Multi-Scanning software upgrade

[jpaulmoore]

I understand that there is a software upgrade available for my Microtek 120 TF that offers a multi-scanning capability. The software that came with my scanner is Silverfast 6 Ai. i am trying to justify the expense of an upgrade for this feature and want to know if multi-scanning will greatly improve my scans? I am not sure that the additional features in the software upgrade would be used, but I would be greatly interested in improving my scans.
Best,
J. Paul

[LSI_Noack]

Dear J. Paul

checkout the demo for a try-before-buy, if you are interested

Best regards
Sonny Noack
- tech support, LSI AG -

[jpaulmoore]

Thanks Sonny for your reply! I will do just that.
Best,
J. Paul

[MicroSF6]

dear sonny,

i have read on the silverfast site that "The old MultiSampling will actually result in blurred images if the scanner does not position accurately enough (which is true for most of the scanners)."

furthermore, your company president has written this:

1. Multi-Sampling (repetition) of the whole scan and realisation in software (this requires the scanner to be mechanically precise enough, so each scan aligns perfectly with the next one. Only very few scanners are precise enough. We have inplemented that method in SilverFast Ai for fe scanners.

2. Multi-Sampling on a single scan-line
A single stationary line of the CCD in sampled several times and averaged. This is the best method, because the mechanical precision of the scanner is not relevant, also the process is faster.

But which of these types of scanning is already used by my Microtek 120tf in silverfast 6 Ai? i am already able to multisample with my silverfast Ai 6 on this film scanner. am i in need of the blurr correction that only Ai studio can provide? you see, a "try-before-buy" experiment with Ai studio would only help me if i knew just what to look for. and surely, there is a logical answer to this question.

[background: only one image (only one file) results from my multsampling scan, regardles of whether i choose 1x or 4x or 8x or 16x multisampling. but 4x mode seems to take 4 times as long (and to make 4 times as many passes) as does 1x mode; 8x seems to take twice as long (and make twice as many passes) as does 4x mode; and so on. indeed, a status window keeps track of progress and indicates, for example something like "multisample number 7 out of 8 now in progress" if i have chosen 8x mode.]

do you think that my system of hardware/software would gain a multisampling advantage from upgrading to Ai studio? i would like to be sure about getting the best scans i can with this scanner.

i have heard that the 120tf should support Multisampling in hardware, at least with the latest firmware (silverfast tested 1.60). does that mean that my machine is capable of what your company president calls "multisampling on a single scan-line"?

i generated an SF LogFile, and it confirmed that my 120tf is running firmware version 1.60. furthermore, i see on the microtek web site that these words appear:

"In addition, the 120tf offers multiple sampling, in which each scanned line is sampled in frequencies of 2, 4, 8, or 16 to minimize signal fluctuation and artifacts."

so it sounds as if the 120tf does support Multisampling in hardware. does this seem true to you? and even if it is true, how can I know if it is actually PERFORMING this each-line multisampling in hardware? how can i confirm this?

i fear that the hardware multisampling that microtek boasts about on their web site might only work when i am using their native microtek scanwizardpro software, but not when i am using silverfast. is this possible? is it possible that when i use the scanwizard pro software my multisampling is line by line in the hardware, but that when i use silverfast ai, my multisampling is scan-after-complete-scan in the software?

thank you very much.

sincerely,

mark woods

[LSI_Noack]

Dear Mark

you already answered most of your questions yourself

So, to cut a long story short:

Method (2) is preferable to (1), because you (as a software) don't need to calculate how the scanned lines correspond to each other from the separate scans; you already know line 123 from the first sampling corresponds to line 123 from the second and third a.s.o sampling scan.

When you use a (1), however, you need to assure that the sampling scans get overlayed correctly to avoid blurring the resulting picture.

Your Artixscan 120 can use (2), i.e. multi-sampling is already supported by its hardware.
This is why already SilverFast Ai (and not only Ai Studio) offers multi-sampling. There's no additional anti-blurring for your scanner availalbe in Ai Studio, if that's what you hoped for.

When choosing a higher multi-sampling level, scanning times will lengthen accordingly. You need to scan multiple times, whether its lines or whole images.

SilverFast and Microtek's Scanwizard software don't work differently here. SilverFast accesses the scanner via Microtek's SDK.
What you might have perceived is that the scanned images are calculated against each other after the final sampling scan ran through. That's alright, how else can you consider all of the samples. But note that the software doesn't need to calculate how the lines in the scans do correspond to each other.

Just a note: There's a firmware revision 1.70 for the Artixscan available.

A <a href="https://www.silverfast.com/sf-download/demo/en.html?product=383">demo version</a> can be downloaded from our website.
If you already do have an Ai 6.4.x or higher for your scanner installed, simply de-serialise it to change it into an Ai Studio demo.

Best regards
Sonny Noack
- Manager Technical Support, LaserSoft Imaging AG -

[Martinalix]

Thanks for update software you solve my big problem

[Johannes]

My opinion:

There is no need to use Multisampling on a Microtek 120 tf, if you use full optical resolution of 4000 dpi. The difference in scan quality compared to the grain-noise of the film is extremely low. Same with Multi-Exposure, which still doesnt work properly, if i remember right. No need for both. Scann with full resolution, which uses all Sensor-pixels and renders a image fine.

Firmware:

I have firmware 1.8 installed on 120tf-hardware. I had 1.6 and updated with help of the german microtek support centre. But no difference in image quality recognized. But i feel a newer Silverfast version as delivered by the CD packed with the scanner is strongly recommended. The scanner works much better with an update.

Silverfast and HDR:

Now i use Silverfast to Scan an image in HDR-Format. Then i use Silverfast HDR (unfortunately to be purchased seperately) to convert to 24 bit JPG or TIFF-File. This makes it possible to change parameters without using the scanners hardware to scan again. Each Slide results in a 500MG-HDR-File in TIF-Format. Then convert to a 24-Bit-File, which takes 250 MB per image (6x7).

[LSI_Luebker]

Dear Johannes,

thank you for your post ,

When scanning with lower resolution the Multi-Exposure works quite well with the Microtek 120tf we generally recommend using 80 % of the max. Physical resolution as it has proven that you get the most reliable result.