Just some stuff about photography


How sharp is your scan ?

in Scanning , Wednesday, May 03, 2017

The object of this review is perhaps not the most enthralling I’ve ever covered. It is, to be exact, an exposed piece of film, a piece of positive “slide” Agfa film in fact. It isn’t even in colour. It is in fact a Resolution Target (USAF 1951), produced by LaserSoft Imaging. You can read all about it here.

There was a special off on these a few weeks back - they quite often happen - and so it was quite a handy retail therapy opportunity. But apart from feeding the inner consumer, you may well ask “what is the point” ?

Resolution target

Resolution Target (USAF 1951)

Well, there are two points. The first is perhaps the most initially compelling, as it gives you a standard reference target with which, following the detailed instructions, you can determine the actual optical resolution of your scanner. This then gives a warm fuzzy feeling and bragging rights over all your scanner owning friends (or of course it might leave you feeling extremely depressed).

Unless you own hundreds of scanners, that’s pretty much the end of it. I own two, so I was kept happily entertained for over 5 minutes!  But, actually, it isn’t the end of it all, because as the instructions explain, there is something else you can use this target for, which turns out to be very enlightening.

The second use is to experiment with different scanner sharpening settings to determine what the optimum settings are. This was quite an eye-opener. I’d always tended to leave sharpening to Photoshop, and some mumbo-jumbo laden plug-in, as I’m too lazy or stupid to really get the details of all this radius, strength and masking stuff. Using the target in conjunction with Silverfast’s USM tool allowed me to quickly determine the best settings for restoring sharpness lost in the scanning process, without overdoing things. It also revealed that the adaptive “Auto Sharpness” mode in USM actually does a pretty good job.  By way of illustration, if I used USM, I tended (for no really good reason) to keep Power in a range of around 100-150. Auto Sharpness, for a full resolution positive 6x9 MF scan set Power at 300 - it goes up to 500. Zooming in at 1:1, it was clear that 100 was too weak, and 300 was not introducing any artefacts. I don’t know yet if this means I’ll switch to Silverfast USM for capture sharpening, but this experience certainly boosts my confidence in it, and it would be quite a time saver over doing it in Photoshop.

So, if you are interested, one, in seeing how much resolution your scanner actually delivers, and two, making the most of that resolution, this Silverfast Resolution Target (USAF 1951) is a good investment. I’m sure it would be equally useful for Vuescan users, there is nothing that ties it to Silverfast software.

Oh, and my results ? 4096 dpi real optical resolution for the Plustek Opticfilm 120, which while less than the advertised 5300 dpi, is pretty much as good as it gets for desktop film scanners. The revered Nikon Coolscan 8000 / 9000 reputedly do hit their advertised 4000 dpi. And apparently, the Opticfilm 120 could do even better with some way of fine tuning focus.  Anyway, 4096 dpi is quite enough for 35mm and 120 film, and this test confirms what I could see subjectively, comparing with scans from my old Minolta Dual Scan Multi Pro.

My Canon 9000F only managed 2048 dpi, less than half the advertised 4800, but that’s pretty much what I expected.

Posted in category "Scanning" on Wednesday, May 03, 2017 at 10:34 PM

The State of Film Scanning

in Scanning , Saturday, November 28, 2015

Prompted by a number of queries both in comments here and more frequently by email, and also by a very well written article on the Plustek OpticFilm 120 scanner, I have decided to write up my feeling on the current state of film scanning.

Xpan norway1506 2 09

First up, an appeal: people, please manage your expectations. Seriously. Film scanning is now right on the edge of a niche within a niche market - film photography - which is either small but stable, or still dwindling, depending on who you believe, and if you consider “Lomography” to be a genuine component of this market (I don’t). Film scanning requires a combination of a complex opto-electronic-mechanical device together with high specialised and complex software. And customers - well, the few that exist - are demanding a perfect solution at an increasingly low cost. Frankly it’s astonishing that any enterprises still find a viable business model in this space. And then when you narrow it down even more to medium format film scanning, it gets crazy.

