Hyper-8mm?

[Cranium]

Hmmm. I know there was a lot of talk of 29.97 crystal for the 6008 years ago, again amongst the music vid and skate folks.
Anyhoo...
Really take a look at some of the stuff shot at 32fps (or as close as you can get to the 30). The 18fps does look good (that's how I've been doing my timelapse projects) but I'd shoot a few rolls with pans and such ~30fps before hacking anything, let alone your Fujica. My concern is it will be subject to similar strobing that you find with progressive video cameras.
It reminds me of a totally impractical idea I once had for low-light (and other effects) filming. An optical coupler, basically the complete OPPOSITE of a Bolex stereo adapter. One lens, beam-split to two c-mounts. Mounts are connected to two identical cameras, crystal sync'd together, but 180deg out of shutter phase. They'd have the exact same imaging, and you could run at 12 fps for 1/25 to 1/20s exposure time, and stick the two sets of images back together in post. Plus, you'd get 5 minutes run time out of 50' carts.
I'd still like to try it, just to see if all of the failure points I foresee would really be bad:
invariably mismatched emulsions.
frame floating
slight exposure differences
misaligned transfer
the fact that I've never seen a "coupler" like this
need syncable cameras, and they'd need to be modified to share a crystal.
I finally decided it's a neat idea. Then I decided 16mm would likely be the way to go at this point!
ah well... another one for the Rubes.
blah blah blah...

[MovieStuff]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:
I'm not sure I agree about the increased resolution.(I realize you wrote 'perceived')</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Hi, Chaz!

Actually, "perceived" is really incorrect. There is definately an increase in resolution if you shoot at 30fps instead of 24. Look at it like this: A single super 8 frame of a sign with small type will be harder to read than a shot of the same sign running at 18fps. This is because there is "accumulative resolution" that is gained by the overlapping of the random grain patterns of each frame. Therefore, the further away from "still frame" you get, the more frames you have overlapping and the more resolution you gain. I think that the reason I didn't see a huge increase in resolution on the 16mm stuff shot at 30fps is because that format already has a huge gain by the sheer size of the frame. However it would seem that super 8 can gain quite a bit of resolution by going to 30fps. At any rate, there was definately an increase in resolution that I saw on the 30fps footage playing back at 30fps compared to the 18fps footage playing back at 18fps.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:

As I am sure you know, when you are capturing, your video camera captures 60 fields per second (2 fields per frame), unless the camera is capable of, and is set to PROGRESSIVE SCAN. So each 'discrete frame' is still captured as 2 fields, regardless if it was shot at 30 or 18fps.</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Yes but that has no bearing here since nothing moved from one field to the next. The WorkPrinter and the DV8 transfer frame by frame, so there is no interlacing artifacts withing each frame. Therefore you get full resolution.
<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:

But, if you capture in progressive scan mode, you WILL realize a sharper image</font></td></tr></table></td></tr></table></BLOCKQUOTE>

This is not really totally true. If you lock your video camera on a tripod and shoot interlaced footage of photo or painting on the wall, it will look NO different than video shot in progressive scan. Why? Because nothing moved from one field to the next. Therefore, even though the frame is made up of two fields scanned sequentially, there are no motion artifacts or loss of resolution as a result of the interlacing. Therefore, since the DV8 and the WorkPrinter captures frame by frame and not in "real time", each frame of film exists as a complete, full resolution image on each video frame, even though the information is spread across two sequential fields.

That is why transferring 30fps footage one-to-one on the WorkPrinter, Rank or DV8 is going to have higher resolution than 24fps film footage spread across 30 frames of video. Not only do you have less "accumulative resolution" with 24fps, but there are going to be frames of video that have pieces of two film frames; neither of which offer full resolution since one field carries a piece of one film frame and the other field carries a piece of the other film frame. You lose "frame discretion" due to the interlacing of the artificial frames generated by the Rank during the 3:2 pulldown process. Shooting and transferring at 30fps totally avoids that issue.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:

But I will mess around with some footage I shot at 32fps, and see what all the fuss is about. I can capture prog scan, but my DVD player and TV are not PS</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Well, again, progressive scan does nothing to help the process here since each film frame being captured is stationary. The only interlacing that is visible is what takes you from one film frame to the next film frame, which will be the same, even if you capture in progressive scan since you will be displaying via standard interlace.

