Newbie looking for some getting started advice

General topics about imaging

heinz
SilverFast User
SilverFast User
Posts: 25
Joined: Sat Oct 15, 2005 3:15 am

Newbie looking for some getting started advice

Postby heinz » Fri Nov 11, 2005 3:11 pm

I got the software and the books and am doing some reading and some experimenting but work keeps interfering, preventing me from going full blast. I got the Scanner calibrated but not much else yet.

I have about a thousand or so Kodachrome transparencies, A Nikon 4000 Super CoolScan, No PhotoShop yet. I'm considering getting Apple Aperture instead.

What I'm stewing about is 1) Do I wait until I know more about SF6 before digitizing my transparencies or 2) do I 'archive' them now as Tiffs and tweak them later with SF6 (assuming that I won't get PhotoShop or Aperture)

I'd appreciate some comments re the pros and cons of either approach. I somehow think I'd feel a little 'safer' if I had my images on disk and then 'go back' to them. Or am I simply increasing workload in the overall.

I want my slides to be 'safe'. I don't know what I'll ultimately do with them other than family viewing, some 8x10 prints and slideshows destined to the family TV, assembling some memories on DVD, etc. Still, I don't want to scan all the images and then find I limited myself because I used settings that in effect are limiting. Disk space is almost no issue as I plan to put all images on an external and burn to DVD for backup.

My 'quiet' hope and desire is that electronic picture frames will increase in size and decrese in cost. That will be my ultimate 'destination', someday, hopefully.

But in the meantime, what to do, what to do, what settings to use if archiving all is the first task.

Thanks in advance for any comments/suggestions you may have.

Heinz

degrub
SilverFast Master
SilverFast Master
Posts: 597
Joined: Fri Dec 21, 2001 1:00 am
Location: Texas

Postby degrub » Sat Nov 12, 2005 4:21 am

Here's what i do

1) get an "antistatic brush" , availible from most photo supply houses. has a polonium , IIRC, thread in it . use it on the slide.

Scan at 4000 PPI, 48 bit color, ICE on , set black and white point, attach scanner profile, 4 or 8x multisample, save as a std tiff file,check for ICE artifacts, rescan without ICE if needed, burn to gold CD, copy to external HD. From that i can use PS to make adjustments.

Otherwise, if you have HDR (used to come with Ai for the 4000) , you can scan to HDR 48 bit at 4000 PPI, etc. , and then work with the image later. See Ian Lyon's tutorials

www.computer-darkroom.com

heinz
SilverFast User
SilverFast User
Posts: 25
Joined: Sat Oct 15, 2005 3:15 am

Postby heinz » Sat Nov 12, 2005 5:37 pm

No HDR included anymore. At least not seperated. There is a 48 Bit HDR Color setting but I'm assuming it's to scan for later use with HDR version of SF?

Q re dpi setting: I have 8000 available but filesize jumps dramatically. 4000 more than adequate or should I go 8000?

TIA

degrub
SilverFast Master
SilverFast Master
Posts: 597
Joined: Fri Dec 21, 2001 1:00 am
Location: Texas

Postby degrub » Sun Nov 13, 2005 4:38 pm

Correct.

Which version of Ai did you purchase ? or was it with the scanner ?

You won't be able to "tweak" later with SF unless you have one of the varients of HDR or DC and use the "for HDR" option. If you have/will have a digital campera, i would get DC(it reads HDR files unless SF has changed something recently, talk to them to confirm) or Photoshop. HDR/DC has some unique features, but PS is more general and there are a large number of add-ons and users. Apple's software is perhps more narrowly focused at this time. See www.luminous-landscape for some discussion.

8000 ppi is interpolated up by SF. Stick with the native 4000 PPI. THere isn't a lot more information in the slide to capture with the CS4000 above 4000 ppi. 8000 PPI would only be usefull for huge blowups (20x or more) of the image.

You will learn scanning by doing, so take a few of the slides and play with the scanning adjustments. Use Ian's tutorials and the books. The difficult part is learning to see what the corrections should be and then figuring out how to use the tools to accomplish that. i have used Dan Margules and Katrin Eismann as guides for seeing. Ai was designed to allow for the best possible image to be produced at scan time and its tools reflect that.

heinz
SilverFast User
SilverFast User
Posts: 25
Joined: Sat Oct 15, 2005 3:15 am

Postby heinz » Sun Nov 13, 2005 6:51 pm

> Which version of Ai did you purchase ? or was it with the scanner ?

The AI Studio version, I.e. full version from LaserSoft.