Please correct my if I’m wrong, but I believe there are three current manufacturers of MF film scanners: Plustek, Reflecta, and Hasselblad. The Reflecta scanners are badge-engineered versions of a white-label device built in Taiwan and sold under several labels, including Pacific Imaging. The Hasselblad scanners are so mind-blowingly expensive that it is hard to imagine them having any customers other than major media companies or very well-funded public institutions. Which leaves Plustek. I don’t really know a lot about Plustek the company, but my impression is of a small but enthusiastic company in a very small pool, not unlike Cosina with rangefinder cameras - oh, wait.

Fifteen years ago the story was very, very different. There were at least 5 manufacturers active in desktop MF film scanners alone (Nikon, Minolta, Polaroid, Microtek, Imacon), with new models and innovations appearing regularly. Add to this several others producing 35mm models only (including Canon and Plustek), and you had a healthy market at a range of price points. There are of course still secondhand models to be found, but the prices tend to be insane, there is mostly no servicing or repair available, and people tend to hang on to scanners that work properly. As said before, these are complex, delicate devices, subject to considerable wear, and don’t last for ever. And drivers compatible with latest operating systems are unlikely to be found. So realistically, you need to rule out the second hand market if you’re planning on staying in film scanning for the foreseeable future. Note, there is another path, which is to buy a second-hand drum scanner. These formerly insanely expensive devices, found only in Pre-Press shops, can now be acquired for something close to a pittance. But wait - they are very complex to install, maintain, and operate, requiring specialist knowledge which is hard to find even on the internet. They are very large, very heavy, and very cumbersome. And as for software - usually dongle protected through something like Apple ADB - forget about compatibility with anything released after around 2005. At the latest. Some brave souls offer bespoke scanning services using these monsters. If you want the ultimate quality, because they really are very, very good, you could go in that direction.

Note though, Tim Parkin, who runs the above linked drum-scanning service, reviewed the Plustek OpticFilm 120 (paywalled), and had this to say (I’m quoting very selectively here): “What’s quite surprising is how good the Opticfilm 120 is compared with the Howtek (4500 Drum Scanner)”, and “The Opticfilm 120 is undoubtedly a very good scanner and if the film holder and focus issues can be addressed this should give results that are dramatically better than the Epson V750 and potentially on a par with the Nikon 9000”. This, with the caveats that he modified the film holder to use two glass inserts to hold the film flat, and that he stated he would not buy an OF 120 with the existing film holders.

This is a real shame, because Plustek clearly put a huge amount of thought and resources into the film holders, which are the best I’ve ever seen, better even than Minolta’s, and provided no fewer than 7 different types with the scanner. It is quite baffling why they do not provide, or at least sell, glass film holders, as based on Tim’s tests, this would elevate the scanner from “best of a poor bunch” to “best desktop film scanner ever”. It may be possible to hack together a holder or two, buying spares and anti-Newton glass inserts, but I’m really not very good at that stuff, not to mention lazy and impatient. Oh, and ugly. Finally, another caveat on the OF 120 is that your chances of getting a glitch-free copy seem to hover around 50%. My first one had a stuck pixel or two in the Infrared channel, and went back for servicing (a total hassle in Switzerland, where Plustek has no official distribution, which is not unusual for small companies, as we’re a small country and not in the EU - but the retailer, Heiniger AG, was very helpful, and eventually I got a brand new scanner). At least we see there the benefits of a warranty and active production.

So, that’s the hardware. The Plustek OpticFilm 120 is, in my opinion (based on 20 years of film scanning), the least-worst choice for a new desktop film scanner. Note, I’ve never tried the Reflecta MF–5000, but the German film scanner info site prefers it over the OF 120. I think they got exasperated with the OF 120’s glitches. To me the Reflecta looks a little clumsy, especially the 35mm holder, and I’m not sure it can scan XPan panoramic format, a must for me. As for the Hasselblad X1/X5, even hiring one for half a day is horrendously expensive. And they don’t even have dust / scratch removal.

