anamorphic lenses take 2 [Archive]

View Full Version : anamorphic lenses take 2

Ruairi Robinson

03-21-2007, 05:31 AM

Hey, I'm sure I'm not the only one who would love to be able to use anamorphic lenses with the red camera (particularly panavision C and E series if the camera gets panavised some day!)

I know it's been talked about before, but I'd love to hear a more detailed explanation for exactly why using anamorphic lenses produces a 3.3:1 ratio image (approx) on a 16x9 sensor area (or on a super16mm camera)

And is would it not be possible to somehow build a mount that either maps the image to fit the full 16x9 sensor, or a 4x3 area of the sensor... or some other more *useful* use of the available screen space...

I hate the idea of having a 3.3:1 image, and having to crop a load off the sides AS WELL as the black bars top and bottom to get a 2.35:1 ratio. Such a waste of resolution...

Is it possible to build a mount with a corrective lens in between or something? that RE/DE-anamorphizes the image? Would you lose clarity/stops etc... Is it a waste of time? Does what I'm asking even make sense? Apologies if I'm talking gibberish here - I'm a director, not a DOP!

The thing is - myself, and I believe a LOT of other people, really really LOVE the look of anamorphic lenses. That particular bokeh stretch, etc. etc. etc. I'd hate to lose that look. I know anamorphics have shoddy focusing, need more light, are heavy, unwieldy, all the rest of it - I don't care. They look fucking great, artifacts and all. That's why they are still so widely used.

If there's any way to make these kind of lenses *useful* on the red camera, I wanna know about it.

Look forward to hearing peoples thoughts.

R.

Evin Grant

03-21-2007, 05:53 AM

IMHO the only viable option is a lighter set of anamorphic lense specifically for 16:9 imagers.

Mark L. Pederson

03-21-2007, 06:08 AM

IMHO the only viable option is a lighter set of anamorphic lense specifically for 16:9 imagers.

I'd put down a deposit on that tomorrow!!

Dominic Jones

03-21-2007, 06:19 AM

Hmmm... Well....

The idea of 1.5:1 anamorphics to suit 16:9 chip systems is nice, and would preserve full resolution, etc, etc, but they won't give you the exact same look as 2:1 optics. To my knowledge there simply aren't any 1.5:1 lenses around, so how much they would differ is a matter of speculation. I for one would love to have a chance to try some out, however.

That said there's a few points I'd like to pick up on in Ruairi's post though, so I'll try to get this into some kind of meaningful order (but I'll probably fail!):

Yes, it probably would be possible (assuming there's room between the rear lens element and the imager, or you don't mind doing the conversion at the front of the lens) to build an adapter that "corrects" the 2:1 optical image for 16:9. You would certainly lose some resolution, and probably some light, and you'd have the issue of perhaps not having such an anamorphic look to the final images as mentioned above. All in all, forget it, would be my advice.

That said, you say you hate the idea of cropping down the sides and having the "black bars" top and bottom. What is your delivery format??

If you're going to theatrical distribution (film or digital) you don't have "black bars" at the top and bottom, you fill the screen vertically and optically stretch the image to fit the wider horizontal ratio. So you just lose the horizontal rez of the crop. Not so bad.

If you're going to DVD/TV/HD-DVD/Blu-Ray/Any other non-theatrical distribution route (internet?!), then you've got way more resolution than you need to fill the screen, so it doesn't matter that you've got to crop and letterbox - you're still getting as much resolution as you possibly can from the delivery format, regardless of what you shot it on - that's where the bottleneck is.

I'm not saying people should rush out and shoot anamorphic, btw, just that if you're cropping to a 'scope ratio, a horizontal crop with anamorphics doesn't lose you much more than a vertical crop with sphericals (a little, but not enough to make that much odds)...