BTW: I checked my emails but I didn't see anything from you. I've had some email hic-ups as of late, so your email could be one of the victims.

Roger

[chaz]

Roger replied: There is definately an increase in resolution if you shoot at 30fps instead of 24. Look at it like this: A single super 8 frame of a sign with small type will be harder to read than a shot of the same sign running at 18fps. This is because there is "accumulative resolution" that is gained by the overlapping of the random grain patterns of each frame. Therefore, the further away from "still frame" you get, the more frames you have overlapping and the more resolution you gain.&lt;&lt;

chaz:Hadn't thought of that. I wouldn't have imagined that it would be noticable, but aparently, it is. I wonder at what point there is no more resolution gained, if one could theoretically shoot successfully at, say, 100, 200, 500 fps?

R: Yes but that has no bearing here since nothing moved from one field to the next. The WorkPrinter and the DV8 transfer frame by frame, so there is no interlacing artifacts withing each frame. Therefore you get full resolution.&lt;&lt;

C: Not sure I agree. You are still capturing interlaced, even if there is no motion. So field one is capturing the odd lines, and field two is capturing the even lines. Upon playback, the process is repeated, and only 1 field is displayed at a time. That's the whole point of progressive scanning, to capture, as well as display both fields at the same time.

&gt;&gt;Originally posted by chaz:

But, if you capture in progressive scan mode, you WILL realize a sharper image

R:This is not really totally true. If you lock your video camera on a tripod and shoot interlaced footage of photo or painting on the wall, it will look NO different than video shot in progressive scan.

C: I did some interlaced/progressive testing prior to my first film capturing. I aimed my Sony at a still, and captured an image, both in interlaced, and progressive. Upon blowing up the image, I found the progressive to have more detail. I also notice a sharper image when shooting in progressive mode. Unfortunately, my Sony only does 15fps progressive, so normal shooting is not usable. But for workprinter-type film capturing, the images are sharper.

R: Well, again, progressive scan does nothing to help the process here since each film frame being captured is stationary. The only interlacing that is visible is what takes you from one film frame to the next film frame, which will be the same, even if you capture in progressive scan since you will be displaying via standard interlace.

C: Technically, it doesn't. But a sharper image will be displayed if the image is captured progressive, and played back to a progressive monitor. Because all the lines are displayed at the same time, not consecutively. That's why High Definition is so good. A partial quote from Joe Kane's 'Video Essentials' page:

20 August 2001
The distribution of Video Essentials (DVD) will draw to a close nearly four years after it was brought to market in October of 1997. It will be making way for a new generation of DVD that uses 1080p/24 High Definition source material and "True Progressive Mastering".

C: It will probably take a few years for HD formats to settle, and for cameras to become affordable for us mortals, but can you imagine a Workprinter hooked up to a 1080p/24 camera, and delivered to a matching 1080p/24 jumbo TV? Good times indeed are ahead of us, my friends.

R: BTW: I checked my emails but I didn't see anything from you. I've had some email hic-ups as of late, so your email could be one of the victims.

C: I am still interested in a condenser lense. I haven't emailed, but will probably order one in a few months, when my normal workload slows. I am looking forward to capturing the remainder of my old family home movies. I will capture progressive, edit and master to 'Frame Based', at 24fps, (films were originally shot 18 fps, but DVD players can do 24fps, so I will encode to MPEG at 24), and burn to DVD. Just got the Pioneer DVR-104.

chaz

[mattias]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">That's the whole point of progressive scanning, to capture, as well as display both fields at the same time.</font></td></tr></table></td></tr></table></BLOCKQUOTE>

when you "display" your footage it will always be shown interlaced on an interlaced display and non interlaced on a non interlaced one, no matter how you shot it. this means that progessive scan and interlaced scan footage (without motion) will look exacly the same if both are shown on the same display, interlaced or not. how the ccd is scanned and how the signal is stored has nothing to do with it. you could use "spiral scanning" and encoding and a static scene would still look exactly the same.

