<
  News     DIFP   Benefits Profiles Account  


forums

Recent Posts



(New threads or threads with new replies - ↑ Links to top of thread)

 Sony Artisan and Windows
Apr 21, 2004 8:38 PM
Posted by John Eakin

 Preferred printing direction (Enhanced Matte) & IP with roll paper
Apr 21, 2004 7:31 PM
Posted by John Vitollo

 Kodak 14n or Fuji S2 for stock?
Apr 21, 2004 7:20 PM
Posted by Andrew Rodney

 Confused about layers in PK Sharpener
Apr 21, 2004 7:11 PM
Posted by Thomas Nash

 Sony Artisan with Monaca Optic XR
Apr 21, 2004 5:05 PM
Posted by Jeff Schewe

 Epson R 800
Apr 21, 2004 4:43 PM
Posted by Sharon Brunson

 Fuji S3 and verticals
Apr 21, 2004 3:18 PM
Posted by Peter Burian

 Applying in camera parameters and RAW files
Apr 21, 2004 3:11 PM
Posted by Peter Burian

 Is the D70 a good deal?
Apr 21, 2004 3:08 PM
Posted by Peter Burian

 Camera for raw
Apr 21, 2004 3:02 PM
Posted by Peter Burian

...see more posts

articles

Weekly Chatter:X-Rite DTP41
Tuesday October 28, 2003

Section 1: X-Rite DTP41






Printers > Fine Art/Photographic Printing > X-Rite DTP41





































AuthorSubject: X-Rite DTP41  
Marco Passanisi
Posted on Mon Oct 6, 2003 1:43 PM

Hi,
I'm not sure of which DTP41 would be best when using GM ProfileMaker, UV or not UV?

My understanding is that the UV version eliminates (or greatly reduces...) the effect of the optical brightener in the paper, but this is just like pretending the brightner is not there.
Isn't it better to measure the brightened patches and let the software handle it with the option "Correct for Optical Brightener"?
If my reasoning is correct then the non UV should be a better choice.
Is this right?

Thanks.

Marco


Andrew Rodney
Posted on Mon Oct 6, 2003 1:51 PM

with ProfileMaker, it doesn't really matter as the software detects the UV and accounts for it. I use several non UV Spectrophotometer's with it and haven 't found any problems. I have a UV DTP-41 (T) but got it before PMP could do this and for other software products that can't deal with UV.


Bruce Fraser
Posted on Mon Oct 6, 2003 1:55 PM

I'd say you want the version without the UV filter. But I'd also advise you to check out Gretagmacbeth's ICColor line of strip readers�I think they're at least worthy competitors to the DTP41. Andrew has both�maybe he'll chime in.


Marco Passanisi
Posted on Mon Oct 6, 2003 2:01 PM

Well, the reason for posting is that I bought one on Ebay, used.
It was listed as DTP41, and I expected (and wanted...) it to be NON-UV. It turned out to be a UV version.

Given that so far I used the manual DTP22 and let the GMProfileMaker handle the brightener, now I don't know if I have to change my procedure for creating profiles with this UV beast.

Can I remove the UV filter from the unit??

Marco


Andrew Rodney
Posted on Mon Oct 6, 2003 2:10 PM

on 10/6/03 1:00 PM, IR - Bruce Fraser wrote:

> Andrew has both�maybe he'll chime in.

I've only used the DTP-41 a few times over the last year or two since
getting the iCColor. One was for doing some transparencies. The other was
for building profiles from other packages besides PMP. There isn't a lot of
3rd party support for the iCColor.


Andrew Rodney
Posted on Mon Oct 6, 2003 2:20 PM

on 10/6/03 1:05 PM, IR - Marco Passanisi wrote:

> Can I remove the UV filter from the unit??

X-Rite can I think.


Gary Smith
Posted on Mon Oct 6, 2003 3:08 PM

I recently got a price of $510 from X-Rite to have the UV cut-off filter removed.


Marco Passanisi
Posted on Mon Oct 6, 2003 3:23 PM

well, at that ridicolous price I will figure out how to remove it myself...

but the question is if the effort is worthy, or if ProfileMaker can generate a profile from a DTP41UV as good as it would from a DTP41.

Marco


Bruce Fraser
Posted on Mon Oct 6, 2003 3:35 PM

You won't be able to use the optical brightener correction, I think, but other than that you'll get pretty much the same results you'd get from the non-UV version.


Marco Passanisi
Posted on Mon Oct 6, 2003 6:35 PM

Well, if someone here, wants to swap an UV for non-UV model, I will be happy to do it.

