Re: LS-2000 help

[Bill Glickman <bglick@xxxxxxxx>]

Thanks Ed, just wanted to be sure.....


> Bill..... Sorry I didnt group them together and reading back I t seams
that
> I DID. Use unsharp msking NOT sharpening 2 different things. I dont have a
> quick rule of thumb answer for unsharp masking so I did some tests. Thats
> probably the best way to see what is appealing to you.
>
> As for the LightJet. I have never sent anything to output on one. I have
> been thinking of doing some of my 6X17's but I have been lazy. I would
think
> that the same rules hold true. Do you have a local Quickprinter that has
> one? Ask them what trhey are using as a starting point.
>
> Ed
>
> > ----------
> > From: Bill Glickman
> > Reply To: panorama-l@xxxxxxxxxxxxxxxxx
> > Sent: Tuesday, August 8, 2000 1:55 PM
> > To: panorama-l@xxxxxxxxxxxxxxxxx
> > Subject: Re: LS-2000 help
> >
> > Ed
> >
> >      Thank you for the response......   just to clarify one thing...
> > during
> > the drum scan, my scanning software offers Unsharp Masking, not
> > sharpening.... I was not aware if you were grouping them in the same
> > category?   If I do not apply any unsharp masking, the images look very
> > weak
> > vs. applying just a bit.  Are you saying NO, do not apply USM, or do not
> > apply sharpening during the scan?
> >
> >        Also, do you ever print to the CSI light Jet 5000?  If so, what
> > settings do you scan for that...LPI, etc.
> >
> >
> > Thanks Ed
> > Bill G
> >
> >
> >
> >
> > > Hi Bill,
> > >
> > > First understand what sharpening does,  Simply put...   To sharpen an
> > image
> > > the software that you use finds a black line  and blends a white line
> > right
> > > next to it (no black lines the software will use a white lineand place
a
> > > black line there). You can see this if you look close after you
sharpen.
> > You
> > > can REALLY see it if you oversharpen. I do not like to scan with
> > sharpening
> > > ON because the software places those artifacts in the origional data
and
> > you
> > > cant go back. So I Scan, Color correct then sharpen IF needed. Or
apply
> > an
> > > unsharp mask. So Dont sharpen when you scan.
> > >
> > > Ok on to the scaling. Scanning at a large size and reducing is the
BEST
> > > CASE. You are effectivly increasing the scan resolution and the output
> > will
> > > be just fine. What I was refering to was the guys that scan an image
at
> > HIGH
> > > DPI 600, 1200, 2400  and then after the image is saved the BLOW UP the
> > image
> > > to 11X14 or 16x20 and wonder why the output looks like poop! What I
was
> > > saying is scan your image to the final size you are going to use. For
> > > example: If you have a 4x5 and you need an 8X10 Final Scan the image
at
> > 200%
> > > not at the max resolution of the scanner and then blow it up. Even
> > though
> > > the image file seems big It does not contain the DATA to increase the
> > size.
> > > Once again you can DECREESE the size and the resolution will increase
> > but
> > > with this method you can end up with 1x2" 100Mb image files.
> > >
> > > 1 final thing If you really really wanna figure out exactly what to
scan
> > > your images at here is THE formula.......
> > >
> > > DPI over LPI Squared +1=256
> > >
> > > Where DPI is the scan scan resolution and LPI is the Line screen the
> > printer
> > > will be using (Epson printers = 150LPI) Square this value and add1.
The
> > > final number = 256 levels of gray. That is the maximum number of
colors
> > the
> > > printer scanner combination can reproduce. If after you figure this
> > formula
> > > and the answer is more than 256 you are scanning at to high a
resolution
> > and
> > > wasting disk space. If the number is UNDER 256 you have not scanned at
a
> > > high enough resolution and there is not enough DATA in the image.
> > >
> > > OK Too complicated.... here is how I do it. Take the LPI of the
printer.
> > > (Epson type printers = approx 150 LPI) and times that by 1.5 which
gives
> > you
> > > 225. Set the scanner to 225DPI and scan away.
> > >
> > > PHEW! I think thats it. I hope I answered more questions then I
created.
> > > Have fun and BUY THAT PHOTOSHOP book its worth the money and your
images
> > > will love you. And NO I didnt write it.
> > >
> > > Ed
> > >
> > >
> > > > ----------
> > > > From: Bill Glickman
> > > > Reply To: panorama-l@xxxxxxxxxxxxxxxxx
> > > > Sent: Monday, August 7, 2000 5:53 PM
> > > > To: panorama-l@xxxxxxxxxxxxxxxxx
> > > > Subject: Re: LS-2000 help
> > > >
> > > > Ed
> > > >
> > > >      Hope you don't mind a few of us tapping into your expertise!
:-)
> > I
> > > > use a Howtek 8000 dpi drum scanner with Trident 4.0 software.  A few
> > > > issues
> > > > your raised....
> > > >
> > > > > Ok just a few guidlines for better scans
> > > > >
> > > > > 1. Turn off sharpening
> > > > > Do that in photoshop during color correction and use SPARINGLY.
> > > >
> > > >       Is this true for drum scanners also?  The images look a bit
weak
> > > > with
> > > > out some USM used.  I try to use low to moderate, but never high.
Is
> > this
> > > > correct?
> > > >
> > > > > 2. Figure out what the FINAL size of the scan and scan that size.
> > > > > (magnification %)
> > > > > For example If you have a 4X5 and you wanna make an 8X10 Scan it
at
> > > > > 200% not at 1200 DPI.....
> > > > > DO NOT!!! I repeat DO NOT!!!  scan at the Max resolution of the
> > > > > scanner and then scale the photo up. That makes the WORST possible
> > scan
> > > > +
> > > > > Its a waste of HD space. Your scans should rarely pass 250 dpi.
(If
> > you
> > > > > wanna know why read the book)
> > > >
> > > >            What exactly do you mean by scaling up?  If I want to
scan
> > an
> > > > image and print at different sizes, say 11x14, 20x24 and 40x48, are
> > you
> > > > saying that if I scan at the 40x48 and then print at 11x14 it will
be
> > an
> > > > inferior scan vs. if I scan it once for each size print output?  If
> > so,
> > > > that
> > > > is very fascinating and not widely known.  Possibly I misunderstood
> > what
> > > > you
> > > > were getting at.
> > > >
> > > > Thanks Ed
> > > > Regards
> > > > Bill G
> > > >
> > > >
> >
> >