Braun MULTIMAG SlideScan 4000 - archiving solutions
(This is an edited , longer version, of an article first published in F2 Freelance+Digital magazine in 2006.)
by Jonathan Eastland.
There will be some among us who still recall the name Braun of Nurnberg. Established by Carl Braun in 1905, It was once the manufacturer of the diminutive and much admired Paxette range of 35mm cameras. Like other German photographic companies of note in the post war decades, Braun also manufactured a range of accessories; its slide projectors were one of the most popular items sold by Wallace Heaton in the 1950s and 60s.
While the Paxette cameras nosed dived into oblivion following the invasion of cheaper imports from the east, Braun's slide projector business remained robust and they are still producing a range of machines for amateur use today. Using its long history of mechanical design and manufacturing techniques for these products, the company launched the Braun 3600 multi scanner in 2004.
That model has now been replaced with the Braun 4000, a compact and small footprint dedicated 35mm film scanner with the capacity to handle up to 100 mounted transparencies using a variety of analogue cartridge or carousel type projector slide holders.
The attraction of dedicated film scanning machines with the ability to grind away in the background while one is otherwise more usefully engaged, has always appealed. A picture library stuffed to the gills with slides needs all the help it can get when budgetry constraints prevent sub-contracting armloads of slides to a.n.other speciaists. While I have had good experiences of this route, I have (a) been reluctant to hand over more treasured originals (few though they are) and (b), batches frequently were in need of reworking.
The Nikon Coolscan 4000 and 5000 ED models are my favourite desk top scanners and have always set the bench-mark of (CCD) film scan quality. But the machines have to be watched, especially when using the automated bulk loaders for slides. Packed together tightly in the casette, old card mounts of which the adhesive has long but unknowingly expired, address labels half torn, bits of selotape and all the other problems associated with the days of shoddy pre-press discipline jamming a batch half way through a session is no joke. It wastes hours of time.
From a mechanical viewpoint, there is no obvious reason why Braun's system using proven cartridge or carousel slide holders, where the slide is adequately separated from its nearest neighbour by a wafer of plastic and thrust into the scanning slot with a metal arm, should prove troublesome. The only downside I could see, (also mentioned in the accompanying instruction booklet), was the same possible nuisance as that applying to use of the Nikon batch holders; slide mounts need to be clean and undamaged and especially, free of sticky label protrusions. Whether I liked it or not, hundreds of images in the library would have to be remounted and in the process, in order to keep key data like captions attached, reannotated.
This problem not withstanding, thousands of slides with mounts in pristine condition exist and once an edit of selected viewing pages had been accomplished I was anxious to see just how hassle free the Braun 4000 could be in service.
The machine footprint is relatively small, about the size of most analogue machines accepting a vertically mounted carousel. The outer casing is virtually all lightweight plastic which is the norm these days and which provoked some thought as to the substance of the chassis on which the few moving parts are mounted. A deeper inspection of this near 3 kilo bundle has not been possible but from what is outwardly visible, such moving parts as there are, are of reasonable quality.
The 4000 is bundled with Adobe Photoshop Elements 2.00, Adobe Album and CyberView X -MS scanning interface software. The latter can be integrated as a TWAIN acquire facility through most popular image manipulation software or used as a stand alone. Either way, installation instructions need to be followed to the letter as, somewhat bizarrely, a driver needed to kick start CyberView is embedded in the scanner.
Loading up the supplied Paximat 50 cartridge took a minute or two; it pays to make sure slides are correctly positioned in the narrow slots and mounted in the usual (for a slide projector) upside down and laterally reversed manner to obtain a right way around right way up screen preview image.
With CyberView finally opened on my e-Mac (this is Mac system 10.1.5 and above or various late Windows software only) and the cartridge correctly positioned, I hit the [scan current file] tab to obtain some initial idea of processing time. The book says 2 minutes for a near 50mb file at its maximum resolution of 3600ppi in 8 bit mode. In fact, it takes a full four minutes from go to a saved file and of this, an inordinate 2 minutes is used for what appears to be a machine ready and calibration process before anything visible happens.
In theory, the capacity to save to file using either of the built-in USB or Firewire options, should have made the process slightly faster, especially when saving directly to the main CPU hard drive. However, I was saving everything on this test to an externally USB mounted Western Digital drive. The routing from scanner via firewire and then out to the main drive, produced a considerable slow down. Saving single files to the e-Mac's own drive reduced the process by almost a whole minute.
CyberView X-MS would not be a first choice of scanning software. While it has many ICM profiles attached for colour negative and black and white emulsions, there are non for reversal apart from [generic 1] and [generic 2]. Neither is there a method of inserting film strips into the scanner; users wanting to scan negs will have to cut them into single frames and fit them in slide mounts. The software is however, as simple and intuitively straightforward to use as older Microtek software and from which it appears, vaguely, to be derived. The manual [back one][scan][forward one] machine controls are emulated on a screen menu bar along with other drop down menus giving basic >preview>scan>image correction facilities.
Colour correction, levels and curves displays are all too primitive to be really useful, though they do provide some control when necessary along with the faster fix using a [more like this] panel of image options. The user can quickly select lighter or darker, more yellow or whatever takes your fancy.
In practice, non of this kind of stuff ever works for me. What I need, as we used to say in the days of the wet darkroom is, 'a good original' . From this, Photoshop, or some comprehensive alternative, will provide all the tools needed to produce a final high quality and processed master file.