Let me know.

Marco


Marco Passanisi
Posted on Tue Oct 7, 2003 3:23 PM

I called X-Rite and they offered to do it for $750!! Well... I said, at that price would you also send me a bunch of flowers??

I did it myself, It's only a circular UV filter that filters the light from the three lamps.

It is glued around the central plastic column where the measuring optics is mounted.

If you feel like opening the instrument that deep down and remember how to assemble it back, then it is easy to remove it.
The assembly is a professional grade assembly, so you won't damage anything if you pay attention.

... and if you damage something, then you can get it repaired for the same price, I guess, so why not trying??

It calibrates perfectly, same as before.

Now I have a NON-UV DTP41.

Marco


Gary Smith
Posted on Tue Oct 7, 2003 3:26 PM

Marco, thank you for being the crash test dummy! I'm used to being the one...

Thanks for the instructions.


Marco Passanisi
Posted on Tue Oct 14, 2003 7:12 PM

Does anybody know if there is a way to make the DTP41 take multiple measurements of the same ptach and send averaged values to the host program (i.e. GM ProfileMaker)?

Thanks.

Marco


Andrew Rodney
Posted on Tue Oct 14, 2003 7:20 PM

-->Does anybody know if there is a way to make the DTP41 take multiple measurements of the same ptach and send averaged values to the host program (i.e. GM ProfileMaker)?

It does that automatically. I think it's something like 8 samples per patch or something like that.


Marco Passanisi
Posted on Wed Oct 15, 2003 1:11 PM

Hi Andrew,
I'm sorry, but this is not possible...
From measuring the testchart I've found out (...not a scientific breakthrough :-) ) that the spectro takes one measurement every 7mm.

Given that it measures 20 patches in about 6 seconds, doing some basic math it comes out that to take 140 measurements (7x20) in 6 seconds, then the filter wheel should rotate at 1400rpm! Simply impossible.

I think that the only way to do it, is to measure the strip more than once and then average the results in Excel or in the GM MeasureTool.

Marco


Andrew Rodney
Posted on Wed Oct 15, 2003 1:25 PM

on 10/15/03 12:15 PM, IR - Marco Passanisi wrote:

> I'm sorry, but this is not possible...

I have no idea what goes on inside but I'm pretty sure you're getting more
than a single sample per patch. I'll ask my mole inside X-Rite.


Marco Passanisi
Posted on Wed Oct 15, 2003 1:40 PM

Thank you Andrew,
I will wait for your inside findings.
What goes inside is a weel that holds the spectral filters and rotates, putting each filter, in turn, between the optics and the sensor.
To take 1 measurement the wheel has to make 1 turn.
So the speed of measurement is imposed by the speed of rotation of the wheel as well as by the speed of acquisition of the sensor.

Marco


Andrew Rodney
Posted on Wed Oct 15, 2003 7:00 PM

We have a few people in X-rite looking for the official answer but I did get this"

"As far as the number of absolute samples I know Bob has passed this on
to engineering. I do know there is no easy way to calculate the result
because the internal logic also throws out readings if they appear next
to a transition spot (gap or patch depending on design.) My shaky memory
says that most of the time more than 1 sample is used and averaged in a
7mm patch but I no longer have the original math lesson laid out by the
firmware engineer".


Andrew Rodney
Posted on Mon Oct 20, 2003 9:07 AM

The official answer on the DTP-41:

The filter wheel in the DTP41 rotates at 12 rps, so we get 12 samples per second. With the default paper speed of 1.2 inches per second or 30.48 mm/sec, this gives us (30.48 mm/sec) / (12 samples/sec) = 2.54 mm per sample. If we ignore the transitions, this means for a 7mm patch we have (7mm patch) / (2.54 mm per sample) = 2.76 samples per patch.

Obviously, we can't ignore the transitions, so some of the readings at the transitions are discarded as Ray points out. This is done on a filter by filter basis, so we end up throwing out some fraction of a sample on each end of the patch. Exactly how many filter readings are discarded depends on the quality of the transitions, so it's hard to pin down the final number of samples for a given patch size. It will typically be around 2 samples for our 7mm example and will be more for larger patches or slower paper speeds. Whatever we end up with is averaged.

I hope this makes everything clear. This material will be on the final exam at the end of the semester.

Daniel J. Beimers
Sr. Digital Systems Engineer
----

Special thanks to all the engineers at X-Rite (especially Daniel). So once again, IR finds the correct people deep inside huge companies to answer your burning questions <g>.


