Not my best Super 8 footage yet.

[Josh Gladstone]

To be fair if the intent was always that the horizontal edges of the sprocket hole were to be the reference for vertical registration and the edges of the film were to be the reference for horizontal, it's legitimate to make that the case now. It still strikes me as horrifyingly sloppy manufacturing, though - especially for something that's so expensive.

 

P

 

I don't see it as sloppy with such a small negative. It's a clever engineering solution that utilizes the physical properties of the small negative size to its advantage.

[Carl Looper]

To be fair if the intent was always that the horizontal edges of the sprocket hole were to be the reference for vertical registration and the edges of the film were to be the reference for horizontal, it's legitimate to make that the case now. It still strikes me as horrifyingly sloppy manufacturing, though - especially for something that's so expensive.

 

P

 

Yes, one might call it "sloppy" manufacturing of the film stock. But that is the genius of the Super8 registration design - that it is unaffected by any such "sloppy" manufacturing. That is what the clip is demonstrating - that the registration design works regardless of such "sloppiness" - and by mechanical means - ie. without any "help" from optical stabilisation techniques.

 

The rocking chair effect.

 

So the theory is that the film is twisting in the gate a little, as a function of the perf weave. But I'm not convinced this is the case. A cursory analysis of the scan from screen captures suggests the image is rocking in sync with the right edge of the film, indicative that it is also rocking in sync with the invisible left edge.

 

If this is the case then the scanner is not yet properly aligning itself with the edge of the film. It would be the scanner introducing a little rocking.

 

Indeed the more I think about it the more the rocking chair theory can't possibly be correct. Previous scans of Logmar footage - stabilised in software - show definitively that the image is not rocking with respect to the camera mask. And indeed, the same is visible in this scan. So Purcell's theory is just completely incorrect.

 

It's the scanner that is introducing the rocking. It is not aligning itself properly with the edge of the film.

 

C

Edited August 22, 2015 by Carl Looper

[Tyler Purcell]

The scanner is already aligned to the left side of the film. That is what we are looking at in this scan.

I was referring to left side with the perf's moved to the right side like you suggested, it was a theoretical situation.

 

What we're looking at has nothing to do with any exotic image stabilisation. The scanner, like a projector is aligned to the left side of the film.

Again, who cares about a moving perf… I care about a stable image, one without the rocking chair syndrome. That's what the audience see's, not the perf and not the side of the frame.

 

A cursory analysis of the scan from screen captures suggests the image is rocking in sync with the right edge of the film, indicative that it is also rocking in sync with the invisible left edge.

Actually it's not what so ever. I see the image itself wobbling all over the place and the right side "bellowing" left and right, inconsequential -out of sequence- of the image wobble.

 

Naa, this is clearly gate wobble, nothing to do with the scanner and everything to do with the original captured image in-camera.

[Carl Looper]

I was referring to left side with the perf's moved to the right side like you suggested, it was a theoretical situation.

 

 

Again, who cares about a moving perf… I care about a stable image, one without the rocking chair syndrome. That's what the audience see's, not the perf and not the side of the frame.

 

 

Actually it's not what so ever. I see the image itself wobbling all over the place and the right side "bellowing" left and right, inconsequential -out of sequence- of the image wobble.

 

Naa, this is clearly gate wobble, nothing to do with the scanner and everything to do with the original captured image in-camera.

 

What on earth does "gate weave " mean?

 

In this clip we have a tripod shot and we can see the content of that shot is not rocking relative to the camera mask (the frame boundary)

 

We know from the camera that the camera mask does not rock relative to the gate. And we know the gate does not rock relative to the camera body. And we know the camera is not rocking relative to the scene (it's on a tripod)

 

So where is the rocking occurring?

 

It isn't occuring in the camera.

 

It's occuring in the scanner. The scanner is not aligning itself properly to the edge of the film. We can see that the right edge of the film is rocking in sync with the rocking of the camera mask - so this tells us the edge of the film and the camera mask (+ gate + camera + world) are all stationary relative to each other.

 

So if the scanner aligned itself to the edge of the film it would also be aligning itself with everything else that is not moving relative to that edge (mask, gate, camera).

 

And the rocking would therefore vanish.

 

If we want to fix this rocking then we need to ensure the scanner properly conforms to the Super8 registration system - which is to say that it properly aligns itself to the edge of the film - in terms of both translation and rotation, ie. in the same way a camera and projector do it - or the optical equivalent.

