Keyboard Shortcuts
Likes
Search
Average a measurement with an i1 Question
|
For anyone who is using the Xrite i1 for measuring targets, or Chris. I was reviewing Chris's(your) great instructions for QCDN. I like these because they are written using numbered steps (thank you!), which my brain prefers. The instructions mention to measure a target three times with an average, and send to the Data Tool (I only just began using this.) My question is how does one "read three times" and then "average".? If I read each row of the target, multiple times, does it simply override the last measurement, or does it somehow average the reading? Any guidance on achieving the average via multiple readings of the target would be helpful.? I typically measure the target once, but I am interested in attempting to work with an average to see if it yields a better result.??
?
Thanks,
DM |
开云体育DM, I was wondering where I said measure 3x and average on my instructions, but you must be referring to Xrite instructions correct? I don’t measure anything 3x and average. But I do blur/average every step on a 21 step printed target and read that blur averaged step once and log it into a plain text file in Word. Then I pull that .txt down to Data Tool and Export Data from Data Tool. I used to use a densitometer to measure the step wedge, but when I pitted my densitometer measures against Photoshop’s LAB the curves plot the same so I quit using my densitometer to measure my step wedge and now just do it in Photoshop. I also don’t use two decimal points in LAB. Mostly a waste of time. And sometimes I fudge points because a blip can indicate a coating anomaly and not a process response anomaly. It used to be that I blur averaged every step then set a white and black point for Step 0 and 100 but for two reasons I don’t do that anymore either: one is it slightly alters the relationships in-between steps 5 and 95 and two, the curves tend to follow the same trajectory anyway when plotted, too. For the record, lately I have returned to a one-part profile for gum printing for a number of lengthy reasons I won’t go into, the main one being there is no need for all that LK and LLK ink in a negative for gum. For one-coat processes like new cyanotype or palladium, LK and LLK do give more grunt in the shadows but overkill for gum. But I think you are doing cyanotype if I am not mistaken. Chris ? From:
[email protected] <[email protected]> on behalf of DM Witman via groups.io <dmwitman@...> For anyone who is using the Xrite i1 for measuring targets, or Chris. I was reviewing Chris's(your) great instructions for QCDN. I like these because they are written using numbered steps (thank you!), which my brain prefers. The instructions mention to measure a target three times with an average, and send to the Data Tool (I only just began using this.) My question is how does one "read three times" and then "average".? If I read each row of the target, multiple times, does it simply override the last measurement, or does it somehow average the reading? Any guidance on achieving the average via multiple readings of the target would be helpful.? I typically measure the target once, but I am interested in attempting to work with an average to see if it yields a better result.?? ? Thanks, DM |
|
Hi Chris,
?
Thanks for the detailed explanation, and for the tips about decimal points. Your explanation of the blur/average completely makes sense to me. I'll have to try it after the semester ends.? I've used targets that were created in an attempt to mitigate for hand-coated anomalies. Would be interested in doing a comparison of results.?
?
PS: I have two versions of instructions, perhaps one of them is older???
?
Thank you!
DM |
开云体育DM, I’ve had to revise the handout probably a dozen times, and especially now that the software is called QTP-DN not QCDN and needed new screen shots. ? Also every time a student makes a mistake or something isn’t clear, I revise my steps, so it is as student-friendly and as pared down as it can get. I can’t imagine how to make it simpler than this. I will be doing a digital negative weekend workshop in a couple weeks for faculty and students combined, only those who choose to be there, so I’ll get to test it on photo faculty, too. That’ll be a hoot. ? At the risk of Richard yelling at me on this list again, I attach the newest handout, though Richard would prefer me not to share my directions in case I “mislead someone and he has to correct the error” so caveat emptor. My step-by-step way of teaching and using the software is the way I have worked and taught now for six years (QTR) and several years (QCDN) and is also informed by Bill Schwab’s videos. ?Co-authoring the QuadToneRIP book with Ron Reeder taught me the bare bones behind these systems. ? Ron was definitely a 2-part profile person and only was slightly convinced in other modes just before he died. He felt 3-parts used too much non-performing ink (LK, LLK). When I first learned QTR, not from Ron but Sandy King, Sandy taught me a 1-part profile. I studied the differences between 1, 2, and 3 parts and have come to the conclusion that for certain processes there is no need to use tons of LK and LLK ink in a negative. My lumen negatives are 1-part , K and Y following the same “trajectory,” and I’m working now with a 1-part gum bichromate profile, too. All this to say I do actually create 1, 2, and 3 part starter profiles but those 1 and 2 parts start from ink descriptor txt files, installed as normal, and then the resulting quads can be brought into QTP-DN and mapped like any other profile. With an ink descriptor file there are variables you can change (gamma, highlight, shadow, overlap, gray value) but that is beyond a beginning college-level alt student so I no longer go there. I would say 25% of my students want to learn profiling. 5% ink descriptors. Chris ? From:
[email protected] <[email protected]> on behalf of DM Witman via groups.io <dmwitman@...> Hi Chris, ? Thanks for the detailed explanation, and for the tips about decimal points. Your explanation of the blur/average completely makes sense to me. I'll have to try it after the semester ends.? I've used targets that were created in an attempt to mitigate for hand-coated anomalies. Would be interested in doing a comparison of results.? ? PS: I have two versions of instructions, perhaps one of them is older??? ? Thank you! DM |
|
Would it be advisable to note that this is for version 3+ of Richards QTP-DN? Bob Hartung On Sun, Mar 23, 2025 at 11:06?AM Christina Z. Anderson via <christinazanderson=[email protected]> wrote:
|
开云体育Good point, Bob. I assume any of my students who are going to purchase it this semester will be purchasing the most recent, and the old one had a different name (QCDN), and I also assume people using QCDN will update to QTP-DN (paying $85 to keep software alive is a small price to pay to a software designer), but you know what they say about “assume.” Mine is 3.2. Chris ? From:
[email protected] <[email protected]> on behalf of Bob Hartung via groups.io <rwhart3675@...> Would it be advisable to note that this is for version 3+ of Richards QTP-DN? ? Bob Hartung ? On Sun, Mar 23, 2025 at 11:06?AM Christina Z. Anderson via <christinazanderson=[email protected]> wrote:
|