2/3 chip

[muneshfilm]

Hey just wanted to say that the new film by Coppola TETRO was shot on a 2/3 chip camera, the sony CineAlta F900 at 1920 x 1080 resolution. It looks great. It was not a RED camera althought they did use a RED for third camera pick ups. I am sure that the 2/3 Scarlet will be up to the task of some fine images, especially knowing the reputation of RED. There has been a lot of disscution about the ability of Scarlets 2/3 being up to task for use as a cine camera.
Watch in HD.

[ame="http://www.youtube.com/watch?v=XJ_XTIsMKig"]YouTube - Tetro: Official Trailer[/ame]

[thequads]

interesting.. 2/3rds digital in black&white really hurt this film. The deep DOF and lack of color made the set's look bland and unintersting... it also made the lighting look very staged.

[antman]

I think an hv30 with a handy35 could have done just as well in black and white.

[muneshfilm]

I highly agree with you Antman. As for Thequads statment about "bland and uninteresting", i couldn't disagree with you more. B/W pulls the eye towards geometrical shapes and lighting. It's very different from colour photography. But its your opinion and thats valid. My word is not the last word. But the filmmakers here wanted a crisp sharp and high contrast look, like B/W photography. Anyways the point was that i believe that our cameras (HV*0) can achieve great images simuliar to Tetro. We dont always need strong DOF all the time. Its a tool. Orson Welles's films are champions at deep photography in B/W. We should focus on storytelling. If Coppola is able to use 2/3 then what is our excuse. Its as if we are waiting for the perfect image and camera. Whatever that means. Technology changes all the time. All films do not have to look the same. We are independants and we should push the boundaries, or else we always have excuses for making our films.

[Duke]

The theoretical advantages of a Scarlet 2/3" sensor are greater DOF control (not an automatically narrower DOF), greater low light sensativity, and at 3k a higher resolution as compared to the average prosumer camera (usually a 1/3" chip.)

However, I'm still waiting for test shot comparisons. If they're really releasing this thing soon why haven't we seen more test shots of a 2/3"? Here's what I'm wondering:

a) light sensativity is partially controled by the size of the photosite on the sensor. With the higher resolution, will the photosites actually be larger and therefore more sensative?

b) DOF is controled by sensor size, amount of zoom and aperture size. The aperture can be controlled and the sensor is twice as big, but the zoom on the fixed scarlet is 8x. Will the DOF control of the fixed lens Scarlet be about the same as an A1 (20x) set at about 3/4 zoomed in?

c) The 3k Scarlet sensor has a Bayer filter making it effectively a 2.2k sensor. That's a little better than a full raster 1920 sensor like the PMW-EX1, but will it be enough to really notice a difference?

Inquiring minds want to know! Sorry guys, but I don't trust the real fanbois to be able to give an unbiased answer, even if they think they can give an unbiased answer.

[muneshfilm]

True in everything you said Duke, but the bottom line is that we already have great cameras. There can be better cameras of course and there WILL be: 1 year to 10 years and maybe beyond. There will always be new toys. But the toys do not make the filmmaker great. Its always nice to have them. Our eyes are always on what is comming out than what we can achieve artistically now. When the new toys show up then we can choose which one suits us, but it will not magically create great art because its sensor is 2.2 and not 3k, or that that its only an 8x zoom. When the time is right and Scarlet is realised, then we can choose the camera that is best for us. Fanboy signing out.

[Erik Bien]

As for B, once upon a time we thought we knew the specific focal lengths for both ends of fixed-lens Scarlet's zoom, but those figures have been written out of the official FAQ in subsequent drafts, so we're back to wait and see.

As for A and C, While not official or endorsed by RED, this chart pretty well summarizes what we know about the new sensors in terms of resolution and pixel pitch.

Measuring the sensitivity of a digital sensor is not entirely straightforward: when overexposed and the photosites can take no further charge, the signal hard clips, but when underexposed, there's a gradual descent into noise. Exactly where you decide the noise becomes too objectionable determines your useable dynamic range. Even if RED had already published recommended ISOs or dynamic range estimates for the new cameras, we'd probably still have to wait and see how they actually perform under poor light compared to other cameras.

[Halsu]

a) light sensativity is partially controled by the size of the photosite on the sensor. With the higher resolution, will the photosites actually be larger and therefore more sensative?

As far as i've understood, the photosites will be larger than in in current 1/3" HD cameras, but about the same size as in 1/2" HD cameras like the EX1.