So the real questions remaining to be answered were these;
Can the Braun 4000 handle the special colour dyes and often deep shadow detail found in Kodachrome slides and does the automated part of the system work on long runs without a hitch? Could I really set it up with a carousel of a 100 slides and trot down the road to the Chinese for dinner?
Worth noting is the fact that CyberView X-MS offers Digital ICE, ROC and GEM in its drop-down menus. When first running the software, check these options remain unticked in their respective boxes; something I omitted to do in my haste to prove the machine's durability. The consequence of this was not only that it took almost four hours to scan and process the first 50 slides, but that, being all Kodachromes, some very odd things happened to their final appearance. ICE does not like conventional silver halide black and white emulsion; neither does it much like Kodachrome, which is essentially a black and white film with the special colour dyes added in processing. On close inspection, the majority of images so processed did not appear to lose time's accumulated foreign matter; their original colour was changed by ROC to something definately not Kodachrome and GEM got rid of what it perceived as noise so well, the final files were all but useless.
On the second run, I made previews of all 50 slides, a six minutes process. This would have enabled me to run a more selective full scan operation from thumbnails displayed in a screen scroll bar. Sticking to the objective was the point however, and while a thumbnail function might prove useful at a later date, there was no immediate indication any of the preview calibrations would carry over to the full scan in multi-scan mode.
The silent delay between each end of file saved to the start of scanning when the machine gave some audible indication of activity was a little disconcerting and in the beginning I found the multi-scan process stopped because the e-Mac had gone to sleep. A fiddle with preferences and energy saving devices seem to resolve the issue; the Braun ground on, in spite of there seeming to be no lasting solution to the issue of the Mac randomly and inexplicably dozing off when I wasn't looking.
Given the scanner's limited d-max specification (3.4), I was somewhat surprised by the high quality appearance of the raw files. Part of the reason for this may be because of the way the scan is made, from a static slide held firmly in place by a spring loaded holder while the cold cathode flourescent smoothly traverses the emulsion, its light focused directly to a static 3 line CCD. The focus plane remained accurately fixed over several hundred slides as was evident from the the appearance of sharply focused edge to edge grain in each image.
On the whole, they looked pretty good and I was finally satisfied after two days of solid multi scanning that the 4000 could indeed be a useful asset to the business. Only one aspect remained to be checked.
Like older Microtek/Polaroid Sprintscan machines, the Braun 4000 scan covers a 37.5mm square area. This is almost exactly the dimensions of a Hasselblad 16S slide and I was keen to see if I could use the bonus facility on this machine. I have missed it since my own Microtek died of pixel failure a couple of years back.
Using a customised GePe plastic mount which requires accurately cutting the transparency to fit, I managed to make the Braun's single scan option work in conjunction with the supplied Paximat 50 cartridge. It would not work with the older Leitz type; the mount spacing is too wide for the finer calibration of the Braun's loading arm, so the machine's auto process missed the slide mount by a good millimeter each time it began the reload for a final scan. Ideally, in scanning a large number of slides of this size, some time should be spent in preparation, if only to search out and locate the thinner types of plastic mount. In the end the scan I hoped for materialised with corners cut off, probably by a part of interior machinery. In this instance, there was no great loss to the image, but I can see future plans for extensive scanning with this size, somewhat curtailed.
SCAN COLOUR AND APPEARANCE
I have often wondered if the inventors of Kodachrome, Mannes and Godowsky, ever gave thought to the possibility of their unique emulsion becoming a favourite for hundreds of documentary photographers, and the images they captured on the emulsion being destined for the printed page.
In the decades that have passed since it was launched by Kodak in 1936, Kodachrome's unique qualities have baffled the best minds in the printing business in attempting to replicate these qualities with ink. The best known examples were seen regularly in the National Geographic magazine, yet despite that journal's success in transferring the Kodachrome image to plate, few other houses - LIFE magazine excepted - seemed to get it right.
In today's digital world, with a hefty reliance on image stock libraries by ad agencies, book publishers and the popular press, spotting a Kodachrome smudge is a rare event. Part of the reason, I suspect, is that many stock agency contributing photographers use dedicated film scanners of the type discussed here, without proper regard to the protocols required to capture the original aesthetic of the image medium. This is not the end users fault, as often, when and if optional ICC or iCM profiles are offered in the interface software, they may not always be accurate. Nor is it the only reason why the Kodachrome effect is so difficult to replicate with CCD scanners.
More sophisticated software allows the user to create custom profiles of the emulsion in use. When possible, this should be done for each emulsion batch about to be scanned as there are often subtle factory differences in the colour quality of film produced on different coating dates. Obviously, this is a logistic nightmare for photographers and agencies with large collections comprised of different stock in need of scanning, so it is hardly suprising there is such a plethora of mediocre print quality on the streets. This argument also powers a great weapon used subliminally by manufacturers to advertise and promote the benefits of colour consistency delivered by their digital capture cameras. But we will save that debate for another occasion.
Kodachrome has a particular oddity, in comparison to subtractive type, layered E-6 emulsions. First, it's a thin base b+w panchromatic film; when exposed in the camera, no colour information latently layered awaits processing. The colour is added by the introduction of special dyes after the b+w image has been developed.
If you have ever spent time projecting Kodachromes on to a screen, you may have noticed that in comparison to subtractive layer reversal types. Kodachrome images often resonate an apparent third dimension for sharply defined artifacts within the motif - the extent and quality of which differs depending on image type, (landscape or still life for example.) and the quaity of glass used to expose the image. But even Kodachromes made with more lowly estimated objectives manage to manifest a different level of image quality compared with other emulsion types.
