6.2.0r1 works!

All the problems with PIE film scanners only

nmoss
SilverFast User
SilverFast User
Posts: 63
Joined: Tue Sep 11, 2001 1:00 am

6.2.0r1 works!

Postby nmoss » Tue Jan 20, 2004 3:02 pm

Hi,
I've complained often enough about the poor color performance of the 6.x versions of SilverFast compared to the 5.5.x series on my PIE 1800 scanner. Now it looks like you've fixed it! For the first time in SF6 my final scans using the twain driver on a Windows 98 PC are a close approximation to the preview scans. It looks like I will finally be able to switch over from 5.5.2r02 (which is still a terrific program).

Now, what is that yellow/brown stripe down the right hand side of images? It's not there in scans with version 5.5.2

Thanks

NMoss

User avatar
LSI_Noack
LSI Staff
LSI Staff
Posts: 1432
Joined: Wed Oct 22, 2003 1:30 pm
Location: Kiel, Germany
Contact:

Postby LSI_Noack » Thu Jan 22, 2004 5:41 pm

Dear nmoss

seems I wouldn't want to know what these yellow-brown stripe on your images are... >;-) SNCR

Now seriously: Whether under Win XP (SP1) nor under Win 98 (SE) I can find experience these artifacts.

Do you have this with all images scanned?
Have you tried cleaning your scanner (this often fails)?

nmoss
SilverFast User
SilverFast User
Posts: 63
Joined: Tue Sep 11, 2001 1:00 am

Postby nmoss » Fri Jan 30, 2004 2:34 am

Well, I was a little hasty. It seems that the greens are remaining in good shape when the final scan is created but blues are becoming very pale and reds are going pink. If it's any help, the dust and scratch preview window also looks as bad as the final scan. The preview is fine.

I just downloaded what it claimed to be SF6.2.2 but it runs as 6.2.1.

This is a PC with Windows98SE and lots of memory. Twain driver. PIE 1800 scanner working great with SF5.5.2 but non functional will all versions of SF6

I will try and send you an image depicting the problems.

NMoss

User avatar
LSI_Noack
LSI Staff
LSI Staff
Posts: 1432
Joined: Wed Oct 22, 2003 1:30 pm
Location: Kiel, Germany
Contact:

Postby LSI_Noack » Wed Feb 11, 2004 10:51 am

Dear nmoss

until now it seems I haven't received any mail from you
(info@SilverFast.com, subject "to noack - pie 1800 color prob").

Does that mean that the latest version (6.2.0r3) has solved this issue?

If not, however, please do send me problem description with supporting screenshots and an SF_Logfile so that I can trigger a bug entry if your problem can be reproduced by the testing center.

TIA, Sonny

nmoss
SilverFast User
SilverFast User
Posts: 63
Joined: Tue Sep 11, 2001 1:00 am

Postby nmoss » Wed Feb 11, 2004 5:23 pm

Hi Sonny,

Excuse me for the delay, I fell off my chair when I saw your note.
You will remember that all my problems are with Negafix using version 6.X
on a PIE 1800 slide/negative scanner. Windows 98SE and the Twain version of SF. I'm using ThumbsPlus to acquire the scans and organize the images, Photopaint to work on them. I did send you an Email with a composite image attached showing the excellence of SF 5.5.2 and the failings of 6.2.2 using this arrangement. Maybe it didn't get through, it certainly wasn't returned to me as undelivered.
Anyway, my silence is in some ways indicative of my resignation over this issue. I have tried the r3 release and found it to be somewhat better than the r2, which was better than the r1. However, it's still not right and I am beginning to become suspicious of my system as it would seem that you would be deluged with messages if these problems are happening to other users.
The problem may be in some sort of equalization step that is completely changing the colors of the scanned image, relative to the prescan. I say that because I can approximate the effect by taking a good-looking image in a graphics program and performing an autoequalize on it. Improvements are noticeable in the more recent releases of SF6.2, but I am afraid that variations in my usage are revealing different degrees of malfunction. As I mentioned in my message, the greens seem to be doing better but the blues and reds are not good. I can improve things by setting the autotolerance to 0 (what does that do anyway?) and completely disabling mask removal in the histogram window.
Also, there is a very pronounced yellow/brown vertical stripe down both sides of the image at the edges. This was also a problem in SF5.5 and I was about to send the scanner back to PIE (even had the return authorization) when SF5.5.1.r22 came out and the problem went away. I suppose it is still a problem with the scanner but the program was able to compensate for it. The images are very granular, too much contrast producing saturation I guess.
As you can imagine, I am tired of trying to get the 6.X versions to work, especially as I am still very happy with 5.5.2. To be honest, I haven't had much luck with the dust and scratch removal and the gane function is better applied to the scanned image with real image manipulation software so there is not much additional functionality missing for me.
However, in response to you query and interest, I will try to doocument the problem very carefully. Did my last image get through, or is there an officila way to submit sample images?

Thanks for your interest

Nick Moss

User avatar
LSI_Noack
LSI Staff
LSI Staff
Posts: 1432
Joined: Wed Oct 22, 2003 1:30 pm
Location: Kiel, Germany
Contact:

Postby LSI_Noack » Wed Feb 11, 2004 6:18 pm

Dear Nick

I really hope we can solve this issue. I did sent you a private message with my direct mail address where you can send your images and further information, too.

TIA, Sonny


Return to “PIE”

Who is online

Users browsing this forum: No registered users and 2 guests