But it's not the photosite size that counts the most, it's the sensor area.

http://hv20.com/showpost.php?p=212753&postcount=28

...this of course assuming all other things are equal, which they are not. Noise levels and dynamic range vary greatly between cameras that have the exact same sensor size and photosite count.

b) DOF is controled by sensor size, amount of zoom and aperture size. The aperture can be controlled and the sensor is twice as big, but the zoom on the fixed scarlet is 8x. Will the DOF control of the fixed lens Scarlet be about the same as an A1 (20x) set at about 3/4 zoomed in?

Sensor size doesn't really affect DOF at all. Only the amount of zoom and aperture do - okay, the final target resolution affects DOF too, as you can see out-of-focus areas easier in a higher resolution image, but that's semantics ;-).

Anyway, the reason large sensors seemingly have shallower DOF is just that with bigger sensors, you need a longer lens to arrive at the same framing.

So, your question is kinda meaningless: when fully zoomed in, the 20X A1 will probably have pretty similar DOF characteristics as 10X Scarlet 2/3", as both will have roughly the same focal length.

But the framing will be different of course. When compared at the same framing (= different focal length but same "zoom level"), 2/3" camera will have shallower DOF.

c) The 3k Scarlet sensor has a Bayer filter making it effectively a 2.2k sensor. That's a little better than a full raster 1920 sensor like the PMW-EX1, but will it be enough to really notice a difference?

PMW-EX1 is a single CMOS bayer sensor camera too. In addition to bayer pattern not giving full nominal resolution, all decent digital cameras (including 3CCD) have an OLP (Optical Low Pass) filter in front of the sensor, which blurs the image significantly below the nominal resolution. This filter is needed to avoid moiré and aliasing.

So yes, due to it's oversampling, Scarlet should be able to produce much sharper and cleaner HD video than any of the current prosumer (and most pro) HD cameras. It should look pretty similar pixel-by-pixel as EX1 scaled to 720p.

Inquiring minds want to know! Sorry guys, but I don't trust the real fanbois to be able to give an unbiased answer, even if they think they can give an unbiased answer.

Well, i'm a fanboi i guess, but my facts should be correct.

[Duke]

As far as i've understood, the photosites will be larger than in in current 1/3" HD cameras, but about the same size as in 1/2" HD cameras like the EX1. But it's not the photosite size that counts the most, it's the sensor area.

If that was true then all the Panny fanbois are wrong and neither the HMC150 nor the HPX170 are really good in low light as advertised because they have the same 1/3" sensor but bigger photosites.

I think it just seems that way to you because a larger sensor allows bigger photosites. (Note in my quote I said 'partially controled by the size of the photosite') There are other factors.

If the Scarlet has bigger photosites than the average 1/3" sensor, great. Maybe we will get good low light out of it. I hope so.

...this of course assuming all other things are equal, which they are not. Noise levels and dynamic range vary greatly between cameras that have the exact same sensor size and photosite count.

So true which is why I want to see examples and comparison shots.

Sensor size doesn't really affect DOF at all. Only the amount of zoom and aperture do - okay, the final target resolution affects DOF too, as you can see out-of-focus areas easier in a higher resolution image, but that's semantics ;-). Anyway, the reason large sensors seemingly have shallower DOF is just that with bigger sensors, you need a longer lens to arrive at the same framing.

Your talking in circles on this one. If the glass and aperture remained the same, but the sensor was bigger it would have a narrower DOF. The three factors interplay together.

So, your question is kinda meaningless: when fully zoomed in, the 20X A1 will probably have pretty similar DOF characteristics as 10X Scarlet 2/3", as both will have roughly the same focal length.

Ahhh, there's the rub. The fixed Scarlet isn't slated to have a 10x zoom. It's touted as an 8x zoom, so the A1 will have the same or narrower DOF with 16x zoom (about 3/4 of the way) on up to 20x zoom.

But the framing will be different of course. When compared at the same framing (= different focal length but same "zoom level"), 2/3" camera will have shallower DOF.

True as far as it goes. But... If you back up twice as far with an A1 and use twice as much zoom the talent will look very much the same. Admittedly, in that instance it is the backgrounds that will be in completely different proportions.

However, the purpose of the narrow DOF is to blur that background and get the audience to not pay attention to the background, but instead to focus on the subjects.

If you want the broad scope of a battlefield, like in your war movie, that might not be what you are after. However, if you are working on a romance that zoom in and reframing can give the extra feeling of intimacy that you are after. Reframing isn't necessarily a bad thing, just another tool.