Since the K-14 development process and formula has been kept a closely guarded secret for decades, I have never been able to establish what it is that gives Kodachrome its unique appearance, apart that is, from the introduction of the special colour dyes. That alone however, is not enough to produce the special depth effect Kodachromes display.
I have no proof, but I would hazard a guess the K-14 development process includes a high edge acutance developer for the b+w image, before the colour dyes are added. Old darkroom hands will be familiar with similar types of developer for standard b+w emulsions, producing a negative which prints with (all other factors being to text book standard) with excellent apparent sharpness using a condenser type enlarger.
Add some special colour effects to the captured b+w image and you have all the ingredients of something different from run-of-the-mill emulsions.
Dedicated desk top CCD film and most consumer level flat bed scanners are limited in their capacity to achieve the kind of density depth necessary to extract all the data contained in an accurately exposed Kodachrome slide. Even some laser drum and higher end flat bed scanners struggle to replicate this type of image in all their highly saturated glory. They are far happier with the lower contrast types of image made on chromagenic negative and can easily turn in an acceptable scan from E-6 reversal emulsion types.
On that note, one might reasonably ask what is the point of investing a relatively low budget figure into any machine that, specifically, is incapable of producing the result you want? To that, there are many permutations of an answer, but often it is underwritten by a price point beyond which we cannot or will not go. That accepted. we can move on, confident that some tool types being discussed will generally be useful and earn their keep to churn out acceptable and reproduceable digital images.
Of all the dedicated film scanners I have had the pleasure of reviewing or testing, the Braun 4000 comes closest to maintaining a near colour emulation for the Kodachrome palette using the machine default settings. Once the raw scan is manipulated however, some of that colour palette originality will be lost to a greater or lesser degree depending on what effects are applied to the digital version.
One other pleasant surprise obtained from this machine was its capacity for consistency, especially in the area of lens autofocus which seems to lock-on to grain with reasonable accuracy during extended batch scanning. In the case of hundreds of Kodachromes now scanned on the 4000, the sharpness of visible image grain is not as precise as I woud like; it lacks the sharply defined edge visible on an original slide examined under a lupe, but it achieves a well earned 9/10 on the score board, compared with a 7 or 8 for some other products. This aspect is probably more important to me than the slightly marginal and possibly exagerrated claims for the extended dynamic range of some other machines but which produce an inferior focused image.
In many instances, the softer focused image often pulls up well with some subtle un-sharp masking, yet rarely, in my experience, achieves the kind of look I am aiming for; sharp yes, but er....obviously artifacted. Combine this with poor colour data extraction and you may see what I mean. The reworked image lacks lustre.
Un-sharp masking evolved in the wet pre-press darkrooms of yesteryear as a device that could be used to heighten the apparent sharpness of a colour image while at the same time subtly reducing the original's contrast to a level suitable for mechanical printing. The digital tool found in many software image manipulation programmes doesn't effect the same result and it is often over applied, frequently by photographers submitting work for publication. Ask most quality pre-press operators what they prefer and the answer will be 'no unsharping....please!' It isn't difficult to understand why photographers might want to sharpen up their images, usually because they look so unsharp when examined at higher screen magnifications. If you want to see what they will really look like at the appropriate scaled image size, view the image on screen at 25%. At this magnification level, unless the image is scaled up to a very large size (beyond A3), it will be difficult to ascertain any defects.
In the case of the Braun 4000 Kodachrome scans, un-sharpening, modestly applied, can have an improving effect on the reproduced image and if you want to see how the effect appears on screen, the following measurements should be set in the dialogue box of the tool when it is opened; 149%, radius = 1.5, threshold = 5. This is the effect applied by most pre-press houses before an image is sent to plate, but the frequency with which it is applied will depend on image reproduction size, line screen size, paper quality and inking levels. Does it work for ink-jet printing, traditional digital photo printing or dye sublimation printing? Yes, but some experimentation is required, especially for ink-jets where the effect can be more pronounced than when used with other print mediums.
FILM VERSUS DIGITAL CAPTURE FOOTNOTE
Image content is what I should be looking at.
A recent end-of-2006-year UK national daily pubication offered its usual insight into what its editors considered the top of the crop pictures of the year, providing an opportunity for enthusiasts like me to compare their choices with many other similar publications falling on the mat at this time.
I will not dwell on the mostly mediocre choices, several of which garnered top slots in all publications. What was interesting, is that few of the images reproduced appeared to have been made on film. You could almost put safe money on the medium format digital captures, less easily on images scored by APS-C type or smaller sensors.
The overall appearance however, while offering superb clinical clarity and a high degree of resolved detail - especially on print reproduced using gravure methods - was, in my humble opinion, bland. The repros lacked the punch, grit and high colour contrast often associated with filmed originals drum scanned to perfection.
'This debate', some will argue, 'is a dead in the water duck.' , for a whole raft of perfectly acceptable reasons and not least of which, most likely, is ' who on earth wants to spend time (and money) attempting, not very successfully, extracting every ounce of data from film when digital capture in a lot of cases is capable of producing at least as good as, if not better, repro, without the inconvenient hassle of analog to digital conversion ?'
For younger generations, the debate mostly doesn't get slated. It's not even a blip on the radar, we are so far down the 'acceptable' digital capture road. But that is all part of the problem. Lots of enthusiasts have been sold the lemon, a product which, while its manufacturers rave about the latest technology, produces a pretty mediocre end result in many cases.
