Lasersoft Imaging released Silverfast 8 towards the end of August. Unfortunately, they don’t yet support my main scanner, although they do support my CanoScan 9000F, but they have just released a public Beta of Silverfast 8 HDR. Since most of my time with Silverfast 6.6 is spent using HDR, this was welcome news.

Since it has come during a bit of a lull in both photography and especially scanning, I haven’t really had much reason to try it, but yesterday evening I thought I’d give it a go. Note, this article is written under the influence of a combined throat infection and heavy cold.

The big thing about Silverfast 8 is the user interface redesign, but that’s not the only point. However, it really dominates the update, so here it is.

The Silverfast 8 HDR Studio user interface

and here it was:

The Silverfast 6 HDR Studio user interface

Silverfast 8 introduces a modern, compact, unified user interface which, although remaining a little idiosyncratic, is a huge improvement.

I haven’t run anything approaching a full session, so I’ll just list a few early impressions. These are taken from running on MacOS X 10.6.8.

Positives:

- hugely improved UI. Massive step forward
- installs and runs following normal guidelines, including access to preference panels, etc. Uses standard OS toolbar.
- detachable tool panel, so you can “roll your own” UI to some extent
- ability to turn various edits on and off in preview (like Aperture or Lightroom)
- ability to run Silverfast 8 and Silverfast 8 HDR concurrently - I think. I’m not 100% sure as my trial of Silverfast 8 for CanoScan 9000F has expired, but I can open both launch screens at the same time. I can also run SF 8 HDR and SF 6 HDR (or AI Studio) at the same time.

Negatives (remembering that this is a Beta):

- allows quit without warning to save edited images
- the colour cast slider seems to have vanished. Now the level is set in Preferences only

Neutral:

- the image manager, Silverfast VLT, which works as a front end to Silverfast HDR 6.6, is gone.  This is not necessarily a bad thing as it is somewhat buggy and has some very poor design choices. However as a way of building up Job Manager lists is was pretty good. Maybe it will return.
- seems stable. No crashes so far.

Generally all the tools remain the same, including the superlative colour correction tools, but they’re easier to use and understand.

All in all it looks encouraging. Let’s just hope Lasersoft come up with a pricelist which takes into account that it’s not 2001 anymore, otherwise selling a product like this into a dwindling market is going to be pretty challenging.

With Mac OS X Lion and new MacBook Airs released today after weeks of speculation, today was a good day to bury bad news in the tech world.  So Lasersoft, God bless ‘em, chose today to announce, at long, long last Silverfast 8, the next major iteration of the venerable Silverfast: 

SilverFast, the most popular scanner software in the world, is released as brand-new version SilverFast 8. After many successful years, SilverFast will be available completely renewed towards the end of August for the most important scanners of all major manufacturers.

I guess Silverfast 7 was dropped in the Baltic Sea or something, since we’re jumping straight from 6.x to 8, or maybe it’s just in recognition that nobody could possibly make us wait this long for a new version.

So, we’re promised such joys as a new user interface (let’s hope they didn’t hire PhaseOne’s designer), multitasking (gasp) and much goodness. The bit that leaves me a bit worried is “for the most important scanners of all major manufacturers”.  The place of the Minolta Dimage MultiScan Pro in that august assembly must be less than guaranteed.

Well, I for one am looking forward to this more than OS X Lion.

Just a quick update on my experience to date with the ScanScience kit.  I’ve now tried scanning negative film (35mm Kodak Ektar 100), and the benefits are much clearer, although mainly due to the fluid mount keeping the film flat.

Here, the top scan is using the standard filmstrip holder, and the bottom one is using a fluid mount.  These 100% screen shots are from the edge of the film: the scans were manually focussed at the centre of the strip, and there the difference is minimal.

The colour differences are not significant - I didn’t take much notice of this during the scan, and just let Silverfast do its thing. You can’t see it here, but the fluid mount scan has slightly more detail and micro contrast. Nothing overwhelming though.

So the conclusion is in this case that it is well worth the trouble - but the advantages over a carefully prepared dry glass holder, for 35mm film, remain small.

