photoblogography - Just some stuff about photography

Aperture gets the boot

never trust a hippy. Or an MBA

in Apple Aperture , Saturday, June 28, 2014

So, Apple has admitted, in some roundabout way, that Aperture is finished. It hasn’t been announced on their website, indeed, it is still promoted and up for sale. No, somehow the news has been leaked in a very uncharacteristic way.

Nobody should be surprised by this. If Apple is consistent about anything, it is in dropping products or complete product lines without any second though for its customer’s investments. But with Aperture, it’s worse. Much worse. With Aperture, it’s customers are not just financially invested, but also creatively invested. This is the drawback of both non-destructive editing and proprietary digits asset management: you need to be able to trust that the software supplier you depend on is committed to long-term support. It was always a risky bet with Apple, and now it has proven to be the case. Apple is focused purely on short-term gain in increasingly dumbed-down disposable consumer electronics, and is fundamentally an untrustworthy partner. It is telling that the company has made no formal statement on its website, and shown absolutely no concern at all for its customers’ plight, and instead offers the insulting idea that it’s a fair swap to lose years of work for some rubbish piece of iCrap gloss in “the Cloud”. And in some cases, those customer’s entire business model rested on trusting Apple as a reliable partner.

I’m pretty sure that within the Aperture team there are people who well realise just how badly customers have been let down, but with Apple’s corporate Iron Curtain firmly in place, we’ll never know. And for the rest, from Tim Cook and all his anonymous MBA cohorts, we’re collateral damage. Long term I suspect it will be their loss.

So what went wrong ? When Aperture hit the market in 2005 it was unexpected and revolutionary. It was also a massive resource hog, and expensive - $599 - and that didn’t help it gain early market share. People assumed it was a Photoshop rival, and perhaps even Steve Jobs did too, with his antipathy towards Adobe, but it wasn’t. It was something completely new, an application designed specifically for the needs of photographers in the digital age. Looking at it as a Photoshop rival obscured the real marvel of Aperture, its photo management and cataloging tools. There’s still nothing to beat it on that front, in fact as far as I know, nothing even comes close.  Then there was the non-modal UI, which some people had (and still do have) a really hard time getting their heads around.  Basically, with Aperture, you choose the context to work in (Project, Album, Book, Website, Light table, Print) and all editing and management tools are available at all times. This is quite the reverse to all other applications, including Lightroom, where the workflow is firmly object(photo)-centric.

Aperture 2

This is a key and unique feature in Aperture: here, as I lay out a selection of images on a light table, perhaps to plan a print series, I can tweak each photo’s settings in-situ as I work. The UI is completely non-modal. Were it not for the Apple Iron Curtain, whoever devised this would be celebrated in the design community. And no, it wasn’t Jonny bloody Ive. Or indeed St Steve.

So Apple had a fantastic application on their hands. But apart from high profile launch events, they essentially put no effort at all into explaining it, or marketing it. When Adobe brought out their rival application, Lightroom, which was certainly a major step forward from Photoshop for digital photo workflow, but much, much less imaginative and ambitious than Aperture, they rolled out the full force of their marketing tools, including getting legions of industry stars and opinion-formers on board, they kept up contact between engineers and users - hell, we even knew who the key engineers were - and they maintained open Beta programs for each new version. Aperture, crippled by general Apple arrogance towards customers, had no chance. The few opinion-formers they got on board seemed to be used solely as marketing mouthpieces, whereas Adobe avoided the whole control-freakery scene. As time went on, Lightroom got caught up in the Adobe bean counters insistence on yearly upgrade fees, and so started to acquire bloat and useless features, without much improvement to the core application. Also, anybody coming to Lightroom from Aperture cannot help but feel manacled by the step-by-step workflow, which reminds me of the 1990s UI “room” concepts championed by Kai Krause. Nothing like Aperture’s unconstrained creativity. Lightroom felt like an engineer’s idea of what photographers wanted. Aperture felt like a photographer’s invention which could maybe do with a touch more engineering input, especially at version 1.