How can I be sure of this?
I am one of those hoarders of stuff. And this includes (some might say, sadly.) more than 40 years worth of tear sheets of reproductions of my work produced over that time and published in a broad range of journals, newspapers and books. Irrespective of image content, I can look back at material reproduced by a variety of means at varying levels of quality. I don't do this very often and that ceratinly has an effect on my perception of what is good and bad when it comes to reviewing this aspect.
In my view, there is no question that medium to large format analog originated reproductions, have often scored in the 9s and 10s - even repros from second rate unmasked contact duplicate transparencies reach high numbers. By far the best quality in general is manifest in continental books and journals with a long history of gravure printing on high quality papers and originating on drum scanners. Some of the best of this can be seen on tear sheets dating from the mid 1980s to the 1990s. As we get into the 21st century, more prolific use of cheap litho and run-of-the-mill web -offset printing lowers the tone.
For smaller formats, the quality differential is less apparent for these types of reproduction except where images are reproduced at whole page or larger scales. Then apparent quality for Kodachromes and Fuji Velvia is often outstanding and they simply get better on heavier, higher quality and more expensive coated paper stocks. Colour negative originals also stand out where time and effort has been exercised in extracting the best possible data to work with in post processing.
Of course, the 6, 10, and 12 megapixel small format digital capture devices are capable of producing excellent results, but as I have said so many times before, it's different. The feel and appearance of printed digital capture lacks the special characteristic film delivers to the page. No where is that more evident in this 40 year collection of paper than from Kodachrome originals.
The eventual demise of that medium, whenever it comes, will be mourned by thousands of photography enthusiasts and for me, it will render my Leica cameras more or less useless for all but b+w film use. Sadly, Kodachrome's passing will not mean an automatic replacement with the new generation of Leica's digital capture devices either; there are other brands which, in my aesthetic opinion, render a digitally captured image with more appeal.
For the moment however, Kodachrome is still with us. If I have a New Year resolution at all, it will be to expose more of this stuff than I did in 2006, using the Braun 4000 to bring its unique quality to the page whenever possible.
BRAUN 4000 SPECIFCATIONS
Compatible for Braun/CS/Universal and LKM Magazines.
- Digital ICE TM – Hardware based Dust-and Scratch removal with infrared sensors and automatically Software correction
- Digital ROC TM – Color Restoration; Image Quality will be restored
- Digital GEM TM – Grain Management; minimize grain to restore the image’s sharpness
Image Sensor: Linear Array 3-line Color CCD
Scanning mode: 48 Bit
Optical Resolution: 3600x3600 dpi
Lamp: Cold Cathode Fluorescent lamp
Scanning preview: 15 seconds
Scanning Speed (ICE off):
Color Scan 1800 dpi: 60 seconds
Color Scan 3600 dpi: 120 seconds
Max. Scanning Area: 37.5 x 37.5 mm
Dynamic Range: 3.4 Dmax.
Batch Scanning: directly from magazine
PC Interface: IEEE 1394 / USB 2.0
Output connector: FireWire (IEEE1394) & 1x USB 2.0
Viewer: Slide Viewer with backlight to examine slide
Slide Editor:
Single Editor with one-touch button and soft-break
One touch button: for one touch scanning
Movement key: forward and backward magazine Control
Sleeping Mode: 1 hour
Weight: 2.9 kg
Dimension: 300x290x125mm (LxWxH)
Min. System Requirement:
Mac: OS X 10.1.5 or higher
PC : Pentium III or higher with at least 256MB RAM (512 MB RAM recommended) and Microsoft Windows 2000 or higher
Copyright: Jonathan Eastland January 2006, 2007.
Visitors, users and viewers of the foregoing content may copy and re-use it in other internet content sites on condition the source of all material so used is acknowledged with the attachment of the following.
www.ajaxnetphoto.blogspot.com 2007.
www.ajaxnetphoto.com 2007.
This original content may NOT be used in any print media made available for commercial resale.
The products and companies named in this website content are trademarks , registered trademarks or servicemarks of their respective owners or licensed users.
by Jonathan Eastland.
There will be some among us who still recall the name Braun of Nurnberg. Established by Carl Braun in 1905, It was once the manufacturer of the diminutive and much admired Paxette range of 35mm cameras. Like other German photographic companies of note in the post war decades, Braun also manufactured a range of accessories; its slide projectors were one of the most popular items sold by Wallace Heaton in the 1950s and 60s.
While the Paxette cameras nosed dived into oblivion following the invasion of cheaper imports from the east, Braun's slide projector business remained robust and they are still producing a range of machines for amateur use today. Using its long history of mechanical design and manufacturing techniques for these products, the company launched the Braun 3600 multi scanner in 2004.
That model has now been replaced with the Braun 4000, a compact and small footprint dedicated 35mm film scanner with the capacity to handle up to 100 mounted transparencies using a variety of analogue cartridge or carousel type projector slide holders.
The attraction of dedicated film scanning machines with the ability to grind away in the background while one is otherwise more usefully engaged, has always appealed. A picture library stuffed to the gills with slides needs all the help it can get when budgetry constraints prevent sub-contracting armloads of slides to a.n.other speciaists. While I have had good experiences of this route, I have (a) been reluctant to hand over more treasured originals (few though they are) and (b), batches frequently were in need of reworking.