Back in October, a blog post by Ctein on The Online Photographer first alerted me to existence of 3rd party fluid mounting kits for many scanners.  Specifically, he mentioned ScanScience‘s kits, based on their Lumina fluid.

I was intrigued enough to check this out, and eventually ordered a kit usable on my Minolta Dimage Scan MultiPro from ScanScience. Since they are in Canada, and the fluid cannot be carried by air, it takes quite a while to arrive, and it’s also better to order a reasonably generous supply.  So there’s quite an element of flying blind here, but my feeling was, well Ctein knows what he’s talking about, what the hell.

Anyway, it arrived a few days ago, and I started playing around with it. Precision scanning is both fiddly and something of a black art. My initial impression of fluid mounting is that it certainly increases fiddlyness, and also adds several more ways in which you can screw up a scan ... or indeed a scanner, in the worst case scenario.

My initial attempts were not too successful, and I had to dismantle a 6x9 glassless holder to fit the fluid mount assembly as recommended. But after a few tries I started to get the hang of it. The main challenges are getting the right amount of fluid onto the various surfaces, and avoiding dust and dirt contamination.

The basic idea with fluid mounting is to keep the film flat, and to avoid optical path degradation which arises from various factors in dry scanning - I won’t repeat the explanations here, you can find them at ScanScience and various other resources.

The following images show the best result I’ve obtained so far. Comparing a couple of sections of a 4800dpi XPan scan of Ektachrome E100G, wet mounted and dry mounted, shows some advantage, at 100% magnification, for wet mounting.

At the top you can see the wet mount version on the left, dry mount on the right [CORRECTED!]. Colour differences are down to slightly different settings in Silverfast HDR - ignore these. The images have suffered a little in JPG compression, but looking at the rightmost telegraph post, and the background forest, you can see a touch more resolution. But it’s not exactly jaw-dropping.

The next example is a bit more convincing:

Here, the lower scan is wet mounted.  Certainly there is a touch more resolution here. Or maybe I focussed better… who knows ?

ScanScience claim a number of things, including:

- Better edge to edge sharpness: hmm. Probably, but with the “sandwich” mounting technique I’ve evolved over the years, and the relatively deep depth of field of the Minolta, I’m not - so far - seeing any benefit.

- Better contrast and detail: as shown above, yes, but we’re splitting hairs, to be honest.

- Extended dynamic range and saturation: Nope. Well, not for E6 slide film, at least. The Minolta covers the range of E100G and other low(er) contrast films quite happily, and has no issue at all with higher contrast.  However, this benefit may apply more to negative colour and black & white. I’ve yet to try this.

Hides dust & scratches: no. Absolutely not. If anything it makes things worse due to more places for dust to get in, and if dust gets into a fluid layer, it’s pretty much game over, time to remount.

So, in summary, I’m seeing minor improvements which don’t really appear to justify the cost and time.  But it’s not quite as simple as that. First of all, one thing is clear: to get any benefit at all, the image has be well exposed and sharp. Secondly, it seems that the benefits are more towards negative and larger formats.

There’s actually a lot of discussion out there on the interwebs about wet mount scanning, and by & large I’d say the overall impression is of mixed results. In particular, this discussion thread confirms my findings. Pity I didn’t do a bit more research…

At the moment I would tend more to reserving it only for “top picks”, but it’s early days yet.

Recently Lasersoft added support for Ektar 100 to their NegaFix tool for inverting negative scans. I’d been wanting to try out this film for a while now, and in fact I had a few rolls sitting ready. So I pulled my Ricoh GR-1v out of retirement and tried it out.

I scanned the film in the CanoScan 9000F, using SilverFast AI with SRDi dust and scratch removal activated, and just went for the default settings for Ektar 100.

The shot was quite tricky with mixed lighting, but Silverfast pretty much nailed it. The scan is very close to the 1-hour photolab print. Certainly there’s plenty more that could be done, but I was more interested in where the baseline was. Having tried to scan various other negative films last year, I wasn’t expecting much. But I was pleasantly surprised.

The resolution of Ektar 100 is quite astonishing, and if I were to switch to negative film, this would be the one.  However, I’m going to stick with slide film.  It has its own drawbacks, but none of the interpretation issues of negative film, or indeed the lack of immediacy.