Now, the software. And again, we need to manage our expectations here. I’m a Silverfast (by Lasersoft) user, and have been for a very long time. I’m under no illusions that Silverfast is perfect - it is software, after all - but I do believe it delivers the best combination of scan quality and workflow efficiency. Note, you do need to consider the software/hardware combination together. It is possible that for other combinations, you’ll get different results. However, Lasersoft are closely involved with the initial and ongoing development of the OpticFilm 120, to the extent that it has a Silverfast badge on the front, and the shipping box is dominated by Silverfast PR. So, I expect that if Silverfast is going to be optimal for any scanner, it’s this one.

Silverfast was stuck on Version 6.x for many years. It had grown into a bit of a monster, with newer features seemingly bolted on at random, and it was crippled, on the Mac, by being based on PowerPC APIs. It ran under Rosetta, but when Apple dropped that as fast as they could (typical a***h*le behaviour from Cupertino), we were stuck. It took Lasersoft a long time to respond, in the light of the parallel dramatic downturn in demand for scanning solutions, but eventually they delivered the completely reengineered Version 8. Well, v8 is still idiosyncratic - it wouldn’t be Silverfast otherwise - but in my opinion, the current v8.5 is their best ever.

A lot of people complain about Silverfast. Now, I’ll admit to long-term user bias, but I’m not fan boy. I’ve been using Apple kit since 1990, and I’m certainly no Apple apologist. I also have professional UX/UI Design experience, and I’m neither known for my willingness to compromise, nor to suffer fools gladly. But given all this, and agreeing that Silverfast has a bit of a learning curve, I cannot understand why people are so critical of it. It makes a very complicated set of tasks, i.e reliably capturing an image imprinted as a positive or negative on a physical transparency, into an optimal, accurate, colour-managed digital copy quite easy. It’s colour editing tools are far superior to Photoshop, when considered specifically for film scanning, although probably the subset of users which can understand and benefit from this is also getting smaller. Yes, it requires some learning time. So does Photoshop. Yes, the documentation for v8 is crap. So is Photoshop CC’s. It’s an industry trend, you don’t get manuals any more (but with a small outlay, you can buy Mark Segal’s excellent guide). Lasersoft customer support is included in the price. It’s a bit haphazard, and trying to find the link to submit a question through the web page is a stroke-inducing experience, but if instead you use the “Request Online Support” feature in the application, you get a very quick response. The heavily moderated forum is a marketing disaster area in my opinion, and it is really frustrating to see what, frankly, seems to be team of dedicated, approachable and responsive people making such an unholy mess of promoting themselves. But, baseline, Silverfast works, and works well, and if you invest a little time into understanding it’s more advanced features, you might well be surprised at how much it can do.

It would not be fair to not mention the alternatives to Silverfast. Well, realistically, there’s only one: VueScan. Yes, Reflecta has its proprietary CyberView (not terribly good apparently), and there are long lost options like Binuscan (a seriously weird product). But now there is only VueScan. I’ve had a few public run-ins with Vuescan here, and indeed one of those articles is, to my astonishment, the most visited of all my posts. For some scanners, VueScan is the only option under modern operating systems, and it is certainly the case that you can obtain equally good results with Vuescan in your workflow as with Silverfast, but in my opinion, and experience, it’s more cumbersome, much more hit & miss, and requires a lot more post-processing. But, doubtless, if you’ve used VueScan for many years you’ll be comfortable with it, and you’ll find Silverfast weird. VueScan costs less money than Silverfast, that’s an undeniably fact. Allegedly VueScan also has good customer support. This is debatable - if VueScan’s sole owner, designer, programmer, marketer and support guy, Ed Hamrick, takes a dislike to you, you’re screwed, despite being happy to take your money. Unfortunately Ed is a little cranky. Clearly he is also to commended for his Herculean efforts, but then again, if tomorrow he decides to retire, where does that leave VueScan?