The Nikon Coolscan 4000 and 5000 ED models are my favourite desk top scanners and have always set the bench-mark of (CCD) film scan quality. But the machines have to be watched, especially when using the automated bulk loaders for slides. Packed together tightly in the casette, old card mounts of which the adhesive has long but unknowingly expired, address labels half torn, bits of selotape and all the other problems associated with the days of shoddy pre-press discipline jamming a batch half way through a session is no joke. It wastes hours of time.
From a mechanical viewpoint, there is no obvious reason why Braun's system using proven cartridge or carousel slide holders, where the slide is adequately separated from its nearest neighbour by a wafer of plastic and thrust into the scanning slot with a metal arm, should prove troublesome. The only downside I could see, (also mentioned in the accompanying instruction booklet), was the same possible nuisance as that applying to use of the Nikon batch holders; slide mounts need to be clean and undamaged and especially, free of sticky label protrusions. Whether I liked it or not, hundreds of images in the library would have to be remounted and in the process, in order to keep key data like captions attached, reannotated.
This problem not withstanding, thousands of slides with mounts in pristine condition exist and once an edit of selected viewing pages had been accomplished I was anxious to see just how hassle free the Braun 4000 could be in service.
The machine footprint is relatively small, about the size of most analogue machines accepting a vertically mounted carousel. The outer casing is virtually all lightweight plastic which is the norm these days and which provoked some thought as to the substance of the chassis on which the few moving parts are mounted. A deeper inspection of this near 3 kilo bundle has not been possible but from what is outwardly visible, such moving parts as there are, are of reasonable quality.
The 4000 is bundled with Adobe Photoshop Elements 2.00, Adobe Album and CyberView X -MS scanning interface software. The latter can be integrated as a TWAIN acquire facility through most popular image manipulation software or used as a stand alone. Either way, installation instructions need to be followed to the letter as, somewhat bizarrely, a driver needed to kick start CyberView is embedded in the scanner.
Loading up the supplied Paximat 50 cartridge took a minute or two; it pays to make sure slides are correctly positioned in the narrow slots and mounted in the usual (for a slide projector) upside down and laterally reversed manner to obtain a right way around right way up screen preview image.
With CyberView finally opened on my e-Mac (this is Mac system 10.1.5 and above or various late Windows software only) and the cartridge correctly positioned, I hit the [scan current file] tab to obtain some initial idea of processing time. The book says 2 minutes for a near 50mb file at its maximum resolution of 3600ppi in 8 bit mode. In fact, it takes a full four minutes from go to a saved file and of this, an inordinate 2 minutes is used for what appears to be a machine ready and calibration process before anything visible happens.
In theory, the capacity to save to file using either of the built-in USB or Firewire options, should have made the process slightly faster, especially when saving directly to the main CPU hard drive. However, I was saving everything on this test to an externally USB mounted Western Digital drive. The routing from scanner via firewire and then out to the main drive, produced a considerable slow down. Saving single files to the e-Mac's own drive reduced the process by almost a whole minute.
CyberView X-MS would not be a first choice of scanning software. While it has many ICM profiles attached for colour negative and black and white emulsions, there are non for reversal apart from [generic 1] and [generic 2]. Neither is there a method of inserting film strips into the scanner; users wanting to scan negs will have to cut them into single frames and fit them in slide mounts. The software is however, as simple and intuitively straightforward to use as older Microtek software and from which it appears, vaguely, to be derived. The manual [back one][scan][forward one] machine controls are emulated on a screen menu bar along with other drop down menus giving basic >preview>scan>image correction facilities.
Colour correction, levels and curves displays are all too primitive to be really useful, though they do provide some control when necessary along with the faster fix using a [more like this] panel of image options. The user can quickly select lighter or darker, more yellow or whatever takes your fancy.
In practice, non of this kind of stuff ever works for me. What I need, as we used to say in the days of the wet darkroom is, 'a good original' . From this, Photoshop, or some comprehensive alternative, will provide all the tools needed to produce a final high quality and processed master file.
So the real questions remaining to be answered were these;
Can the Braun 4000 handle the special colour dyes and often deep shadow detail found in Kodachrome slides and does the automated part of the system work on long runs without a hitch? Could I really set it up with a carousel of a 100 slides and trot down the road to the Chinese for dinner?
Worth noting is the fact that CyberView X-MS offers Digital ICE, ROC and GEM in its drop-down menus. When first running the software, check these options remain unticked in their respective boxes; something I omitted to do in my haste to prove the machine's durability. The consequence of this was not only that it took almost four hours to scan and process the first 50 slides, but that, being all Kodachromes, some very odd things happened to their final appearance. ICE does not like conventional silver halide black and white emulsion; neither does it much like Kodachrome, which is essentially a black and white film with the special colour dyes added in processing. On close inspection, the majority of images so processed did not appear to lose time's accumulated foreign matter; their original colour was changed by ROC to something definately not Kodachrome and GEM got rid of what it perceived as noise so well, the final files were all but useless.
On the second run, I made previews of all 50 slides, a six minutes process. This would have enabled me to run a more selective full scan operation from thumbnails displayed in a screen scroll bar. Sticking to the objective was the point however, and while a thumbnail function might prove useful at a later date, there was no immediate indication any of the preview calibrations would carry over to the full scan in multi-scan mode.
The silent delay between each end of file saved to the start of scanning when the machine gave some audible indication of activity was a little disconcerting and in the beginning I found the multi-scan process stopped because the e-Mac had gone to sleep. A fiddle with preferences and energy saving devices seem to resolve the issue; the Braun ground on, in spite of there seeming to be no lasting solution to the issue of the Mac randomly and inexplicably dozing off when I wasn't looking.