But essentially Aperture was completely out of place in Apple’s product line-up. A deep, non-glitzy application that demanded, but rewarded, serious commitment on the behalf of its users. Certainly no iOS or AppStore fluff. Indeed, although the Iron Curtain lets out no whispers, I really wonder if Aperture’s genesis lies outside of Apple, much like Final Cut. Perhaps Aperture also was initially a Macromedia project, and therefore might even share some DNA with Lightroom. Pure speculation, and I guess we’ll never know. Or care.

So now what ?

Well, Aperture is still working, and Apple has committed to maintenance support for at least one further OS X iteration. But the end is irrevocable, and that means that any work expended from now on in Aperture is wasted. There’s much talk of migrating to Lightroom, or whatever, but let’s be clear: you can migrate your metadata - ratings, stars, keywords - but that’s it. You cannot migrate any develop settings, or your whole library structure, your projects, albums, smart albums, light tables, books, etc. You might be able to migrate your keyword hierarchies, which for power users is a big deal. Certainly you can migrate these to Media Pro. I have 51,225 photos in Aperture. If I were to set aside the time to recreate all the non-destructive edits in Lightroom, I might as well give up photography.

So what are the alternatives ? I’ve tried most of them: my mainstream history goes something like this: Olympus Studio -> Adobe Camera RAW 1.0 -> CaptureOne 3.6 -> Iridient Raw Developer -> Lightroom 1 -> Aperture 2 -> Aperture 3. Along the way I’ve tried out pretty much all other options available on the Mac At present I use Aperture 3 for everything, except for Sigma Merrill files which I develop in Iridient Developer (which has indeed recently become much closer integrated with Aperture).  Prior to Lightroom, indeed prior to digital, I used what was called iView MediaPro, and is now called PhaseOne MediaPro, to catalog and manage my library. I’ve carried on using it for scanned files and Sigma files alongside Aperture, and I still consider a great tool. Indeed, many years ago I speculated that a merger between CaptureOne and MediaPro would be a combination to beat. Eventually PhaseOne did acquire MediaPro, but frankly they haven’t done a lot with it.

To replace Aperture 3 we need to consider two aspects (at least). RAW development, and Digital Asset Management. There are only two real options which cover both parts. Lightroom, which is fully integrated, and CaptureOne/MediaPro, which is more like a bunch of bits flying roughly in the same direction. Although I have no particular axe to grind with Adobe, and indeed have used (and paid for) Photoshop since v2.0, InDesign since v1.0 (and Pagemaker before that, indeed before Adobe acquired it), and a whole host of other Adobe apps, I just don’t find Lightroom very inspiring. But I can’t deny that it does the job, and is probably the sensible choice. CaptureOne, on the other hand, seems to be more driven by photographers than marketing, and v7 has a management component which clearly inherits conceptually from MediaPro, although just how C1 and MediaPro are supposed to be “integrated” still puzzles me. CaptureOne went through a very bad patch after v3.x. Version 4 was very late, all-new and something of a disaster. But now at v7 it seems to have matured.

I guess in the coming weeks, I’ll try the latest versions of both on a small library and make my choice based on what I actually see, which is how I came to choose Aperture. Ill write more about this in the coming days / weeks, maybe.

But what about Apple? I’ve been an Apple customer for over 20 years. I bought my first Mac (a Powerbook Duo) on a university discount scheme. I’ve never been a fanboy, although I got close to it in the dark years of the 1990s, and I’ve always seen pros and cons to buying Apple. These days it’s more through inertia and have a considerable investment in software and peripherals that I stick with Apple. And on the whole, indeed it does “just work”. In Management Powerpoint Bullshit speak, companies talk of being in a stakeholder, partnership relationship with their customers. Apple are not. Apple, in 2014, see customers purely as cash machines. Their total wall of secrecy and refusal to engage - and it was not always so, not by a long way - makes them, as a company, amoral and totally untrustworthy. Unfortunately the whole industry is going that way. We’re a long, long way from the Revolution In The Valley.


Do you fake it ?

film, that is

in Film , Friday, November 25, 2011

The background current of film pushing against the digital torrent seems to be continuing unabated. An notable new twist is the increasing interest in, or at least marketing push, of film emulation software, of the likes of Alien Skin Exposure or DxO Filmpack. Personally I’m not that interested in faking it - I don’t see much value in disassociating the result from the process, and anyway I’m not that impressed with the results. I can understand the value to illustrators and publishers, in particular for some of the more extreme effects like aged 1962 Agfa consumer prints, but in general if you want it to look like Ektachrome, why not use Ektachrome ? It’s not that hard!

