Multi Pass Scanning?

[photogdave]

Anyone out there using multi pass scanners? I have the Minolta Scan Multi II, which is a great scanner for me because of the multiple formats it offers. However, I can't tell if doing multi pass scanning actually makes a difference. Does anyone have any hard and fast evidence in favor of or against multi pass scanning?

 

[jbf]

Honestly.... i saw little to no difference with the Nikon Coolscan 5000 ED that we have at school. Just takes longer. I mean obviously there is some difference for shadows... especially if you scan 16bit. But to be honest, it adds so much more time onto each scan that for me it is not worth it.

When I scan my negatives I usually do anywhere from 10-20 some odd rolls at a time. Sometimes more. So I just want a good scan in the quickest amount of time.

I get great tonal range out of the scans I do without all of the extra passes so I think there's no need.

 

[dmr]

Theoretically it should help to cancel out noise.

In reality, I really can't see much difference.

For a good presentation print I will scan at 4 samples, however. It won't hurt. 🙂

 

[Brennotdan]

I have seen test images somewhere online, where the shadow detail was very slightly improved with each pass. For me, it isn't worth even doubling the time for a scan. I just scan at the highest resolution possible at 16 bit.

edit: I have also done my own test with a three pass scan, expecting miracles and could not see enough of a difference at all for triple the scan time

 

[photogdave]

Interesting replies - thanks!
Any differences between B&W and color?

 

[Hephaestus]

I don't know how much shadow detail is increased but I have found that taking an extra pass or two can reduce the appearance of noise and scratches.

Ryan

 

[Chriscrawfordphoto]

I use the Nikon 8000ED and gave up multipass scanning soon after buying it 6 yrs ago. It didn't do a thing for black and white or color negatives. No difference at all. For color transparencies that are underexposed or that have a lot of very dark areas, it does reduce noise a little but the scan times are outrageous, so I only use it if I have scanned a film and decide it could use it...then I rescan. But the difference is not great.

 

[MikeL]

I'm with Christopher here, I haven't seen much difference. Scanning as a positive and inverting seems to recover shadow detail better than multipass when I screw up an exposure. I could be wrong, but not worth the time. I haven't had 16-bit make/save a photo for me either though.

 

[Brennotdan]

I like having the post processing latitude with a 16 bit image. I don't think it is necessary with a well exposed neg...but I do it anyway.

scanning as a positive ey? I've never tried that, I'll have to experiment.

 

[gregg]

The theory that multi-pass cancels more noise doesn't meet reality for me. Just the other day I ran my Minolta Dimage Scan Elite II at 1, 2, 4, 8 and 16 passes on a Fuji Neopan 400 shot that showed shadow detail over a shoulder to see the differences.

There may have been a MINOR difference in the shadow detail between 1 and 16 passes when examined at a 400% zoom, but not viewable in an 8x10 print or in a jpg output.

Here they are:

1 Pass
2 Pass
4 Pass
8 Pass
16 Pass

 

[RF-Addict]

I have the Nikon 9000 and I cannot see ANY differnce between one pass and multi-pass scanning. I keep thinking that there is something I'm doing wrong because every scan expert out there tells me that I should do multi scanning, but I just cannot see the difference. I scan mostly slides and I try to make sure that they are well exposed, which they mostly are - noise is not an issue for me at all.

 

[stevew]

Tried Multipass scanning.

Tried Multipass scanning.

On Nikon Coolscan V seemed maybe slightly less noise but also slightly less sharpness. Wasn't worth it for me.

 

[dmr]

If anybody's interested, here are four threads, all related, about scanning negatives as positives, complete with examples, histograms, a poll, circles and arrows and a paragraph on the back of each one, explaining what each one was, to be used as evidence ... 🙂

http://www.rangefinderforum.com/forums/showthread.php?t=18536
http://www.rangefinderforum.com/forums/showthread.php?t=18582
http://www.rangefinderforum.com/forums/showthread.php?t=18644
http://www.rangefinderforum.com/forums/showthread.php?t=18739

 

[amateriat]

dmr: That one works best over Thanksgiving, but I love you for the referernce anyway. 🙂

At any rate: the problem with you guys is two-fold: First, most of you have a handle on proper exposure and, where applicable, development; second, you own at least reasonably-good scanning gear. Under these circumstances, you're not likely to find much advantage in mulitple passes. I found it useful with a few terribly-dense Kodachromes I scanned for a client, plus some rather old Tri-X negs I had developed indifferently half a lifetime ago. With most anything else, it's largely unnecessary. But I'm glad it's there!


- Barrett

 

[dmr]

dmr: That one works best over Thanksgiving, but I love you for the referernce anyway. 🙂

I was wondering if anybody would get that! 🙂

Don't forget your shovels and rakes and implements of destruction! LOL! 🙂

 

[Bob Michaels]

27 color glossies with circles and arrows and a paragraph on the back seems to remind me of song I play every Thanksgiving in memoriam to an event on that day long ago.

Seriously, my Viet Nam era military induction was briefly delayed because of a conviction for littering. But the delay was enough time for me to re-enroll in college and get a deferment.

 

[Bob Michaels]

I tried multi pass scanning years ago and could see no real difference other than increased time.

I have always used Vuescan and find that the "long pass option" will dig out some shadow details. It seems to be sort of like HDR in how it works. But I use that very infrequently.

 

[Trius]

I have no experience with this, but I wonder ... for those who say there is no real advantage, are we talking about viewing at web resolution, or the result in a large print from a scan? (Frankly, I'd rather have an optical print than a digital/inkjet print from a scan, but that's another story.)

I actually see enough difference in Gregg's examples to make me think that for hard copy output, there could be an advantage. But a low web resolution, it's hard to make a final conclusion.

So for me the real question is: What is the final output?

 

[amateriat]

Trius: In my case, at least, I'm referring to results in the final print, as well as viewing high-resolution files evaluated on-screen. For we viewing, it'll make little difference except in extreme situations. The thing I keep having to remind myself of os that the last two film scanners I've owned (both Minoltas, my previous Dimage Scan Dual IV and my current Dimage Scan Elite 5400), while equipped for up to 16x multi-pass–actually single-pass multisampling, in their case–scanning, their sensors were sufficiently advanced as to rarely need that option invoked, even with the occasionally-blown exposure. You'd need a seriously dense neg or 'chrome, and, not to crow too proudly, there aren't too many of those in my archives. (Doesn't mean I never made duff shots; it just means I had the good sense to chuck 'em with reasonable dispatch.)


- Barrett

 

[mfogiel]

Multipass scanning only makes sense when scanning contrasty slides, it is of no practical use for negatives. From experience I've noted that 80% of the improvement can be had doing just a double pass.