So yes, due to it's oversampling, Scarlet should be able to produce much sharper and cleaner HD video than any of the current prosumer (and most pro) HD cameras. It should look pretty similar pixel-by-pixel as EX1 scaled to 720p.

That would be excellent. I suspect that the 3k resampled to 2k will be like the difference between a EX1 (at 1920) and an A1 at (1440). More details due to more data. [On the other hand I've seen A1 and EX1 footage cut together and it's not that noticable unless you're looking for it.] I just want to see the samples. I've long bemoaned the fact that we aren't seeing any Scarlet samples for comparison purposes.

Don't get me wrong, I want the Scarlet to succeed. I will most likely be buying one. However, so many things are interdependent that I really want to know the strengths and weaknesses before I plunk my money down.

We probably all agree that it's not just specs on resolution, low light, or DOF control, but how they fit together and how easy they are to use. Tell Jim J. to show the clips so we can quite speculating. Cheers

[Duke]

True in everything you said Duke, but the bottom line is that we already have great cameras.

And have had great cameras for a few years now, which is why I chose to make my movie with an 'inexpensive' A1 (which so far has compared quite well with all other cameras in shoot outs), and will put my money in front of the camera instead.

[Halsu]

If that was true then all the Panny fanbois are wrong and neither the HMC150 nor the HPX170 are really good in low light as advertised because they have the same 1/3" sensor but bigger photosites.

Okay, to be more specific, i should have said combined photosite area when talking about sensor area.

I think it just seems that way to you because a larger sensor allows bigger photosites. (Note in my quote I said 'partially controled by the size of the photosite') There are other factors.

Yep - i simply roughly calculated the sensor area versus number of photosites to get the ballpark of photosite size.

If the Scarlet has bigger photosites than the average 1/3" sensor, great. Maybe we will get good low light out of it. I hope so.

Even if they were exactly the same size, it would outperform the 1/3" sensors, because of supersampling.

So true which is why I want to see examples and comparison shots.

We all do ;-)

Your talking in circles on this one. If the glass and aperture remained the same, but the sensor was bigger it would have a narrower DOF. The three factors interplay together.

No, they do not, and no, the DOF would not change.

Changing sensor size has the exact same effect as cropping/resizing image in post (assuming you have enough resolution to crop etc.). It does NOT affect DOF in any other way.

Ahhh, there's the rub. The fixed Scarlet isn't slated to have a 10x zoom. It's touted as an 8x zoom, so the A1 will have the same or narrower DOF with 16x zoom (about 3/4 of the way) on up to 20x zoom.

The zoom level is really pretty irrelevant, the only thing that really matters is the amount of DOF at any given framing.

20X fully zoomed in is pretty much useless in most shooting situations anyway, especially indoors.

True as far as it goes. But... If you back up twice as far with an A1 and use twice as much zoom the talent will look very much the same. Admittedly, in that instance it is the backgrounds that will be in completely different proportions.

To get somewhat similar image (DOF and framing), you would need to move both the subject AND the background by the same amount. In other words, you need twice as big a room.

And, if you i.e. add a 2X tele extender to the Scarlet to get to the exact same shot as A1, whaddya know, you again have shallower DOF.

That would be excellent. I suspect that the 3k resampled to 2k will be like the difference between a EX1 (at 1920) and an A1 at (1440). More details due to more data.

The difference should be bigger than that, if i.e. Red One footage is comparable to Scarlet.

I've long bemoaned the fact that we aren't seeing any Scarlet samples for comparison purposes.

There are some sample shots using the Scarlet 2/3" sensor on the Red forum. Haven't you seen them?

Also, there's plenty of 3K Red One footage around the web, it should give you a pretty good idea of what to expect.

Don't get me wrong, I want the Scarlet to succeed. I will most likely be buying one. However, so many things are interdependent that I really want to know the strengths and weaknesses before I plunk my money down.

We're in the same boat. My confidence level is pretty high though, after working with Red One in multiple projects.

We probably all agree that it's not just specs on resolution, low light, or DOF control, but how they fit together and how easy they are to use. Tell Jim J. to show the clips so we can quite speculating. Cheers

I have a hunch there won't be any new clips or other info until we're quite close to the release date...

[Duke]

I saw you assertion that the 3k Red One images should be the same, but it's a different sensor, different density, different photosite size, so there will be some differences.

I just did a quicky search and couldn't find any before I ran out of time. If you can post a link to and 2/3" sample images I'd appreciate it.