Michael Reichmann recently reviewed DxO Filmpack, and didn’t lose the opportunity to give film a bloody good kicking.

I respect Michael’s experience, although I have some reservations about the direction he’s been heading in since - apparently - money became no object. His photography seems very inconsistent these days, which is a pity. Ten years ago it could be inspirational. Now, despite his protests to the contrary, all he really seems to do is to test cameras, just with a limitless travel budget. Anyway, my point is that there are other photographers who I respect who seem to have a rather different take - from famous ones like Michael Kenna, to emerging stars like Bruce Percy, “alternative” web gurus like Kirk Tuck, Robert Boyer, and seemingly the entire readership of Great British Landscapes.

I could point to Bruce in particular as a clear example that Michael is just plain wrong. Using film - Velvia and Portra I believe - seems to have helped him to develop a very distinctive and personal style. Do his photos suffer from any of film’s perceived weaknesses ? I don’t think so. In fact, when you see so many landscape photographers piling on contrast, blocking out shadows and pushing contrast to (usually, unwittingly) squash down to get that Velva effect, it is a touch ironic. Especially when the same ones spend hours hurling invective at each other in flame wars on who’s (digital) camera has the greatest dynamic range. Then again I don’t much care for Velvia - classic Velvia that is - myself.

Reichmann again “My second impression is to once again confirm how truly poor film based imaging is / was compared to todays’ digital capture. Using a variety of images I went through every available colour transparency and negative emulsion looking for one that appealed to me more than the original processed with my usual workflow. Not a single one even came close.”. Well I beg to differ. Unless pixel peeping comes into, I can easily recall a handful of classic Michael Reichmann film images. I can’t say that so much of his digital work has stick in my memory. Maybe it’s because of the diluting effect of the avalanche of images.

From my own perspective, the image below is one I took a very long time ago, on Kodachrome 64, before I was really into photography. I’ve been trying to recapture that quality of light ever since. The closest I’ve got on digital, I think, is with the Olympus E-1’s Kodak sensor.

Damoy pink 1

But digital seems to be unable to record my impression of subtle gradations such as those in this sky. It has a tendency to turn pinks into yellows or indigos, or just sees blue. Digital doesn’t get it. Probably it has something to do with white balance software. Possibly - probably even - it is representing the “truth”.  I’d never argue that film is better than digital. Then again I’d never argue the opposite. But dismissing out of hand just makes so sense, in the context of anything either than throw-away photography.


Why is Vuescan struggling ?

Worth what you pay for it ?

in Unsolicited, rabid opinions , Tuesday, August 23, 2011

August 2014: Apparently this post is the most popular on my site. And by far. Which is a bit sad for both of us.  I’m tempted to take it down, but I’ll leave it for historical interest. Maybe you’ve come here to observe a bit of wild-eyed flaming of Ed Hamrick and Vuescan. Well, enjoy, but please note it is water well under the bridge, indeed we can’t even see the bridge any more. Ed & I had an email conversation and it all ended up perfectly amicable even if we agreed to disagree.  These days I sometimes use Vuescan. I took some time to understand it better, I managed to calibrate it, and when my scanner, or Silverfast, causes problems, it’s always useful to be able to fall back on Vuescan for issue solving. And if your workflow would be to take uncorrected linear gamma scans direct into Photoshop, for some reason Vuescan’s 48bit output is much more malleable than Silverfast’s 48bit “HDR”. Oh, and you might also like to see Ed’s original riposte.

Earlier today, I came across a piece of negative marketing of a type which always irritates me. This was from Ed Hamrick, of Hamrick Software, author of Vuescan, offering Silverfast users a free upgrade to Vuescan Pro if they promise never to use Silverfast again and to send him their Silverfast serial number. This is already sounding ethically dubious, and possibly worse, but then he goes on to roundly rip Silverfast to pieces, while saying what a nice guy Karl-Heinz Zahorsky, the CEO and founder of LaserSoft is. All this under the strawman banner “Why is LaserSoft struggling ?”

