Jump to content

Blackmagic's new Cintel scanners...C and S Drives


Recommended Posts

Blackmagic discontinued their previous scanner and now have a C-Drive (capstan) and S-Drive (sprocket) scanner both w HDR. Their previous model was said to be poor at handling warped 16mm film. Is the current C-Drive scanner the same at handling warped film like the previous model or did they improve warped film handling?

Thanks

Link to comment
Share on other sites

  • Premium Member

I think this is very old news. The capstan drive and HDR are all part of the 2018/2019 update package. 

Yes, it's better at handling warped film, but the gate is not designed to handle it very well.

Its not an archival scanner, it's a camera negative scanner. 

  • Like 1
Link to comment
Share on other sites

  • Site Sponsor

Also the "HDR" works by rewinding the film and doing a second pass and not by double flash while scanning like the Scan Station.

It is what it is.

Pretty poor for 16mm.

Effective Resolutions

3840 x 2880 - Super 35

3390 x 2864 - Standard 35

3390 x 2465 - Anamorphic 35

1903 x 1143 - Super 16

1581 x 1154 - Standard 16

  • Like 2
Link to comment
Share on other sites

  • 2 weeks later...
On 7/20/2021 at 12:52 PM, Robert Houllahan said:

Also the "HDR" works by rewinding the film and doing a second pass and not by double flash while scanning like the Scan Station.

It is what it is.

Pretty poor for 16mm.

Effective Resolutions

3840 x 2880 - Super 35

3390 x 2864 - Standard 35

3390 x 2465 - Anamorphic 35

1903 x 1143 - Super 16

1581 x 1154 - Standard 16

 

For HDR you should have 3 exposures minimum 0,+1,-1. What type of exposure for HDR does the Scan Station do?

Rewinding and rescanning might lead to registration issues. Or does the Cintel have rock solid registration?

Link to comment
Share on other sites

  • Premium Member
2 hours ago, Daniel D. Teoli Jr. said:

For HDR you should have 3 exposures minimum 0,+1,-1. What type of exposure for HDR does the Scan Station do?

Rewinding and rescanning might lead to registration issues. Or does the Cintel have rock solid registration?

Actually for HDR you only really need TWO flashes. One flash normal and one flash with a slightly higher exposure, to raise the blacks up. Then they combine the image so you have more detail/less noise, in the blacks. The HDR modes are not designed to deal with highlights because that's not really where the problems lie, it's mostly in the black detail where the issues are. 

2 hours ago, Daniel D. Teoli Jr. said:

Rewinding and rescanning might lead to registration issues. Or does the Cintel have rock solid registration?

You'd think, but the HDR scanners all have very good registration. They also aren't overlapping the entire image, they making a mask and overlapping only parts of the image from what I can tell. The Arriscan has the best because it's pin registered triple flash full RGB. But again, very slow. They also do HDR without moving the film, they flash the same frame multiple times before it moves to the next frame.

I think the Scan Station does the same thing "real time" HDR, rather than rewinding and running again. 

Link to comment
Share on other sites

37 minutes ago, Tyler Purcell said:

Actually for HDR you only really need TWO flashes. One flash normal and one flash with a slightly higher exposure, to raise the blacks up. Then they combine the image so you have more detail/less noise, in the blacks. The HDR modes are not designed to deal with highlights because that's not really where the problems lie, it's mostly in the black detail where the issues are.

This is not entirely correct. The second flash isn't for the "darks" it's or the densest area of the film, which for negative means the highlights. You get much more to work with in the highlights with HDR, with a gentler roll-off on the high end, in our experience. We recommend HDR regardless of the film - camera original pos or neg, or even print. We've seen benefits to HDR with all of these formats, despite the accepted "common knowledge" that it's not a benefit with print.  

(for example: if you have a faded color print that's gone magenta, and you do a single and two-flash scan of it, then bring those files into Resolve and line the frames up, switching between them you'll see more color is there (you can see the trace on a vectorscope get bigger in the HDR scan), which aids in recovering the fade. It's not perfect, but it's a better starting point than single flash scans.

 

37 minutes ago, Tyler Purcell said:

The Arriscan has the best because it's pin registered triple flash full RGB.

Arriscan is 2-flash. The original scanner took two images of each frame, the XT uses the Alexa sensor's dual-gain feature, which acts like a two-flash HDR scan in one image. As far as I'm aware the only widely used commercially made scanner that does 3-flash is the Director.

Xena might do this too, but I'm not sure. Rob will tell us. 

37 minutes ago, Tyler Purcell said:

I think the Scan Station does the same thing "real time" HDR, rather than rewinding and running again. 