 

C

Edited August 23, 2015 by Carl Looper

[Carl Looper]

I see the image itself wobbling all over the place and the right side "bellowing" left and right, inconsequential -out of sequence- of the image wobble.

 

 

Just to clarify.

 

The right edge of the film (that white line) oscillates left and right and this tells us the width of the film is varying. We know this because the scanner is aligned to the left edge of the film (which we can't otherwise see).

 

The image (the picture within the camera mask) is staying relatively put with respect to the scanner frame. But it is otherwise rocking.

 

The right edge of the film is also rocking, in addition to it's variation left and right. The variation left and right does not interest us as it has no affect on the registration. But it's rotation does interest us. If it's rotating in sync with the camera mask it means these two (mask and edge) are not rotating relative to each other. And this is the case. They are rotating in sync with each other, therefore they are not rotating relative to each other. In short they maintain the same angle.

 

So if they are rotating relative to the scanner, it also means the scanner is rotating relative to the edge and mask.

 

So if the scanner is properly aligned to the edge of the film (in terms of rotation), the rotation we otherwise see will cancel out and we won't have any more rocking. So this needs to be done.

 

C

Edited August 23, 2015 by Carl Looper

[Carl Looper]

Ok, I've done a very close analysis using screen captured frames from the vimeo clip, comparing them in Photoshop, and I can verify that the right edge does indeed rotate relative to the scanner frame. It's not an optical illusion (or not just an optical illusion:) And this in itself verifies that the scanner is not properly aligning itself to the left edge of the film - I think it's a reasonable assumption to make that the left edge and right edge of the film are parallel (or parallel enough).

 

But furthermore I was able to verify that the camera mask and film edge are indeed in sync with each other (in terms of rotation). In other words if the scanner were to properly align itself with the edge of the film (the left edge) then it would resolve the rocking issue.

 

C

Edited August 23, 2015 by Carl Looper

[David Cunningham]

The initial "issue" is most definitely occurring in camera because the registration is pin-based and following a flawed sprocket hole pattern. This IS gate weave. The film is moving side to side in the gate. The scanner is trying but not yet perfectly able to follow the same weave pattern necessary to keep the frames stable relative to each other.

 

This is not really a "fault" of anyone but merely an unexpected side effect of using proper pin registration with a film format never designed for it. Super 8 is truly a low end film format with high levels of fault tolerance built into it. It's just the way super 8 is. Eventually the scanner will be able to compensate for it.

 

Btw... If you were to print this negative or shoot something with a logmar on reversal and then project it using a high end projector like the Elmo 1200 series you WOULD see this weave in the projection because the projectors are not pin registered but edge guided. The edge of the film would stay stable and thus the frame would weave back and forth.

[Carl Looper]

The initial "issue" is most definitely occurring in camera because the registration is pin-based and following a flawed sprocket hole pattern. This IS gate weave. The film is moving side to side in the gate. The scanner is trying but not yet perfectly able to follow the same weave pattern necessary to keep the frames stable relative to each other.

 

This is not really a "fault" of anyone but merely an unexpected side effect of using proper pin registration with a film format never designed for it. Super 8 is truly a low end film format with high levels of fault tolerance built into it. It's just the way super 8 is. Eventually the scanner will be able to compensate for it.

 

Btw... If you were to print this negative or shoot something with a logmar on reversal and then project it using a high end projector like the Elmo 1200 series you WOULD see this weave in the projection because the projectors are not pin registered but edge guided. The edge of the film would stay stable and thus the frame would weave back and forth.

 

The film in the camera is left edge guided. This scan is left edge guided (but not very well). The perf is weaving with respect to the left edge. The pin is only providing for vertical registration. The left edge is providing for horizontal registration.

 

The right edge in the scan is rotating. Therefore the scanner is not rotationally aligned to the left edge (assuming the left edge is parallel to the right edge). The scanner not using a second edge point?

 

The slit width is varying which is why the right edge is varying. The perf stays in sync with the right edge, suggesting the perforator was aligned to the right edge (for some reason).

 

The rotation of the right edge stays in match with rotation of the camera mask - so the film is not twisting in the gate.

 

C

Edited August 23, 2015 by Carl Looper

[Carl Looper]

None of this is about blaming anyone. It's about finding the most appropriate solution.

 

The scanner was originally using the perf for sideways (horizontal) registration, but the perfing of the filmstock was discovered to be weaving (a lot) - so the scanner was having trouble using the perf (obviously).

