costas_lymbouris

Each version of the Dimage Scan Utility is a complete package, so you only need the latest version.

Rather than a video card, you will get more performance with extra RAM upto 2 gig, its a better way to spend the money. Also consider a second hard drive if your system only has a single drive, put the PS scratch file on it.

I would say its like unfolding a newspaper that has been folded over many times to fit into your pocket. You can still read it, but you are not going to get rid of the creases.

Not sure if this is what you are looking for, but might help

 

shift control N to create a new empty layer

move it to the top of the stack

shift control alt E will merge the content of all underlying layers into the new one, whilst preserving the old layers

 

you cna then then run the blur filter on the new layer and adjust opacity to get the effect you want

Make sure you visit the Minolta website and download the latest version of the scanner software which is DiMAGE Scan Version 1.1.5

 

http://www.minoltaeurope.com/magic_frameEoI.shtml?/pe/head_pe.htm?cgi-bin/db4web_c/db/minoeu/custsupp/support_stage.d4w?language=english

 

When you use it, restrict yourself to making image changes to the functions under the exposure tab. Once you know what results it can deliver then by all means look at 3rd party software.

Sounds reasonable - if you give nearly all the memory to Photoshop then there is not much left for other things such as the operating system.

 

You need to work out how nuch memory all the processes running on your system need and make sure at least that much is available. In addition anytime Photoshop needs to access a file, run a process or something else, it uses system memory so you need to allow a bit for that as well.

 

Given you have 1.8 gig I would suggest you allow about 500Meg for system processes and see how you get on with that. If performance is good allocate a bit more to Photoshop and so on till you notice a difference.

I have had my 5400 since summer of 2003. I mainly scan colour negatives and a few slides and a bit of B&W. Never had the problems that pop up here regularly. With the current version of the drivers it works just fine with the Minolta software.

 

Whatever you are scanning, you need something like the built in light diffuser Minolta called the grain dissolver to deal with grain when you scan at high resolutions.

It's possible to do it in one file

 

Shift Control N to create a new layer

 

position the new layer at the top of the stack if not already there

 

Shift Control Alt E will merge all the underlying layers into your new one

 

Sharpen the new layer.

If as you say command-prompt would be acceptable, then download a copy of ImageMagick for free from the link, the windows executable.

 

http://www.imagemagick.org/www/download.html?

 

When you have installed, you will find there is an executable called convert.exe which will do this (and a lot more) using a parameter called strip. Instructions will be in the installation subdirectory www

I assume you mean 300+films need to be scanned rather than negatives.

 

As the previous posters have said, ICE saves considerable time with colour film but if most of your films are B&W then ICE will not work on them. So if most of your stuff is colour, then the 5400 is worth the extra money; if your scanning B&W ICE need not be considered.

I suppose there are many Minolta websites, but if you look on the european site at the link below and select download/software you will eventually get to the right place to download it. The web site will not let you get there via a direct link.

 

http://www.minoltaeurope.com/magic_frameEoI.shtml?/pe/head_pe.htm?cgi-bin/db4web_c/db/minoeu/custsupp/support_stage.d4w?language=english

There might be a new one on the way, then again some companies do not stock if they cannot compete on price.

 

The link below shows about a dozen suppliers, the cheapest is Amazon and www.cameras2u.com (their web site says in stock) for 462.15, thats 60 cheaper than Jessops.

 

http://www.pricerunner.co.uk/computing/peripherals/scanners/154572/prices

Quote

 

but this precludes me from using any of the image correction tools

 

 

Its the image correction tools that are giving you the choppy histograms, DO NOT USE THEM. Make your corrections using only the controls in the exposure tab and you should see an improvement in the histogram produced. Also turn on the grain dissolver, you cannot use ICE on B&W but the grain dissolver can help with your grain problem.

Have a look at this recent review (link below) covering 8 LCD monitors, 17 inch and includes details of colour performance, for which they recommend the ViewSonic VX715 at about $400. No good for games but gets the accolade for digital photo work.

 

http://graphics.tomshardware.com/display/20041123/index.html

 

I dont have one so cant say anymore.

.

Francois

 

have you had good results with Ilford Smooth Pearl Paper for B&W and did you use the Ilford or one of the generic Epson profiles for B&W, in which case which one?