On top of all this, people who say VueScan is easier (or even easy) to use must be living on a different planet - the one the user interface was designed on. Working out how to tell it where you want to save a scan to is complicated enough, but say you want to color calibrate your scanner: in Silverfast, you do this: load the calibration slide, press IT8 calibrate. That’s it. In VueScan: in Input menu, set “Profile Scanner”. Load the calibration slide. Preview. On the internet, find your calibration slide reference data file. Download it. In Color Menu, under Scanner IT8 data, press the “@“ button (which in Vuescan apparently means “file path”), located the downloaded file. Use another “@“ button to tell it where to save the profile. Press Scan. Oh, that’s it! (probably). And all that profile stuff hides in the same menu where you can set colour balance using crude sliders. Ok, I’m not saying none of this stuff works. Clearly, it does. But people saying that VueScan is “easier to use” ? Give me a break - cheaper, yes. Wider scanner compatibility, yes. Good enough output, certainly. But easier to use - well, possibly it’s a more satisfying challenge for digital imaging geeks, but for people for whom scanning is necessary evil on the path to getting good prints, I’m very doubtful.

VueScan SF menus

The File Output menus in VueScan (left) and Silverfast (right).  I know which one I find more intuitive. Of course neither follow anything approach OS UI guidelines, for no good reason I can see.

Of course it is all highly subjective. The upside is that we have two very competent competing solutions in the market. The downside is that commercial realities impose compromises on both.

So, to the people who’ve been mailing me in the past, and those who will do so in the future, all I can tell you is make your own choices, but be grateful you still have choices to make. Buying a software license for Vuescan or Silverfast is not just giving you access to the software, but encouraging the respective makers to continue. Same for hardware - Plustek and Reflecta (and clones) still make new, up to date film scanners for 35mm and Medium Format. Buying one gets you new hardware, a warranty, and customer support. And the right to nag their marketing teams to do better. Given the state of the film photography market, I’d say things are in better shape than we might have expected a few years back.

Posted in category "Scanning" on Saturday, November 28, 2015 at 07:27 PM

Opticfilm 120 revisted

in Scanning , Wednesday, December 17, 2014

The day I got back from Colombia, just after I stumbled out of bed, very jet-lagged, the postman delivered a large box. Inside it was a replacement Plustek Opticfilm 120. Back in October I had discovered that the month-old original was producing a long streak in the infrared channel, contaminating the “iSRD” dust and scratch removal. Plustek tech support identified the cause as dust inside the optics, and said that the scanner needed to be returned for servicing. Unfortunately Plustek do not have formal distribution in Switzerland, so it had to go back to the dealer, under warranty.  It took a while, but this wasn’t too noticeable as I was away for over 3 weeks. And eventually I received a completely new scanner, directly from Taiwan.

Apart from this issue, I was satisfied enough with the first copy. But the second seems actually to be better. Looking at film grain, the focussing, which was ok with the old one, is a little better. And the iSRD now works fine, also, so far (touch wood) with no alignment problems (possibly also thanks to improvements in Silverfast v8.2). Multisampling still doesn’t work, due to slight alignment (or possibly blooming) issues. But in any case, I don’t see any improvement in density with slide film.  The single sampling DMax seems quite adequate in this case. Possibly it is more effective with negative film - I’ll try again one day.

Anyway, at least this justifies one key argument in favour of the Opticfilm 120 over an old Minolta or Nikon scanner - warranty, dealer and manufacturer support.

I’ve been able to quickly deal with a small backlog of film to scan - editing digital files from Colombia will have to wait.


Bachalpsee, Grindelwald, Switzerland

For now I haven’t got much planned, film photography-wise.  My stocks of E100G are almost exhausted.  Hopefully they will stretch until the first rolls of Ferrania’s new slide film turn up.

Posted in category "Scanning" on Wednesday, December 17, 2014 at 07:40 PM

The Plustek OpticFilm 120 film scanner

in Scanning , Monday, September 22, 2014

Following major problems with my Minolta Dimage Scan MultiPro scanner over the previous months, I recently decided to switch to a new Plustek OpticFilm 120 medium format film scanner (or “OF120” for short). My principal motivation for this is to scan film, not to make tests, but since several people have already asked, and there is very little information on the web about the OF120 which does not quickly degenerate into foam-at-the-mouth ranting, I’ve decided to dedicate a little time to provide some comparisons between the two scanners.