 

The scanner was then adapted to use the edge of the film (just like Super8 projectors). And this solved the horizontal registration (just like Super8 projectors).

 

But it hasn't been done very well yet, because we still have some rocking (unlike Super8 projectors).

 

This rocking was theorised as caused by the pin and perf weave interacting - which is quite a plausible theory - but to be sustainable there would have to be evidence of the edge rotating independantly of the camera mask. However they are rotating in sync.

 

The solution then is to put some more work into the scanner aligning itself better with the left edge of the film - in terms of rotation (just like Super8 projectors).

 

C

Edited August 23, 2015 by Carl Looper

[Perry Paolantonio]

 

Yes, one might call it "sloppy" manufacturing of the film stock. But that is the genius of the Super8 registration design - that it is unaffected by any such "sloppy" manufacturing. That is what the clip is demonstrating - that the registration design works regardless of such "sloppiness" - and by mechanical means - ie. without any "help" from optical stabilisation techniques.

 

Here's what I don't understand about your argument. I've been shooting Super 8 for 25 years. I have always seen horizontal weave in the film. That's with film that was projected, scanned, and telecine'd. It's demonstrably worse with modern scanners that don't do what the ScanStation is doing now (edge alignment for horizontal positioning), because they're more precise and that exposes the inconsistently placed perfs. But horizontal weave in Super 8 has always been there, with just about any camera/projector setup I've ever used.

 

When I was in college, I shot Super 8 on probably a dozen different cameras (ranging from my own crappy Chinon to just about every Nizo ever made, Canons, Nikons, etc), and projected them on everything from small consumer projectors through really nice onces like the Elmo GS1200. They all had horizontal weave, even on shots where we had the camera locked down on a beefy Sachtler fluid head tripod.

 

I just don't understand how you can say the scanner is the problem here, when the problem is clearly with the manufacturing of the film. And don't get me wrong, I'm not bashing Super 8 - I love the format and still shoot in it. But it was conceived of as a cheap and convenient consumer format. It was never meant to have the level of precision you'd expect from 16 or 35, and that's just fine. It is what it is.

 

But placing all the blame on the manufacturers of scanners and saying that the film itself is fine simply isn't correct, because this is an issue that's always been there - even when projected.

 

-perry

[Carl Looper]

Hi Perry,

 

in the clip being shown here the image area is not side ways weaving (relative to the scanner frame). It is, however, rocking a little (rotational motion).

 

I spent my entire child hood shooting Super8 (late 70s) - tons of it - but only projected on a projector, and it looks just like what we're seeing here: no sideways weave.

 

Of course, back then they didn't have such wildly weaving perfs, but if they did, it would look exactly the same as this scanned clip: no sideways weave, because back then (as much as now) the projectors (just like the cameras) use the edge of the film for registration - not the goddamned perf.

 

In other words it never mattered if the perf was weaving. And it shouldn't matter now. And as we can see in this clip, it doesn't matter.

 

I'm not trying to blame anyone for anything. The scanner team have resolved the original issue: they are now using the edge of the film for registration (just like projectors did and still do), instead of trying to use the goddamned perf (which projectors never used and still don't use).

 

However the rocking issue remains to be resolved. This is not to blame anyone for anything - it's simply to say that the solution to the problem here can't be solved in the camera because the camera doesn't know which way the scanner is rocking the image.

 

C

Edited August 24, 2015 by Carl Looper

[Perry Paolantonio]

I think there are a couple things being argued here that are getting all tangled up together.

 

First, let's talk about the horizontal weaving. Perforation edge to film edge distance varies in Kodak Super 8, frame by frame, on a 5-frame cycle. That is, frames 1 and 6 are the same distance from the edge of the film, but frames 1 and 2 are not. When the camera is edge guided (pin registration doesn't really come into play with this if the pin is small enough not to move the film), a scanner that doesn't also edge-guide will result in an image that weaves back and forth. This is what the ScanStation produced up until the past month or so, when software edge guiding was added.

 

We agree on this, no?

 

Second: Rocking - I'm not convinced the rocking exists in the scan. I can say this because we've scanned a bunch of footage from the Logmar and have never seen it in our scans. The only time we've seen this was in the test films Friedemann Wachsmuth scanned with us and posted to Vimeo a few months ago, but that footage was post-processed and stabilized by him. The gentle rocking you see there IS NOT in the original scan, which I am looking at right now. Nor is it in the subsequent scan we did of the same film, after getting the software update from Lasergraphics that fixes the horizontal weave issue. No rotation. It's not there. I also looked at some other Logmar footage we've scanned for other clients, and there is no rocking on their scans.