/matt

[crimsonson]

SOrry I have to disagree with Roger about the the quality of progressive vs interlace scanning of a still image.

because:

1. Cameras have built in filtering and image enhancing algorithms. This is to reduce line artifacts (jaggies, saw tooth). it is always on. It is adaptive but nevertheless does not prevent a slight degregation of image quality.
With a true progressive camera - the filtering are not necessary.

2. Inter-line flicker: moving or not when a horizontal line occupies two fields and there is a vast contrast difference - the image will appear to shake (jitter). This is worst when showed in a non interlaced monitor (aka regular TV).


That being said how fast is the difference? It may no be the stark when viewed in a 320 TV lines but in CRTs and broadcast monitors it is more obvious.

[MovieStuff]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by crimsonson:

1. Cameras have built in filtering and image enhancing algorithms. This is to reduce line artifacts (jaggies, saw tooth). it is always on. It is adaptive but nevertheless does not prevent a slight degregation of image quality.
With a true progressive camera - the filtering are not necessary.</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Theoretically not necessary but that doesn't mean it isn't present in a true progressive scan camera. ALL cameras have some sort of image enhancement, whether progressive scan or interlaced. Again, people with different cameras have tried the progressive vs interlaced test of a stationary subject and have found the progressive image to be softer, so my point is that -if image processing and number of pixels being used are the same- there should be no perceptible difference in an image that is displayed interlaced whether it originated via interlaced or progressive scan.

The differences you speak of (if displayed interlaced) have nothing to do with the actual technique of capturing still images interlaced vs progressive scan but, rather, what particular features/image enhancement a particular camera might have.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by crimsonson:

2. Inter-line flicker: moving or not when a horizontal line occupies two fields and there is a vast contrast difference - the image will appear to shake (jitter). This is worst when showed in a non interlaced monitor (aka regular TV).</font></td></tr></table></td></tr></table></BLOCKQUOTE>

If the number of lines being used is the same, the the jitter will be present on interlaced display, even if the original image was created using progressive scan. Again, once it is in the computer, there is no "interlacing". The computer has no idea if the horizontal line you speak of was scanned in two parts (fields) and then combined into a single image or if the horizontal line was scanned progressively to form the same image. The interlacing is applied only for display on NTSC video; it is not an inherent part of the image stucture.

Now, if the progressive scan has more resolution and that increased resolution were displayed progressively, then I would agree. But if the image is being displayed interlaced and the same number of lines were used to capture the image for both interlaced or progressive, the only difference you might see is the particular camera's own unique image processing characteristics, as illustrated by Chas's experiment with both interlaced and progressive scanned still images. I mean, it's the same camera with the same number of lines being scanned, right? It can ONLY be the way the image is processed internally since the final image is being viewed interlaced in the end.

But all this is really academic as it relates to the original discussion. The bottom line is that if you capture each frame of film (progressive or interlaced) and display it at 30fps with no pulldown that would interpolate and interlace frames, then you get a smoother, higher resolution picture than stretching 24fps across 30fps by creating 6 artificial frames; none of which constitute a "complete" frame with full resolution.

The specific ins and outs of one camera's ability to produce better images in progressive vs another model's ability to do the same via interlace is a totally different discussion that has more to do with idiosynchrocies of design than conceptual technique, I think, if the target is non-moving and the camera locked off. Again, if all image processing and the number of pixels and lines being used are the same, an interlaced display of both images will look the same.

Roger

[MovieStuff]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:
I wonder at what point there is no more resolution gained, if one could theoretically shoot successfully at, say, 100, 200, 500 fps?</font></td></tr></table></td></tr></table></BLOCKQUOTE>

I think that 30fps for super 8 is probably about it and still have it look like film. Again, if you shot at 60fps and projected at 60fps or spread those frames across 60 fields per second, it would look like video, motion wise. 30fps 16mm doesn't get quite as dramatic a "boost" as 30fps super 8, I'm guessing because the grain on 16mm is so fine to begin with. Any gain in resolution in super 8 is more apparent since we're so used to seeing so little, compared to larger formats. I swear, the 30fps S8 I've been looking at looked just like a 16mm print telecined.