I recommend you first read these two in-depth reviews by qualified experts.

Mike Pasini
Tim Parkin

My own report will be entirely subjective.

My primary “use case” for the OF120 is scanning Xpan format film (nominally 65 x 24mm), and the vast majority of that is reversal film, principally Ektachrome E100G, Provia 400X, Provia 100F and Velvia 100F. I have some 6x6 and 6x7 120 reversal film (and a little 612 if we consider the Belair), but currently I do not shoot medium format. I do not shoot much negative film, although I’ve just shot 5 rolls of 35mm Portra 400, 2 in the Xpan and 3 in my Minox 35ML. They’re not yet processed. I’ve also used Portra and Ektar in the past, but generally I’ve preferred reversal film. For black & white you’ve come to the wrong place: I’ve only ever shot C41 B&W, and small amounts of Agfa Scala. If you want to know how well Tri-X or HP5 or whatever scans, you’ll have to send me a sample (but please ASK FIRST, in the comments below: I’ve got a day job too).

A few other ground rules:

  • I use Silverfast. I might try Vuescan just out of curiosity at some point, but I won’t put any time into resolving any issues, if there are any. Ed Hamrick’s antipathy towards the OF120 doesn’t encourage me much. He probably had a fit when he saw the Silverfast logo on the front :-)
  • In “productive work”, I use the Silverfast Archive solution. I know some people don’t see the point. Fine, variety is the spice of life, and I have no wish to either defend or evangelise my personal choices.

So, for me it boils down to this:

  • Can the OF120 match, or come acceptably close, to the scan performance of the MultiPro on XPan slides ?
  • Can it do this reliably, with crashing, and with requiring constant nursing ?
  • Is it reasonably fast ?

If you want to stop reading now, the answers so far are, in summary, yes, yes, and yes.

1. The Plustek Opticfilm scanner

First impressions are good. The scanner arrives in a large box, well protected. Half of the box real estate is given over to Silverfast marketing, which is some indication of the co-operation between the two companies. Inside the box is the scanner, a 6x7 Silverfast IT8 target, a Plustek DVD, a Silverfast 8 DVD, two skimpy getting started manuals, and seven boxes containing the film holders. There are film holders for mounted 35mm, 35mm filmstrip (with two slots for multiformat 35mm such as XPan), 6x6, 6x7, 6x8, 6x9 and 6x12 medium format. All are glassless. All except for the mounted 35mm are adjustable, I assume in order to account for the different spacing between frames on various MF cameras. The closing mechanism is via two magnetic clasps. Loading film is very easy, and all in all these are by far the most impressive, best engineered scanner film holders I’ve ever seen. Film appears to be held tight and is well tensioned. Whether or not this is good enough for curled film remains to be evaluated. Note, I can’t see any technical reason why a 6x17 holder could not be added. You could load 6x17 into the 6x12 holder, but there’s no obvious way to tell Silverfast to scan it. Although there is a potential workaround, which I’ll try out. There are no technical specifications provided whatsoever. Not even the usual admonitions not to use the equipment in the shower or use it to fry eggs. Or indeed the mandatory CE notices…

The design of the scanner itself is quite minimalist. It’s larger than the Minolta, but about the same weight. It features small Silverfast and iSRD badges on the front. Mechanically it seems pretty sound.

IMG 1116

Black is the new Beige

2. Setup

Setting up the OF120 on a Mac is basically a matter of plugging it in. The Plustek installation disk has an installer which appears to do something, but I have no idea what, as the driver is apparently included in Silverfast. It would be nice if Plustek had included a slightly more detailed manual on the DVD, but they didn’t. Anyway, there’s not actually a lot to discuss. Next, Silverfast is installed. The version on my disk is 8.0.1r18, which is seriously out of date. The current version is 8.0.1r54, and it’s a very good idea to update this before going any further.