Registration isn't really a huge concern. If you can see the perfs, you can register two frames to a very high degree of accuracy, probably more accurate than mechanical pins, which is what the ScanStation does. As a continuous motion scanner, it takes two images of each frame while the frame is in the gate, but the frame is never at the same position for those two. They are aligned after the images are scanned, and that's how they're able to do 2-flash HDR. As you might imagine, it halves the running speed of an SDR scan because the frame can only be in the gate for so long. 

Edited by Perry Paolantonio
  • Upvote 1
Link to comment
Share on other sites

  • Premium Member
4 minutes ago, Perry Paolantonio said:

The second flash isn't for the "darks" it's or the densest area of the film, which for negative means the highlights. You get much more to work with in the highlights with HDR, with a gentler roll-off on the high end, in our experience.

Technically you're totally right, but if you do the base scan protecting the below mid gray, when you add the HDR pass, it recovers what was lost in the highlights. So the end result isn't highlights necessarily, it's the fact your black's are more protected. IF you scan it that way, we've found it delivers an exceptional image. I agree, if you have an HDR scanner, always use HDR mode. 

8 minutes ago, Perry Paolantonio said:

Arriscan is 2-flash. The original scanner took two images of each frame, the XT uses the Alexa sensor's dual-gain feature, which acts like a two-flash HDR scan in one image. As far as I'm aware the only widely used commercially made scanner that does 3-flash is the Director.

I meant triple flash RGB not HDR, discussing how slow it is. 

It actually is a 7 flash scanner R+B+G + R+B+G (HDR) and UV. Tho supposedly it can be run without doing triple flash and without the pin registered gate. But I have never seen the other mechanics or modes, we only use it for pin perfect scans. 

We have the XT scanner and that's how it works. Delivers great images when it works. 

13 minutes ago, Perry Paolantonio said:

Registration isn't really a huge concern. If you can see the perfs, you can register two frames to a very high degree of accuracy, probably more accurate than mechanical pins, which is what the ScanStation does. As a continuous motion scanner, it takes two images of each frame while the frame is in the gate, but the frame is never at the same position for those two. They are aligned after the images are scanned, and that's how they're able to do 2-flash HDR. As you might imagine, it halves the running speed of an SDR scan because the frame can only be in the gate for so long. 

Yep exactly.

Doesn't the Scan Station have an external registration capturing element, which it uses to real-time (via GPU) fix the files before dropping them to the drive? IE; it doesn't need to have the perfs in frame to register? 

Link to comment
Share on other sites

3 minutes ago, Tyler Purcell said:

Doesn't the Scan Station have an external registration capturing element, which it uses to real-time (via GPU) fix the files before dropping them to the drive? IE; it doesn't need to have the perfs in frame to register? 

No. it uses the perfs in the captured image for registration, though it doesn't require the *entire* perf. For formats like 35mm only the inside edges of the perfs are visible. See: https://www.gammaraydigital.com/blog/lasergraphics-scanstation-65k-maximum-resolutions  for examples of what the scanner "sees" in full overscan with all the common gauges,  on the 6.5k sensor. 

It's all done on the image. frame registration is a trivially easy process using machine vision when you have something like perforations, which are well defined, to use as reference points. They might do this on the frame grabber in an FPGA, or they might do it in CPU or GPU, I have no idea. Even on 14k images like we're working with now, the time it takes to convert a full res image to greyscale, find the perfs, align the perfs to a fixed location on the X/Y axis with rotation, and then apply that translation to the actual image, is just a few milliseconds on a decent CPU.

They might be doing this on the GPU, if the image is already there (that would just make sense), but I think they're mostly using the GPU to accelerate scaling, and applying color corrections, because that's faster than CPU, usually. 

Link to comment
Share on other sites

  • Premium Member
4 hours ago, Perry Paolantonio said:

It's all done on the image. frame registration is a trivially easy process using machine vision when you have something like perforations, which are well defined, to use as reference points. They might do this on the frame grabber in an FPGA, or they might do it in CPU or GPU, I have no idea. Even on 14k images like we're working with now, the time it takes to convert a full res image to greyscale, find the perfs, align the perfs to a fixed location on the X/Y axis with rotation, and then apply that translation to the actual image, is just a few milliseconds on a decent CPU.

Very interesting. 

I thought they had a laser perf detector system that could also read the perf location and then used that data to help guide the software, not just for triggering the camera. 

Do you know if they bought a pre-existing software that does this work and then simply modified it to work with their software, or was this an in-house development. We're trying to find something to help with registration on our scanner and unfortunately, as you know all the good tools are grossly expensive. It seems like someone must have a very basic program out there which can allow the user to define a perf location (X-Y coordinate) and then re-frame all of the other DPX files in that folder (in this case the entire scan) to match the first frame's perf location. Seems pretty easy. 

