The new Muller HDS scanner, what do you think about that?

[aj]

About resolution, these cameras use bayer pattern which means the images in reality are 1/3 in resolution of what is advertised. However, I would choose areascan (bayer) cameras over linescan cameras because of image stability and less possible artifacts.

Or use a mono camera and take RGB separate frames and merge them.

That is what a camera does The sensors are usually integrated on one imager-chip.

[RCBasher]

About resolution, these cameras use bayer pattern which means the images in reality are 1/3 in resolution of what is advertised. However, I would choose areascan (bayer) cameras over linescan cameras because of image stability and less possible artifacts.

Or use a mono camera and take RGB separate frames and merge them.

That is what a camera does The sensors are usually integrated on one imager-chip.

Not exactly. There are normally two types of colour camera, those which use a Bayer pixel filter pattern on top of a mono sensor and those which use 3 mono CCDs with three sensor sized filters. All sensors are fundamentally mono anyway. There is at least one other type, Foveon if I recall correctly, which has layered filtering very much like film.

As Andreas mentioned, there is a loss of definition (relative to the claimed sensor pixel count) with Bayer filtering but as I mentioned already, probably only about 20% in reality. My system can do RGB sequence flashes for use with a mono sensor, but the question is, is there enough system stability between the exposures and can the post combining process be sufficiently accurate to "reclaim" that 20%? IMHO, probably not worth the effort, just buy a Bayer camera with 20% more pixels!

[Justin Lovell]

We tested many different scanners and had our socks blown off by the lasergraphics scanstation. Got all the bells and whistles to be running up to 5k scans in 8, 16 and 35. The engineers at Lasergraphics are incredible. Couldn't find another scanner that gave the performance and image quality that it does!

They just released an update at nab to increase scanning speeds from 2k 30fps and 5k 15fps to 2k 60fps and 5k 30fps. Yeah. That's right.

Www.framediscreet.com

[RCBasher]

We tested many different scanners and had our socks blown off by the lasergraphics scanstation. Got all the bells and whistles to be running up to 5k scans in 8, 16 and 35. The engineers at Lasergraphics are incredible. Couldn't find another scanner that gave the performance and image quality that it does!

They just released an update at nab to increase scanning speeds from 2k 30fps and 5k 15fps to 2k 60fps and 5k 30fps. Yeah. That's right.

http://Www.framediscreet.com

Some questions:
1) How much is the Scanstation? I suspect we are comparing Ford Fiesta prices with Ferrari prices. The later may be more powerful but the former has probably a better quality ride and lower noise levels ;)
2) Can you post your test results? Just a crop from an uncompressed single frame (e.g. 16bit TIFF) will do, taken at the maximum transport speed....it is important to check for motion blur on a continuous transport machine at the claimed high speeds. Also a dark frame from your test, taken at the same exposure settings as for a typical film.
3) Do you know what area scan sensor is used?

For the test results, I'd like to see something like this this crop from a b&w 16mm frame:
https://www.dropbox.com/s/ov7gkz8sxkpgs ... g1crop.tif
Open it up in Photoshop or the like and apply are large amount of gamma (like 0.25) and watch the detail appear out of the blacks.
I already did it here:
https://www.dropbox.com/s/4g10jo97t5ydn ... g4crop.tif

BTW, that was taken with a 2K sensor (nearly, it was 1920 wide) and 40fps/16bit is quite comfortable, 96fps/16bit possible if you have the computing and storage grunt.

Doing the same on a CMOSIS 2K sensor just gave a grey nothing in the background:
https://www.dropbox.com/s/fd982jkla100n ... p.tif?dl=0

[Will2]

We tested many different scanners and had our socks blown off by the lasergraphics scanstation. Got all the bells and whistles to be running up to 5k scans in 8, 16 and 35. The engineers at Lasergraphics are incredible. Couldn't find another scanner that gave the performance and image quality that it does!

The guys at Cinelicious say they did the same crazy tests and found the Scanity to provide the best results. They were particularly interested in 16mm 4K scans. May have to do with multiple scans at different exposures combined like HDR. Those tests were a couple years ago I think.

http://www.dft-film.com/products/scanity.html

[Andreas Wideroe]

Costwise the scanity is about $1million, the scanstation 5k is about $200K, the Müller scanner is about $20K... so I guess to some extent it is safe to say you get what you pay for, but on the other hand, a new and more budget friendly scanner might give you awesome images for the bucks and we all see what direction the technology is taking us in.