The first thing I then did was to unwrap the gorgeous IT8 target and go about calibrating the scanner. Then comes the first stumble: on first run, Silverfast defaults to the 35mm holder. It does not recognise the inserted holder as it does for the Minolta.  Therefore, if, as a seasoned Silverfast user, you just pile in and press Auto IT8, things are not going to go according to plan. After a brief movement of puzzlement, I selected Image > ScanMode > Tranp. 6x7.  From then onwards, Silverfast performs its favourite party trick with the usual aplomb, and hey, presto, you have a colour-profiled scanner.

The OF120 is quieter in operation than the Minolta. The ticking of the Minolta stepping motor is quite absent. At the start of the scan, an abrupt “clunk” issues from the OF120, and then in the main it just whirs fairly peacefully.

3. Results

My first trial (I’m trying to avoid the term “test” here) involved my number one priority, XPan scanning. A pre-sales query to Plustek tech support in Europe produced a detailed reply explaining how to do this: the film size is not explicitly supported, so use the 35mm filmstrip holder, then tell Silverfast it’s the 6x9 holder. As a bonus, you can actually load two strips in, and batch scan them. As a downside, the film needs to be cut into 2-frame strips, which is a touch unfortunate as in recent years I’ve been using 3-frame strips in the Minolta. Anyway, the film slides snugly into place (none of the fiddling around you need to do with the Minolta holders) and the holder snaps shut with a reassuring clunk.

I’ve read scare stories about scanning speed. Well, I didn’t sit here with a stopwatch, but I’m extremely familiar with how long the MultiPro takes to scan an XPan frame, and the OF120 is in the same ballpark. I might even dare to say it’s a bit quicker. I’ve tried various combinations of iSRD (infra-red dust removal, the OF120’s alternative to DigitalICE and multi-exposure, and scanned at 5300dpi, close to the the Minolta’s native 4800dpi for the centre area. In each case the more complex combinations slow things down, but to the same degree as for the Minolta. Interestingly, the hardware scan times don’t vary much, unlike with the Minolta, but Silverfast sits there processing for longer.

So, let’s get to the important bit: the results. Here’s my test frame, an XPan shot taken earlier this year, and scanned at 4800dpi, 16 bits per channel, with 4x Multisampling and D-ICE on the Minolta.

Xpantest venice straight

I scanned the same frame on the OF120, at 5300dpi, 16 bits per channel, with iSRD (no Multiexposure. I’ll get to that later).

The nice thing about this frame is that it features the vital Brick Wall need to test stuff. Sadly no cats are currently present, which I know will be seen as a major failing in my methodology. It will just have to do.  Let’s look at a couple of details of the frame. In both cases there is no sharpening involved. First of all, the plaque on the bridge:

Minolta venice crop1

Minolta at 100%

Plustek venice crop1

OF120 at 100%

Not a lot in it really, is there ? Note, for the Minolta there was some contrast added in Silverfast. For the OF120 I left it flat.

Now let’s look at a shadow area, under the bridge to the left:

Minolta venice crop2

Minolta at 50%

Plustek venice crop2

OF120 at 50%

Again, not much in it. The shadow detail is about the same, despite the fact that the Minolta scan benefitted from multi-sampling and the OF120 did not. The two scans have slightly different black points.

On this example, I’d say that the OpticFilm 120 is a close match to the MultiPro. While I might be just a little disappointed that it isn’t better, in terms of resolution the lemon is squeezed dry anyway at this point. There isn’t any more detail to extract from 35mm film.

Here’s a second example, from a very different part of the world.

Xpantest2 srd

This time I’ve taking a sample from the middle of the frame:

Xpan antarctica minolta detail

Minolta at 100%, 4800dpi, 8x multi-sampling, 48bit HDR scan, manual focus, no sharpening

Xpantest2 srd detail

OF120 at 100%, 5300dpi, 48bit scan, no sharpening

This time around the Minolta undeniably has the edge, although do note that printing this file at 360dpi resolution would give an output size of 97 x 38cm. The largest XPan print I’ve made is 80 x 30cm. Nevertheless, my feeling is the Opticfilm 120 can, in theory, do a little better than this.