Link to comment
Share on other sites

  • Site Sponsor
4 hours ago, Perry Paolantonio said:

It's all done on the image. frame registration is a trivially easy process using machine vision when you have something like perforations, which are well defined, to use as reference points. They might do this on the frame grabber in an FPGA, or they might do it in CPU or GPU, I have no idea.

Real-Time perf stabilization is done in GPU on all of these non mechanically pin registered machines.

BT at Co3 told me that they had done reg tests on the Scannity, the Scan Station the Spirit 4K and the Arriscan and that the Arri was the best registration of all of the systems they tested.

I was told that LG does all of this in CUDA language on the GPUs thus the speed, on our SSP the Scanner app is actually a 32bit one.

Xena uses OpenGL and Quadro GPUs so same result just slower than the LG machines which are sold as the fastest as a main sales point.

Most of the computer and imaging parts are off the shelf for these new machines with some customer hardware and then mostly software to glue the pieces together and make them all work.

 

Link to comment
Share on other sites

13 minutes ago, Tyler Purcell said:

Very interesting. 

I thought they had a laser perf detector system that could also read the perf location and then used that data to help guide the software, not just for triggering the camera. 

Do you know if they bought a pre-existing software that does this work and then simply modified it to work with their software, or was this an in-house development. We're trying to find something to help with registration on our scanner and unfortunately, as you know all the good tools are grossly expensive. It seems like someone must have a very basic program out there which can allow the user to define a perf location (X-Y coordinate) and then re-frame all of the other DPX files in that folder (in this case the entire scan) to match the first frame's perf location. Seems pretty easy. 

Lasergraphics (as far as I know) has never used a laser perf detector.  The original director was mechanically pin registered but the current one is sprocketless. The scanstation has been sprocketless and has done optical perfection detection from the beginning. Which I know because we have the first scanstation they shipped.  
 

Open CV is what you’re looking for. It’s free, fast and designed specifically to do things like object detection.  Lasergraphics software is custom. I don’t know what they use under the hood for perf detection but that kind of thing is available in APIs from frame grabbers, could be implemented in an fpga or could be done in software in their app. 

  • Like 1
  • Upvote 2
Link to comment
Share on other sites

3 minutes ago, Robert Houllahan said:

Real-Time perf stabilization is done in GPU on all of these non mechanically pin registered machines.

BT at Co3 told me that they had done reg tests on the Scannity, the Scan Station the Spirit 4K and the Arriscan and that the Arri was the best registration of all of the systems they tested


we are doing the perf detection in CPU. On a 14k image detecting the perfs and calculating the offset takes us less than 20ms. Granted we’re not scanning at 30fps. For that you’d need GPU almost certainly. Or you’d do it on the frame grabbers FPGA. 
 

Digital registration should improve with resolution. We’ve done some tests with the northlight (mech pin) and scanstation and found the scanstation was more stable. With faster scans they may have to simplify the file more than with a slower scanner, and that might lead to less accuracy. I’ll post some tests done on our scanner in a few weeks using opencv. 

Link to comment
Share on other sites

  • Site Sponsor
6 minutes ago, Perry Paolantonio said:


we are doing the perf detection in CPU. On a 14k image detecting the perfs and calculating the offset takes us less than 20ms. Granted we’re not scanning at 30fps. For that you’d need GPU almost certainly. Or you’d do it on the frame grabbers FPGA. 
 

Digital registration should improve with resolution. We’ve done some tests with the northlight (mech pin) and scanstation and found the scanstation was more stable. With faster scans they may have to simplify the file more than with a slower scanner, and that might lead to less accuracy. I’ll post some tests done on our scanner in a few weeks using opencv. 

Having been through this for years the GPU is generally where everyone has landed for machine vision based registration, the Xena allows you to select GPU or CPU and CPU is a significantly slower method that kills scan speed, and that is why LG does it in CUDA. I think it is the only way to get that job done at the speeds they can achieve with scan station, I don't think the frame grabber has the processing capability to do this in a fgpa there is just not enough processing at least for fast scanning.

As for the Northlaaaggg it is a line array and they drag the stage along for scanning the image, a not perfect mechanical arrangement, the Arriscan is an Area Scan sensor so the film and camera are held still for capture. Plus German fussiness.

Edited by Robert Houllahan
Link to comment
Share on other sites

  • Premium Member
On 8/4/2021 at 3:28 PM, Perry Paolantonio said:

Open CV is what you’re looking for. It’s free, fast and designed specifically to do things like object detection.  Lasergraphics software is custom. I don’t know what they use under the hood for perf detection but that kind of thing is available in APIs from frame grabbers, could be implemented in an fpga or could be done in software in their app. 

Thanks Perry good info! Much appreciated. 