Given the scanner's limited d-max specification (3.4), I was somewhat surprised by the high quality appearance of the raw files. Part of the reason for this may be because of the way the scan is made, from a static slide held firmly in place by a spring loaded holder while the cold cathode flourescent smoothly traverses the emulsion, its light focused directly to a static 3 line CCD. The focus plane remained accurately fixed over several hundred slides as was evident from the the appearance of sharply focused edge to edge grain in each image.
On the whole, they looked pretty good and I was finally satisfied after two days of solid multi scanning that the 4000 could indeed be a useful asset to the business. Only one aspect remained to be checked.
Like older Microtek/Polaroid Sprintscan machines, the Braun 4000 scan covers a 37.5mm square area. This is almost exactly the dimensions of a Hasselblad 16S slide and I was keen to see if I could use the bonus facility on this machine. I have missed it since my own Microtek died of pixel failure a couple of years back.
Using a customised GePe plastic mount which requires accurately cutting the transparency to fit, I managed to make the Braun's single scan option work in conjunction with the supplied Paximat 50 cartridge. It would not work with the older Leitz type; the mount spacing is too wide for the finer calibration of the Braun's loading arm, so the machine's auto process missed the slide mount by a good millimeter each time it began the reload for a final scan. Ideally, in scanning a large number of slides of this size, some time should be spent in preparation, if only to search out and locate the thinner types of plastic mount. In the end the scan I hoped for materialised with corners cut off, probably by a part of interior machinery. In this instance, there was no great loss to the image, but I can see future plans for extensive scanning with this size, somewhat curtailed.
SCAN COLOUR AND APPEARANCE
I have often wondered if the inventors of Kodachrome, Mannes and Godowsky, ever gave thought to the possibility of their unique emulsion becoming a favourite for hundreds of documentary photographers, and the images they captured on the emulsion being destined for the printed page.
In the decades that have passed since it was launched by Kodak in 1936, Kodachrome's unique qualities have baffled the best minds in the printing business in attempting to replicate these qualities with ink. The best known examples were seen regularly in the National Geographic magazine, yet despite that journal's success in transferring the Kodachrome image to plate, few other houses - LIFE magazine excepted - seemed to get it right.
In today's digital world, with a hefty reliance on image stock libraries by ad agencies, book publishers and the popular press, spotting a Kodachrome smudge is a rare event. Part of the reason, I suspect, is that many stock agency contributing photographers use dedicated film scanners of the type discussed here, without proper regard to the protocols required to capture the original aesthetic of the image medium. This is not the end users fault, as often, when and if optional ICC or iCM profiles are offered in the interface software, they may not always be accurate. Nor is it the only reason why the Kodachrome effect is so difficult to replicate with CCD scanners.
More sophisticated software allows the user to create custom profiles of the emulsion in use. When possible, this should be done for each emulsion batch about to be scanned as there are often subtle factory differences in the colour quality of film produced on different coating dates. Obviously, this is a logistic nightmare for photographers and agencies with large collections comprised of different stock in need of scanning, so it is hardly suprising there is such a plethora of mediocre print quality on the streets. This argument also powers a great weapon used subliminally by manufacturers to advertise and promote the benefits of colour consistency delivered by their digital capture cameras. But we will save that debate for another occasion.
Kodachrome has a particular oddity, in comparison to subtractive type, layered E-6 emulsions. First, it's a thin base b+w panchromatic film; when exposed in the camera, no colour information latently layered awaits processing. The colour is added by the introduction of special dyes after the b+w image has been developed.
If you have ever spent time projecting Kodachromes on to a screen, you may have noticed that in comparison to subtractive layer reversal types. Kodachrome images often resonate an apparent third dimension for sharply defined artifacts within the motif - the extent and quality of which differs depending on image type, (landscape or still life for example.) and the quaity of glass used to expose the image. But even Kodachromes made with more lowly estimated objectives manage to manifest a different level of image quality compared with other emulsion types.
Since the K-14 development process and formula has been kept a closely guarded secret for decades, I have never been able to establish what it is that gives Kodachrome its unique appearance, apart that is, from the introduction of the special colour dyes. That alone however, is not enough to produce the special depth effect Kodachromes display.
I have no proof, but I would hazard a guess the K-14 development process includes a high edge acutance developer for the b+w image, before the colour dyes are added. Old darkroom hands will be familiar with similar types of developer for standard b+w emulsions, producing a negative which prints with (all other factors being to text book standard) with excellent apparent sharpness using a condenser type enlarger.
Add some special colour effects to the captured b+w image and you have all the ingredients of something different from run-of-the-mill emulsions.
Dedicated desk top CCD film and most consumer level flat bed scanners are limited in their capacity to achieve the kind of density depth necessary to extract all the data contained in an accurately exposed Kodachrome slide. Even some laser drum and higher end flat bed scanners struggle to replicate this type of image in all their highly saturated glory. They are far happier with the lower contrast types of image made on chromagenic negative and can easily turn in an acceptable scan from E-6 reversal emulsion types.
On that note, one might reasonably ask what is the point of investing a relatively low budget figure into any machine that, specifically, is incapable of producing the result you want? To that, there are many permutations of an answer, but often it is underwritten by a price point beyond which we cannot or will not go. That accepted. we can move on, confident that some tool types being discussed will generally be useful and earn their keep to churn out acceptable and reproduceable digital images.