I have done some initial fiddling around trying to insert AN glass over the film, to improve flatness. Results so far were either “no change” or “worse”. But my methodology was hardly robust.

4. Some issues

I’ve spent far too long on this write-up already, but there are some issues I’ve come across which merit a brief mention.

Multiexposure alignment

This is currently the biggest problem I have. Multiexposure, which is required according to Plustek to reach their quoted DMax of 4.1, doesn’t work. The two exposures do not align properly. I have reported this to Lasersoft, and received a quick reply from Arne Ketelhohn, who states “In case there is a large offset in the infrared or ME images you can increase the detection offsets in the preferences’ special tab. The ME setting would have to be set in the scansoftware”. Ok, but while there is indeed a preference to set the maximum correction range from ME alignment, there’s no way I can see to apply the offset itself, and believe me I’ve looked. Unfortunately Arne did not reply to my follow up. Possibly Lasersoft were/are busy with Photokina, but I’m not letting them off the hook. They need to fix this. Note that ME alignment was always a bit haphazard with the Multipro, which is why I preferred to use the slower but equally effective and 100% reliable multi-sampling.
UPDATE, 23-09-2014I did get a reply from Arne Ketelhohn, who explained that there is no manual adjustment for ME, you just need to try different values of offset to get it calibrated.  I spent about an hour on this, I eventually found that a value of 6 gave a nearly flawless result, however looking closely a few “ghosts” could be found.

iSRD alignment

Same story as above, although strangely it has so far only cropped up in “64bit HDRi” files. In this case, there is a manual offset slider (two in fact, horizontal and vertical) and these can be used in Silverfast HDR to fine-tune alignment. iSRD seems to be pretty much on a par with DigitalICE, with the plus point that it is tuneable. Or maybe that’s a drawback…

Frame detection

A lot of people have reported problems with frame detection. I had one issue with the software not scanning all the way to the edge of the holder. A reset of both software and scanner fixed it. It only happened once, no big deal.

Fuji “pepper grain”

And a real blast from the past here: the old Fuji Provia pepper grain issue pops up again. Not really a problem with the scanner, and probably only a pixel-peeping issue. However I now wish I’d kept one of my Scanhancers to see if it works with the Opticfilm. I guess probably not, but pepper grain dissolving was one its benefits.

iSRD edge artefacts

In some extreme cases iSRD gives stair-step artifacting at high-contrast edges. I’ve only found this once so far on a 6x6 Velvia slide. Note that DigitalICE suffers from similar issues, which again the Scanhancer resolves. Since in Silverfast, iSRD can be fine-tuned and masked, there’s probably a workaround to this, it just requires a little time.
CORRECTION, 23-09-2014The stair-steps were Operator Error. I had the detection slider set far too high. A “normal” value is about 5.

5. Initial conclusions

It is very important to bear in mind that my conclusions are heavily biased by my personal needs and quality thresholds. I also just want things to work, I have no desire at all to spend all my scant free time “testing”, debugging and resolving issues. I just want to be able to scan my films at a high enough quality to frame them, show them around, and take them to potential exhibitors as samples of my work. The largest I print is A2 size on an Epson 3800 (on on A3 height panoramic sheets). Much as I loved my MultiPro, it was getting far too frustrating coaxing it along all the time.

In a nutshell, so far the OpticFilm 120 meets and perhaps slightly exceeds my expectations. So far it, and Silverfast 8, have been stable and reliable.

If today I had the choice between on OpticFilm 120 and a new Sony DimageScan MultiPro designed for OS X 10.9 with full tech support, I’d buy the MultiPro.  But given the choice between fading MultiPros and Coolscans, requiring maintaining second computers with old operating systems, and increasing amounts of attention, and a new Plustek OpticFilm 120 designed for OS X 10.9 with full tech support, I took a deep breath and went with Plustek. So far I don’t regret that decision.