I had previously had written:

"Yes but that has no bearing here since nothing moved from one field to the next. The WorkPrinter and the DV8 transfer frame by frame, so there is no interlacing artifacts withing each frame. Therefore you get full resolution."

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:
Not sure I agree. You are still capturing interlaced, even if there is no motion. So field one is capturing the odd lines, and field two is capturing the even lines. Upon playback, the process is repeated, and only 1 field is displayed at a time. That's the whole point of progressive scanning, to capture, as well as display both fields at the same time.</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Ah, but you really don't! Progressive scan doesn't happen all at once. It scans down from the top to the bottom and a new scan replaces the old one in the same pattern. So there really are "fields" in progressive scan. They follow each other instead of being interlaced; but they are fields none the less and the total duration of a single field at 24P is longer than the duration required for two interlaced fields at 30 frames (60 fields) per second. If the subject doesn't move between interlaced fields, there is no practical visual difference between a progressive scanned frame and an interlaced frame. The scan of the two images is so fast that the eye will see them practically the same.

I'm not saying that progressive scan is the same as interlaced when shooting a live moving image; that's a huge difference. I am saying that, if the end result is displayed via interlaced fields, then it makes no difference if the original frames were interlaced or progressive scanned if the target was a still frame of motion picture film where nothing moved during "exposure" of the two fields. Computer edit programs don't operate interlaced. The images are merely solid frames made up of pixels and the computer doesn't know if every other line of pixels was originated via progressive scan or interlaced. They're just "frames". The interlacing is applied AFTER the fact by the display device.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:
I did some interlaced/progressive testing prior to my first film capturing. I aimed my Sony at a still, and captured an image, both in interlaced, and progressive. Upon blowing up the image, I found the progressive to have more detail. </font></td></tr></table></td></tr></table></BLOCKQUOTE>

I don't doubt it but I have little belief that it really had anything to do with using progressive scan. In other words, it isn't that the progressive scan is sharper as much as it is the interlaced isn't as sharp as it should be. I've known people that have done the same experiment and have seen either no difference or have found the progressive frames to be NOT as sharp. So I think your success has more to do with the design of your camera than anything inherent in the progressive can mode if the subject isn't moving and the same number of pixels are used for both progressive scan and interlaced. They really should not look any different once they are in the computer, since the computer doesn't "see" interlaced or progressive as being any different from each other.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:
a sharper image will be displayed if the image is captured progressive, and played back to a progressive monitor. Because all the lines are displayed at the same time, not consecutively. </font></td></tr></table></td></tr></table></BLOCKQUOTE>

Agreed! Progressive shooting and display is better but, again, the lines are NOT displayed all at once. They are, in fact, displayed consecutively in progressive from top to bottom. The lines are not displayed all at once and the time it takes to display a complete 24P frame is longer than it takes to display two interlaced fields of NTSC video.

If a subject moved during display of the two fields, then that would make a huge difference. But that's not what we're talking about here. Again, once it's in the computer, there is no "interlacing". It's just a bunch of still images being knocked about. The interlacing is applied simply to accomodate the needs of NTSC display.
Look at it this way: Let's say you set up your camera on a tripod and shot progressive scan footage of a photo on the wall. Then, in Photoshop, you removed all the odd lines and saved that file. On a copy you removed all the even lines and saved that as a file. Now you have odd and even fields taken from a single image. If you then combined those two images together, they would create a whole image again. Now, the funny thing is that, if you had started with footage of a still photo that had been shot interlaced, it would take the same steps to separate the image into two fields. The reason? Computers don't see video frames (or any other images) as having "fields". They're just a bunch of still frames (progressive while on the computer!) that are interlaced only when displayed on an NTSC monitor.

I think your progressive scanned frames look better simply because your camera is designed that way; not because copying still frames via progressive scan will inherently increase the resolution. There really shouldn't be a bit of difference. I believe you that there is, but there really shouldn't be if the same number of pixels are being used to make up the final image in the computer.