Of all the dedicated film scanners I have had the pleasure of reviewing or testing, the Braun 4000 comes closest to maintaining a near colour emulation for the Kodachrome palette using the machine default settings. Once the raw scan is manipulated however, some of that colour palette originality will be lost to a greater or lesser degree depending on what effects are applied to the digital version.
One other pleasant surprise obtained from this machine was its capacity for consistency, especially in the area of lens autofocus which seems to lock-on to grain with reasonable accuracy during extended batch scanning. In the case of hundreds of Kodachromes now scanned on the 4000, the sharpness of visible image grain is not as precise as I woud like; it lacks the sharply defined edge visible on an original slide examined under a lupe, but it achieves a well earned 9/10 on the score board, compared with a 7 or 8 for some other products. This aspect is probably more important to me than the slightly marginal and possibly exagerrated claims for the extended dynamic range of some other machines but which produce an inferior focused image.
In many instances, the softer focused image often pulls up well with some subtle un-sharp masking, yet rarely, in my experience, achieves the kind of look I am aiming for; sharp yes, but er....obviously artifacted. Combine this with poor colour data extraction and you may see what I mean. The reworked image lacks lustre.
Un-sharp masking evolved in the wet pre-press darkrooms of yesteryear as a device that could be used to heighten the apparent sharpness of a colour image while at the same time subtly reducing the original's contrast to a level suitable for mechanical printing. The digital tool found in many software image manipulation programmes doesn't effect the same result and it is often over applied, frequently by photographers submitting work for publication. Ask most quality pre-press operators what they prefer and the answer will be 'no unsharping....please!' It isn't difficult to understand why photographers might want to sharpen up their images, usually because they look so unsharp when examined at higher screen magnifications. If you want to see what they will really look like at the appropriate scaled image size, view the image on screen at 25%. At this magnification level, unless the image is scaled up to a very large size (beyond A3), it will be difficult to ascertain any defects.
In the case of the Braun 4000 Kodachrome scans, un-sharpening, modestly applied, can have an improving effect on the reproduced image and if you want to see how the effect appears on screen, the following measurements should be set in the dialogue box of the tool when it is opened; 149%, radius = 1.5, threshold = 5. This is the effect applied by most pre-press houses before an image is sent to plate, but the frequency with which it is applied will depend on image reproduction size, line screen size, paper quality and inking levels. Does it work for ink-jet printing, traditional digital photo printing or dye sublimation printing? Yes, but some experimentation is required, especially for ink-jets where the effect can be more pronounced than when used with other print mediums.
FILM VERSUS DIGITAL CAPTURE FOOTNOTE
Image content is what I should be looking at.
A recent end-of-2006-year UK national daily pubication offered its usual insight into what its editors considered the top of the crop pictures of the year, providing an opportunity for enthusiasts like me to compare their choices with many other similar publications falling on the mat at this time.
I will not dwell on the mostly mediocre choices, several of which garnered top slots in all publications. What was interesting, is that few of the images reproduced appeared to have been made on film. You could almost put safe money on the medium format digital captures, less easily on images scored by APS-C type or smaller sensors.
The overall appearance however, while offering superb clinical clarity and a high degree of resolved detail - especially on print reproduced using gravure methods - was, in my humble opinion, bland. The repros lacked the punch, grit and high colour contrast often associated with filmed originals drum scanned to perfection.
'This debate', some will argue, 'is a dead in the water duck.' , for a whole raft of perfectly acceptable reasons and not least of which, most likely, is ' who on earth wants to spend time (and money) attempting, not very successfully, extracting every ounce of data from film when digital capture in a lot of cases is capable of producing at least as good as, if not better, repro, without the inconvenient hassle of analog to digital conversion ?'
For younger generations, the debate mostly doesn't get slated. It's not even a blip on the radar, we are so far down the 'acceptable' digital capture road. But that is all part of the problem. Lots of enthusiasts have been sold the lemon, a product which, while its manufacturers rave about the latest technology, produces a pretty mediocre end result in many cases.
How can I be sure of this?
I am one of those hoarders of stuff. And this includes (some might say, sadly.) more than 40 years worth of tear sheets of reproductions of my work produced over that time and published in a broad range of journals, newspapers and books. Irrespective of image content, I can look back at material reproduced by a variety of means at varying levels of quality. I don't do this very often and that ceratinly has an effect on my perception of what is good and bad when it comes to reviewing this aspect.
In my view, there is no question that medium to large format analog originated reproductions, have often scored in the 9s and 10s - even repros from second rate unmasked contact duplicate transparencies reach high numbers. By far the best quality in general is manifest in continental books and journals with a long history of gravure printing on high quality papers and originating on drum scanners. Some of the best of this can be seen on tear sheets dating from the mid 1980s to the 1990s. As we get into the 21st century, more prolific use of cheap litho and run-of-the-mill web -offset printing lowers the tone.
For smaller formats, the quality differential is less apparent for these types of reproduction except where images are reproduced at whole page or larger scales. Then apparent quality for Kodachromes and Fuji Velvia is often outstanding and they simply get better on heavier, higher quality and more expensive coated paper stocks. Colour negative originals also stand out where time and effort has been exercised in extracting the best possible data to work with in post processing.
Of course, the 6, 10, and 12 megapixel small format digital capture devices are capable of producing excellent results, but as I have said so many times before, it's different. The feel and appearance of printed digital capture lacks the special characteristic film delivers to the page. No where is that more evident in this 40 year collection of paper than from Kodachrome originals.