 

Thanks

Mendel, I understand the large difference in the resolution between 5400 down to 2700; you are getting one quarter of the data. Where I need high resolution I take it, where I dont think I will ever use it, I leave it. What you have pointed out is there is also a halfway stage and where appropriate I will incorporate this into the workflow for those images that warrant it ? thanks. It's not black or white and another option for images I might have second thoughts about is to scan at 5400 then down res to 120Mbytes or so if I change my mind. Obviously better/faster than scanning a second time.

 

My concern has been with capturing the highest level of colour fidelity and good tonal range because I know with the 5400 resolution is never going to be an issue for me. Preserving end-to-end colour fidelity (probably the wrong word) from negative to print (and archive) is the more challenging activity and after several false starts I believe I am now achieving this through the workflow I outlined.

But Mendel, it depends on how good the source material is and what you want to do with the output. When I first got the scanner I scanned everything at 5400dpi because I could. However I just did not need this amount of data unless I was going to print bigger than A5 or needed to do a lot of corrections to an image. Much of the old source material taken handheld contains little more information scanned at 5400 than 2700dpi so I scan most of it at 2700dpi for archive and rescan at 5400 dpi those frames I want to do something special with.

 

After a bit of experimenting and discussion with others on a couple of BBs (going back 9 months now) we came to the conclusion that the Minolta scans everything at 5400dpi, then down samples to the chosen resolution. Could be wrong but I think this is still the case. I found no difference between letting the scanner down sample and doing it myself when trying to produce a 50Mbyte 16-bit file.

 

I am looking at 3 copies of an image on my screen, a 5400dpi scan (217Mbyte file) at 100%, a downsampled version of the 5400dpi scan to give a 110Mbyte file at 2 times magnification, and a 2700dpi scan (56Mbyte file) at 3 times magnification so detail is roughly the same size. I can see more detail in the downsampled 110 Mbyte file than I can in the 2700dpi scan. But I would probably have to print these images at A3 size to see the difference and anything I print larger than A5 gets special treatment anyway. The 5400dpi scan contains a little more detail than the downsampled image at 110Mbytes, but I cannot see any difference in a slightly larger file of 133Mbytes.

 

So what to do, I think you can get into a rut with a workflow plus the material shot since I bought the scanner could benefit from a different treatment. I have been shooting with the scanner capability in mind so have been using prime lenses more frequently. I do not like to mess with the raw scan files, used to do that when I first started scanning and found I was forever scanning the same films over and over again whenever I improved the wotkflow. Having revisited this again, I will do a bit more experimenting and maybe scan more of my current films at 5400 dpi and downsize what would have previously been scanned at 2700dpi to 120 Mbyte files but think that more than half of what I scan will still be at 2700dpi. I am not going to do anything with it that warrants it being bigger; I just want digital images for review and to print for A4 and A5 sized portfolios. I am quite happy to go back and scan again the few frames out of every film that deserve special treatment when I come to edit them.

 

Just my thoughts and always keen to learn better and simpler ways of doing things,

 

thanks

Costas

Graham - a bit off topic but here goes, just my thought on this.

 

I think we all develop a workflow that meets our needs and after a year of using the 5400 I have pretty well settled on what suits me. There is no right or wrong here as long as what you do yields results you are happy with, so I will just try to explain what I do and why.

 

In scanning, I am trying to reduce the amount of post scan work I do to a minimum; so I deal with each individual frame prior to scan.

 

Rough crop, adjust exposure settings to get neutral whites/greys when appropriate, set focus point for autofocus and scan. I scan everything at 2700 dpi 16 bit positive/negative, output resolution 720dpi, and no profile assigned. This gives me a 55Mbyte file suitable for printing A5 or A4. If an image is going to need a lot of post processing or heavy cropping or I know I will print it at A4, then I will scan it at 5700 dpi which gives an image file of 220Mbytes. Applies to less than 10% of the images scanned recently.

 

These are stored on my hard drive and at the end of the month copied to an external hard drive for backup.

 

I then process the image files later, scan again any I think could benefit from a different treatment such as different exposure or 16-bit linear if shadows and highlights both blown. I print those I want hard copy and save the others in my processed archive, flattening but keeping them in 16 bit. Being in 16 bit means I can tweak them again later without getting distortion.

 

I use a working space called BetaRGB and relative colorimetric intent, this has the same white point of D50 as the Minolta profiles so the effect is to just move the scanned image data into a working space profile suitable for editing, without any remapping and so minimise quantization errors. BetaRGB was designed by Bruce Lindbloom as the smallest working space containing all colours that most commercially available films can reproduce; so by using this there should not be any clipping or loss due to out-of-gamut colours. This is a smaller space than the ProPhoto RGB or WideGamut RGB but much larger than Adobe RGB. Worth visiting brucelindbloom.com for his view on working spaces and to download a copy of BetaRGB.