Marco Passanisi
Posted on Tue Oct 21, 2003 11:22 AM

Andrew,
thank you very much for collecting the info on this matter.
At this point I'd still like to know if there is a way to set a slower paper speed.

Thanks.

Marco


Andrew Rodney
Posted on Tue Oct 21, 2003 11:40 AM

on 10/21/03 10:25 AM, IR - Marco Passanisi wrote:

> At this point I'd still like to know if there is a way to set a slower paper
> speed.

I know no way of doing this from the user level. It's possible the software
may be able to do this.


Marco Passanisi
Posted on Thu Oct 23, 2003 3:33 AM

Can you throw in some suggestions on the following problem?

This is the story:

I wanted to increase the number of patches in a single column in order to shorten the profiling cycle-time.
I did this by simply cutting and pasting the TIFF testcharts that come with GM ProfileMaker and making 50 patches columns on a verctically oriented 13x19 sheet.
The testchart was not rescaled, up/down/sampled or whatever.
I, of course, modified accordingly the reference .txt file.
Using the X-Rite ToolCrib I updated the strip definition into the instrument, so that it would expect a 50 patches column.

This is the problem:

When I try to read the new testchart I get a tremendous rate of reading failures to a level that it's no more an advantage (timewise) to have 50 patches columns.
The MeasureTool simply asks to read the same column over and over...
The column goes through the DTP41 flawlessly and I don't see any evident reason for the reading failure. All the patches are perfectly placed, shaped and sized.

I'm stuck.

I quickly designed a few columns with 30 patches and they measure nicely.
Should I try 35 then 40 then... until I find the upper limit?
The instrument should be capable of reading up to 100 patches.

I initially thought it might be a limitataion of the GM software but when I tried to output the raw data into Excel, I was getting the same failures, and again ok with 20 and 30 patches columns.
Any idea??

Thanks.

Marco


Andrew Rodney
Posted on Thu Oct 23, 2003 7:40 AM

on 10/23/03 2:35 AM, IR - Marco Passanisi wrote:

> Any idea??

Print the same target several times (rotate them on the page too). Measure
and average all the sheets.


Marco Passanisi
Posted on Thu Oct 23, 2003 10:55 AM

Hi Andrew,
sorry but I don't quite get your answer...

Which target should I print several times? And what's the point if I can't measure it?

I'm sure I'm misunderstanding your reply.

I redesigned in PS a new testchart with 50 patches columns. This new testchart doesn't get read by the instrument.

The standard target, which reads nicely is 20 patches, another custom I made with 30 patches columns still reads nicely, the 50 patches fails...

Marco



Andrew Rodney
Posted on Thu Oct 23, 2003 6:45 PM

on 10/23/03 10:00 AM, IR - Marco Passanisi wrote:

> Which target should I print several times? And what's the point if I can't
> measure it?

I thought you wanted more samples per patch. If you measure the targets
multiple times, that's what you're getting.


Marco Passanisi
Posted on Fri Oct 24, 2003 10:55 AM

No Aandrew,
sorry for the misunderstanding, I said I wanted more patches in a single column so that I can scan the entire target in a shorter time. So I re-arranged the GretagMacbteh testachert (that has the columns made of 20 patches) into a testachrat that has 50 patches per column.
When I try to read the 50 patches columns the DTP41 can't read them...

Marco


Andrew Rodney
Posted on Fri Oct 24, 2003 7:20 PM

on 10/24/03 10:00 AM, IR - Marco Passanisi wrote:

> When I try to read the 50 patches columns the DTP41 can't read them...

There has to be some SDK from both companies that detail how to do this but
it may not be trivial.





Other Sections Available if you login:
calendar

April 2004
S M T W T F S
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18
19
20
21
22
23
24
25
26
27
28
29
30
 
May 2004
S M T W T F S
            1
2 3 4
5
6
7
8
9 10 11 12
13
14 15
16
17
18
19
20
21
22
23
24 25 26 27 28 29
30 31          
 

Upcoming Events


MAKE YOUR WORKflow (A D-65 event - led by IR expert Seth Resnick)
    4/18/04 at 9:00 AM to 4/24/04
The Digital Portfolio, (at John Paul Caponigro's studio, Maine)
    4/19/04 at 9:00 AM to 4/23/04
Article - Book Review
    4/24/04 at 5:00 PM
Santa Fe Workshops - Master Photoshop
    4/26/04 at 9:00 AM to 4/30/04
Article - Tutorial
    4/30/04 at 5:00 PM

...see more upcoming events

imagingrevue Members receive special discounts on most events listed in calendar

  


customer.support@ imagingrevue.com
Privacy Policy
Terms of Service