 

The two examples I've seen of this are the one at the top of this thread and the one Friedemann posted in May or June. We know Friedemann's version included a post-scan stabilization pass. We know Moises did not stabilize this film based on the initial post in this thread. Do we know if any post-scan stabilization happened at Pro8mm? Can Moises confirm that?

 

Third: Edge guiding. The Scanstation uses the left (perf-side) edge for horizontal registration. The software places it along a perfect vertical line. If the film were to pass through the camera at a 60 degree angle, the software would align it to a 90 degree angle (not that that could happen, just using an extreme example to illustrate). Using the right edge of the frame is entirely unnecessary, because the left edge is being used to correct for both position and rotation at the same time. This was confirmed with Lasergraphics this morning.

 

 

-perry

 

 

 

[Carl Looper]

I think there are a couple things being argued here that are getting all tangled up together.

 

First, let's talk about the horizontal weaving. Perforation edge to film edge distance varies in Kodak Super 8, frame by frame, on a 5-frame cycle. That is, frames 1 and 6 are the same distance from the edge of the film, but frames 1 and 2 are not. When the camera is edge guided (pin registration doesn't really come into play with this if the pin is small enough not to move the film), a scanner that doesn't also edge-guide will result in an image that weaves back and forth. This is what the ScanStation produced up until the past month or so, when software edge guiding was added.

 

We agree on this, no?

 

Second: Rocking - I'm not convinced the rocking exists in the scan. I can say this because we've scanned a bunch of footage from the Logmar and have never seen it in our scans. The only time we've seen this was in the test films Friedemann Wachsmuth scanned with us and posted to Vimeo a few months ago, but that footage was post-processed and stabilized by him. The gentle rocking you see there IS NOT in the original scan, which I am looking at right now. Nor is it in the subsequent scan we did of the same film, after getting the software update from Lasergraphics that fixes the horizontal weave issue. No rotation. It's not there. I also looked at some other Logmar footage we've scanned for other clients, and there is no rocking on their scans.

 

The two examples I've seen of this are the one at the top of this thread and the one Friedemann posted in May or June. We know Friedemann's version included a post-scan stabilization pass. We know Moises did not stabilize this film based on the initial post in this thread. Do we know if any post-scan stabilization happened at Pro8mm? Can Moises confirm that?

 

Third: Edge guiding. The Scanstation uses the left (perf-side) edge for horizontal registration. The software places it along a perfect vertical line. If the film were to pass through the camera at a 60 degree angle, the software would align it to a 90 degree angle (not that that could happen, just using an extreme example to illustrate). Using the right edge of the frame is entirely unnecessary, because the left edge is being used to correct for both position and rotation at the same time. This was confirmed with Lasergraphics this morning.

 

 

-perry

 

 

 

 

Yes, we agree that the sideways weave has been resolved. That's what I've been saying the entire time.

 

In terms of rocking I'm not talking about Wachsmuth's clip. As you say, that clip was handled in a post-scan stabilisation (prior to the sideways weave resolution). And the rocking there is entirely synthesised, as a very low frequency rocking.

 

I'm talking about this clip by Moises.

 

I don't know if any post-scan stabilisation has been done here - I was of the understanding that there wasn't any post-scan stabilisation - that this clip was simply demonstrating that the sideways registration had been resolved - which it has.

 

But I could very well be wrong there - so we'll need to check with Pro8mm.

 

When analysing frames I found that the visible right edge of the film is rotating. Between two arbitrary frames I measured a rotation angle of 0.13 degrees (in the clips frame of reference). That doesn't sound like much, but it is a lot and corresponds to a significant visible rocking of the image. But if LaserGraphics are correcting for rotation, then the remaining reasons are:

 

1. post-scan stabilisation being done (inadvertently reintroducing rocking)

2. gate weave (film twisting) in the camera

 

I've only been addressing option 2: that it is not gate weave in the camera.

 

C

 

ps. note on the pin registration. The height of the pin is the height of the sprocket hole. The width of the pin is many times smaller.