The eventual demise of that medium, whenever it comes, will be mourned by thousands of photography enthusiasts and for me, it will render my Leica cameras more or less useless for all but b+w film use. Sadly, Kodachrome's passing will not mean an automatic replacement with the new generation of Leica's digital capture devices either; there are other brands which, in my aesthetic opinion, render a digitally captured image with more appeal.
For the moment however, Kodachrome is still with us. If I have a New Year resolution at all, it will be to expose more of this stuff than I did in 2006, using the Braun 4000 to bring its unique quality to the page whenever possible.
BRAUN 4000 SPECIFCATIONS
Compatible for Braun/CS/Universal and LKM Magazines.
- Digital ICE TM – Hardware based Dust-and Scratch removal with infrared sensors and automatically Software correction
- Digital ROC TM – Color Restoration; Image Quality will be restored
- Digital GEM TM – Grain Management; minimize grain to restore the image’s sharpness
Image Sensor: Linear Array 3-line Color CCD
Scanning mode: 48 Bit
Optical Resolution: 3600x3600 dpi
Lamp: Cold Cathode Fluorescent lamp
Scanning preview: 15 seconds
Scanning Speed (ICE off):
Color Scan 1800 dpi: 60 seconds
Color Scan 3600 dpi: 120 seconds
Max. Scanning Area: 37.5 x 37.5 mm
Dynamic Range: 3.4 Dmax.
Batch Scanning: directly from magazine
PC Interface: IEEE 1394 / USB 2.0
Output connector: FireWire (IEEE1394) & 1x USB 2.0
Viewer: Slide Viewer with backlight to examine slide
Slide Editor:
Single Editor with one-touch button and soft-break
One touch button: for one touch scanning
Movement key: forward and backward magazine Control
Sleeping Mode: 1 hour
Weight: 2.9 kg
Dimension: 300x290x125mm (LxWxH)
Min. System Requirement:
Mac: OS X 10.1.5 or higher
PC : Pentium III or higher with at least 256MB RAM (512 MB RAM recommended) and Microsoft Windows 2000 or higher
Copyright: Jonathan Eastland January 2006, 2007.
Visitors, users and viewers of the foregoing content may copy and re-use it in other internet content sites on condition the source of all material so used is acknowledged with the attachment of the following.
www.ajaxnetphoto.blogspot.com 2007.
www.ajaxnetphoto.com 2007.
This original content may NOT be used in any print media made available for commercial resale.
The products and companies named in this website content are trademarks , registered trademarks or servicemarks of their respective owners or licensed users.
Labels: Braun, Kodachrome, Photography, Scanning
posted by Ajax at 1.1.07
7 Comments:
Very useful article.
Note that Reflecta offered an identical model to Braun - the Digit Dia 4000. This has recently been upgraded to the Digit Dia 5000 with claimed scan speed improvements.
I used the 4000 with VueScan and Silverfast and found both gave better results than Cyberview. For critical stuff I prefer Coolscans, but for automated batch scanning this works so much more reliably than the sf-210 feeder on the Nikons.
Thank you for the very informative article! I also recently found another scanner that looks identical to the Braun called Pacific Image PS3650. B&H is selling it in NY for about $800 compared to the $1,600 for the Braun. However, Braun claims to have an "anti-jam" feature. Do you know anything about this feature? Can you please also tell me more about how frequently the Braun jammed using cardboard mounted slides?
Have just posted a follow-up on the Braun 4000 regarding batch scanning and the problem of jamming when using old or damaged mounts.
New post can be found at -
http://ajaxnetphoto.blogspot.com/2007/06/braun-4000-scanning-extra-1.html
Jonathan
Thanks Jonathan! Looks like the problem slides are those that have mount seperations problems. All mine have been kept in plastic sleeves over the years and they appear to be in excellent condition except for slight yellowing. Will give it a go and tell you how it goes.
I've got several thousand slides to scan, but they're all mixed. Different emulsions, some colour (wide variations of colour balance), some monochrome (positive), in a variety of different mounts. Would the use of the Braun 4000 be likely to cause more hassle than wading through them manually on my existing Nikon Coolscan III?
BRAUN SCANNING - 3
Cyberview appears to be able to handle a mix of emulsions without problem. However, I prefer to sort originals into their respective profiles, then load up and begin. This way, when it comes to post processing, I know what I am supposed to be dealing with on the colour profile throughout a sesssion, which might take several days. Frankly, if you are scanning batches for eventual uploading to a library site, the Braun quality is good enough. I've found - so far after scanning on a wide variety of tools - not a single client has come back and said 'can you redo this'. My satisfaction seems to be the only issue here and for the once in a blue moon image, I'd fork out for a really good drum scan. Otherwise, the Braun and copycat scanners with the same specs or better seem to give an excellent result.
Where hundreds or thousands of originals are involved, there's no option; any machine that proves to be reliable and durable and AUTOMATIC is first on the list.
Jonathan
Thanks for this. I've used the Reflecta DigitDia 5000 and VueScan for a c. 12,000 slide library of mixed ages and emulsions. The only jams were 3 very old and very chewed Kodachrome cardboard mounts, (out of about 1500 Kodachromes). ALL the rest just worked perfectly. It really is reliable and a great timesaver. That has to count for a lot in my book. Quality is good too, but that depends on getting the software settings correct for each emulsion type.
Post a Comment
<< Home