Re: filmscanners: GEM, ROC compared to Vuescan
"Jack Phipps" [EMAIL PROTECTED] wrote: When you do a scan using the Nikon interface do you have the "CleanImage" (Nikon's old name for Digital ICE) in the "Mode: On (Sharpen)" or "Mode: On (Normal)"? The mode Nikon calls "Mode: On (Normal)" actually blurs the image slightly. You should run in "Mode: On (Sharpen)". This should solve your problem with the sharpness. Er, pardon me if I'm stating the obvious, but doesn't that just mean that the software is compensating for the loss of sharpness it's generating? If ICE doesn't affect image sharpness, why would "sharpen" be necessary? Rob PS Jack, I wish I could use ICE, because I can actually live with the slight loss of sharpness. But I *can't* use ICE because Nikon refuse to fix the jaggies problem in Nikonscan. It makes Nikonscan and ICE worthless on the LS30. I know that isn't your problem, but it's a shame not to be able to use ASF's product. Thank God for Vuescan is all I can say.
RE: filmscanners: GEM, ROC compared to Vuescan
Thanks for the examples, Henry. I agree that Ed should separate the filters - I have a non-IR scanner, and am not a user of Vuescan yet, but this function would probably tip me into the camp. But ONLY if it doesn't soften my image as much as your samples indicate.. Mark T. At 06:02 PM 15/02/01 -0600, you wrote: From: "shAf" [EMAIL PROTECTED] My 1st experience with Ed's "scrub" was disappointing with respect to edges ... it wasn't like a "gaussian blur", but more like a "median" filter. I agree. It seems to me that the grain reduction does soften the image. I showed that on a webpage that I sent to this list a couple of weeks ago. Here it is again: http://www.geocities.com/hr1066/vuescan66.htm I still wish that Ed would make the grain reduction a separate control so that it wasn't combined with the cleaning options. snip == Mark Thomas [EMAIL PROTECTED] http://www.adelaide.net.au/~markthom
Re: filmscanners: GEM, ROC compared to Vuescan
[EMAIL PROTECTED] wrote: But ONLY if [vuescan] doesn't soften my image as much as your samples indicate.. I think most of the softening is the dust removal algorithms at higher settings, but it's hard to know. Only separating the features would make it possible to tell. Rob
Re: filmscanners: GEM, ROC compared to Vuescan
From: "Rob Geraghty" [EMAIL PROTECTED] I think most of the softening is the dust removal algorithms at higher settings, but it's hard to know. Only separating the features would make it possible to tell. If you look at the first example on my page you will see that it was scanned with version 6.5 (the last version before the new grain reduction and cleaning algorithms) and it was done using the Scour (strongest) setting. It *doesn't* have any softening. By the way, these scans were done using the Minolta Scan Elite which has the infrared channel. In the past Ed has said that his cleaning algorithms do *not* soften the image if there is an infrared channel. As you can see from the scans though, version 6.6 does soften the image when Scrub and Scour are used. Whether that is because the cleaning algorithm is broken in some way or because of the grain reduction algorithm, I don't know. It is the main reason I would like to have separate controls though. _ Get your FREE download of MSN Explorer at http://explorer.msn.com
Re: filmscanners: GEM, ROC compared to Vuescan
But ONLY if [vuescan] doesn't soften my image as much as your samples indicate.. I think most of the softening is the dust removal algorithms at higher settings, but it's hard to know. Only separating the features would make it possible to tell. Rob One could go back and try dust removal using Vuescan 6.5 (or some release before 6.6, which is when the film grain reduction was added to vuescan). -- Bob Shomler http://www.shomler.com/gallery.htm
