizes available in different inks. For example- Give me the smallest size, particularly in the highlight inks, but then give me your largest size for the skeletal black for the best dmax. But you can leave each ink non-variable, make sense? Doable? Tyler
2005-08-20 by john dean
What I'd really like... not vareiable, but to have different dots sizes available in different inks. For example- Give me the smallest size, particularly in the highlight inks, but then give me your largest size for the skeletal black for the best dmax. But you can leave each ink non-variable, make sense? Doable? Tyler Wo. That is like what I used to do when we made subtle duo tones and tri tones for Heidelburg offset reproduction. Small overlapped dots in the highlights less dots and more singular ones in the deepest shadows.
2005-08-21 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, "Tyler Boley" <tyler@t...> wrote: > --- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> ... > snip... > > The targets you mention are the ICC profiling targets that usually > build > > on linearised inkchannels. Linearisation targets exist in many shapes > > CMYK-CcMmYK-Ncolor and they have ranges of 11 to 41 patches per > > inkchannel. the lowest number can be used for printsystems that > actually > > address the heads separately but on a CMYK driver that is linearised as > > CMYK but drives a CcMmYK printer it is better to use the 21 patches so > > there are 10 steps measured per head. With the 4800 that has 3Ks you > > could go up to 31 steps and that has been used for Colorspan models > that > > have 3 ink dilutions of CMYK each = 12 heads. > > My impression of the process is a bit different, correct me if I'm > worng. Individual ink components of a color, C and c for example, or > the 3Ks you mention, are generally not addressed individually in the > linearization process. Some control is usually given with respect to > how the transitions will happen, but then the color (all inclusive > "C", or"K") is linearized as one. How exactly the total patch count > may divide up between those indivicual components may not be very tidy. > I'm sure this varies from RIP to RIP. I recall Roy asking if we > thought there would be an advantage to individul ink linearization, > and wonder the same thing myself. > ..... > > no doubt. But the partitioning step needs to be considered as a major > part of the overall performance picture. How many patches and how much > work needs to be done at the end can have a lot to do with how well > behaved the ramp is to begin with. > That few have trouble with QTR speaks well of Roy's existing > partitioning scheme. Setting it up to pull 7 inks in and out without > major problems must have been fun, or not! > Tyler The 7 inks for K7 worked just fine. The partitioning code has been the same since day one. I thought about more levels of linearization but I'd probably do it after the partitioning rather that before with each ink. But it's interesting that with the K7 inks its noticibly (but not much) better to use 2880dpi with only one dropsize rather than 1440dpi which is using two sizes. I think its because the 2X difference in drop size is as big as the difference between ink densities. Roy
2005-08-21 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, "Tyler Boley" <tyler@t...> wrote: > --- In QuadtoneRIP@yahoogroups.com, "Roy Harrington" <roy@h...> wrote: > ... > > Another factor that I'm beginning to think is also more important is the > > variable dotsize of the printers. My main printer has been the 7500 > which has > > only one drop size and linearity has always been very smooth. But > my smaller > > printers are all variable dot and I think that may be a large factor > in their more > > bumpy linearity per ink. > > Yes! Seeing that here too... > Basically, with the x600s and SP, variable dot can't be linearized > with over two partitions. At least I haven't been able to do it. From what I've gathered so far the transitions between dropsizes is easily as crucial as the transitions between inks. Unless both are done smoothly as based on the relative densities/dropsizes it awful difficult to compensate later on in the linearization. > They're going to have to make it happen for the x800s though, somehow, > with the K3s. > What I'd really like... not vareiable, but to have different dots > sizes available in different inks. > For example- Give me the smallest size, particularly in the highlight > inks, but then give me your largest size for the skeletal black for > the best dmax. But you can leave each ink non-variable, make sense? > Doable? No exactly. You have to select one set of dropsizes per print. But there's a lot of potential and need for selecting which dropsizes to use where. One thing that I've figured out lately is that too many small dots are bad. They tend to show banding -- mixing the sizes is important, even when the dots may show a little more. (Banding is worse than dots). Roy > Tyler
2005-08-21 by Tyler Boley
--- In QuadtoneRIP@yahoogroups.com, "Roy Harrington" <roy@h...> wrote: ... > From what I've gathered so far the transitions between dropsizes is easily > as crucial as the transitions between inks. Unless both are done smoothly > as based on the relative densities/dropsizes it awful difficult to compensate > later on in the linearization. That's certainly what I'm seeing here. Of course once again Epson seems to able to utilize their screening, dithering, ramping dots sizes, etc to best effect. The 4800 samples I have here, in ABW mode, the transitions and dot structure seems very smooth with a 3 part K. Ot course they are dithering in a bit of 3 more inks, which helps I'm sure. > > They're going to have to make it happen for the x800s though, somehow, > > with the K3s. > > What I'd really like... not vareiable, but to have different dots > > sizes available in different inks. > > For example- Give me the smallest size, particularly in the highlight > > inks, but then give me your largest size for the skeletal black for > > the best dmax. But you can leave each ink non-variable, make sense? > > Doable? > > No exactly. You have to select one set of dropsizes per print. But there's a > lot of potential and need for selecting which dropsizes to use where. Ah well, with limited knowledge it's easy to have bright ideas. > One > thing that I've figured out lately is that too many small dots are bad. They > tend to show banding -- mixing the sizes is important, even when the dots > may show a little more. (Banding is worse than dots). Interesting, with all but one of SP's dither modes (including the best one for 1440) I got banding with 2880. Of course 2880 uses one dot size, ultra-micro. This would support your observations. Another thing effecting smooth ink transitions in SP, the pass options. Between dither and pass options, things get a bit mysterious. Can you effect 2880 banding with QTR dither options? Tyler
2005-08-21 by Ernst Dinkla