Ruairi Robinson

03-21-2007, 06:40 AM

The idea of 1.5:1 anamorphics to suit 16:9 chip systems is nice, and would preserve full resolution, etc, etc, but they won't give you the exact same look as 2:1 optics. To my knowledge there simply aren't any 1.5:1 lenses around, so how much they would differ is a matter of speculation. I for one would love to have a chance to try some out, however.

Preserving full resolution is not SUCH a big deal if we have a 4k imager to start with. Anything 3k and over is going to look pretty good on a cinema screen. (2k, and in particular HD always looked a little soft to me in the cinema)

And as you say, you won't get the same look with 1.5:1 anamorphics. People don't use anamorphic lenses for FUN - they use them because they like the specific look. (or they get a bigger/cleaner negative area on 4x3 35mm film area, which is irrelevant when we are talking about the red camera, and a 16x9 chip!)

Yes, it probably would be possible (assuming there's room between the rear lens element and the imager, or you don't mind doing the conversion at the front of the lens) to build an adapter that "corrects" the 2:1 optical image for 16:9. You would certainly lose some resolution, and probably some light, and you'd have the issue of perhaps not having such an anamorphic look to the final images as mentioned above. All in all, forget it, would be my advice.

Well... that's what I wondered. Since you have extra glass for the anamorphic element anyway, it sort of sounded stupidly innefficient. Can't hurt to ask though...

That said, you say you hate the idea of cropping down the sides and having the "black bars" top and bottom. What is your delivery format??

35mm film for theatrical distribution, HD, pal, ntsc. All of them. But theatrical to start with.

If you're going to theatrical distribution (film or digital) you don't have "black bars" at the top and bottom, you fill the screen vertically and optically stretch the image to fit the wider horizontal ratio. So you just lose the horizontal rez of the crop. Not so bad.

You are of course correct been a while since I dealt with stuff shot super16/scope so I'd forgotten exactly how it works... Still, extra horizontal info ain't much use to anyone since scope is so wide anyway (if it was vertical at least you could use the extra image for reframing!)

If you're going to DVD/TV/HD-DVD/Blu-Ray/Any other non-theatrical distribution route (internet?!), then you've got way more resolution than you need to fill the screen, so it doesn't matter that you've got to crop and letterbox - you're still getting as much resolution as you possibly can from the delivery format, regardless of what you shot it on - that's where the bottleneck is.

That's true, but it's still stupidly innefficient, and there are no benefits to working this way I can think of, apart from preserving the look.

I'm not saying people should rush out and shoot anamorphic, btw, just that if you're cropping to a 'scope ratio, a horizontal crop with anamorphics doesn't lose you much more than a vertical crop with sphericals (a little, but not enough to make that much odds)...

Would it be possible for someone to build a lens with the lens properties of 2:1 optics, but for a 16x9 sensor? To have the same optical characteristics as the Panavision style c/E series scope lenses? for example - the 2:1 vertical bokeh stretch, and the extremely narrow DOF, and I suppose the coatings that create that unique flare... etc.

to be clear - nobody gives a flying fuck any more about the "advantages" of shooting scope to get a bigger negative area for widescreen photography, etc.

It's ALL about the look.

R.

Dominic Jones

03-21-2007, 06:59 AM

That's true, but it's still stupidly innefficient, and there are no benefits to working this way I can think of, apart from preserving the look.
Well, sure - but (a)that's the highest resolution of the delivery medium, so it's only even more inefficient to start with a higher resolution frame and (b)...

to be clear - nobody gives a flying fuck any more about the "advantages" of shooting scope to get a bigger negative area for widescreen photography, etc.

It's ALL about the look.
That's my point exactly!!! :biggrin:

You are of course correct been a while since I dealt with stuff shot super16/scope so I'd forgotten exactly how it works... Still, extra horizontal info ain't much use to anyone since scope is so wide anyway (if it was vertical at least you could use the extra image for reframing!)
You could quite happily use the extra horizontal information to reframe horizontally, so you still have that advantage, as with a vertical crop (but it a different axis, of course)...