Re: filmscanners: GEM, ROC compared to Vuescan
From: Bob Shomler [EMAIL PROTECTED] One could go back and try dust removal using Vuescan 6.5 (or some release before 6.6, which is when the film grain reduction was added to vuescan). Again, I refer you to this page I set up that does just what you suggest: http://www.geocities.com/hr1066/vuescan66.htm _ Get your FREE download of MSN Explorer at http://explorer.msn.com
RE: filmscanners: GEM, ROC compared to Vuescan
Bob writes ... But ONLY if [vuescan] doesn't soften my image as much as your samples indicate.. I think most of the softening is the dust removal algorithms at higher settings, but it's hard to know. Only separating the features would make it possible to tell. Rob One could go back and try dust removal using Vuescan 6.5 (or some release before 6.6, which is when the film grain reduction was added to vuescan). Some of the posts to this thread seem to be confused as to what is "softening" the image. It is NOT Vuescan's "dust removal" algorithm ... it IS the "grain reduction" algorithm which Ed has added to "scrub" and "scour". My impression is any qualifier (weak - strong) for "dust removal" no longer exists (... altho I don't know how Ed applies this to non-IR scanners ...), ... but there are 2 degrees of grain reduction. Selecting "dust removal" without "grain reduction" is done by selecting "clean", altho there is no method of selecting "grain reduction" w/o "dust removal". shAf :o)
Re: filmscanners: GEM, ROC compared to Vuescan
Related: why in general does Ice-based dust removal soften the image. The purpose of the IR channel is to identify the particular spots that are dust / scratches. Shouldn't the software only affect those areas leaving the rest of the image alone and sharp? John M.
RE: filmscanners: GEM, ROC compared to Vuescan
John Matturri writes ... Related: why in general does Ice-based dust removal soften the image. The purpose of the IR channel is to identify the particular spots that are dust / scratches. Shouldn't the software only affect those areas leaving the rest of the image alone and sharp? The IR channel is not strictly speaking a picture of dust/defects ... it also contains a component if the image itself. The ASF trademarked ICE algoritm apparently has a problem with this weakness of the IR channel (comments?, Jack) ... whereas Ed's Vuescan algoritm does not. shAf :o)
RE: filmscanners: GEM, ROC compared to Vuescan
Digital ICE from Applied Science Fiction should not soften the image. It identifies defects and corrects them. Even if there is a single strand of hair on a models face, it should still be there after the defects are removed. I have an example of an old Marilyn Monroe image taken by George Barris. There is hair or lint that is in the original record. It is still there. Only the surface defects are corrected. We don't "guess" at what is under the defect or color it in for surrounding pixels, we correct the surface defect. Jack Phipps Applied Science Fiction -Original Message- From: John Matturri [mailto:[EMAIL PROTECTED]] Sent: Friday, February 16, 2001 12:04 PM To: [EMAIL PROTECTED] Subject: Re: filmscanners: GEM, ROC compared to Vuescan Related: why in general does Ice-based dust removal soften the image. The purpose of the IR channel is to identify the particular spots that are dust / scratches. Shouldn't the software only affect those areas leaving the rest of the image alone and sharp? John M.