Now, as far as I know, LaserSoft has never engaged in such tactics. It promotes its own wares, sometimes well, sometimes less so, but it never, ever rubbishes the competition. Hamrick then follows up with a gratuitous analysis of LaserSoft’s “problems” and an “unedited list” of more than 1000 largely illiterate one-liner comments of converts to Vuescan, the majority of which seem to have very little clue of what they’re talking about - and Hamrick knows it. Frankly, these people are not Silverfast’s customer base.

Hamrick goes on to pick apart various aspects of Silverfast, and Lasersoft. Now, Lasersoft sure aren’t perfect, but if you’re going to start slinging mud, you’d better make sure of your target.  A few choice examples:

Anyone who primarly does reflective scans can buy a good printer/scanner/copier for $100, and anyone still scanning film can use the Epson V700 to do this

Sure, Ed. You’re right, and pretty much every reasonably experienced film photographer in the world is wrong. The V700 is ok. In fact, for large format it’s probably the only reasonable option. But for optimal 35mm scans ? Come on!!  And this “anyone still scanning film” ... well, yeah. Guess what. They’re using Silverfast. 

Let’s see what a in-depth review of the V700 has to say:  “Finally one can say that the Epson Perfection V700 Photo is good for digitizing normal vacation pictures and similar images even directly from the film. For applications without professional requirements the scanner is very well suitable. Professionals, whom the V700 actually addresses with the possibility to scan medium formats and large formats, won’t however be satisfied with the picture quality”.  I think I’ll skip Ed’s advice on this one.

This leads to a dilemma - the market for the scanners SilverFast supports is shrinking rapidly, and even the least expensive printer/scanner/copiers are more than good enough for 99% of reflective scanning.

Well, Hamrick may believe that “printer/scanner/copiers are more than good enough for 99% of reflective scanning”. So what ? Silverfast is designed for photographers and pre-press. Possibly there little ROI in providing a cheap enough version for casual users of all-in-one, Walmart special offer copiers. 

What can LaserSoft do, other than try to reduce costs by laying off engineers and delay new product development? It’s been 5 years since Intel Macs were introduced, and LaserSoft still hasn’t released a universal binary version of SilverFast.

Again, so what ? First, how does Hamrick know so much about LaserSoft’s business ? If they are reducing costs, they’re hardly alone.  As for the Universal Binary, up until recently it has been of no use.  LaserSoft, correctly, point out that scan times and limited by scanner performance.  So post processing may be a touch faster with a Universal Binary, but frankly I’m not convinced.  This is really typical software geekery.

The majority of LaserSoft revenues used to come from bundled software sales

I believe it still does

Epson Scan is better than SilverFast

Totally unsubstantiated wild claim. Epson Scan is better than Vuescan!!!

Canon sells many more printer/scanner/copiers than high-end flatbed scanners

Yes, Ed, you’ve made it clear that you’ve missed the point. Stop digging.

Plustek and Reflecta scanners aren’t very good

Really. Why do you support them then ? And pro photographers such as Mark Segal beg to differ. Have they turned down your bundling offer ?

VueScan is a 5 MByte download, SilverFast 8 is a 170 MByte download

Well yes… but that does include the video guides and documentation. Documentation, Ed. Heard of it ? I haven’t downloaded Silverfast 8 yet, but Silverfast 6 is around 25Mb. Bigger than Vuescan, yes, but there’s quite a lot more in it.

It goes on, and quite frankly is astonishing. Did Zahorsky run over his dog or something ?  But anyway, we finally get on to this little claim:

VueScan produces better scans

Well, now I’m listening. Especially as I’m a licensed Vuescan user. I gave up at around version 5, where the appaling UI and bizarre behaviour finally drove me away. So let’s see if Version 9 has improve things. Honestly, if it gives better results, I’m not proud.

The Test

So, I downloaded Vuescan 9, although I had a bit of trouble getting past a website which insisted on pushing “Vuescan Mobile” at me. Let the customer decide, Ed, please.

First impressions were pretty familiar. It’s still got a design only a geek could love, full of weird UI elements and oddities. But at least they line up and the labels don’t overflow any more.

