View Full Version : NegPos vs. Silverfast HDR
I'm a little new to this, but I am really interested in the concept of these two programs. Firstly, let me ask those of you who use either of these a question. Is what I need to get started with both of these only a 16bit linear raw scan? This means I wouldn't have to do any color correcting at the time of scanning? I just basically have to set the scanning software to deliver a truely linear 16bit Raw scan? So I can just scan away and then get to "processing" the scans whenever I want, without having to have a scanner around. Is this any different than letting the scanner software "process" the image, other than I can do the "processing" at my leisure? What are the real differences between NegPos and HDR? What is the real differnces between these two and doing the processing at the time of scanning with the scanning software? In an early post there was mention by Julian that Kirk found Silverfast to be better once the calibration was done? Can either of you expand on this? What does the calibration entail, and is it necessary for scanning color negatives? For those of you who use either of these, do they seem to work better than the scanning software? Are there any other software programs that function in a similiar way? If these work as good as they sound, then all I would have to do at the scanning stage is learn how to get a truely 16bit linear raw scan from whatever scanning software I am using. Thanks in advance for you help.
Silverfast uses something called Negafix to do their color conversions. With Silverfast I usually scan in 48-bit HDR mode which turns off Negafix and most (all?) other filters, so should be linear, and then I use NegPos (or ColorNeg now) to do the color conversion in CS2. Calibration in Silverfast is using IT8 targets--there should be documentation about the process on their site. I prefer Negpos to Negafix out of the box, since the results look better to me. I haven't calibrated Silverfast or NegPos, yet.
Vuescan will also let you scan to linear raw scan and then reprocess that scan in different ways without having to rescan. What is most useful if have limited memory or huge files is that it can downsize the scan on the fly, so that you have a raw scan that is more more manageable in size. This is great for overscanning, esp. in 16 bit color. You can even take the raw scan into Photoshop and use it directly, inverting and adjusting with Photoshop commands. I assume you could pump it into NegPos as well.
I use both Silverfast and Vuescan - Vuescan generally does a better job, but there are some negatives where Silverfast does a better job.
You won't get the kind of results from "calibration" for negatives that you will get from color transparency film calibration, its two different animals. In one instance (trans) you are asking your scanner and computer to agree that what the computer receives agrees with the target or source. (coke can red on film = this coke can red on hard drive or add a further conversion for coke can red on monitor) If you calibrated with the same film stock and have your lab details nailed down you can reproduce transparency colors pretty exactly.
Color neg is different. The orange mask varies in value with the image density with exposure and you have the extra step of conversion from negative to positive colors plus a whole different playing field when you shoot under different lighting conditions. These extra steps really screws with color accuracy. What you can get with transparency shot under controlled lighting conditions is "accurate" color. What you get from color negative is "subjective" or "pleasing" color. Color negative color is interpretive. In other words in one case you're asking your computer to "make me a copy of this color transparency just like it looks" in the other you're asking for "make this color negative look like what I saw". Your computer does not know what you saw!! Of course getting your computer moving in the direction with at least knowing how your film stock -should- convert in an ideal lighting situation is a good thing but its not going to be "dead right", not as good as color transparency calibration can be.
Your idea of gathering raw 16 bit scans and working on them later is sound. I've worked like this as needed for years with scans from smaller color negative film. The problem with larger film is the size of high res. 16 bit files - they get huge! I guess if you have the latest and greatest computer and loads of memory you can get away with it. The advantage is that you can learn -only- Photoshop and do all your work there. Each time a new scanner or driver software comes out you don't have a new program to learn. The disadvantage is that you have to do more work. And then there's the possibility that your chosen scanner won't give you back the raw file you want.
It could be that you're making this too complcated. It could be that you'll save bundles having someone with a drum scanner making you big cheap raw 16 bit files and you do the correction. I'm not sure what the questions are behind your question, if there are any? If you're thinking of taking the plunge into scanning and digital printing I suggest that you buy one of the popular flatbeds and just start scanning -- you'll figure out what works for you and what doesn't. If you're starting from scratch with no equipment or software you will eventually buy Photoshop and you will eventually buy a scanner -- so just jump in!
Anyway, I hope this helps.
Thank you all for your responses thus far.
I guess I mainly want to know what the difference is between using silverrfast AI to "process" an image versus silverfast HDR. By the way, I have used silverfast AI extensively to scan on Epson Flatbeds and a Howtek 4000 drum scanner. My understanding with Silverfast AI is that whatever adjustments you make, and whatever you do in the Negafix section, is basically applied to the RAW scan. So if you choose a 24bit scan, that original raw scan undergoes these adjustments and ends up as an adjusted 24 bit scan. So my question with silverfast HDR is that does it function on a different principle? For example, is the user interface, such as negafix and other controls the exact same as in silverfast AI? If so, it seems to me that the only difference between the two is that AI applies the adjustments at the time of scanning and HDR allows you to apply "scanning adjustments" at a later time. So, there is really no difference in final image color or quality. Is this right? Now, does NegPos (ColorNeg) basically function in the same way as HDR? I know it has different controls and options, but is it the same principle? And finally, if HDR is basically the same as using silverfast AI with negafix in color and quality, then does silverfast or NegPos seem to do a better job handling Color negatives. I am fextremely familiar with AI and negafix, but have not yet had the opportunity to use NegPos. And in response to Henry, Yes, sending out for great 16bit linear Raw scans from drum scanners has crossed my mind. Also, I just want to know the differences between these and silverfast AI? Mainly because I love the idea of being able to simply slap a negative on a scanner and press the scan button and worry about "processing" it later. When using a drum scanner and silverfast AI, I sometimes scan and then when I get a good look at the scanned image, I realize that the scan could be better in this way or that...I can't make good judgements on the silverfast prescan screen. This means that I end up rescanning the image, and it is often 30-45minutes per scan on these howteks. But if I can just scan the Raw image and not have to worry about rescanning, my time would be much less in front of this monitor. Thanks again guys.