Tyler Boley wrote: >--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> ... >snip... > > >>The targets you mention are the ICC profiling targets that usually >> >> >build > > >>on linearised inkchannels. Linearisation targets exist in many shapes >>CMYK-CcMmYK-Ncolor and they have ranges of 11 to 41 patches per >>inkchannel. the lowest number can be used for printsystems that >> >> >actually > > >>address the heads separately but on a CMYK driver that is linearised as >>CMYK but drives a CcMmYK printer it is better to use the 21 patches so >>there are 10 steps measured per head. With the 4800 that has 3Ks you >>could go up to 31 steps and that has been used for Colorspan models >> >> >that > > >>have 3 ink dilutions of CMYK each = 12 heads. >> >> > >My impression of the process is a bit different, correct me if I'm >worng. Individual ink components of a color, C and c for example, or >the 3Ks you mention, are generally not addressed individually in the >linearization process. Some control is usually given with respect to >how the transitions will happen, but then the color (all inclusive >"C", or"K") is linearized as one. > I thought that I described that in similar words. >How exactly the total patch count >may divide up between those indivicual components may not be very tidy. > > Whether it exactly divides up to equal linearisation points per channel doesn't matter much. But if you check for example the linearisation correction curves of an Epson 9000 then I wonder how 3 points per head would be sufficient. And that printer isn't using different droplet sizes per head. That the linearisation in the end has to set 256 shades right in all setups doesn't mean that there will be no differences in its task between different B&W inksets/printers. The extreme case of 7 non linearised head curves bulging upwards including 2 droplet size curve nicks per head + 6 points where the curves have to join, will not happen I guess because the partitioning and overlap in the partioning point cuts droplet sizes, smooths both droplet size transfers and the partitioning point tranfers. The upper part of most of the non linearised head curves is cut off anyway as the density isn't increasing there. Where the smoothing happens, where some droplet sizes vanish is hard to predict, we partion on the max density per channel and where it fits the black channel density. A bit of a black box there, which may be better controlled by 50 linearisation points than with 20. That the partitioning has an important role to play is clear to me. The trend towards K7 inks isn't one I see as a solution and I think that using more heads + less dilutions of black ink + effective use of droplet sizes and actually using the mix you get from heads weaving (as intended in the printer design) will in the end deliver less banding, better linearisation, more consistency in time. I have written this before in relation to BO printing and on the K7 threads. There are indications that Epson did change the weaving for the Adv. B&W driver in the K3 printers to give better B&W printing. So far we didn't address that aspect much (Paul Roark did however in his distribution of the quad inks, then more or less directed by the limits of the Epson driver), and thought it would be perfect to add black dilutions where the head quantity increases. Let's say it is like putting the 4 wheels of a car in line, has some advantages but not in all aspects. >I'm sure this varies from RIP to RIP. I recall Roy asking if we >thought there would be an advantage to individul ink linearization, >and wonder the same thing myself. > > I do not think that's wise either. Adding the separation curves after linearisation introduces the imbalance of the curves themselves. Profiling to the perceptual curve could smooth that out of course. It will also take more time to make curves for new ink/paper combinations. I have been on that route and I do not want to go back. > > >>21 should be good for B&W systems K2-K3-(K4). >> >> > >I have on occassion seen an advantage to doing many more here with 4, >and wound up at 80. I've seen a higher count like this fix problems >between say 95% and 100%, more than once. > > It must be quite difficult to get the readings right for 80 steps but averaging readings on more targets before computing helps. Something I did throw out in my first message: the steps in the linearisation target/software are fixed on for example 5% increase. A target of 30 steps with some step variation where it counts would keep the target number down but increase the linearisation quality. But like you mention the partitioning differences in quad setups may make it difficult to see where it counts to have a higher frequency in steps. The shadows are without doubt a rewarding spot to have more steps for measurement. Maybe a separate profiling target could get another step frequency related to the perceptual values. It will diminish the iterative function of the measurements but make the last step more suited for the task. It probably would require higher step frequency in the linearisation target where it builds on to keep the steps you print for the profile making more related to the first readings (and less on the interpolated parts of the linearisation curve). > > >>But for a K7 set it means >>3 measure points per ink channel (head) which is too low in quantity. >> >> > >The above may apply here again. It seems, under the curcumstances, we >are somewhat at the mercy of the partitioning scheme. If it's very >well behaved and designed for the ink set and media, a good overall >linearization (with an adequate patch count as you are discussing) >should work well. >In fact this seems to be what we are all doing for now anyway. The >downside is that a less then stellar partition scheme can be bad >enough that overall linearization will not fix it. >With what most people are using, that doesn't seem to happen much. > > > > >>The CMYK linearisation of the Wasatch SoftRip has 14 patches per color, >>on the Cc and Mm inkchannels of the 9000 it means 7 patches per C,c,M,m >>head and I think that's about the lowest you can go. >> >> > >So it does allow you to linearize each color component individually? > > No, it takes the C+c channel together but limits the linearisation to 14 patches so 7 patches for C and c individually if they are distributed equally to both sides. With (extra) linearisation done with other profiler software you could add patches but internally it is not possible. > > >>The ICC profiling done with the 21 step target on top of the >>linearisation done with the same 21 step target may work as an >> >> >iterative > > >>measurement as well and so average the result a bit (it produces >> >> >another > > >>curve of course) but whether that is sufficient for B&W sets above K4 >>isn't clear. I think there's improvement possible. >> >> > >no doubt. But the partitioning step needs to be considered as a major >part of the overall performance picture. How many patches and how much >work needs to be done at the end can have a lot to do with how well >behaved the ramp is to begin with. >That few have trouble with QTR speaks well of Roy's existing >partitioning scheme. Setting it up to pull 7 inks in and out without >major problems must have been fun, or not! > > I have no 7 ink experiences as I have written before but having done my share of experiments I think that there are more ways to get smooth prints. The linearisation etc of QTR is very user friendly and has been sufficient for me so far. The almost custom quad inks + two different hue toners distributed in another way on the 9000 that I have were easy to get in line. The prints are not the smoothest around given the printer and number of k inks and one toner but I have done my best to get the linearisation right and customers are satisfied. That printer with its limited nozzle count and differences between head output learned me something about partioning and linearisation. I'm trying to get a picture of what is possible in linearisation. Where finer step differences still contribute to the result or get in conflict with the consistency etc of the spectrometer and the repeatability etc of the printer. The partitioning is important, the distribution of the inks, the number of heads and nozzles as well. I'm not ultimately seeking higher target patch numbers but whether the linearisation and consistency of printing can be made better, in repeatability, banding, detail etc as well. Setting the target step number against 256 shades is one thing, setting the number against the number of heads/nozzles/inks involved is another thing. If all 7 heads were used for BO they even out their differences and the linearisation curve would be a simple one I guess. Not to mention the one droplet size the printer is forced to then. Ernst
2005-08-21 by Ernst Dinkla