First run: although it did find my networked multi-mode printer/scanner, it failed to find my USB connected Canoscan 9000F. After a relaunch, it found my Minolta film scanner as well. It never did find the Canon. Probably because it’s not a “printer/scanner/copier”.

I went to pick up my old serial code, and entered it. It didn’t work, but that was just a guess really, because I got no feedback. Ok, so I need to get an updated serial number. Fine. That worked, well enough, but the user experience has already deviated well away from smooth. I wonder if the average printer/scanner/copier user would have worked it out ?

Ok, fine. Let me at those awesome results.  I loaded up a slide.  And clicked on “Preview”.  And Vuescan, way off in a little corner, tells me it is “Calibrating”.  I wait for minute or so, then it shows Busy 0%, eventually after, 2 minutes or so it shows Busy 100%. This goes on for a while. It starts again: Busy 0 to 100% another 2-3 minutes. No attempt to tell me what is going on, and no attempt to show a standard system activity bar.  During this time the application is locked up. And then it starts again - busy 0%.  What is it doing, calibrating R, G & B channels ? No idea. Anyway, the claim of “speed” is already wearing thin. Nope, not RGB, because it’s started again. And again.

Finally, a dialog. Please insert the film holder. So I did. But the scanner does not grab it. There’s something not right here.  Everything locks up. Great. I shut down the scanner, force quit Vuescan. And try again. I’m tenacious.

It starts up again, can’t find the Minolta. Shut down. Starts up again, finds Minolta. Finally I get it do a prescan.  It contrives to make the usually quiet-ish Minolta sound like a garbage truck in a tin can factory. Very noisy AF, very noisy prescan and no faster than Silverfast.

The prescan area is too big, which reminds me I’ve always been very suspicious of Vuescan’s handling of the Minolta’s hi-res area (4800dpi for a 35mm strip, 3200 for 120 film).  Using the Scanhancer, it seems there’s no way to get a decent preview, which Silverfast has no trouble with.  Also, nothing approaching Silverfast’s tuning tools. Not even remotely. However, there is one big plus, potentially: the option to use Multi Exposure at the same time as Multi Scanning (which Lasersoft have always said is of little benefit).

Vuescan Preview

Vuescan preview

Silverfast preview

Silverfast preview

I eventually found the “advanced” settings. Not exactly intuitive, but well at least that’s consistent. And I get things set up as I want, and start a scan. Is it faster ? No, of course it isn’t: scan time is scanner limited. Output is very dark, very compressed histogram. However shadows are exceptionally clean - although later when I ran the same slide through Silverfast, it was equally good.

The UI remains exceedingly clunky and uninspiring, and if Silverfast 6 has its annoyances, VueScan just responds with a different set. Some things, for examplre setting preferences, are marginally more simple with Vuescan, but other things, for example prescan colour correction, or manual focus, are way, way worse.

Vuescan’s web site features testimonials from Smart Computing, PC World, Computer Shopper, Mac Guild, etc. Although to be fair Amateur Photographer praised it highly. But Silverfast features reviews by pro photographers such as Mark Segal (who was complimentray about the Plustec scanner which Hamrick dismisses), John Barclay, Timothy Grey, etc. No PC geeks here.

I could probably could make Vuescan work for me, especially if I invested in Sascha Steinhoff’s book.  Vuescan is not bad. For a casual user it’s a better investment than Silverfast, which in its consumer, dumbed down mode is too complex for the target market but also too light on features. VueScan is much cheaper. For advanced users it can also deliver scans just as good as Silverfast. Probably. But it will make you work much harder and it is missing all the refinements of Silverfast.  Generally I’d say there’s a pretty even split out there between Silverfast and Vuescan fans.

But it’s the negative, dishonest marketing that really leaves a bad taste. Another Hamrick quote is “they don’t ask for my advice, and free advice is worth what you pay for it”. Is a free Vuescan upgrade worth what you pay for it too ? So what is this Vuescan upgrade free ? Why such aggressive marketing ? Why is Vuescan struggling ?

I wouldn’t cut off my nose to spite my face if Vuescan really was better,  but the fact is I’ve had years of great results, friendly support and trouble free operation from Silverfast, and I’m not going to switch.  Silverfast has lots of flaws, and probably it is a touch too expensive. But frankly, looking at similar image products from, say, Adobe, or Nik, it certainly isn’t outrageously priced. And personally I don’t find that an annual price-gouging upgrade to be a benefit.