I don't know any more to tell you.
I think some of your questions could be answered at the Silverfast site. Don't they have a trial version of the various parts of Silverfast?
I could imagine using HDR workflow the way you're thinking with good results. As far as I know that is its intention or design. I can't make out that much from the Silverfast preview either, and if I was waiting for 30-40 minutes for a scan I'd want it right too!
Many moons ago I used a Polaroid 35mm scanner as you are describing. For whatever reason it gave really crappy color results with some color neg. films. So I'd bypass the Polacolor or whatever it was called and just saved 16 bit raw scans and corrected in PS. Afterwhile I started doing this nearly all the time with that scanner because the results were consistently good. I could scan a whole batch of same set-up shots and run actions to get them real close by saving the first files settings. Of course I was dealing with relatively small scans.
If nothing else comes up here, and you do pursue this, I'd like to hear what you find out.
"If so, it seems to me that the only difference between the two is that AI applies the adjustments at the time of scanning and HDR allows you to apply "scanning adjustments" at a later time. So, there is really no difference in final image color or quality. Is this right?"
Basically, yes. As far as I know there are only two time-saving features of HDR, the HiRepp tool (good for quickly opening very large files) and SC2G, which is a b&w conversion tool. Otherwise you use the same tools in HDR as you do in Ai, the only difference being that with HDR you're applying the tools to an image that has already been scanned. And of course in HDR there is no prescan screen and there are only two prescan tools (scaling and resolution) because you're applying the tools to an image that has already been scanned.
I don't know if this answers your question or not, if not let me know though I don't claim to be an HDR expert, I had enough trouble figuring out basic Ai without getting heavily involved in HDR as well.
Thank you all,
I would love to be using the trial software, but I am actually between scanners right now. I'm trying to decide what scanner to purchase for my 8x10 color neg scanning. I'm thinking about software and workflow because that will probably have an affect on my choice. If I end up getting a drum scanner, I might just end up with software like trident, which is supposed to be really good, But I have been coming across a lot of talk about NegPos (ColorNeg) and Silverfast HDR...so I wanted to familiarize myself with them before I made any choices. I have used quite a few scanning softwares, and so far I have found Silverfast AI to be the best for me, but I see room for improvements. And Yes, I will make sure to keep you posted if I end up trying them both or found out any good info.
Thank you for answering that question. Now I know that SilverfastAI and HDR basically deliver the same result through the same controls. If i end up staying with silverfast, I will definitely get HDR because it would be absolutely wonderful for the way I like to work. Thanks
If anyone can chime in about the benefits of ColorNeg and how it compares to Silverfast AI or HDR, I would be most appreciative. I know that the best thing I can do is test them for myself, which I will probably end up doing, but right now I don't have the hardware and would love to hear ya'lls thoughts. Thanks againg everyone.
If anyone can chime in about the benefits of ColorNeg and how it compares to Silverfast AI or HDR, .
negpos allows you to calibrate the conversion from neg to pos individually. Ai has a series of canned conversions
I've been reading about this conversion calibration and researching NegPos for about the last 8 hours. It all sounds really interesting. Are your results with it that good? How well does it perform without doing the conversion calibration...is it basically on par with Silverfast? When you did the calibration, did you follow the instructions exactly? And if so, how difficult is it? What scanner do you use and what software do you use to get your 16bit linear raw scan? Would a silverfast 48bit HDR scan work for NegPos, or do you recommend using Vuescan to get it? Sorry for all the questions, but there doesn't seem to be too many people who have worked with it and calibrated it to their needs. Thanks.
Anymore NegPos or ColorNeg users who can chime in? Also, it might be of interest to know that I emailed David Dunthorn about the possibility of a Mac version and this was his reply:
"I've started work on a Mac version and am learning the Mac system. Notes like yours will help spur me along to get this done!"
So, anyone who would like the possibility of using ColorNeg (NegPos) on a Mac operating system should visit the website at http://www.c-f-systems.com/Plug-ins.html#colorneg and send him an email attesting to the fact that there are people out here who would be interested. For $67, it is one of the cheapest softwares you can buy og its kind and it will work with a scan from any scanner and software combination that will deliver a 16 bit uncorrected linear scan. That's a deal.
Question: With NegPos (or whataver the new version is called) - can the process be batched up to process multiple images at a time?
I have yet to use NegPos, but my understanding is that while you could batch scan the 16 bit linear files, you can't actually batch process them in NegPos. But, I believe you are able to save your "adjustment" settings in NegPos so that you could apply the same conversion to all of the negatives very quickly, but individually. I'm not positive, but that is my understanding of it. Hopefully a NegPos user could verify this. Anyone?
Vuescan would work. I use the twain on my minolta film scanner as it gives a good linear scan and more accurate manual focussing than vuescan. The calibration is easy but tedious. I use portra 160/400/800 in rollfilm so I only did 3 calibrations. Negpos is the only way I can get in line histos after the conversion. This gives you a good base to then colour correct in photoshop. If you save your raw scan, you then have an untouched archive version of your scan to rework later if you so choose.
Powered by vBulletin® Version 4.2.1 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.