Thank you re clarifying/confirming that SF tools are meant for "during scan tweaking" which basically leaves out the option/thought of scanning en-masse (archiving) and then going back to them.

User avatar
RAG
SilverFast Master
SilverFast Master
Posts: 761
Joined: Wed Jan 12, 2005 7:59 am
Location: Sonoma County, California

Postby RAG » Sun Nov 13, 2005 10:29 pm

Heinz,

Another bit of advice that dove tails with what degrub has already told you, is that you should set your output size to the largest image you expect to need when creating your archive scans.

Use the following as a guide for enlarging and image. If you set your effective image resolution to 4000, but you want to increase the size of the output image by 200% (2 times) your scanner will have to scan at an effective resolution of 8000. At this resolution the quality of your image will drop because you are forcing your scanner to interpolate pixels. To avoid this drop in quality always keep the resolution within your scanners maximum "OPTICAL" resolution. In this example you would set your effective resolution to 2000, which would allow your scanner to scan the image at 4000 to create an image that is 200% or 2 times as large. [Formula: size increase times effective image resolution = actual scan resolution]

Having all of this under your hat, I should also tell you that the method of presenting your final image plays a role in all of this as well. If you are using a desktop printer at home to create your prints then you actually only need images that are maximum 300ppi/dpi in size (This is the size professional printers use, most home printers do fine with 240ppi/dpi). If you are going to use the images for screen display 96ppi will meet your needs.
Member in good standing - NAPP
A picture is worth a thousand words! :-)

heinz
SilverFast User
SilverFast User
Posts: 25
Joined: Sat Oct 15, 2005 3:15 am

Postby heinz » Mon Nov 14, 2005 1:41 am

Well, I'm in the midst of 'trying' :-)

Confused re the dpi/ppi though.

I have my output set to 2560x1600 as I figure that some images will be used on my 30" display.

If I set 400dpi the enlargement is greater than when set at 4000 dpi. Resulting filesize is the same. Am I correct in assuming that both settings in effect provide the same results, I.e. are equal?

What I'm currently doing is using the the defaults SF started out with with the exception of dpi and output size.
I.e.
48 bit Color, No filter, Image type: Standard
Original and Scale% I control with Output settings of 2560x1600 (pixels)
Q-Factor is set to 1.5, Screen 1050 lpcm

I use the Automatic Image optimization after the pre-scan and occasionally adjust brightness prior to Scan. I have it set to multi-scan: 8

Images are saved as Tiffs.

How far off am I ?

degrub
SilverFast Master
SilverFast Master
Posts: 597
Joined: Fri Dec 21, 2001 1:00 am
Location: Texas

Postby degrub » Mon Nov 14, 2005 1:45 am

As RAG said, look at your output size. if you want to print at 8x10 inch, then 4000 ppi/300 ppi will give you about 13x over the original (1x1.25 for slide) , more than enough for an 8x10. If you want larger than that, look at fractal interpolation or other algorithms. Scans at 8000 PPI will not change the result .

User avatar
RAG
SilverFast Master
SilverFast Master
Posts: 761
Joined: Wed Jan 12, 2005 7:59 am
Location: Sonoma County, California

Postby RAG » Mon Nov 14, 2005 5:26 am

Heinz,

DPI stands for dots per inch and PPI pixels per inch. DPI is what printers use and PPI is what display screens use.

It seems like you are talking about the image dimensions of 2560X1600 Vs what I was talking about earlier. I was talking about "Scale" times your DPI or PPI setting = the DPI or PPI your scanner will actually use during the scan process.

<img src="/img/forum/scale.gif">
Member in good standing - NAPP

A picture is worth a thousand words! :-)

heinz
SilverFast User
SilverFast User
Posts: 25
Joined: Sat Oct 15, 2005 3:15 am

Postby heinz » Mon Nov 14, 2005 8:38 pm

Back to the books it seems :-)

What I'm trying to wrap my brain around is to get a full/native image for the Cinema display (2560x1600) and with the same scan use the resulting image to print snap-shot size and to use the image in slideshows that are destined to be displayed on a TV screen.

Max printing might be an 8x10 but perhaps with some cropping/enlarging.

I'm thinking that if I plan for the 2560x1600 it will scale down to smaller displays, which is better than scaling it up/stretching it?

The only cropping I'm doing at time of scan is to eliminate the round corners, I.e. I take the max.

I think what is helping to confuse me is that if I look at the image with Preview on the Mac and select "Actual Size" the 300 dpi is larger than the 4000 dpi version scan. While the specs show 2560x1783, the image does not cover the display. It does though when used as a desktop background.