Would it be possible for someone to build a lens with the lens properties of 2:1 optics, but for a 16x9 sensor? To have the same optical characteristics as the Panavision style c/E series scope lenses? for example - the 2:1 vertical bokeh stretch, and the extremely narrow DOF, and I suppose the coatings that create that unique flare... etc.
Well, the coating could of course be duplicated, but all of the other properties are due to the nature of the 2:1 optics - so with 1.5:1's you'd get all of those properties, but reduced in magnitude. As I said before, quite how much that will affect the overall look the lenses produce is a matter of speculation unless or until someone builds some...

Ruairi Robinson

03-21-2007, 07:08 AM

Well, you could quite happily use the extra horizontal information to reframe horizontally, so you still have that advantage, as with a vertical crop...

Sure, but obviously extra headspace vertically is much more useful (especially if you are framing for multiple formats, etc. And if there's a kick in the camera move that you have to iron out by stabilising etc. extra vertical headspace is again much more helpful...

R.

Ruairi Robinson

03-21-2007, 07:10 AM

http://www.vantagefilm.com/en/news/index_28.shtml

Vantage film have a lens filter that recreates that horizontal blue flare... haven't tested it though

Dominic Jones

03-21-2007, 07:21 AM

Sure, but obviously extra headspace vertically is much more useful (especially if you are framing for multiple formats, etc. And if there's a kick in the camera move that you have to iron out by stabilising etc. extra vertical headspace is again much more helpful...

R.
Yeah, vertical headroom is more handy, certainly - just pointing out the possibilities!!

I've read about various streak and anamorphic "effect" filters, but never used any. I mentioned in another one of the anamorphic debate threads that I'm slated to shoot a feature on HD for a vertical crop to 2.35:1 this summer, so if that goes ahead I may well test some or all of the available filters. I'll report back if I do...

Of course, they're not going to simulate things like the discrepancy in horizontal and vertical DoF, and the distinctive Bokeh of anamorphic optics, so they're not going to get you all the way there...

Ruairi Robinson

03-21-2007, 07:27 AM

Yeah, vertical headroom is more handy, certainly - just pointing out the possibilities!!

I've read about various streak and anamorphic "effect" filters, but never used any. I mentioned in another one of the anamorphic debate threads that I'm slated to shoot a feature on HD for a vertical crop to 2.35:1 this summer, so if that goes ahead I may well test some or all of the available filters. I'll report back if I do...

Of course, they're not going to simulate things like the discrepancy in horizontal and vertical DoF, and the distinctive Bokeh of anamorphic optics, so they're not going to get you all the way there...

Yeah, I mention the filter only for the sake of discussion. To be hoest, the blue flare thing is about the least interesting of the artifacts of anamorphic lenses as far as I'm concerned. They are painfully overused to "jazz up" shitty ads on tv far too much...

But I'm very fond of the way DOF looks on close ups... What can I say - I'm a fan of Blade Runner...

Dominic Jones

03-21-2007, 07:31 AM

What can I say - I'm a fan of Blade Runner...
Oh, ho ho hooooo, you've hit the mother-load there!!

If I'm ever sat in the pub with a bunch of mates and some poor, unsuspecting soul mentions Blade Runner, you can literally hear the groans of "Don't get him started" half way down the street!!!

What a beautifully shot film. That's it. I'll leave it there...!
:biggrin:

Paris Remillard

03-21-2007, 07:49 AM

Check out Babel. Prieto shot the Japan section, minus a couple of low light shots, on 'Scope. But since the film was released as 1.85, he just cropped off the sides. They wanted the look of scope for the same reasons you mention, oval highlights, flares, DOF, but not the full width of the frame. There was an article about it in American Cinematographer a couple of months ago.
Anyway, you've said that rez doesn't matter that much at these levels(of resolution) anyway, so, you'll certainly get the look if you just crop off the sides.