Edited August 24, 2015 by Carl Looper

[Perry Paolantonio]

I didn't have a ton of time to do a search through old footage, but I grabbed some reels of my own family's home movies, which happen to be sitting on a shelf at the office. I picked a random reel and scanned through it. All of the film I looked at was Kodachrome II Type A, which was manufactured from 1965-1974. This one was manufactured in 1969, according to the date code burned into the film.

 

Here's a link to a few seconds of the film as a 2k ProRes file, zooming in on just the perf and film edge (which is why it's a bit soft, since there's some digital blow-up happening at this zoom level): https://www.dropbox.com/s/3wivwy3s5hz8h4l/1969kodachrome.mov?dl=0

 

As you can see, the perforation is moving relative to the film edge. it is *not* as extreme as what we see on better cameras, but there's definitely inconsistent perf placement. The pattern is actually a little different than what we see in most batches, which result in a sawtooth wave repeated every 5 perfs. What we see here is that perfs 1 and 6 are the same, and are off from 2, 3, 4, and 5, which seem to be the same. That is, in a 5-punch perforator, the first one is off, while the next 4 are the same.

 

What we see in modern film looks as if the punch is at a slight angle relative to the edge of the film, thus the sawtooth pattern.

 

With older footage, perf inconsistencies are less common, but they are there. They manifest in slightly different patterns than what we see in modern stock, and in many cases with older film, they're basically perfect. This says to me that the manufacturing of the film has gotten sloppier than it once was, but that it was never 100% perfect and that this issue has been there pretty much since the beginning.

 

-perry

[Perry Paolantonio]

When analysing frames I found that the visible right edge of the film is rotating. Between two arbitrary frames I measured a rotation angle of 0.13 degrees (in the clips frame of reference). That doesn't sound like much, but it is a lot and corresponds to a significant visible rocking of the image. But if LaserGraphics are correcting for rotation, then the remaining reasons are:

 

1. post-scan stabilisation being done (inadvertently reintroducing rocking)

2. gate weave (film twisting) in the camera

 

We agree on this. You can clearly see the rocking motion in the clip Moises posted. This is not something I've seen in any other logmar clips that we've scanned, so I agree one of these two options is probably the culprit. I can't think of anything else that might cause that effect.

 

-perry

[Carl Looper]

With older footage, perf inconsistencies are less common, but they are there. They manifest in slightly different patterns than what we see in modern stock, and in many cases with older film, they're basically perfect. This says to me that the manufacturing of the film has gotten sloppier than it once was, but that it was never 100% perfect and that this issue has been there pretty much since the beginning.

 

There's no need for it to be perfect because cameras and projectors don't use the perf for sideways registration. If they did use it for sideways registration it would require cameras and projectors to be continually shifting the filmstrip sideways on every frame, be it by a little (as with past film stock) or a lot (as in some batches of current Super8 film) - both of which would be increasing the risk of instability. It is so much more simple and elegant to use the edge of the film as the sideways reference - which is what camera/projector designs adopted.

 

C

Edited August 24, 2015 by Carl Looper

[Carl Looper]

One thing to keep in mind is that the way scanner registration works, and the way that a post-scan image stabilisation system works, are both based on the same underlying principles - the use of machine vision techniques.

 

This has revolutionised scanning systems insofar as it allows material from the archive, that might have seen better days, to see the light of day again. The film might have shrunk, or acquired torn edges, or damaged perfs, etc, but by the genius of scanner technology, the film is able to be transported through a scanning system without risking further damage to it, by using "optical" registration instead of a "tactile" one. This has allowed such material to become visible again, where before it would have had to sit in the archive in an indefinite state of wishful thinking.

 

But that is not quite the point I wanted to make (as important as it is).

 

While we can digitise film without damaging it, the scanner's optical registration system only needs to be generic.

 

The principles guiding scanner design are, from the beginning, a "fix it in post" system. Even when the scanner uses edge guidance it is still using a "fix it in post" system to implement that. There is nothing wrong with this - indeed it is the genius of scanner design that it can actually be reprogrammed in this way. It is the genius of the digital age that the same hardware can do different things - by simply changing the software.

 

All it means is that the evolution of the system is moved into the software domain where one doesn't have to build new hardware in response to something that needs doing. However, being in the software domain doesn't mean attention to details should be any less so. It just means one doesn't have to go back to the machine shop.

 

Now in general we don't need the scanner to do anything more than a generic task: basically to do what projectors do - it doesn't need to do anything else. But in saying that we're already in a domain where doing just such a thing is not a trivial task. There are all sorts of algorithms and knowledge in play doing just this - and the potential for various assumptions to prove incorrect - but fortunately, being software, it is also correctable ie. without requiring alteration to the hardware (and the much larger costs that such can entail).