Link to comment
Share on other sites

  • Site Sponsor
On 8/4/2021 at 1:29 PM, Tyler Purcell said:

Doesn't the Scan Station have an external registration capturing element, which it uses to real-time (via GPU) fix the files before dropping them to the drive? IE; it doesn't need to have the perfs in frame to register? 

The Scan Station has to be able to see the perforations with the imaging sensor to register the perfs and stabilize the film, there is no external sensor or mechanism. The transport is driven by a servo controlled capstan which has a optical encoder wheel attached to it, the film is driven at constant speed and then the machine vision perf stabilization does the rest.

Link to comment
Share on other sites

25 minutes ago, Robert Houllahan said:

The Scan Station has to be able to see the perforations with the imaging sensor to register the perfs and stabilize the film, there is no external sensor or mechanism. The transport is driven by a servo controlled capstan which has a optical encoder wheel attached to it, the film is driven at constant speed and then the machine vision perf stabilization does the rest.

How does it do with severely warped film?

Link to comment
Share on other sites

On 8/4/2021 at 12:35 PM, Tyler Purcell said:

Actually for HDR you only really need TWO flashes. One flash normal and one flash with a slightly higher exposure, to raise the blacks up. Then they combine the image so you have more detail/less noise, in the blacks. The HDR modes are not designed to deal with highlights because that's not really where the problems lie, it's mostly in the black detail where the issues are. 

You'd think, but the HDR scanners all have very good registration. They also aren't overlapping the entire image, they making a mask and overlapping only parts of the image from what I can tell. The Arriscan has the best because it's pin registered triple flash full RGB. But again, very slow. They also do HDR without moving the film, they flash the same frame multiple times before it moves to the next frame.

I think the Scan Station does the same thing "real time" HDR, rather than rewinding and running again. 

When you say slow, what is the difference between Arri slow scan and one of the faster scanners?

Link to comment
Share on other sites

On 8/4/2021 at 1:40 PM, Perry Paolantonio said:

No. it uses the perfs in the captured image for registration, though it doesn't require the *entire* perf. For formats like 35mm only the inside edges of the perfs are visible. See: https://www.gammaraydigital.com/blog/lasergraphics-scanstation-65k-maximum-resolutions  for examples of what the scanner "sees" in full overscan with all the common gauges,  on the 6.5k sensor. 

It's all done on the image. frame registration is a trivially easy process using machine vision when you have something like perforations, which are well defined, to use as reference points. They might do this on the frame grabber in an FPGA, or they might do it in CPU or GPU, I have no idea. Even on 14k images like we're working with now, the time it takes to convert a full res image to greyscale, find the perfs, align the perfs to a fixed location on the X/Y axis with rotation, and then apply that translation to the actual image, is just a few milliseconds on a decent CPU.

They might be doing this on the GPU, if the image is already there (that would just make sense), but I think they're mostly using the GPU to accelerate scaling, and applying color corrections, because that's faster than CPU, usually. 

When you are doing 14k for 70mm what size file do you get for a feature length film? 

Looked at your link,, but still not sure how the overscan works for registration. Can you scan films without a section of sprockets in the scan or are some sprockets always included in the scan for registration and they have to be cropped out?

Edited by Daniel D. Teoli Jr.
Link to comment
Share on other sites

  • Site Sponsor

A ScanStation can scan at 30fps or 15fps in HDR mode, the Arriscan in HDR is about 1fps. maybe slower, I am not sure about the speed on the new Arriscan XT but it does have a full immersion liquid gate option.

Edited by Robert Houllahan
Link to comment
Share on other sites

  • Premium Member
12 hours ago, Robert Houllahan said:

the Arriscan in HDR is about 1fps. maybe slower, I am not sure about the speed on the new Arriscan XT but it does have a full immersion liquid gate option.

Yea right around there, I haven't timed ours, but it's no faster than the Imagica's, when using HDR mode. 

Link to comment
Share on other sites

  • Premium Member
13 hours ago, Daniel D. Teoli Jr. said:

How does it do with severely warped film?

Pretty good honestly, one of the better machines out there thanks to the gate design and the capstan drive. 

Our little Film Fabriek also has a gate for warped film and can run it capstan without frame detection. It needs to be heavily post stabilized, but it does work. We've run film with destroyed sprockets and the results look like a normal piece of film. 

Link to comment
Share on other sites

  • Premium Member
12 hours ago, Daniel D. Teoli Jr. said:

When you say slow, what is the difference between Arri slow scan and one of the faster scanners?

As Robert said, 30FPS for SDR scan on the Scan Station in 4k is pretty normal, IF you have stupid fast storage. 

We average around 14 fps 4k 10 bit. 

The "slow" scanners are 1fps or less, depending on the machine. But in SDR mode, you can get upwards of 3fps. 

  • Upvote 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...