Martin Drew

03-21-2007, 07:51 AM

I have drawn up a graphic to show how the different ratios work out when you window a 2:1 anamorphic lens. Hopefully that will make it a bit clearer.

http://www.reduser.net/forum/uploaded/59_1174488521.gif

So at 2540p, active pixels would be 2540 x 3035 for anamorphic 2.39:1 ratio
and at 4K, active pixels would be 2304 x 2753 for anamorphic 2.39:1 ratio

(By the way why I am doing this and not working?)

M

Dominic Jones

03-21-2007, 08:09 AM

Nice images Martin, good work (or lack of it!!)...

Martin Drew

03-21-2007, 08:34 AM

Given that at 4K, the active pixels would be 2304 x 2753 for anamorphic 2.39:1 ratio. There is an argument for saying the quality would be sufficient for outputing at 2K. You will lose the wide end of the anamorphic lens though by a whopping 1.4x

Martin Drew

03-21-2007, 08:37 AM

(or lack of it!!)...

Yeah. Really got to knuckle down now!! Keep getting distracted though :whistling:

M

Ryan Manes

03-21-2007, 09:36 AM

Didn't Fincher use anamorphic Lenses on the Viper for Zodiac?

John Allardice

03-21-2007, 10:03 AM

Martin Drew

03-21-2007, 10:29 AM

Nope, the viper has the facility to re-map its pixel array to fill a 16x9 frame with a 2.35 image. Its essentially the same as shooting spherical, but you don't loose the 280-odd lines of vertical resolution by cropping to achieve your aspect ratio.

J

But of course that is only possible because it has pixels to spare. When you crop on the Red you will be doing the same thing if you shoot 4k and your end target is 2K or 1080p

M

Ruairi Robinson

03-22-2007, 09:54 AM

But of course that is only possible because it has pixels to spare. When you crop on the Red you will be doing the same thing if you shoot 4k and your end target is 2K or 1080p

M

hmm... so if I was a big movie director doing a massive summer movie, and saying "I'll shoot on red if I can get anamorphic lenses working efficiently"

Would I be getting the same answers?

How much does it cost to build a bloody lens in the first place, give or take?

ALSO - we keep hearing that the sensor is replaceable/upgradeable - does that mean it would be possible to install a 4x3 sensor? (same size as the arriD20...!) so you can use the full frame of scope lenses PROPERLY...

Martin Drew

03-22-2007, 10:19 AM

Anything is possible... depends on how deep your pockets are. If you were JJ you would get your own sensor designed.

It should be straightforward to design and build an anamorphic adapter with the appropriate distortion to correct for the sensor mismatch either for using a spherical lens or an anamorphic lens but it isn't going to be cheap, probably makes more sense shooting on film in this case.

M

Dominic Jones

03-22-2007, 10:37 AM

It should be straightforward to design and build an anamorphic adapter with the appropriate distortion to correct for the sensor mismatch either for using a spherical lens or an anamorphic lens but it isn't going to be cheap, probably makes more sense shooting on film in this case.

It's also not going to be great for your optical path either...

The D20, whilst it has a high MP count sensor, scales output down to remove any artifacts from the colour filter array, so even with the crop you'll get more pixels with the Red than the Arri.

That said, the D20 will have sub-sampled pixels with full colour sampling and (presumably at least) less aliasing etc (i.e. all the advantages of sub-sampling pixels from a larger array), so it's not an apples-for-apples comparison.

But just FYI...

Stephen Williams

03-22-2007, 10:43 AM

But of course that is only possible because it has pixels to spare. When you crop on the Red you will be doing the same thing if you shoot 4k and your end target is 2K or 1080p

M

Hi Martin,

Not quite the Viper is 27.6 megapixel to start with! ( 3 x 9,200,00)

Stephen

Ruairi Robinson

03-22-2007, 10:49 AM

Anything is possible... depends on how deep your pockets are. If you were JJ you would get your own sensor designed.