As witnessed by the badges on the front, the OF120 was born joined at the hip with Silverfast Ai Studio 8. In my opinion this is a good thing, but as with every other first world problem these days, there are astonishingly rabid opinions expressed about this on the internet, many of which seem to emanate from uninformed, rigid and closed minds. Although personally I like Silverfast, and indeed Lasersoft, I’m far from blind to their shortcomings, and I’ll be making a considerable nuisance of myself getting defects recognised and resolved.  Lasersoft really, but really needs to get with the program and engage constructively with its remaining user base. Satisfied Silverfast users are Lasersoft’s biggest asset, it really is remarkable that they do not appear to understand this.

Ok. Enough already. Any questions ?

Posted in category "Scanning" on Monday, September 22, 2014 at 10:07 PM

Carry On Scanning

in Scanning , Sunday, August 31, 2014

Since my last post mourning the apparent demise of my faithful Minolta film scanner, I have tried every kind of arcane trick know to the Internet, and a few more besides, to bring it back to life. It is sometimes possible to get it to revive, but there’s no pattern to it. I managed to extract a full-blown, medium format 16x sampling megascan from it, too, but soon after it relapsed. I have to face facts, I’m wasting far too much precious time on this.

One reason why it has been so much the focus of my attention - apart from a 15 year film archive, which can always benefit from my improving scanning skills - is my current project to refine a set of Antarctic landscape panoramas.  I’m trying to get the colour profile exactly as it should be, which to my way of seeing needs to be delicate, slightly subdued, but still allowing the often astonishing colour to speak.  But not the overblown, digital look that plagues so much photography (Adobe Lightroom default profiles have to take of the blame for this). Of course, photographing on reversal film means that I’ve pretty much defined the look before it gets anywhere near a computer, but there are still opportunities and decisions to be made in the scanning and post-processing stages.  The ideal is to transfer what I see on the light table on the screen, and then to print, but that’s very hard to achieve, especially without a drum scanner. And when I’m engaged in a long stretch of batch scanning, sometimes my initial post-processing attempts are not ideal. For example:

Xpan antarctica06 06 old

I’m not sure what I was thinking of here. The contrast is too strong, and the delicacy of the colours in the ice is lost. I’ve also pushed the sky and sea too much towards neutral.

The revised version is much closer to the Ektachrome, although with less density. In the processing, “less is more” certainly applied. Note, in both cases, reducing down to web sized JPGs is introducing some exaggerated tone transitions, especially in the sky.

Xpan antarctica06 06

Fortunately, using the Silverfast archive workflow I can go back and re-work the post processing without needing to do new scans. Unfortunately, for most of my Antarctic scans I used the Scanhancer to try to eke out the last bit of pixel-peeping quality, and this has not worked out to well. The coupled increase in exposure times seems to have greatly exaggerated shadow noise, possibly due to an ageing scanner CCD, and a few near invisible scratches on the Scanhancer itself have resulted in bands of shadowing on the scans, which was not immediately noticeable, but which are almost impossible to fix.

So going back to the scanner quandary, unless I decide to give up, I have three choices: try to get the Minolta fixed, which seems unlikely, track down a good, working Minolta DSMP or Nikon Coolscan 9000 at a sensible price, or take a chance on a Plustek OpticFilm 120. Although the inter webs are full of whining about the Plustek, two reviewers who actually have some track record have been less negative: Mike Pasini (“we achieved our finest scans of the test images we’ve ever managed. But it wasn’t easy.”), and particularly, Tim Parkin, who is something of a scanner guru (“the OpticFilm is definitely has the potential to be a great scanner and I can only recommend if you have the wherewithall to play around with creating a custom film holder”). Another strong argument is that the OpticFilm is currently in production and support by a company for which scanning is a major business activity. Well, I’m going to dither for a little longer, but I’m leaning towards the OpticFilm. Especially as it supports 6x12 film format and alledgedly could be persuaded to scan 6x17.

Posted in category "Scanning" on Sunday, August 31, 2014 at 06:33 PM