Roy, you wrote: > >But it's interesting that with the K7 inks its noticibly (but not much) better to use >2880dpi with only one dropsize rather than 1440dpi which is using two sizes. >I think its because the 2X difference in drop size is as big as the difference >between ink densities. > > > Which to me means that the K7 inkset is already stretching the ink partitioning too far. Better linearisation of the droplet transfer points should deliver equal quality with less ink dilutions. There's also no advantage in consistency by having one droplet size over the entire range. In practice this means the smallest droplet and that one is more prone to visible dotgain when humidity, coating qualities etc vary. John Dean described what was used for duotone offset printing and that had much to do with dotgain increase of small dots at 30% and up so a coarser dot formation is used there to diminish the area/circumference ratio and by that dotgain/bleeding. Resolution suffers then but it would be hit by bleeding as well and at least the density can now be controlled. Solving the dotgain issues of stochastic printing in offset wasn't possible without strict control of all the variables and in practice hybrids of halftone and stochastic printing were introduced to make the process more stable. In the shadows the dotgain isn't reduced by diluted inks either, the effect is as it is with BO printing at that point. Ernst
2005-08-21 by Tyler Boley
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: ... > I think that using > more heads + less dilutions of black ink + effective use of droplet > sizes and actually using the mix you get from heads weaving (as intended > in the printer design) will in the end deliver less banding, better > linearisation, more consistency in time. I have written this before in > relation to BO printing and on the K7 threads. Are you speaking of using different density K inks in parallel rather than partitioned series, or some combination of the two? We discussed this once before, I thought the BO users could benefit from a the more complex dither that would result. I never got the chance to test it. Life is getting too short for all this testing. Actually, can you assign two inks to the same part of the scale in QTR? ... > >I'm sure this varies from RIP to RIP. I recall Roy asking if we > >thought there would be an advantage to individual ink linearization, > >and wonder the same thing myself. > > > > > I do not think that's wise either. Adding the separation curves after > linearisation introduces the imbalance of the curves themselves. > Profiling to the perceptual curve could smooth that out of course. It > will also take more time to make curves for new ink/paper combinations. > I have been on that route and I do not want to go back. Believe me, with all due respect to Dan Culbertson, neither do I <G>. ... > >I have on occassion seen an advantage to doing many more here with 4, > >and wound up at 80. I've seen a higher count like this fix problems > >between say 95% and 100%, more than once. > > > > > It must be quite difficult to get the readings right for 80 steps but > averaging readings on more targets before computing helps. Yes, it will automatically average multiple readings of the same chart, and warn you if you should reread a strip due to a suspicious difference. But even with little averaging when in a hurry, I've only seen very occasional errors. Also, repeated linearizations are iterative, which helps. But I know what you mean and was wary of potential problems at first with large patch counts. > Something I did throw out in my first message: the steps in the > linearisation target/software are fixed on for example 5% increase. A > target of 30 steps with some step variation where it counts would keep > the target number down but increase the linearisation quality. Seems like a great idea, I'm sure 80 for me is overkill just to get more linearization approaching 100%. Tyler
2005-08-22 by Ernst Dinkla
Tyler Boley wrote: >--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: >... > > >>I think that using >>more heads + less dilutions of black ink + effective use of droplet >>sizes and actually using the mix you get from heads weaving (as intended >>in the printer design) will in the end deliver less banding, better >>linearisation, more consistency in time. I have written this before in >>relation to BO printing and on the K7 threads. >> >> > >Are you speaking of using different density K inks in parallel rather than partitioned series, >or some combination of the two? >We discussed this once before, I thought the BO users could benefit from a the more >complex dither that would result. I never got the chance to test it. Life is getting too short >for all this testing. >Actually, can you assign two inks to the same part of the scale in QTR? > > I've used a quad setup on the 9000 that had the Eboni black + the middle grey ink in the Mm channels and was linearised like that and in the Cc channels, the darkest grey + lightest grey ink in Mm and linearised like that again. Some partitioning was laid on top of that, (the highlights were done by the lightest range only, mixing of the two ranges was of course mainly in the mid ranges, 2% of the darkest grey went up with the black) trying to get as close to native linearisation and an overall linearisation etc added afterwards. Wasatch SoftRip. That's a simple example of what I propose. BTW, Paul Roark's PS curves and ink mixes must have had a similar structure. The crudest example is having 4 black inks in a C86 for BO, that will give you a nozzle quantity that comes near 360 if I recall it correctly. On a K7 printer and with QTR you could make one K4 range and a K3 ink range, linearise them and mix them with the slider. The 5 partitioning points of the two ranges are all overlapped, and you get weaving of two heads on any spot. Depending on the droplet sizes available (resolution) and the partitioning the droplet size transfer points could overlap as well. If you use the normal K7 inks then you would expect more visible dots in the highlights but I think it will not be much more visible. You will loose the possibility to mix toner in but that's not available in the K7 set anyway. No need to change inks. Ink distribution may benefit from the original CMYK weaving positions though, for example the K3 in the CcY lines the K4 in the KkMm lines. In a way these overlapping quads resemble the B&W films with two or more emulsion layers. I have no idea how the K7 inks are distributed on the ink positions, if they use the two main lines as sketched above and given that the overlap in the partitions is already relatively wider as one head only covers about 14% of the total range, the result may be close to having two ranges next to one another. The ink distribution something like K7-K, K6-C, K5-k, K4-c, K3-M, K2-Y, K1-m. If that's happening already then I shouldn't worry about the K7 concept. No, I don't think you can assign two inks to the same part in QTR, I have asked that once and got an answer by Carl or Daniel but couldn't figure it out or my question wasn't understood. I think it would be good if QTR allowed the use of multiple heads for the same task. Ernst
2005-08-22 by John Moody
Roy,
I thought the 2-bit VSD mode had three drop sizes? The bits would be no-dot, 00, then 01, 10, and 11, right?
It appears the relative drop size ratio is printer and mode dependant as well. Some Epson stuff here. http://www.epson.co.jp/e/technology/sankei_4.htm
Best regards,
John Moody
-----Original
Message-----
From: QuadtoneRIP@yahoogroups.com
[mailto:QuadtoneRIP@yahoogroups.com]On Behalf
Of Roy Harrington
Sent: Saturday, August 20, 2005
9:48 PM
To: QuadtoneRIP@yahoogroups.com
Subject: [QuadtoneRIP] Re: QTR
51-step linearisation
--- In QuadtoneRIP@yahoogroups.com, "Tyler Boley"
wrote:
> --- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla
...
> snip...
> > The targets you mention are the ICC profiling targets that
usually
> build
> > on linearised inkchannels. Linearisation targets exist in many
shapes
> > CMYK-CcMmYK-Ncolor and they have ranges of 11 to 41 patches per
> > inkchannel. the lowest number can be used for printsystems that
> actually
>; > address the heads separately but on a CMYK driver that is
linearised as
> > CMYK but drives a CcMmYK printer it is better to use the 21
patches so
> > there are 10 steps measured per head. With the 4800 that has 3Ks
you
> > could go up to 31 steps and that has been used for Colorspan
models
>; that
> > have 3 ink dilutions of CMYK each = 12 heads.