M

Heh, I met him a couple of weeks ago :)

It should be straightforward to design and build an anamorphic adapter with the appropriate distortion to correct for the sensor mismatch either for using a spherical lens or an anamorphic lens but it isn't going to be cheap, probably makes more sense shooting on film in this case.

I'm pretty well sold on the all-digital workflow though... Won't miss dealing with stock/development and waiting for dailies... and grading can be all-digital too...

If red has a "good enough" colour/contrast ratio, and produces purrdy pictures, being able to shoot in scope is really the only barrier left as far as I'm concerned... though it's not a deal breaker or anything...

Matt Uhry

03-22-2007, 10:52 AM

Dominic Jones

03-22-2007, 10:56 AM

Ah, now that's interesting. I didn't know they'd made them for vista - what mount are they??

Be great to see some tests, even if the lenses aren't that great - just to see how much of the anamorphic "look" they retain at 1.5:1...

Martin Drew

03-22-2007, 11:17 AM

Not quite the Viper is 27.6 megapixel to start with! ( 3 x 9,200,00)

Not sure what you are disagreeing with here Stephen. As I said...pixels to spare. It still ends up at 1920 x 1080 though. Even taking into account the bayer res loses in the Red you should still have enough resolution to take full advantage of the 1080p format.

M

Martin Drew

03-22-2007, 11:21 AM

Hi Anamorphic Fans,
A friend of mine has some of the remains of Paramount's Camera dept. including a set of 1.5x anamorphics built for Vistascope

I found this. Sold in January on Ebay for $76

http://cgi.ebay.com/Anamorphic-lens-Vistascope-movie_W0QQitemZ260071572801QQihZ016QQcategoryZ4690 QQcmdZViewItem

Ruairi Robinson

03-22-2007, 12:14 PM

Not sure what you are disagreeing with here Stephen. As I said...pixels to spare. It still ends up at 1920 x 1080 though. Even taking into account the bayer res loses in the Red you should still have enough resolution to take full advantage of the 1080p format.

M

I'm concerned much less with the resolution loss, than with losing the sides of the frame on wider lenses... defeats the purpose of using a wide lens!

R.

Michael Hastings

03-22-2007, 01:14 PM

I have two anamorphic lenses made by Todd AO. They are listed as HISPEED ANAMORPHIC ZOOM made by TODD AO VIDEO. They are modified Fujinon 7x7 Wide angle TV zooms - it is the regular 7mmx7 with this pretty massive anamorphic adapter on it which is 130mm on the front and the back end goes over the 72 or 77mm front end of the Fujinon zooms. One is built on a Sony B4 mount and the other, which I partially dismantled, was on an Ikegami B3 mount.

They have their own film style focus mechanism on them. I have a few grand in them - original concept was to get anamorphic with the older non 16x9 video cameras. I don't have the time to really figure them out but if someone here is interested let me know and I will send pictures and we could strike a deal. It seems to me that they could be used either with the current lens or used to modify another lens - since the one I dismantled appears to work kind of like the anamorphic adapters for the DVX100 did.

I found this. Sold in January on Ebay for $76

http://cgi.ebay.com/Anamorphic-lens-Vistascope-movie_W0QQitemZ260071572801QQihZ016QQcategoryZ4690 QQcmdZViewItem

chuck colburn

03-22-2007, 01:44 PM

Squeeze me?

http://cgi.ebay.com/Set-of-3-LOMO-Anamorphic-lenses-ARRI-Moviecam_W0QQitemZ150103419712QQcategoryZ4691QQrdZ 1QQcmdZViewItem

http://cgi.ebay.com/Set-of-TOYO-anamorphic-lenses-ARRI-movie-camera-PL_W0QQitemZ150103418878QQcategoryZ4691QQrdZ1QQcmd ZViewItem

Matt Uhry

03-22-2007, 01:56 PM

That Vistascope lens is not a camera taking lens, it's either a (unlikely) front attachment for a spherical lens or (likely) a attachment for a projector. Still if you find some optics that squeeze 1.5 that could become parts for a system if you are so inclined.