Re: filmscanners: GEM, ROC compared to Vuescan
"Bob Shomler" [EMAIL PROTECTED] wrote: One could go back and try dust removal using Vuescan 6.5 (or some release before 6.6, which is when the film grain reduction was added to vuescan). I still have several versions prior to 6.6. I'll see if I can give it a go. Rob
Re: filmscanners: GEM, ROC compared to Vuescan
"John Matturri" [EMAIL PROTECTED] wrote: Related: why in general does Ice-based dust removal soften the image. The purpose of the IR channel is to identify the particular spots that are dust / scratches. Shouldn't the software only affect those areas leaving the rest of the image alone and sharp? Vuescan's lowest setting on dust removal doesn't seem to affect image sharpness. ICE on my LS30 *does* affect the whole image. I suspect that ICE "sees" the grain of the film (or rather the "apparent" grain :) as some sort of defect and tries to correct it. As I mentioned elsewhere, we'll have to wait and see if ICE3 has fixed this problem. Rob
Re: filmscanners: GEM, ROC compared to Vuescan
"Jack Phipps" [EMAIL PROTECTED] wrote: Digital ICE from Applied Science Fiction should not soften the image. Jack, are you talking about the current version of ICE, or the version implemented on the Nikon LS30, LS2000 and Minolta Scan Elite? If you're talking about the new version we'll have to take your word for it because I don't think anyone outside of ASF or Nikon has seen it in action. If you're talking about the version on the scanners I mentioned above, it most definitely DOES soften the image, at least on the Nikons - I haven't tried the Minolta. I have an LS30, and I have used ICE enough to be able to state that for a fact. Rob
RE: filmscanners: GEM, ROC compared to Vuescan
When you do a scan using the Nikon interface do you have the "CleanImage" (Nikon's old name for Digital ICE) in the "Mode: On (Sharpen)" or "Mode: On (Normal)"? The mode Nikon calls "Mode: On (Normal)" actually blurs the image slightly. You should run in "Mode: On (Sharpen)". This should solve your problem with the sharpness. The grain shouldn't affect function of Digital ICE. Jack Phipps Applied Science Fiction -Original Message- From: Rob Geraghty [mailto:[EMAIL PROTECTED]] "John Matturri" [EMAIL PROTECTED] wrote: Related: why in general does Ice-based dust removal soften the image. The purpose of the IR channel is to identify the particular spots that are dust / scratches. Shouldn't the software only affect those areas leaving the rest of the image alone and sharp? Vuescan's lowest setting on dust removal doesn't seem to affect image sharpness. ICE on my LS30 *does* affect the whole image. I suspect that ICE "sees" the grain of the film (or rather the "apparent" grain :) as some sort of defect and tries to correct it. As I mentioned elsewhere, we'll have to wait and see if ICE3 has fixed this problem. Rob
filmscanners: GEM, ROC compared to Vuescan
Has anyone that has a scanner with GEM and ROC compared their results with Ed Hamrick's recently added features in Vuescan that reduce grain and restore color? How do they compare? _ Get your FREE download of MSN Explorer at http://explorer.msn.com
RE: filmscanners: GEM, ROC compared to Vuescan
Henry writes ... Has anyone that has a scanner with GEM and ROC compared their results with Ed Hamrick's recently added features in Vuescan that reduce grain and restore color? How do they compare? I agree we need someone to make a ^direct^ comparison. However, I was initially impressed with GEM and its ability to soften grain while keeping an edge intact. My 1st experience with Ed's "scrub" was disappointing with respect to edges ... it wasn't like a "gaussian blur", but more like a "median" filter. shAf :o)
RE: filmscanners: GEM, ROC compared to Vuescan
From: "shAf" [EMAIL PROTECTED] My 1st experience with Ed's "scrub" was disappointing with respect to edges ... it wasn't like a "gaussian blur", but more like a "median" filter. I agree. It seems to me that the grain reduction does soften the image. I showed that on a webpage that I sent to this list a couple of weeks ago. Here it is again: http://www.geocities.com/hr1066/vuescan66.htm I still wish that Ed would make the grain reduction a separate control so that it wasn't combined with the cleaning options. On Feb. 5 Ed said: "Except that VueScan's dust removal works better than Digital ICE, VueScan's "Restore colors" works better than Digital ROC, and VueScan's Clean function works better than Digital GEM. I'm biased of course, and people should come to their own conclusions, but these assertions are simple things to verify by simply trying them." The original message can be seen here: http://phi.res.cse.dmu.ac.uk/Filmscan/2001/Feb/0277.html Ed also wants feedback on this. I'm hoping that someone with a scanner that has GEM and ROC can tell us how they compare to Vuescan. Until the new Nikon scanners are available that might mean that the only scanner with GEM and ROC is the Minolta Scan Multi II. Is that right? _ Get your FREE download of MSN Explorer at http://explorer.msn.com