Have a look at the $30K Cintel scanner from Blackmagic. It is still not shipping though, but only a few years ago, a release like that would have shook the scanner world seriously. 4K for $30K (yeah yeah UHD).

There is also a great diversity in sensors and technology. The scanity uses line array sensors combined, so called TDI with 96 lines in a row. This gives then greater dynamic range and better image stability compared to a single line sensor. I'm not sure what the Scanstation uses except it is a areascan sensor. The Müller scanner suggests using Point Grey cameras or Imperx cameras I think running over USB 3.0.

My choice today on the higher end market would be the Scanstation 5K. Versatility, feedback from users aswell as the design/simplicity of operating it would be major pros in my book.

On the other hand, wouldn't it be awesome if we could build a Filmshooting scanner - open source scanner - modulare?

/Andreas

[vintagefilm]

On the other hand, wouldn't it be awesome if we could build a Filmshooting scanner - open source scanner - modulare?

Something like this?

openscanner-960.jpg

This is version 1. We have done a lot of redesign work, and will soon be building a version 2. I would be happy to share details if anyone is interested. The website is openscannerproject.com but it is empty at this point. I am crushed with work right now. I have done a lot of work, but there is a lot more that needs to be done.

[Andreas Wideroe]

Hi Grace!
Very interesting. We've put a lot of consideration into taking one of our old FDL-60s apart, remove everything except transport and put in LEDs and 4-8K machine vision camera (cameralink or coaxpress). However, I don't like the idea of using 30 year old equipment. There are so many great new technical devices on the market now. I know several people who have made their own "high-end" scanners and it should be perfectly doable. I have the domains www.openfilmscanner.com/net/org for this purpose. ;) Perhaps we should join forces eventually and invite others in?

Best,
Andreas

[RCBasher]

The opensource scanner does not look too different to the FilmFabriek Muller scanner in principle. Question is, what is the point? I have supplied a good many of my lighting and capture software solutions to other amateurs around the globe and whilst a large percentage have managed to get up and running with virtually no assistance, one or to have struggled with the very basics of mounting a magnet and sensor or other relatively simple task. So my point is, who has the skills and workshop facilities to build such a scanner for themselves - even if all the designs are opensource?

Now let's say someone decides to set up a small production line to machine the parts and sell a kit? They will not do it for free and machining in small quantities is not cheap. Not everyone has the skills to put it together, so maybe someone puts it together and sells it as a scanner? Not many would be sold on an annual basis and they would still have to put food on the table. When you take a sensible view of it, the Muller scanner (and the like) start to look good value, especially as it is completely free to use whatever camera you like (2K, 4K) and upgrade it easily in the future. You are, in effect, buying the opensource proposal where someone has already done the hard bit that most people would find difficult to do themselves.

As for the Blackmagic Cintel at supposedly $30K, let's see how much it really is when loaded with the essential "optional" bits and also how low the price remains should it wipe out the competition. It is clearly being heavily subsidised at that price if it will be as good as they claim it will be....or they expect to sell many hundreds of them per year to drive down the manufacturing cost.

[MovieStuff]

The opensource scanner does not look too different to the FilmFabriek Muller scanner in principle. Question is, what is the point? I have supplied a good many of my lighting and capture software solutions to other amateurs around the globe and whilst a large percentage have managed to get up and running with virtually no assistance, one or to have struggled with the very basics of mounting a magnet and sensor or other relatively simple task. So my point is, who has the skills and workshop facilities to build such a scanner for themselves - even if all the designs are opensource?

I totally agree. Open source isn't like posting a recipe on Pinterest. If you don't know how to use power tools, don't understand electronics, don't understand codecs, and optics are simply magical to you, then you will have a very hard time trying to build your own unit without some degree of supervision from a knowledgeable person. Years ago in 2001-2001, when I was first open for business and still struggling to find my footing, I used to sell the condenser lens assembly as a separate item for do-it-yourself customers. I figured it would be good for PR and I wanted to be helpful. It was a huge disaster. People were just putting them in front of unmodified projectors and wondering why there was an image of the super bright projection lamp included in their picture. Worse, despite my warnings, many actually looked directly into the condenser lens while their original projector lamp was blasting away at full brightness through an empty gate. I had people risking blindness out of ignorance. I mean there was, literally, not a single person that knew how to implement the lens properly nor how to manage synch issues nor any of the other dozens of disciplines required to build even the most modest film transfer unit. Ultimately, I stopped selling components because I spent far more time on the phone instructing customers how to build their systems than it took for me to build the transfer unit they were trying to avoid buying in order to save money. So, in effect, I was financing their learning and building experience. It was a bad experience.