Matt Uhry
www.mattuhry.com

Kenn Christenson

03-22-2007, 02:08 PM

I have seen 1.5x anamorphic adapters on ebay that were made for still photography.

You could always add this lens to the front of probably an 80mm prime -

http://cgi.ebay.com/Kowa-Prominar-Anamorphic-35-1-5x-Lens_W0QQitemZ110106128683QQcategoryZ4702QQrdZ1QQc mdZViewItem

I shot a 16mm short in a similar fashion, but I definitely don't recommend it.

Nick Shaw

03-23-2007, 10:14 AM

The D20, whilst it has a high MP count sensor, scales output down to remove any artifacts from the colour filter array, so even with the crop you'll get more pixels with the Red than the Arri.

That said, the D20 will have sub-sampled pixels with full colour sampling and (presumably at least) less aliasing etc (i.e. all the advantages of sub-sampling pixels from a larger array), so it's not an apples-for-apples comparison.

The D20 may downsample (1.5x in 'TV mode' to go from 2880 pixels wide to 1920) but that certainly does not eliminate artefacts. Attached is a crop of a grab from a test I did with a D20 last week, captured 10-bit uncompressed straight off the camera, not via tape. It shows very noticeable Bayer artefacts.

No Bayer sensor is immune to those artefacts, but hopefully with the RED doing a 2:1 downscale to give 1080p, they should be greatly minimised, as each Bayer cluster of four pixels will be scaled to a single 1080p pixel. At least that's my assumption.

Nick Shaw

03-23-2007, 10:22 AM

Following on from my previous post (sorry this is off topic for this thread, but I'm jumping off from the mention of the D20) it was interesting to use the D20, and get a look at the competition! The guy at Arri Media said they didn't see the RED as competition (time will tell on that one) but he applauded what Jim is doing.

I was pretty impressed with the D20. It was a VFX shoot and I was able to pull pretty good keys from the footage. Not as good however as the key I pulled from David Stump's greenscreen shot after I down-scaled that to 1080. I definitely believe the results would have been much better if I'd had my RED in time for the shoot.

Joshua Provost

03-23-2007, 10:49 AM

Not quite the Viper is 27.6 megapixel to start with! ( 3 x 9,200,00)

You don't count all three sensors in a 3CCD design to compute the resolution. It's 9.2 megapixel, 1CCD or 3CCD.

Michael Hastings

03-23-2007, 11:14 AM

If you do pixel offset, which most modern cameras do, you can get a 1.4x increase in resolution or a real effective 12.8 megapixel. P.S. It's not just interpolation. I don't know if the Viper uses pixel offset or not. anyone?

You don't count all three sensors in a 3CCD design to compute the resolution. It's 9.2 megapixel, 1CCD or 3CCD.

Martin Drew

03-27-2007, 03:51 AM

Hi Anamorphic Fans,
A friend of mine has some of the remains of Paramount's Camera dept. including a set of 1.5x anamorphics built for Vistascope (8perf horizontal, 1.5 squeeze) Productions. These 4 prime lenses are from the 50's and probably are more "interesting" and less "high performance", but I'll shoot some tests and we can see...

Matt.

It just ocurred to me, these lenses are designed for 8 perf horizontal so the gate size is going to be massive. You will only see a tiny portion of the centre of the image on Red sensor (about 50% of the original FOV), so you aren't going to get any wideangle shots.

M

Johan Lindgren

03-27-2007, 08:47 AM

Remember: No anamorphic front adapter will ever give you those streched bokeh out of focus ovals, that you are all after.

Stephen Williams

03-27-2007, 09:17 AM

You don't count all three sensors in a 3CCD design to compute the resolution. It's 9.2 megapixel, 1CCD or 3CCD.