>
> My impression of the process is a bit different, correct me if I'm
> worng. Individual ink components of a color, C and c for example, or
> the 3Ks you mention, are generally not addressed individually in the
> linearization process. Some control is usually given with
respect to
> how the transitions will happen, but then the color (all inclusive
> "C", or"K") is linearized as one. How exactly the
total patch count
> may divide up between those indivicual components may not be very
tidy.
> I'm sure this varies from RIP to RIP. I recall Roy asking if we
> thought there would be an advantage to individul ink linearization,
> and wonder the same thing myself.
>
.....
>
> no doubt. But the partitioning step needs to be considered as a major
> part of the overall performance picture. How many patches and how much
> work needs to be done at the end can have a lot to do with how well
> behaved the ramp is to begin with.
> That few have trouble with QTR speaks well of Roy's existing
>; partitioning scheme. Setting it up to pull 7 inks in and out without
> major problems must have been fun, or not!
> Tyler
The 7 inks for K7 worked just fine. The partitioning code has been
the same
since day one. I thought about more levels of linearization but I'd
probably
do it after the partitioning rather that before with each ink.
But it's interesting that with the K7 inks its noticibly (but not much)
better to use
2880dpi with only one dropsize rather than 1440dpi which is using two
sizes.
I think its because the 2X difference in drop size is as big as the
difference
between ink densities.
Roy
2005-08-22 by Roy Harrington
John, That's basically correct but when I did the 2200 I found the biggest dot outputing too much ink so I ended up using just the 2 smaller sizes. The dropsizes are all very printer and resolution dependent -- they are all different. Thanks for the link. Looks like the R800 info. Most of the other Epson docs are very vague about drop sizes. Roy --- In QuadtoneRIP@yahoogroups.com, "John Moody" <moodymz3@y...> wrote:
> Roy, > I thought the 2-bit VSD mode had three drop sizes? The bits would be > no-dot, 00, then 01, 10, and 11, right? > It appears the relative drop size ratio is printer and mode dependant as > well. Some Epson stuff here. > http://www.epson.co.jp/e/technology/sankei_4.htm > > Best regards, > John Moody > > -----Original Message----- > From: QuadtoneRIP@yahoogroups.com [mailto:QuadtoneRIP@yahoogroups.com]On > Behalf Of Roy Harrington > Sent: Saturday, August 20, 2005 9:48 PM > To: QuadtoneRIP@yahoogroups.com > Subject: [QuadtoneRIP] Re: QTR 51-step linearisation > > --- In QuadtoneRIP@yahoogroups.com, "Tyler Boley" <tyler@t...> wrote: > > --- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> ... > > snip... > > > The targets you mention are the ICC profiling targets that usually > > build > > > on linearised inkchannels. Linearisation targets exist in many shapes > > > CMYK-CcMmYK-Ncolor and they have ranges of 11 to 41 patches per > > > inkchannel. the lowest number can be used for printsystems that > > actually > > > address the heads separately but on a CMYK driver that is linearised as > > > CMYK but drives a CcMmYK printer it is better to use the 21 patches so > > > there are 10 steps measured per head. With the 4800 that has 3Ks you > > > could go up to 31 steps and that has been used for Colorspan models > > that > > > have 3 ink dilutions of CMYK each = 12 heads. > > > > My impression of the process is a bit different, correct me if I'm > > worng. Individual ink components of a color, C and c for example, or > > the 3Ks you mention, are generally not addressed individually in the > > linearization process. Some control is usually given with respect to > > how the transitions will happen, but then the color (all inclusive > > "C", or"K") is linearized as one. How exactly the total patch count > > may divide up between those indivicual components may not be very tidy. > > I'm sure this varies from RIP to RIP. I recall Roy asking if we > > thought there would be an advantage to individul ink linearization, > > and wonder the same thing myself. > > > ..... > > > > no doubt. But the partitioning step needs to be considered as a major > > part of the overall performance picture. How many patches and how much > > work needs to be done at the end can have a lot to do with how well > > behaved the ramp is to begin with. > > That few have trouble with QTR speaks well of Roy's existing > > partitioning scheme. Setting it up to pull 7 inks in and out without > > major problems must have been fun, or not! > > Tyler > > The 7 inks for K7 worked just fine. The partitioning code has been the same > since day one. I thought about more levels of linearization but I'd > probably > do it after the partitioning rather that before with each ink. > > But it's interesting that with the K7 inks its noticibly (but not much) > better to use > 2880dpi with only one dropsize rather than 1440dpi which is using two sizes. > I think its because the 2X difference in drop size is as big as the > difference > between ink densities. > > Roy > > > > > > > > > > _____ > > YAHOO! GROUPS LINKS > > * Visit your group " QuadtoneRIP > <http://groups.yahoo.com/group/QuadtoneRIP> " on the web. > > * To unsubscribe from this group, send an email to: > QuadtoneRIP-unsubscribe@yahoogroups.com > <mailto:QuadtoneRIP-unsubscribe@yahoogroups.com?subject=Unsubscribe> > > * Your use of Yahoo! Groups is subject to the Yahoo! Terms of > Service <http://docs.yahoo.com/info/terms/> . > > _____
2005-08-22 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > Tyler Boley wrote: > > >--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > >... > > > > > >>I think that using > >>more heads + less dilutions of black ink + effective use of droplet > >>sizes and actually using the mix you get from heads weaving (as intended > >>in the printer design) will in the end deliver less banding, better > >>linearisation, more consistency in time. I have written this before in > >>relation to BO printing and on the K7 threads. > >> > >> > > > >Are you speaking of using different density K inks in parallel rather than partitioned series, > >or some combination of the two? > >We discussed this once before, I thought the BO users could benefit from a the more > >complex dither that would result. I never got the chance to test it. Life is getting too short > >for all this testing. > >Actually, can you assign two inks to the same part of the scale in QTR? > > > > > > I've used a quad setup on the 9000 that had the Eboni black + the middle > grey ink in the Mm channels and was linearised like that and in the Cc > channels, the darkest grey + lightest grey ink in Mm and linearised like > that again. Some partitioning was laid on top of that, (the