Grrr, I just feel so stupid about this.

User avatar
RAG
SilverFast Master
SilverFast Master
Posts: 761
Joined: Wed Jan 12, 2005 7:59 am
Location: Sonoma County, California

Postby RAG » Mon Nov 14, 2005 10:39 pm

Heinz,

Please don't feel stupid this is a complex subject.

Let's see if I can help by telling you the reason the file size is larger is because there are more pixels in the image when you scale it up to a larger size. The expressions PPI or DPI define how many pixels or dots exist in a given area which in both of these cases is an inch. If you have 300 dots or pixels in one inch and your image is 600 inches (I am staying linear for simplicity) you would have a total of 180000 pixels (number of pixels in an inch times the number of inches), correct? If this is correct it is also correct that the file size must increase.

A simple analogy might help here, if you have bricks that are 4 inches by 4 inches in size, and each one weighs 2KG, it will only take two of them to fill an area of 4 inches by 8 inches at a total weight of 4KG (File size). Now, if you attempt to fill an 8 inch by 8 inch area using the same bricks it will take a total of 4 bricks weighing 8KG (File size). This resulted in double the number of bricks used and doubles the weight.

Does this help at all?
Member in good standing - NAPP

A picture is worth a thousand words! :-)

heinz
SilverFast User
SilverFast User
Posts: 25
Joined: Sat Oct 15, 2005 3:15 am

Postby heinz » Mon Nov 14, 2005 11:44 pm

Thank you soooo much for hanging in :-)

Filesize difference I have no problem with. If specs change, something has to give. But what about 2 different approaches?

I did two scans, one at 300 and the other at 4000 dpi. Please take a look at the settings at: http://homepage.mac.com/heinz/2005/page7/page7.html

(didn't know how to embed the images here :-)

Are they truly the same? Similar files size, hence it must be similar amount of bricks (information)? Is there an advantage to using one over the other or am I completely off track.

User avatar
RAG
SilverFast Master
SilverFast Master
Posts: 761
Joined: Wed Jan 12, 2005 7:59 am
Location: Sonoma County, California

Postby RAG » Wed Nov 16, 2005 6:40 am

Heinz,

Though the file sizes are approximately the same your second example would probably be better for creating an 8 X 10 print.

In your fist example you are effectively down sampling the image and using smaller "bricks" as well. At this size you would be forced to increase the size in Photoshop in order to create the 8 x 10. What will happen is that your smaller "bricks" will be spread out over a larger area leaving gaps that will have to be filled in by one of Photoshop's algorithms.

Remember that ?Original? and ?Output? are the current and desired areas you wish to fill with image information. ?Scale? represents the amount of change between the two areas, and DPI or PPI represent the number of dots or pixels in an inch. Once you scan an image at a particular resolution you have to add more pixels if you want to make it larger or take pixels away if you want to make it smaller. Generally speaking making an 8 x 10 image into a 4 x 6 or 3 x 5 has better results than doing the opposite.

Play with a couple of images and you will see.
Member in good standing - NAPP

A picture is worth a thousand words! :-)

heinz
SilverFast User
SilverFast User
Posts: 25
Joined: Sat Oct 15, 2005 3:15 am

Postby heinz » Wed Dec 07, 2005 2:01 am

Thought I'd post a little update :-)

Have scanned a few hundred Kodachrome & Ektachrome by now. Don't really know why I'm doing what yet but the results are acceptable and if nothing else I'll have a "contact sheet" in iPhoto to find specific slides to scan again.

I decided to scan at 300dpi to a resolution of 1680x1050 for my new iMac.

I basically mess around with setting until they look good :-))

Slow process but sofar reasonably good.

I can hardly wait until I get to negatives. Pleasantly surprised at how the slides held up. They are all 77/78 vintage as are the negatives.

Have some 110 stuff to scan too.

Was it worth it to get SilverFast? YES!!!

User avatar
RAG
SilverFast Master
SilverFast Master
Posts: 761
Joined: Wed Jan 12, 2005 7:59 am
Location: Sonoma County, California

Postby RAG » Wed Dec 07, 2005 2:40 am

Hey,

Glad to hear you are having fun, and making progress!!!

What type of scanner do you have that will allow you to scan 110 negatives? I have some 110 myself, but I don't have a holder for scanning them at this point.
Member in good standing - NAPP

A picture is worth a thousand words! :-)


Return to “Imaging in general”

Who is online

Users browsing this forum: No registered users and 2 guests