 

Beyond such generic tasks we can move any such tasks to more dedicated systems (After Effects etc) where the focus is just on the scan data (rather than how to acquire it) and on what can be done with the data (quite a lot can be done).

 

Now it may be that even the generic tasks end up being a bit beyond the capabilities or responsibilities we might be putting onto the scanner. But we don't want to throw the baby out with the bathwater. The scanner hardware is brilliant. The way forward is forward. Moving forward as they say. Where else would we go?

 

Whether done in the scanner stage or in a subsequent post stage, we need to properly characterise what the camera and scanner are doing at a generic level. I think the best way forward is to just make some test charts and put such through the camera/scanner pipeline. We can then do precise digital analysis of such and verify where each transform is happening.

 

It's entirely resolvable. It's just working out the best (simplest most elegant) way to resolve it that is being explored.

 

C

Edited August 25, 2015 by Carl Looper

[Tyler Purcell]

It's entirely resolvable. It's just working out the best (simplest most elegant) way to resolve it that is being explored.

Then give us an example of a clip without this problem.

[Carl Looper]

Then give us an example of a clip without this problem.

 

Hi Tyler - I know you're not interested in resolving this problem so I'm not sure what kind of contribution you are hoping to make here.

 

So far your solutions are:

 

1. Don't shoot Super8.

2. Shoot digital and fake a film look.

 

Both of which are fine solutions but they are both meaningless in the context of a Super8 forum - don't you think?

 

The way forward, which is the paragraph immediately preceding the one you've just quoted is:

 

"Whether done in the scanner stage or in a subsequent post stage, we need to properly characterise what the camera and scanner are doing at a generic level. I think the best way forward is to just make some test charts and put such through the camera/scanner pipeline. We can then do precise digital analysis of such and verify where each transform is happening."

 

Perhaps you might like to read the rest of the post before continuing

 

C

Edited August 25, 2015 by Carl Looper

[Carl Looper]

I might just say that Tyler's theory of camera gate weave (now that I understand that term) was entirely plausible. And may still be, although evidence I've acquired so far suggests it doesn't hold. But it's only based on a few sample frames, so I could very well be wrong.

 

I also made a mistake in too hastily discounting such a camera gate weave theory, by associating the camera mask to image relationship as indicative of no such weave - but of course, the camera mask to image relationship will always have no weave, whether there is gate weave or not. Indeed I've mentioned this in the past in relation to Wachsmuths test - that there is nothing to be celebrated in the camera mask to image relationship. It will always be perfect.

 

Rather the evidence for camera gate weave is to be found in the mask to film edge relationship. Are they rotating independantly of each other or are they in correlation? If rotating independantly then it's camera gate weave (most likely caused by the perf weave). If correlated then it must be:

 

1. scanner 'gate' weave, or

2. post-scan stabilisation weave (if such occured).

 

Before jumping to hasty conclusions we have to separate out the translational motion of the right edge from it's rotational motion. The translational motion is irrelevant in terms of assessing camera gate weave because neither the camera nor scanner are using this edge for anything. Its the left edge of the film we really need to be looking at but in the absence of such we can use the right edge as a proxy (ie. making the assumption that the left edge and right edge are parallel enough). And if the edge we can see (the right edge) is exhibiting rotational motion (as it is in this scan) we have to separate out whether it's occuring in the scanner, or in some unknown post-scan stabilisation that might have been applied.

 

This is the first part of analysis - in and of itself such rotation tells us that there is either scanner or post-scan weave (because we should not see any such rotation), but it's not yet telling us whether camera gate weave isn't also occuring.

 

The second part of the analysis involves checking if the camera mask to edge relationship is correlated or not. This will tell us whether there is, or isn't, also camera gate weave.

 

It is this second analysis I've carried out - if only with a few frames. There was found a clear correlation between rotation of the camera mask, and rotation of the right edge. Of course the correlation could be a complete fluke. Or rather, two complete flukes, because I did two tests in which this was the result.

 

C

[Tyler Purcell]

You're a funny guy Carl!

 

You will never meet someone my age, more into film then I am. But that's beside the point.

