Version 0r6 even worse

[pike105]

Do these people at Lasersoft ever test their software before putting it into the market? Probably not.

Trying to get rid of the bloody cross hair, I installed Version ... 0r6.

Result: IT8 Calibration creates total nonsense. It does not find anything, not the grid, not the bar code. I tried with different 3 targets. See example.

In total:
Negafix useless (Nikon Scanners)
ME useless
IT8 not working
incompatible with monitor-profiles of LUT-type (only matrix-profiles work)
No response from "support"
No response from Mr. President

<img src="/img/forum/clipboard01.jpg">

[LSI_Noack]

Dear pike105

while I found two cases concerning different questions and they still needed a response (given now) - I unfortunately failed to extract some case for your email address you're using in this forum from our support system addressing the question raised in this thread.

Changes to the IT8 calibration functionality have already been done and will come available soon with the next update of SilverFast.

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

[dnobel]

Do these people at Lasersoft ever test their software before putting it into the market? Probably not.

I love SilverFast. After working in frustration with a number of different scan programs a couple of years ago, I finally tried SF in conjunction with an E6 IT8 target on my Nikon LS 50, and achieved very close to perfect colour and tone on my very first scan preview, even before manual tweaking. For all the program?s warts, I have found it to be the best out there for my purposes, and I have been using it ever since.

Having said that, pike105, I feel your pain here. I too am frustrated with LaserSoft's tendency to throw in various "improvements" with updates, said new features often being poorly implemented, undocumented (other than by marketing hooha), and non-optional. And even their major new features too often arrive sans documentation--not even a read-me addendum--so that even when they work properly, we aren?t told how to access and use them. Far to frequently, all I have to go on is the promo blurb on the web site. For a leading program at these price points, this is simply unacceptable; LaserSoft needs to recognize this and correct the situation ASAP.

Trying to get rid of the bloody cross hair, I installed Version ... 0r6.

If a new feature is not a core function, please ALWAYS make it optional (i.e. defeatable).

Result: IT8 Calibration creates total nonsense. It does not find anything, not the grid, not the bar code. I tried with different 3 targets. See example.

And NEVER introduce changes to core functions before the new implementation is rock solid. I also experienced the new IT8 process as pretty well a crippled disaster. This amounts to betaware. I have also experienced major problems around the Auto Frame Detection which makes it worse than useless for me: It causes the SF preference file to become fatally corrupted for use within PhotoShop.

Nevertheless, I have been able to get the IT8 calibration to produce an accurate profile, although the process isn't fun. The barcode recognition doesn't work, but it is simple to manually download the correct calibration text file to the hard drive and access it there when prompted.

Here is how I am managing it:

-- Always calibrate in stand-alone, not plug-in mode.

-- As a matter of course, delete the SF preference file by holding down Shift-Option-SF when launching SF to calibrate. (Failure to do this has yielded some wacky average delta E readings after calibration.)

-- Do NOT rotate the preview window. Leave it in the default orientation, i.e. portrait mode, so that the top of a 35mm transparency target faces right.

-- Calibrate at the same gamma setting (set under Options? > General) in which scanning is done: for me this is 2.10.

-- If at first you don't succeed, keep trying. Manually correct the sampling mask if necessary and then click Start again?repeat if necessary.

-- At the end of the process, I check the delta E reading that comes up. I find that on my setup, a correct slide calibration reports an average delta E of 1 or 1.1.

In total:
Negafix useless (Nikon Scanners)
ME useless
IT8 not working
incompatible with monitor-profiles of LUT-type (only matrix-profiles work)
No response from "support"
No response from Mr. President

1) I have not had problems using Negafix. What are you experiencing?

2) ME is the single most effective improvement I have found in SF since I started using the program a couple of years ago. The increase in dynamic range and decrease in noise has been simply astonishing, far superior to the results I got with multiple scan mode and achieved in far less time. I have had zero issues with ME. What are your specific issues?

3) As mentioned, I get the IT8 cal to work, but it doesn't work well or consistently. Crippled but not entirely broken for me.

4) What problems are you having with your LUT monitor? I use a NEC 2180UX which I calibrate via LUT and have not experienced identifiable issues with SilverFast. Since ColorSync is the mediator between the monitor and SilverFast, I?m not sure I understand how SF could be the problem.

5) Yes, support communications leaves something to be desired?o.k., a LOT to be desired!

6) Can't comment on this, never having tried to contact El Presidente.