highlights > were done by the lightest range only, mixing of the two ranges was of > course mainly in the mid ranges, 2% of the darkest grey went up with the > black) trying to get as close to native linearisation and an overall > linearisation etc added afterwards. Wasatch SoftRip. That's a simple > example of what I propose. BTW, Paul Roark's PS curves and ink mixes > must have had a similar structure. > The crudest example is having 4 black inks in a C86 for BO, that will > give you a nozzle quantity that comes near 360 if I recall it correctly. > On a K7 printer and with QTR you could make one K4 range and a K3 ink > range, linearise them and mix them with the slider. The 5 partitioning > points of the two ranges are all overlapped, and you get weaving of two > heads on any spot. Depending on the droplet sizes available (resolution) > and the partitioning the droplet size transfer points could overlap as > well. If you use the normal K7 inks then you would expect more visible > dots in the highlights but I think it will not be much more visible. You > will loose the possibility to mix toner in but that's not available in > the K7 set anyway. No need to change inks. Ink distribution may benefit > from the original CMYK weaving positions though, for example the K3 in > the CcY lines the K4 in the KkMm lines. In a way these overlapping quads > resemble the B&W films with two or more emulsion layers. There's certainly lots of ways to make curves for the inkset. A 7-way partition is probably the most obvious but the interleaving of 2 separate sets makes some sense. In fact one person I know is interleaving K with 3 grays from K7 for one set and K with 3 piezos for the other set. Then you can blend for different tones. But two sets from the K7 would work and maybe you may have a point on it being better for linearization. > > I have no idea how the K7 inks are distributed on the ink positions, if > they use the two main lines as sketched above and given that the overlap > in the partitions is already relatively wider as one head only covers > about 14% of the total range, the result may be close to having two > ranges next to one another. The ink distribution something like K7-K, > K6-C, K5-k, K4-c, K3-M, K2-Y, K1-m. If that's happening already then I > shouldn't worry about the K7 concept. There are just 7 shades, partitioning overlaps a lot as it is. But if more overlapping would probably happen with the interleave idea. > > No, I don't think you can assign two inks to the same part in QTR, I > have asked that once and got an answer by Carl or Daniel but couldn't I'm not sure what you want either but there is now a COPY_CURVE function that can dup any curve for multiple inks. > figure it out or my question wasn't understood. I think it would be good > if QTR allowed the use of multiple heads for the same task. Not sure if it's a benefit or not. A counter-argument is that you would have less diversity of dropsizes. I think that diversity of dropsizes is particularly beneficial to eliminating banding. Actually its most important to use drops that are big enough to merge between lines when the density gets darker. Roy > > Ernst
2005-08-23 by Ernst Dinkla
> > >Not sure if it's a benefit or not. A counter-argument is that you would have >less diversity of dropsizes. I think that diversity of dropsizes is particularly >beneficial to eliminating banding. Actually its most important to use drops >that are big enough to merge between lines when the density gets darker. > >Roy > > With some printers that have few nozzles per head it may be worth it. For example the C86 and similar ones. For BO printing or a semi BO with one lighter ink. Or if you want the smallest droplet of (older) printers on purpose. I agree that droplet size diversity should be kept as much as possible. Which brings up the subject again of higher patch numbers in linearising to control the more complex hardware of multiple heads, inks and 2-3 droplet sizes. Ernst
2005-08-23 by Olivier
I would like to add/inquire about another factor in the smoothness/banding issue. In a technical article from Epson, the company claims : " Halftone printing In order to simulate varying size halftone dots in computer printers, dithering is used, which creates clusters of dots in a "halftone cell." We recommend using Error Diffusion when printing halftoning. The more dots printed in the cell, the darker the gray. As the screen frequency gets higher (more lines per inch), there is less room for dots in the cell, reducing the number of gray levels that can be generated. Epson's (LPI) lines per inch conversions are as follows: 720 dpi equates to 240 lpi 360 dpi equates to 120 lpi " Question 1 : Am I correct assuming the cell is thus 3x3 e.g. 9+1 shades of grey available per cell. I now refer to an article by Rags Garner (http://www.rags-int- inc.com/PhotoTechStuff/Epson2200/). The purpose of this article is to find out what should be the image resolution vs printer dpi settings to output the best possible pattern. I did not bother in the past and simply set image res at 360 and printer dpi at 1440 for lack of information and common acceptance. Rags findings are that the best image res/dpi settings would be 288ppi/1440dpi based on a 2200 with 4x24 (96) nozzles e.g. 5x5 cell (???). Besides and image of 288ppi would turn to be 144lpi and would be slightly below average human "resolution" thus leading to some detail loss or pixelisation. Question 2: This is the first time I ever come to this settings ? What do you think ? It seems to contradict Epson 3x3 cell. This would assume a 5x5 cell yieling 25+1 possible tones, I would tend to multiple this (less 1 for pure white, not drop) by the 3 variable drop sizes to make 75+1 shades available per cell. It's seems that human eye can perceive about 100 grey tones which would turn the number of ink dilution of 3 more than enough. Question 3 : I'm wondering about the K7, but if the above is not too ridiculous, K7 would not be so usefull (with only gray inks) and linearisation on a 21stepwedeg sould suffice to prevent banding and provide adequate smoothness? I'm a very beginner, I was just initially looking for the proper ppi/dpi settings to print correctly and was wondering about the best linearisation process. Last, while Epson and QTR do not document the screening methods (I hope this is the proper term in English, in French it's "tramage") whether it is AM or FM (or a mix of both should also have an impact propably in the hilglight rendering. Variable dot size call for AM method (dot at same distance, different size)so precise setting ppi/dpi would be of even higher importance (I feel). I'm pretty sure I confuse a lot of things : your advise in the settings and the best quadtone combination would be welcome.