 

Kodak has done a great job producing some fine new stocks. Camera companies have spent boat loads of money making new cameras. Lab's have re-worked gates and developed their own telecine/scanning solutions. A lot of money has been put into this format and the image looks great! However, there is one nagging defect that keeps it from taking that next big leap.

 

I'm a film guy and I want my camera negative to be perfect, not need cleanup work in post production. The format will never progress with an inherent problem like this existing and the only solution is transferring it to digital formats for stabilization. We need a stock and cameras which are made better and don't have these problems. Then anyone, anywhere who transfers Super 8 to digital or even blows it up to 16/35, won't see these problems ever again.

 

My frustration is that fixing things digitally doesn't solve any problems. Re-inventing the format, doesn't either. The solution is simple, it's small scale and it's something that will keep the format exactly what it is today, without these problems.

[Carl Looper]

You're a funny guy Carl!

 

You will never meet someone my age, more into film then I am. But that's beside the point.

 

Kodak has done a great job producing some fine new stocks. Camera companies have spent boat loads of money making new cameras. Lab's have re-worked gates and developed their own telecine/scanning solutions. A lot of money has been put into this format and the image looks great! However, there is one nagging defect that keeps it from taking that next big leap.

 

I'm a film guy and I want my camera negative to be perfect, not need cleanup work in post production. The format will never progress with an inherent problem like this existing and the only solution is transferring it to digital formats for stabilization. We need a stock and cameras which are made better and don't have these problems. Then anyone, anywhere who transfers Super 8 to digital or even blows it up to 16/35, won't see these problems ever again.

 

My frustration is that fixing things digitally doesn't solve any problems. Re-inventing the format, doesn't either. The solution is simple, it's small scale and it's something that will keep the format exactly what it is today, without these problems.

 

Hi Tyler - I know you love film - I read your posts :)

 

Myself - I have two agendas here.

 

One is that I'm building a Super8 to 16mm optical printer, in which I'll be using mechanical means to do the registration. But I'm also using digital means to assess the mechanical system and make requisite adjustments to the mechanical system, and elaborate digital control of the mechanical system (using the same machine vision algorithms used by scanners, After Effects, etc) because I dig both systems: the mechanical and the digital. And I'm also a software developer so I can rewrite the software till the cows come home.

 

Now I also treat scanning as that which is required to distribute material otherwise shot on film, across digital channels. So for me a scanner needs to have at least a generic registration system at least as good as traditional Super8 projectors. But I don't want the scanner developers pulling their hair out. If they can't do it, that's fine - I'll just write my own generic software to do the registration. It's not a big deal. But if the scanner can do it - then I don't have to do it.

 

I just don't know yet (conclusively) that the scanner side of the equation is at least as good as a traditional Super8 projector. I've watched tons of Super8 (on a projector) and it just seems to me scanner registration is not yet quite right yet, ie. in terms of what a traditional Super8 projector would deliver. But maybe I'm wrong. It's just a question of finding out.

 

So far I can't convince myself that the "sloppy" perf weave is actually producing any registration problem (eg. causing camera gate weave).

 

I can't yet conclude that this is a "nagging defect" or "inherent problem".

 

I just think there's some work to be done in fine tuning the scanner. The scanner can be understood as the digital equivalent of a film projector - for the screening of films over digital channels. It's not a "fix it in post" machine. A film sitting on a reel is not a film. A film is what is up on the screen, and the scanner is one machine for putting the film up there on the screen. If the scanner putting a film up on a screen is a "fix it in post" machine then so too are film projectors.

 

As a "fix it in post" machine one problem a film projector can be understood as solving is the problem of a film sitting in a can, where nobody can see it. Its solves that problem. But in doing so it must also solve registration.

 

C

Edited August 25, 2015 by Carl Looper

[Tyler Purcell]

Videotape your projector with a super 8 shot that's locked on a tripod.

 

I've done this and it wobbles all over the place. Mind you, I only have one super 8 projector right now, an Elmo ST180, but I rebuilt it recently and everything is in tip top shape. I've done the same test with my Kodak 16mm projector from WWII and the rocking chair effect was nonexistent. Sure, it had registration problems, but what projector doesn't?

 

That was a rhetorical question… :)

[Carl Looper]

Videotape your projector with a super 8 shot that's locked on a tripod.

 

That's a really good idea. To avoid perceptual bias could also transfer the same roll via a scanner - and then would have a common frame of reference to compare the pipelines.

 

C

Edited August 25, 2015 by Carl Looper

[Tyler Purcell]

Bingo! :)