Hi Joshua,

Why not? There is far mor data from a 9.2 megapixel RGB than from a 9.2 Bayer.

It will be interesting to compare a 4k RGB scan to a 4.5K Bayer sensor and see which has the most resoloution.

Stephen

Martin Drew

03-27-2007, 09:52 AM

Stephen you are right that there is going to be more resolution from three 9.2 megapixel panels than from one 9.2 mega bayer panel, but I am sure you would accept that you can't just multiply the number of pixels by the number of panels to give you a figure for resolution. Even if there is some spatial offset between the panels it still wouldn't give you 3 x the resolution.

And I reckon I could guess at an answer to your second question. The 4K RGB scan will probably have the upper hand. Graeme previously said that a bayer array can resolve at over 70% of the pixel count. So a 4.5K bayer should give you a minimum of 3.15k resolution. That said the 4K scan won't resolve 4K of course because it will require OLPF.

Getting back to the original thrust of the thread though 4.5K Bayer in the Red with a 2:1 anamorphic lens should have sufficient resolution (2540 x 3035) for downsampling to 2K , even allowing for the necessary cropping.

M

Stephen Williams

03-27-2007, 10:19 AM

Joshua Provost

03-27-2007, 10:57 AM

Why not? There is far mor data from a 9.2 megapixel RGB than from a 9.2 Bayer.

It will be interesting to compare a 4k RGB scan to a 4.5K Bayer sensor and see which has the most resoloution.

Stephen,

You are absolutely correct, all other factors being equal, there is more detail to be had from a 3CCD array than a 1CMOS (or 1CCD) array.

I only mention it because I have seen a lot of people mistakenly assume that you can simply multiply the resolution of each CCD in a 3CCD array and arrive at the resolution. This discounts the fact that the channels are combined to create a single image with the resolution of just a single CCD. I know you know way better than that, but I didn't want to let that slip.

It would be more fair and accurate to say that a 1CMOS has less resolution than its quoted MP spec, than to say that a 3CCD has more resolution than a single individual sensor.

So, in this case the comparison is much closer than 12MP vs. 27MP would indicate. Plus, the RED promises to have a killer debayer algorithm.

Josh

Jeff Brue

03-27-2007, 11:15 AM

The D20 may downsample (1.5x in 'TV mode' to go from 2880 pixels wide to 1920) but that certainly does not eliminate artefacts. Attached is a crop of a grab from a test I did with a D20 last week, captured 10-bit uncompressed straight off the camera, not via tape. It shows very noticeable Bayer artefacts.

No Bayer sensor is immune to those artefacts, but hopefully with the RED doing a 2:1 downscale to give 1080p, they should be greatly minimised, as each Bayer cluster of four pixels will be scaled to a single 1080p pixel. At least that's my assumption.

I'd just to add on a bit to that. I wouldn't say no bayer sensor I'd say no bayer algorithm. I've been hearing of some fairly ingenious algorithms that use motion picture estimation to actual generate the proper demosaic on a frame by frame basis. Again its a thing that could never been done on camera on a realtime HD-SDI out, but if it means 5 seconds a frame on a modern machine...well its just really cool math instead of a bad sensor.

Also in regards to sub sampled color. I work with the Viper and the S2 workflow at the camera house in Hollywood. I have to say the footage is gorgeous on a day in day out basis. The things that it lets vfx do are great, and because you don't have to deal with things like compression, gate weave, film scratches, whatever gets in the way. New applications for post like optical flow can become incredibly more useful.

Thats part of why I'm terriblly interested in the idea of capturing a wavlet compressed bayer pattern like graeme is doing because all of those interesting algorithms will still be eminently possible. Don't get me wrong I have a feeling we'll still get the optical out ports, but Red on paper is going about this the right way.

Jeff Brue
Technical Director
Digital Film Company