Page 1 of 1

Setting output resolution exactly to optical

PostPosted: Fri Apr 11, 2008 11:03 pm
by Tim B
Hi,
I have just installed my new Epson v750 Pro and SilverFast, which is version 6.5.5r5. I am using a PC with Windows XP Pro.

There is confusion over using the Ctrl key which brings up a warning dialogue. This explains new keys to find Highlight and Shadow, but makes no reference to which keys to use to reveal true optical resolution in the Frame options window.

Trial and error has revealed that F5 appears to act as Ctrl and F6 as Shift+Ctrl.

But I can see no way of setting the output resolution to exactly match the Optical reolution of the scanner. Whichever setting I use, it is at least slightly different.
Can this be done? It is driving me crazy!
Thanks for any help you can offer.

Tim

PostPosted: Tue Apr 22, 2008 6:37 am
by LSI_Muenier
Dear Tim,

to choose the optical resolution of a scanner, it is common in SilverFast to move the resolution slider to the last but one position to the right (and choosing a scaling of 100%). Please note that the rightmost position is for a resolution 200% above the optical resolution.

Best regards
Martin

PostPosted: Tue Apr 22, 2008 2:50 pm
by Tim B
Thank you Martin for your reply.

In fact the Epson v750 can operate at various optical resolutions; I presume that your comment would refer to the maximum optical resolution, which in the case of the v750 is 6400 dpi, which is shown when the slider is, as you suggest, in the last but one position.

The problem I was having was I think a function of having set the "Q" factor to more than 1 and using other than 100% scaling. Nothing I did would give me an exact figure for optical resolution. This coupled with the odd behaviour of the "Ctrl" key and no explanation about the F5 and F6 keys, led to my confusion.

I'd be grateful if you would comment on what the "Q" factor actually does and the use of the "Ctrl" key to reveal actual optical resolution, or not in this case.

Thanks again,

Tim

PostPosted: Tue Apr 22, 2008 3:10 pm
by LSI_Muenier
Dear Tim,

the "Q-factor" (or quality factor, e.g. 1.5) makes sure that your scan resolution is automatically higher than the output resolution you aim at - even at 100% scaling. The factor of 1.5 leads to 50% extra resolution, so all the details are scanned that are needed to make them appear when you output at the resolution you have typed into the "Screen" edit field.

Example: You will observe that a screen of 300 lpi and a Q-Factor of 1.5 correspond to a scan resolution of 450 dpi.

The Crtl (or rather F5) then tells you that your scanner will internally only be able to scan at 480 dpi (which is the next higher possible scan resolution). So, SilverFast will resample these 480 dpi image data to the 450 dpi you requested (no problem, of course).

This was for 100% scaling. If you set scaling to 200%, then the resolution field still shows 450 dpi, but your output size (in inches) doubles, and the MBytes are four times as large as before ("square doubling").

F5 now tells you that SilverFast will request 900 dpi from the scanner, which is a resolution the scanner internally can prepare, and no extra resampling is done in SilverFast.

Best regards
Martin

PostPosted: Thu May 01, 2008 9:56 am
by Tim B
Dear Martin,

Thank you very much for your clear and helpful explanation.

Tim

Re: Setting output resolution exactly to optical

PostPosted: Wed Aug 13, 2008 10:01 pm
by agripix
I have the V750 Pro, Ai 6.6.0r1, via Mac OS X 10.4.11.

I'm happy with my scans BUT confused about the correct resolution settings. I have been setting my scanning at 360lpi (540dpi) and the output size (% scaling) to the size of the print needed - for example, 12"x12", 12x16 etc. Files sizes are in the 150MB +/- range. The scans are fine, and the printed output from an Epson 2400 is good.

However, I notice many posters here are scanning with the settings at, say, 6400dpi to maximize the capacity of the scanner. Presumably, the scaling is left at 100% of the negative/positive, otherwise the file size is astronomical - and then the image size is selected later, in the next step in Photosphop etc.?

Am I thinking correctly?

Is there any noticeable difference between a print made from a scan at 360/540 to the desired print size, and a scan at 6400 100% scaled then resized later to the same print size as the first?

Thanks.

Re: Setting output resolution exactly to optical

PostPosted: Thu Aug 14, 2008 10:15 am
by Tim B
Hi, Agripix,

I always scan the neg. or slide at 100% and set the resolution to whatever I require. If I am looking to very big prints I use 3200dpi or maybe 6400 dpi, but the latter produces HUGE files from medium format originals (like 500Mb) so usually my maximum is 3200 (250Mb files)

I then reset resolution in Photoshop to 240 or 300 which gives me a print size of roughly 80x80cm or 28x28 inches from 6x6cm neg.

I don't see much difference in print quality as long as you are avoiding resampling and even then it will depend on your software.

Hope that helps,

Tim

Re: Setting output resolution exactly to optical

PostPosted: Thu Aug 14, 2008 11:03 am
by agripix
Thanks, Tim, for your reply.

I'll try to see if there is a visible difference by doing both.

I have been scanning all my 6x6 6x7 6x9 at 360/540 and setting my output as my estimated maximum print size. Usually something less than 13"x19" (Epson R2400 max size). If I needed to put any up on a website, I just changed image size downwards a bit, and dpi in post processing to 72dpi. Of course, I presume that if I wanted to go beyond my selected print size with that type of scanning process, I would be missing some information as I grew the print size (or used software to recreate the missing information.)

So, as you say, maybe I should always scan at the ~4800dpi and 100% scale to get all the information, and then select all the output (print) sizes after I have a scan file, during post processing ... ? Just playing with the settings here seems to create about the same file size either way ... At the moment, my scanning is a 3-4 minute process, so I'll also be interested to see if time is a factor.

I hope to see a few more responses to this thought.

Cheers, Colin