2005-08-23 by Tyler Boley
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: ... > I've used a quad setup on the 9000 that had the Eboni black + the middle > grey ink in the Mm channels and was linearised like that and in the Cc > channels, the darkest grey + lightest grey ink in Mm and linearised like > that again. Some partitioning was laid on top of that, (the highlights > were done by the lightest range only, mixing of the two ranges was of > course mainly in the mid ranges, 2% of the darkest grey went up with the > black) trying to get as close to native linearisation and an overall > linearisation etc added afterwards. Wasatch SoftRip. That's a simple > example of what I propose.... I think I see what you are doing. Similar setups could be put together with some creativity with SP, but the main tool allowing your idea is the QTR blending (I think?). But I wonder- given the large amount of overlap that must occur with 7 inks (which Roy also mentiones) the K7s with QTR may be doing something actually similar. Am I wrong in deducing that a main advantge of your idea is making use of as many nozzles as possible, and the resulting complex weave? Tyler
2005-08-23 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > > > > > > >Not sure if it's a benefit or not. A counter-argument is that you would have > >less diversity of dropsizes. I think that diversity of dropsizes is particularly > >beneficial to eliminating banding. Actually its most important to use drops > >that are big enough to merge between lines when the density gets darker. > > > >Roy > > > > > With some printers that have few nozzles per head it may be worth it. I can't see number of nozzles having any effect. The only difference between say the 2200 w/96 nozzles and R2400 w/180 nozzles is that the head is larger -- it does 1 inch swath instead of a .53 inch swath per pass. The nozzles are 1/180 inch separation in both cases. It takes multiple passes to fill everything in between. Roy
> For example the C86 and similar ones. For BO printing or a semi BO with > one lighter ink. Or if you want the smallest droplet of (older) printers > on purpose. I agree that droplet size diversity should be kept as much > as possible. Which brings up the subject again of higher patch numbers > in linearising to control the more complex hardware of multiple heads, > inks and 2-3 droplet sizes. > > Ernst
2005-08-23 by Roy Harrington
Hi Olivier, I think the Epson article is just trying to help people who think in LPI to understand comparisons. I don't believe that any of the inkjet drivers really work on a "halftone cell" basis. Its a convenient method to describe the general idea of how multiple grays can be achieved with just dots but its not how it actually works. The Garner article is incorrect in most of its assumptions so all the calculations are not very useful. I can't imagine where the 4x24 nozzle arrangement idea came from -- the 96 nozzles per ink are in a straight line 1/180 inch apart. The nozzle check shows this clearly (note that the whole thing is done in one pass). There are no cells so trying to calculate 4x4s and 5x5s aren't relevant. What he's really shown is just how moire patterns can arise. All Epson drivers that I've seen resample the data at either 360x360 or 720x720 before applying a stochastic (random) dithering algorithm which may or may not also include error diffusion. The simplest explanation for dithering is just that the probability of a dot existing at a particular location is proportional to the value of gray you want there. The trick though is that its pseudo-random in such a way that the dots are more evenly spaced. Roy --- In QuadtoneRIP@yahoogroups.com, "Olivier" <odesmais@y...> wrote:
> I would like to add/inquire about another factor in the > smoothness/banding issue. > > In a technical article from Epson, the company claims : > " > Halftone printing > > In order to simulate varying size halftone dots in computer printers, > dithering is used, which creates clusters of dots in a "halftone cell." > We recommend using Error Diffusion when printing halftoning. The more > dots printed in the cell, the darker the gray. As the screen frequency > gets higher (more lines per inch), there is less room for dots in the > cell, reducing the number of gray levels that can be generated. > Epson's (LPI) lines per inch conversions are as follows: > 720 dpi equates to 240 lpi > 360 dpi equates to 120 lpi > " > > Question 1 : > Am I correct assuming the cell is thus 3x3 e.g. 9+1 shades of grey > available per cell. > > I now refer to an article by Rags Garner (http://www.rags-int- > inc.com/PhotoTechStuff/Epson2200/). > > The purpose of this article is to find out what should be the image > resolution vs printer dpi settings to output the best possible pattern. > I did not bother in the past and simply set image res at 360 and > printer dpi at 1440 for lack of information and common acceptance. Rags > findings are that the best image res/dpi settings would be > 288ppi/1440dpi based on a 2200 with 4x24 (96) nozzles e.g. 5x5 cell > (???). Besides and image of 288ppi would turn to be 144lpi and would be > slightly below average human "resolution" thus leading to some detail > loss or pixelisation. > > Question 2: > This is the first time I ever come to this settings ? What do you > think ? It seems to contradict Epson 3x3 cell. > > This would assume a 5x5 cell yieling 25+1 possible tones, I would tend > to multiple this (less 1 for pure white, not drop) by the 3 variable > drop sizes to make 75+1 shades available per cell. It's seems that > human eye can perceive about 100 grey tones which would turn the number > of ink dilution of 3 more than enough. > > Question 3 : > I'm wondering about the K7, but if the above is not too ridiculous, K7 > would not be so usefull (with only gray inks) and linearisation on a > 21stepwedeg sould suffice to prevent banding and provide adequate > smoothness? > > I'm a very beginner, I was just initially looking for the proper > ppi/dpi settings to print correctly and was wondering about the best > linearisation process. > > Last, while Epson and QTR do not document the screening methods (I hope > this is the proper term in English, in French it's "tramage") whether > it is AM or FM (or a mix of both should also have an impact propably in > the hilglight rendering. Variable dot size call for AM method (dot at > same distance, different size)so precise setting ppi/dpi would be of > even higher importance (I feel). > > I'm pretty sure I confuse a lot of things : your advise in the settings > and the best quadtone combination would be welcome.
2005-08-23 by Roy Harrington
On Tuesday, August 23, 2005, at 10:42 AM, Tyler Boley wrote:
> --- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote:
> ...
>> I've used a quad setup on the 9000 that had the Eboni black + the
> middle
>> grey ink in the Mm channels and was linearised like that and in the Cc
>> channels, the darkest grey + lightest grey ink in Mm and linearised
> like
>> that again. Some partitioning was laid on top of that, (the highlights
>> were done by the lightest range only, mixing of the two ranges was of
>> course mainly in the mid ranges, 2% of the darkest grey went up with
> the
>> black) trying to get as close to native linearisation and an overall
>> linearisation etc added afterwards. Wasatch SoftRip. That's a simple
>> example of what I propose....
>
> I think I see what you are doing. Similar setups could be put together
> with some creativity with SP, but the main tool allowing your idea is
> the QTR blending (I think?).
> But I wonder- given the large amount of overlap that must occur with 7
> inks (which Roy also mentiones) the K7s with QTR may be doing
> something actually similar.
> Am I wrong in deducing that a main advantge of your idea is making use
> of as many nozzles as possible, and the resulting complex weave?
> Tyler
>
Here's a couple of examples of what partitioning looks like for 2, 4,
and 7 grays.
They are crude text graphs but show that at all times there are 2 or 3
inks being used.
Down the page are the K percents, across is amount of each ink used.
