Silverfast scanner software

[Kenneth Sky]

Mark
Thanks for this very informative article. I am about to receive an Epson 4990 which comes with Silverfast SE. I plan to scan my MF negatives on it. Silverfast literature seems to denigrate the SE as entry level for beginners and promotes Ai6 as the most appropriate solution. However, in your article you don't use most of the features such as ICE. Is it neccessary to upgrade?

[mikeseb]

I second the thanks, Mark. I scan MF film with a Nikon 8000. I use Nikon Scan almost entirely with good results, though its interface leaves a lot to be desired.

I have VueScan which I first bought to use with an ancient HP flatbed, since retired; but I don't share the cultlike affection for VueScan I've seen elsewhere online. It is very difficult to use when batch scanning MF film strips with multiple images, and the documentation is paltry at best.

I have considered Silverfast but have been put off by the price/benefit calculation. Not sure what I'll do now but I'm going to give it another look.

[Mark D Segal]

Kenneth and Mike - much appreciated - that is what makes it worthwhile writing.

Kenneth - I don't know SE - never used it. For my scanner the cost of the full shee-bang was not too way out to lunch so I just bought it, in order to have everything they offer. Then I picked and chose what I need. I think the best answer I can give under the circumstances is to compare the features SE provides with the features I illustrated in the article, and if you see stuff there you will really miss, and if the price of the program for your scanner makes it worthwhile, then you should up-grade. Perhaps they offer an up-grade price? I don't know.

As for ICE - if you are scanning alot of old stuff that has become cruddy, this is a wonderful feature to have. It works very, very well on crud. In fact, it is not supposed to work on Kodachrome (at least the version that came with Minolta's 5400 software), but I used it on 50 year-0ld Kodachromes and it worked just fine. When I asked Minolta why it worked when they said it wouldn't, they said that indeed it can but they can't guarantee it, so they were being careful.

Mike - likewise, in your situation, I would have a careful look at the SE versus full program features, and then you'll have a better idea what is worthwhile relative to your needs.

[Kenneth Sky]

Thanks Mark

[Jack Flesher]

Good job Mark, but a question on Q-Factor.

As I understand it, all it is is an output resolution multiplier.  So why not just scan at the direct PPI with a Q-Factor = 1.00 ?

(I think I read somewhere that it is primarily for line art, and the ideal setting is root-2, or 1.4. I assume it helps alleviate moire from the lines?)

[Mark D Segal]

Jack I think what you are suggesting is fine if the final destination of the output is an inkjet printer. One of the most frustrating aspects of Silverfast, as you may have read not-even-so-much-between-the-lines of my article, is the questionable quality of the documentation. Laser-soft itself can be obscurantist, and the next best "official" source is Taz Tally's book ("Silverfast The Official Guide"), which has its strengths and weaknesses. One of the weaknesses is on this very issue. He says three different things about Q-factor in three different places, so it leaves readers in a total quandary about what to do. In one place he says leave it set at 1.0 (page 38 - Quick Start advice). In another place he says use 1.4 (as recommended by Lasersoft Imaging) for commercial printing or 2.0 for images with lots of high contrast edges (page 55). Then in yet another place he recommends an initial "General" default value of 2.0, which can then be re-adjusted (to what???) in the Scan Control window (page 61).    

One gets the impression that for inkjet prints anything between 1 and 2 would be OK provided the final resolution outcome is what you want. After reading all this stuff, the first time I opened the program it was set at 1.5, and I got the resolution and dimension relationships the way I wanted them, and I'm not destined for commercial printing, and the inkjet results were just fine so I simply left it there.

[Jack Flesher]

Thanks for the clarification Mark.  Yes, it is too bad SF themselves is not more forthcoming about what some of their tools actually do!  Anyway, in the case of my Epson 4990 using SF Ai scanning 4x5 color transparancy at 2400 PPI final, it appears that using a Q-Factor of 1.0 generates a slightly better final scan than using the default 1.5@1600PPI.  The difference is so slight it probably would not show in a print though...

[Eric Myrvaagnes]

Thanks for the article, Mark. As always, you are clear and informative. I wish I had had your article when I first started using SF; it could have saved me a lot of time and headache.

I use SF SE on an Epson 4990, and I find that most of the features you use are included in SE. One that isn't there is the Q-Factor, and it sounds as if its absence merely saves me from some additional confusion.

So far, SE has worked well for me, and I have never been tempted to upgrade to AI (although I'm pretty sure a special upgrade offer came with the scanner).

Eric

[Mark D Segal]

Thanks Eric, much appreciated. I also wish I had the article before I wrote it!  

If SE does all you need, you've saved yourself a bundle!