That said, I am totally in support of people building their own units and the cheaper, the better. I think it's good for the industry because it pushes the larger companies to think smaller and more affordable. I can remember when we made Jet Benny and needed it transferred back in 1986, it was nigh impossible to find a Rank house with a Super 8 gate in the USA. The rates were incredibly high with a huge surcharge for renting the Super 8 gate. When I started my business, to be clear, I was lucky. I was in the right place at the right time. But I know for a fact that the success of my own efforts to offer a cost effective alternative gave confidence to subsequent companies like FlashScan, Muller and others to venture into what most people thought was a dead market because there was a false assumption that all home movies had already been transferred to VHS and had then been discarded. So sometimes it takes the little guy to push the big guys into action. But the little guy still needs to know what he's doing. Enthusiasm won't carry the day, I'm afraid. Of course, there is much fun to be had in the learning experience. I have fun almost every day. Almost.

Roger

[vintagefilm]

Andreas-- don't throw away the FDL-60 yet. My next version has a central spine for the light, gate, and camera that can be used within a larger chassis. The idea is to make it modular, and adaptable to any transport. I have a spare Spirit chassis that will eventually be utilized in this way. The Xena scanner is similar, in that Rennie uses a Rank chassis and transport tech for his scanner. Of course he is way advanced from where I am. I would love to cooperate with anyone who is working toward an open-source solution. There are several skills that will be important to the project. Right now, I need someone who can make 3d engineering drawings, and someone who is proficient in C/C# to build an interface.

And Frank, I think you are right about the FilmFabriek scanner. It is a very good commercial solution, and I don't think it is priced too high. But I do see a need for an open source solution that really lays out the tech for all to see. I don't know how many people will actually build one. But there may be interest from archives that can't afford to buy a machine, but who may have someone with the skills to build one. I am inspired by what you have done, Frank, with your lighting designs. Very clever work, you are to be congratulated.

Machines based on projectors seems to work pretty well, but I chose to build a transport that works in a different way, using no drive sprockets or pull down, so the film path can be less dangerous for the film. I looked at the way the FDL60/90/Quadra worked for film handling, and used that sort of idea to go forward. In fact, the first project was building a custom lens/gate for our FDL/Quadra machines, to transfer reg8 and super8. When that was done, I was looking forward to HiDef, and couldn't afford a Spirit or Shadow. I have since gotten my Spirit. So I am looking at this as a 2K/4K project now.

As far as the Blackmagic/Cintel project, the real question is will they actually release it?

Andreas, could we use these forums as the place to discuss all of this?

[vintagefilm]

Roger, as one of your (very) early customers, I am very appreciative of what you have done, and continue to do. I think the Retro machine is a stroke of genius. You have reduced the complexity down to a level that allows most people to make very good transfers. Thank you. Certainly building a scanner is not for most people. But the tech is out there in bits and pieces, and I am just looking to assemble it into a coherent design that I can use in my business. I have benefited from open source projects in the past, primarily Joomla, a web platform. So my effort to make my design open source is a way to pay it forward a bit. And if anyone is interested in helping, let us know!

[MovieStuff]

Roger, as one of your (very) early customers, I am very appreciative of what you have done, and continue to do. I think the Retro machine is a stroke of genius. You have reduced the complexity down to a level that allows most people to make very good transfers. Thank you.

You are welcome but no thanks are needed. I think the main point is that, while my units make it convenient for those on a budget to buy a plug and play unit, a much better unit CAN be built for the same price or less... if you know how. Open source helps fill in the blanks, information wise, but the point that Frank made about skill can't be casually overlooked. And anyone that decides to chair an open source project and be the central "go to" person better have a lot of spare time to answer questions and hold peoples' hands because, believe me, it never ends.

Roger

[RCBasher]

But the tech is out there in bits and pieces, and I am just looking to assemble it into a coherent design that I can use in my business. I have benefited from open source projects in the past, primarily Joomla, a web platform. So my effort to make my design open source is a way to pay it forward a bit. And if anyone is interested in helping, let us know!

Grace, I wish you good luck with the project and if I can help, I will. I would also caution though about what you might be getting into if you go further than just making one for yourself, e.g. going open source. As I tried to indicate in my last post, but perhaps not very well, you could be opening up a minefield of support issues that will drain away your life! The amount of time I spend just on emails for each fully built and working sub-system I have shipped is way disproportionate to what one might think.

Frank

[MovieStuff]

The amount of time I spend just on emails for each fully built and working sub-system I have shipped is way disproportionate to what one might think.

Word.

Roger