UltraChrome 2 grays -- Black and Light-Black
0 0% |
5 2% | k
10 4% | k
15 6% | k
20 8% | k
25 10% | k
30 12% | k
35 14% | k
40 16% | k
45 18% | k
50 20% | k
55 22% | k
60 24% | k
65 25% | k
70 27% | k
75 29% | k
80 31% | k
85 33% | k
90 35% | k
95 37% | k
100 39% | k
105 41% | k
110 43% | k
115 45% | k
120 47% | k
125 49% | k
130 51% | k
135 53% | k
140 55% | k
145 57% | k
150 59% | k
155 61% |K k
160 63% | K k
165 65% | K k
170 67% | K k
175 69% | K k
180 71% | K k
185 73% | K k
190 75% | K k
195 76% | K k
200 78% | K k
205 80% | K k
210 82% | K k
215 84% | K k
220 86% | K k
225 88% | K k
230 90% | k
235 92% | k K
240 94% | k K
245 96% | k K
250 98% | k
K
255 100% |
K
A quad system:
0 0% |
5 2% | Y
10 4% | Y
15 6% | Y
20 8% | Y
25 10% | Y
30 12% | Y
35 14% | Y
40 16% | Y
45 18% | Y
50 20% | Y
55 22% | Y
60 24% | Y
65 25% | Y
70 27% | Y
75 29% |M Y
80 31% |M Y
85 33% | M Y
90 35% | M Y
95 37% | M Y
100 39% | M Y
105 41% | M Y
110 43% | M Y
115 45% | M Y
120 47% | M Y
125 49% | M Y
130 51% | M Y
135 53% | Y M
140 55% |C Y M
145 57% | C Y M
150 59% | C Y M
155 61% | C Y M
160 63% | C Y M
165 65% | CY M
170 67% | Y C M
175 69% |Y C M
180 71% | C M
185 73% | C M
190 75% | C M
195 76% | C M
200 78% | C M
205 80% | M C
210 82% | M C
215 84% | K M C
220 86% | K M C
225 88% | KM C
230 90% | M K C
235 92% | K C
240 94% | K C
245 96% | K C
250 98% | C K
255 100% | C
K
Finally the K7 system:
0 0% |
5 2% | Y
10 4% |m Y
15 6% | m Y
20 8% | m Y
25 10% | m Y
30 12% | m Y
35 14% | m Y
40 16% | Y m
45 18% | M Y m
50 20% | M Y m
55 22% | Y M
m
60 24% | Y M
m
65 25% | M m
70 27% | M m
75 29% | m M
80 31% |c m M
85 33% | c m M
90 35% | c m M
95 37% | c m
M
100 39% | cm
M
105 41% | m c
M
110 43% | m c
M
115 45% | c
M
120 47% | c M
125 49% | c M
130 51% | c M
135 53% | M c
140 55% | M c
145 57% | C M c
150 59% | C M c
155 61% | C M c
160 63% | CM
c
165 65% | M C
c
170 67% | C
c
175 69% | C c
180 71% | C c
185 73% | c C
190 75% |k c C
195 76% | k c
C
200 78% | kc
C
205 80% | c k
C
210 82% | k C
215 84% | k C
220 86% | C k
225 88% | K C
k
230 90% | K C
k
235 92% | CK
k
240 94% | C K
k
245 96% | K
k
250 98% | kK
255 100% | k
K
Roy
-
Roy Harrington
roy@...
Black & White Photo Gallery
http://www.harrington.com2005-08-24 by Ernst Dinkla
Roy Harrington wrote: >--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > > >>>Not sure if it's a benefit or not. A counter-argument is that you would have >>>less diversity of dropsizes. I think that diversity of dropsizes is particularly >>>beneficial to eliminating banding. Actually its most important to use drops >>>that are big enough to merge between lines when the density gets darker. >>> >>>Roy >>> >>> >>> >>> >>With some printers that have few nozzles per head it may be worth it. >> >> > >I can't see number of nozzles having any effect. The only difference between >say the 2200 w/96 nozzles and R2400 w/180 nozzles is that the head is >larger -- it does 1 inch swath instead of a .53 inch swath per pass. The nozzles >are 1/180 inch separation in both cases. It takes multiple passes to fill everything >in between. > >Roy > > > The weaving of more heads certainly will help in smoother patterns. Less nozzles per head can be compensated by more passes and with that speed sacrifice but it doesn't compensate the weaving. I have not checked the number of passes and the resolution possible on the cheapest Epsons like the C86 but I do not expect that they are in the same quality class the 2200/2400 are in. So I suggest that having more heads for the same task can be a sensible thing. Ernst
2005-08-24 by Ernst Dinkla
Roy Harrington wrote: > >Here's a couple of examples of what partitioning looks like for 2, 4, >and 7 grays. >They are crude text graphs but show that at all times there are 2 or 3 >inks being used. >Down the page are the K percents, across is amount of each ink used. > > For the K7 it has even more mixing than I expected. What is the overlap number you use to get the quad result below and for which printer was it used ? Ernst
>A quad system: > > 0 0% | > 5 2% | Y > 10 4% | Y > 15 6% | Y > 20 8% | Y > 25 10% | Y > 30 12% | Y > 35 14% | Y > 40 16% | Y > 45 18% | Y > 50 20% | Y > 55 22% | Y > 60 24% | Y > 65 25% | Y > 70 27% | Y > 75 29% |M Y > 80 31% |M Y > 85 33% | M Y > 90 35% | M Y > 95 37% | M Y >100 39% | M Y >105 41% | M Y >110 43% | M Y >115 45% | M Y >120 47% | M Y >125 49% | M Y >130 51% | M Y >135 53% | Y M >140 55% |C Y M >145 57% | C Y M >150 59% | C Y M >155 61% | C Y M >160 63% | C Y M >165 65% | CY M >170 67% | Y C M >175 69% |Y C M >180 71% | C M >185 73% | C M >190 75% | C M >195 76% | C M >200 78% | C M >205 80% | M C >210 82% | M C >215 84% | K M C >220 86% | K M C >225 88% | KM C >230 90% | M K C >235 92% | K C >240 94% | K C >245 96% | K C >250 98% | C K >255 100% | C > K > > >
2005-08-24 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > Roy Harrington wrote: > > > > >Here's a couple of examples of what partitioning looks like for 2, 4, > >and 7 grays. > >They are crude text graphs but show that at all times there are 2 or 3 > >inks being used. > >Down the page are the K percents, across is amount of each ink used. > > > > > > For the K7 it has even more mixing than I expected. > > What is the overlap number you use to get the quad result below and for > which printer was it used ? > > Ernst > The quad setup was not mine so I don't have all the details. It was for a MIS FS inkset (same densities as piezotones). By just looking at the numbers the overlap number was very small about 5. (You can tell by how much C is still in the mix at 100%). Roy
2005-08-24 by Roy Harrington