The only ones I recall were Izzy the lizard and Enzo the dog. My recollection was that they were scarlet, but 5k instead of 3k.

[Halsu]

I saw you assertion that the 3k Red One images should be the same, but it's a different sensor, different density, different photosite size, so there will be some differences.

Of course the images will be somewhat different, but i.e. resolution and sharpness wise they should be close, red 3K should serve pretty well as a quality approximation.

As far as low light performance goes, that's of course hard to say. One can guesstimate that 2/3" should lose maybe a stop or so due to smaller sensor area, but gain back that same stop due to better, new generation sensor.

The only ones I recall were Izzy the lizard and Enzo the dog. My recollection was that they were scarlet, but 5k instead of 3k.

Enzo was 5K S-35, Izzy was 3K 2/3":

http://reduser.net/forum/showthread....highlight=izzy

[Rikki]

Wonder when Scarlet will appear. It was supposed to be out last year from what I remember.

Quite happy with my combo of an HV20 and my new Panachronic HMC150 in the meantime. Plus having watched the RED One users being basically beta testers for the greater cause I would hate to use a Scarlet right off the bat in a commercial venture and have it screw up.

[Halsu]

Wonder when Scarlet will appear. It was supposed to be out last year from what I remember.

Nope - the first estimates were for spring/summer 2009. Now it looks like there's a few months delay, meaning late fall or winter, so not that bad.

Of course, i'm an optimist ;-)

[Duke]

Also, it depends on what you are calling a release date.

a) sending out a few hand built models to known directors for publicity purposes.

b) sending out a few to writers/editors for publicity.

c) selling them to prior Red owners.

d) selling them to a registered back-log of buyers.

e) everyone can buy them at a store.

Ask a dozen people and I bet you get a dozen answers. I only count when everyone that wants one can buy one unless that backlog is really short. But, if it's like the Red One it won't be but I can always cross my fingers and hope.

[Halsu]

I only count when everyone that wants one can buy one unless that backlog is really short. But, if it's like the Red One it won't be but I can always cross my fingers and hope.

AFAIK, they currently have Ones on stock, order and get an immediate delivery... but i do get what you mean ;-)

In Scarlet's case, they've said it will be mass produced, unlike One. So the Average Joe/Jane might very well have his/her hands on a Scarlet in much more timely fashion than with One's case was in the beginning. Or, that's at least what i hope for ;-)

[Duke]

Nope - the first estimates were for spring/summer 2009. Now it looks like there's a few months delay, meaning late fall or winter, so not that bad.

Of course, i'm an optimist ;-)

Maybe a little too optimistic as fall or winter of 2009 seems to have come and and nearly gone.

In the winter (11/30) of 2009 we're going to get an announcement of when they'll begin releasing them, and it was announced there will be an additional 30 day delay in that release.

I'm still keeping my fingers crossed for sometime in 2010 for a release to the general public, but suspect my prediction of after NAB (April) 2011 might be more accurate.

[Seo]

Heh, it turned 11/30 1 hour and 16 minutes ago on the Pacific. Can't wait....

[Marshallator]

NO! Don't give in to the Red hype machine! You must resist!

It will be interesting to see what they say.

[muneshfilm]

By the way TETRO came out on dvd and blue-ray today if anyones interested in seeing this wonderful film. Like i have said before it was shot on a CineAlta F900..a 2/3 sensor camera.

[jordanService]

Regardless of the camera discussion the preview looks amazing! I really want to see this, I love deep focus, I think it is incredibly difficult to do well because it takes up so much more time on location and set design. You can't fudge the blurry BG.


I might just pick up this film. What was the budget for this? If you do not mind me asking.

[muneshfilm]

According to DarkHorizons.com the production budget was 15 million. There is an interesting pdf you can download on the film or view it online:http://docs.google.com/viewer?a=v&q=...65gMTLZgtnnwcg. This film has gotten some great reviews you can find online.

[jordanService]

Yea, after I posted I realized it was by Francis Ford Capola! No wonder why it looks amazing.

[Duke]

Maybe a little too optimistic as fall or winter of 2009 seems to have come and and nearly gone.

In the winter (11/30) of 2009 we're going to get an announcement of when they'll begin releasing them, and it was announced there will be an additional 30 day delay in that release.

I'm still keeping my fingers crossed for sometime in 2010 for a release to the general public, but suspect my prediction of after NAB (April) 2011 might be more accurate.

Now its summer of 2011.... maybe, or end of 2011... maybe.