 

Preparing for printing: a topic of huge debate. From the limited amount of research I have done it seems Epson printers such as my R800 will sample the image at 720 dpi (Canon is 600 dpi) no matter what you give it. So since the scans have pretty high resolutions anyway, I resize my image to the output size I want using the crop tool, setting a resolution of 720dpi and either one or both dimensions (or use one of my presets). I can then sharpen the image for printing at the output resolution which is 720dpi; this being the accepted wisdom but not something I have ever tested (because it seems to make sense).

 

I find this gives me prints very close to my screen images, including sharpening artefacts before I realised that gross over sharpening is not beneficial when feeding a printer with files at its native resolution.

 

OK, most of my old stuff shot years ago does not benefit much from this treatment although a few gems surface now and then. However having devised a workflow that is capable of excellent technical results, I find that the stuff I have been shooting over the last 6 months or so is giving me a lot of satisfaction, both when viewed on screen and print by myself and friends. And thats why I do it.

When you do not assign a profile for the Minolta software to convert too from its internal capture space, what you get is the widest possible gamut for your positive or negative scan that the Minolta software can output (ignoring the 16-bit linear option which I assume is just a dump from the CCD as it were). When I look at the gamut of the Minolta profiles, they are wider than the working colour space I use, so I tend to archive my files in the form where they have not been converted to a colour space and also archive the Minolta profiles at the same time. That way I know I have just about all the information I am likely to get out of those frames, so should not need to repeat the scans. Essentially you have stored all there is from the positive/negative processing capabilities of the software. You can scan everything 16-bit linear of course in which case it does not matter, but then you got a whole lot of extra work to do when perhaps only 2% or 3% of frames merit it.

 

I cannot see any disadvantages to this approach. The conversion to a working space can take place when the software creates the output file in which case you will loose some information for ever, before the file is even written; or you convert to a working space when you read the file for editing and loose some information at that point. However, you retain the ability to read that original file into a different perhaps wider colour space without incurring a second bout of information loss.

 

Does it matter in real life, probably not but having spent 12 months scanning film and probably have another few years to do, I am not going to repeat the process in 5 or 6 years time when printers and monitors are all capable of a much wider gamut than Adobe RGB.

You need to copy the file MLTF5400n.icc to the folder on the MAC where Photoshop looks for profiles, I only know the name under XP as given above. The above file name opens with the name

 

Minolta DiMAGE Scan Elite (nega input)

 

It is for negative input, read something along those lines in the documentation when I first started using the scanner about a year ago, dont remember which document though.

There are 3 Minolta profiles available if you scan without having the software convert the colours to a standard profile. The one you mention above is for slides/positive images which is why you are having a problem.

 

On a Windows system, if you look in the directory where the 5400 Minolta scanner software is installed, there is a profile directory; something like

 

\DS5400\DS_Elite5400\Profile

 

In that directory you will find 3 files

 

MLTF5400.icc MLTF5400n.icc MLTF5400p.icc

 

Copy these to your system profiles directory

 

C:\WINDOWS\system32\spool\drivers\color

 

You will then find you have one labelled for negative images when you next use Photoshop, and yes it does make a significant difference!

You could try the free version of WiziWYG monitor calibration software, not sure how good it is but should get you near where you want to be.

 

 

http://www.praxisoft.com/pages/products.wiziwyg.html

Check out the Vuescan site, a Konica Qscan is listed amongst the supported scanners. Look under supported scanners then minolta Konica at this link

 

http://www.hamrick.com/vuescan/vuescan.htm

 

You can download a free tryout version to see if it works with your particular model.

There is an excellent set of instructions and actions at The Lights Right web site for producing saturation masks. Link

 

http://www.thelightsrightstudio.com/DigitalDarkroom/PhotoshopTools/TLRSaturationMask.htm

I think there is enough information already posted on this site to indicate that it does not have to take 25 minutes to scan a frame ? be it slide or negative. I average 4 to 5 scans an hour with ICE (2700/5400 resolution does not make a significant difference), but that includes the time for pre-scanning to crop the image, making adjustments to the exposure, and manually setting the autofocus point.

 

Would help if you explained what settings you are actually using in the software, the specification of the equipment you have the scanner attached too, and what interface to the scanner you are using.