It is a shame when the very software developers we depend on to keep the program current unintentionally become serial software saboteurs. Yes, I want improvements and new features, but not at the price of stability and functionality.

Please, LaserSoft?make sure it works before releasing it, and don?t release updates and versions without first including clear instructions for ALL new features not documented in the last downloadable version of the full manual.

[Arftown]

All,

I have exactly the same problem pike105 has described above.

I am using the reflecta DigitDia5000
I have tried to completeley delete the old Silverfast Version incl. registry, then make a clean install etc.

Still not working!

I am not satisfied with this situation and would expect a fix from LSI.

Thanks.

Regards,
Andreas

[pike105]

Hi dnobel,

"1) I have not had problems using Negafix. What are you experiencing?"

If you have a Nikon scanner try negative direct mode. Much better scans in less time. Negafix has a nonlinearity problem resulting in a strange color cast. See even http://www.silverfast.com/faq/1006/en.html


"2) ME is the single most effective improvement I have found in SF since I started using the program a couple of years ago. ( ... ) What are your specific issues?"

Please upload examples. May be my slides and my LS5000 are so perfect, that I cannot see the effect :-)

"4) What problems are you having with your LUT monitor? ( ...) I?m not sure I understand how SF could be the problem."

I am not talking about the LUT being loaded into the graphic adapter.
There are two types of icc-monitor profiles (I) matrix-type, (II) look-up-table-type. (more precise). SF seems to have a problem in the blacks with (II), with Photoshop its fine.

L-Values measured at the input of the graphic adapter (behind monitor-profile) are looking like this:

(Column 16 IT8)

PS
I 24
J 17
K 10
L 5

SF
I 23
J 11
K 3
L 1

[dnobel]

"If you have a Nikon scanner try negative direct mode. Much better scans in less time. Negafix has a nonlinearity problem resulting in a strange color cast. See even http://www.silverfast.com/faq/1006/en.html"

I went to that thread and looked at the scans put up online: No doubt, the NegaFix scans look awful.

Admittedly, I do most of my work with slides and have limited experience scanning negs. I have gotten some acceptable results with Negafix in the past, but did some experimenting recently to see if I could reproduce some of the problems others have reported. My own results vary according to film specs and original exposure. I have gotten some good scans from both NegaFix and Negative: Direct using SF with my L 50.

Having said that, I also see what others are reporting, namely under-exposure with certain films, although so far not the color casts. Fortunately, Neg:dir seems to work pretty well for me when NegaFix doesn't and vise versa.

Don't know if this has any bearing, but I discovered a while ago that having Nikon Scan and SilverFast installs resident on the same disk caused problems with both, mostly with SF's colors and use of ICC profiles. Once I carefully stripped out every scrap of both, including all NS profiles, and reinstalled SF only, results really improved.

I will need to do more work on this before drawing conclusions, but probably don't have time for that at the moment.



" 'ME is the single most effective improvement I have found in SF since I started using the program a couple of years ago.'
Please upload examples. May be my slides and my LS5000 are so perfect, that I cannot see the effect :-)"

Please go here to download the demonstration files I created for this:
http://web.mac.com/anobel/iWeb/Site%203/Photos.html



"I am not talking about the LUT being loaded into the graphic adapter.
There are two types of icc-monitor profiles (I) matrix-type, (II) look-up-table-type. (more precise). SF seems to have a problem in the blacks with (II), with Photoshop its fine."

You are completely beyond me technically on this. How does this affect things in real-world applications? The images I get from SF scans look great to me, but if I'm missing something, please explain.

[dnobel]

pike105,

After doing some further experimenting, I went back and edited my last post in reference to NegaFix, reflecting mixed results. Apologies if this has created any confusion.

Most of my scan experience is with slides, so I am much less knowledgeable hands-on with negatives and issues unique to them. I would have to do a lot more work with many different negative strips of varying speed, manufacture, and exposure to draw any useful general conclusions.

I find it easier working with slides. One of the problems with scanning neg film is lack of a clear reference point when either there are no corresponding prints existing--or where prints were processed by unreliable facilities. When scanning slides, I examine them on a light table under a loupe to compare with the pre-scan--they are the reference for both color and tone.

My sense so far is that among NegaFix, Neg.:Direct, and NikonScan, each has strengths and weakness, and each may be more or less appropriate for someone in a particular set of circumstances. I am partial to the first two because they give me access to SilverFast's pre-scan toolset, which I am used to and find highly effective.