Travels in HDR

with NIK HDR Efex Pro

in Photography , Sunday, August 21, 2011

I’ve always been pretty suspicious of HDR. When Photoshop originally turned up with “merge to HDR” in CS2, I certainly tried it out, but was unable to get anything but the most ghastly results. Certainly nothing that could persuade me that it was a better technique for dealing with high contrast than masking two exposures. Where HDR has been highly and successfully exposed, through sites such as Trey Ratcliffe’s “Stuck in Customs”, all I can say is “de gustibus non est disputandum” - it doesn’t appeal to my tastes, but I can recognise that it can be a valid artistic decision.

However (funny how my second paragraphs often start off with “however”), I have carried on fiddling about with now and again, and have evaluated a fair number of software tools. I finally decided to take the plunge, and buy Nik HDR Efex Pro. Partly because I like Nik software in general, but mainly based on what I could see on Jason Odells, “Luminescence of Nature” web site.  Odell, along with Tony Sweet, shows a series of “natural” HDR landscapes which are far more to my taste than Ratcliffe’s ultravividity, and started to convince me that maybe HDR can be worthwhile.

So, early one morning last week I set off to try it out in practice.  I wanted to see if HDR could provide me with a more satisfying image in a situation where contrast was high, but still just about manageable in a single exposure.

First, here is the single exposure which I find the most acceptable (Olympus E-3, f/11, 0.6s at 1SO 100, +0.3ev):

Lavertezzo, single exposure

Next, an HDR image from HDR Efex Pro, using 5 exposures at 1ev intervals, starting with HDR Efexs’s default setting, and adding a little “structure” and 10 points on the “Method strength” slider:

Lavertezzo, HDR

The differences are not that huge. First of all, I think that the HDR image remains credible, which is the first hurdle.  It also shows more tonal detail in the mid-tones and shadows (the submerged stones, for example). However, it also slightly exaggerates the highlights.  Well, seeing as this was only my second attempt, using a software package with a vast array of adjustments and options, I would say it holds some promise.

The application itself is very nicely done. Easily the best HDR application I’ve tried in terms of ease of use and general workflow. The inclusion of Nik’s U-Point system for targeted local adjustments is a unique selling point, and a very effective tool.

I doubt that I’m going to turn into an HDR maven - although I must confess that I can’t deny a certain cheap thrill sometimes in turning all the sliders up to 11 - but in some circumstances it looks like it can add clear value to the end result.




Media Pro becomes Media Pro

Same again, please

in General Rants , Friday, May 20, 2011

So. It’s finally happened. (Microsoft) Expression Media, ex iView Media Pro, is no more, and PhaseOne MediaPro 1 is with us. And in theory - or it least, in marketingspeak - the integration of Media Pro and CaptureOne, which I wished for at least as far back as 2005, is here to.

Well sort of.  This is what we actually have:



Expression Media ... or is it MediaPro ?



MediaPro ... or is it Expression Media ?

So the difference is… the panels are dark grey instead of light grey, the colours are a taste to be acquired, to be polite, and the icons are nearly invisible. Excellent.

The much trumpeted CaptureOne “integration” is at lipstick-on-a-pig level, and apart from that very little has changed, although if you take a look at the product forum, you might conclude that an impressive range of exciting new bugs has been inserted.

In itself this is hugely disappointing, but on the plus side, the product is at least being developed, and by a focused and far more appropriate owner than Microsoft. I think we’ll have to wait and see what Phase One manage to do over the coming months. If release 1 is a stable (cough) baseline which they will then build on with frequent updates and bug fixes, including some significant new features, fine. If on the other hand they think they’re done here, and can sit back for two years and then release another facelift, well then it’s RIP.

As far as I’m concerned, I would like these features:

- reworked and easily application-interchangeable hierarchical keywording

- proper integration with Capture One, not whitewash

- support for versions, stacks, whatever you call variations on a master image. ANY master image, including scans (this then helps with proper C1 integration

- options for UI colour scheme


Page 4 of 4 pages ‹ First  < 2 3 4