--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > Roy Harrington wrote: > > >--- In QuadtoneRIP@yahoogroups.com, Ernst Dinkla <E.Dinkla@c...> wrote: > > > > > >>>Not sure if it's a benefit or not. A counter-argument is that you would have > >>>less diversity of dropsizes. I think that diversity of dropsizes is particularly > >>>beneficial to eliminating banding. Actually its most important to use drops > >>>that are big enough to merge between lines when the density gets darker. > >>> > >>>Roy > >>> > >>> > >>> > >>> > >>With some printers that have few nozzles per head it may be worth it. > >> > >> > > > >I can't see number of nozzles having any effect. The only difference between > >say the 2200 w/96 nozzles and R2400 w/180 nozzles is that the head is > >larger -- it does 1 inch swath instead of a .53 inch swath per pass. The nozzles > >are 1/180 inch separation in both cases. It takes multiple passes to fill everything > >in between. > > > >Roy > > > > > > > The weaving of more heads certainly will help in smoother patterns. Less > nozzles per head can be compensated by more passes and with that speed > sacrifice but it doesn't compensate the weaving. I have not checked the > number of passes and the resolution possible on the cheapest Epsons like > the C86 but I do not expect that they are in the same quality class the > 2200/2400 are in. So I suggest that having more heads for the same task > can be a sensible thing. > > Ernst I think you are assuming that weaving is a bit more magic than it really is. On say the 2200 in one pass the head puts down a dot up to every 1/360 inch horizontally and every 1/180 inch vertically. (The nozzles are all vertically in a line). So to get 1440x720 dpi there must be 4 passes to fill in the full 1440dpi horizontal times 4 to fill in the full 720dpi vertical. Imagine that the paper is divided into all these 4x4 grids. Each pass of the head can fill just one square of this 4x4 grid so there will be 16 passes needed in the area to fill it all in. The whole function of weaving is to ensure that a different nozzle is used for each of the 16 squares in the 4x4 grid. There's no way to use more than 16 nozzles to fill in a 4x4 grid. The difference between 96 nozzles of the 2200 vs 180 nozzles of the R2400 is that you are working on 180 different 4x4 grids at the same time. But within a single 4x4 its the same basic thing. Conceptually it pretty simple, in practice its a bit of a nightmare of shuffling data. Roy
2005-08-25 by Ernst Dinkla
>I think you are assuming that weaving is a bit more magic than it really is. >On say the 2200 in one pass the head puts down a dot up to every 1/360 inch >horizontally and every 1/180 inch vertically. (The nozzles are all vertically in >a line). So to get 1440x720 dpi there must be 4 passes to fill in the full 1440dpi >horizontal times 4 to fill in the full 720dpi vertical. Imagine that the paper is >divided into all these 4x4 grids. Each pass of the head can fill just one square >of this 4x4 grid so there will be 16 passes needed in the area to fill it all in. The >whole function of weaving is to ensure that a different nozzle is used for >each of the 16 squares in the 4x4 grid. There's no way to use more than 16 >nozzles to fill in a 4x4 grid. The difference between 96 nozzles of the 2200 >vs 180 nozzles of the R2400 is that you are working on 180 different 4x4 >grids at the same time. But within a single 4x4 its the same basic thing. > >Conceptually it pretty simple, in practice its a bit of a nightmare of shuffling data. > > > Roy, I have learned a thing or two along the thread. There's less to be alarmed about than I thought at the beginning. At least I know better what basically is going on and build on that with the knowledge I already had. Getting back to the subject line I wonder what you think about a target in between the 21 and 51 patches and the stepping rate more variable in the range so it controls the shadow and wherever needed better. As discussed with Tyler Boley. Little I can contribute but the target design itself. Ernst
2005-08-25 by hogarth@snappydsl.net
I think I understand what I'm seeing here. It reminds one that using more dilutions of black will give one diminishing returns. But, we as a group are all about getting the very best we can out of our printing. This begs the question: How many is too many? People I trust tell me that there's a clearly visible advantage in going from four inks to six. Is there a visible advantage in going from six inks to seven? eight? twelve? -- Bruce Watson Roy Harrington wrote:
> Here's a couple of examples of what partitioning looks like for 2, 4, > and 7 grays. > They are crude text graphs but show that at all times there are 2 or 3 > inks being used. > Down the page are the K percents, across is amount of each ink used. > > A quad system: > > 0 0% | > 5 2% | Y > 10 4% | Y > 15 6% | Y > 20 8% | Y > 25 10% | Y > 30 12% | Y > 35 14% | Y > 40 16% | Y > 45 18% | Y > 50 20% | Y > 55 22% | Y > 60 24% | Y > 65 25% | Y > 70 27% | Y > 75 29% |M Y > 80 31% |M Y > 85 33% | M Y > 90 35% | M Y > 95 37% | M Y > 100 39% | M Y > 105 41% | M Y > 110 43% | M Y > 115 45% | M Y > 120 47% | M Y > 125 49% | M Y > 130 51% | M Y > 135 53% | Y M > 140 55% |C Y M > 145 57% | C Y M > 150 59% | C Y M > 155 61% | C Y M > 160 63% | C Y M > 165 65% | CY M > 170 67% | Y C M > 175 69% |Y C M > 180 71% | C M > 185 73% | C M > 190 75% | C M > 195 76% | C M > 200 78% | C M > 205 80% | M C > 210 82% | M C > 215 84% | K M C > 220 86% | K M C > 225 88% | KM C > 230 90% | M K C > 235 92% | K C > 240 94% | K C > 245 96% | K C > 250 98% | C K > 255 100% | C > K > > > Finally the K7 system: > > 0 0% | > 5 2% | Y > 10 4% |m Y > 15 6% | m Y > 20 8% | m Y > 25 10% | m Y > 30 12% | m Y > 35 14% | m Y > 40 16% | Y m > 45 18% | M Y m > 50 20% | M Y m > 55 22% | Y M > m > 60 24% | Y M > m > 65 25% | M m > 70 27% | M m > 75 29% | m M > 80 31% |c m M > 85 33% | c m M > 90 35% | c m M > 95 37% | c m > M > 100 39% | cm > M > 105 41% | m c > M > 110 43% | m c > M > 115 45% | c > M > 120 47% | c M > 125 49% | c M > 130 51% | c M > 135 53% | M c > 140 55% | M c > 145 57% | C M c > 150 59% | C M c > 155 61% | C M c > 160 63% | CM > c > 165 65% | M C > c > 170 67% | C > c > 175 69% | C c > 180 71% | C c > 185 73% | c C > 190 75% |k c C > 195 76% | k c > C > 200 78% | kc > C > 205 80% | c k > C > 210 82% | k C > 215 84% | k C > 220 86% | C k > 225 88% | K C > k > 230 90% | K C > k > 235 92% | CK > k > 240 94% | C K > k > 245 96% | K > k > 250 98% | kK > 255 100% | k > K > > Roy > > - > Roy Harrington > roy@... > Black & White Photo Gallery > http://www.harrington.com >