Roger

<font color="#a62a2a" size="1">[ August 07, 2002 08:25 AM: Message edited by: MovieStuff ]</font>

[chaz]

Interesting food for thought, and more research. I suspected my Sony's electronics had something to do with the sharpness increase during prog scan. As I understand it, the Canon cams that do progressive can do 30fps, but mess around with the image electronically, and the user was not pleased with the results.

One last point, if we are talking about an increase in rez at 30 fps film vs. 24, (or 18), and are getting down to the grain, do you suppose the film moves any, at all, from the time it takes to capture one field to the next? Surly the film can't be perfectly still, can it? Even the vibration of the projector mechanism must introduce a small amount of motion? Understand, I am not trying to be a wise guy, just an honest question. When I was adjusting my homemade 'workprinter', I noticed that some of my captures were blurry, as the projector was in the process of moving the film, or the film hadn't stopped yet. An adjustment to the location of the limit switch corrected this.

I need to bone up on this subject more. I am wondering about the timing of the capture of fields. I mean, are both fields captured 'boom-boom, then pause, then boom-boom, or is it an even pattern, boom-boom-boom-boom, etc?

chaz

[MovieStuff]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:
do you suppose the film moves any, at all, from the time it takes to capture one field to the next? Surly the film can't be perfectly still, can it? </font></td></tr></table></td></tr></table></BLOCKQUOTE>

If everything is working correctly it should be rock steady.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:
When I was adjusting my homemade 'workprinter', I noticed that some of my captures were blurry, as the projector was in the process of moving the film, or the film hadn't stopped yet. An adjustment to the location of the limit switch corrected this.</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Relative to the time it takes the video camera to scan, even film running at 6fps in a WorkPrinter (which has no shutter) will have a dwell time that is an eternity. The window of opportunity for the frame capture to be made is quite large. The basic WorkPrinter that operates at only 1fps obviously has a HUGE dwell time. In either case, there should be zero film movement if the advance is working properly.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by chaz:

I need to bone up on this subject more. I am wondering about the timing of the capture of fields. I mean, are both fields captured 'boom-boom, then pause, then boom-boom, or is it an even pattern, boom-boom-boom-boom, etc?</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Boom-boom-boom-boom.

One field follows the other at the same rate, same pattern of top to bottom. The only reason that you don't "see" the change over is due to the phosphors of the CRT holding the first field while the second field scans into place. This is another reason I maintain that shooting an interlaced image of a stationary subject will not look any different than progressive scan if the final image is displayed interlaced. There is never a time when the two fields are NOT up at the same time since the phosphors hold the first field long enough for it to be seen along with the second field. In essence, the display will provide a complete picture of both fields long enough for the eye to "see" it as a complete image before changing to the next pair of fields.

Furthermore, if the image is watched on an LCD monitor, there is no interlacing at all since it has an active matrix that only changes the pixels required to show motion. That is why you can shoot off an LCD panel with a motion picture camera and not get flicker. Video on an LCD monitor often looks more like progressive scan than interlaced video for this reason.

Did you do a transfer test of some 30fps footage yet? Curious how it came out.

Roger

[chaz]

Roger stated: Did you do a transfer test of some 30fps footage yet? Curious how it came out.

No, I have a couple of video projects to get out the door before I fire up the capture rig. I do have the footage on DV, but unfortunately I rendered all to 30fps from 18/32fps. So the stuff shot at 32fps is slomo, and the 18 fps plays normal. Lots of extra frames. (I could use TMPG to remove the redundant frames and render to mpeg, but that would add an extra element into the equation.)

I did this before I realised I may want to save all footage to DV tape at its original, 1 film image per 1 video frame. I now save it both ways. I will probably render the 18fps stuff to a 24fps avi, and encode to mpeg to burn to disc, in addition to 29.97 avi for DV tape backup.

But I am eagerly anticipating the look of the 32fps stuff played back at 30. I doubt that the speed change will be very noticeable playing back 32fps at 30. I will be able to make a good comparison, as I shot footage at the beach using both speeds with the same roll, and the k-40 looks stunning. Brilliant colors, nice colorful bathing suits (sorry, no bikinis, just nieces & nephews) and a blue sky with some clouds.

chaz