DVX is not HD - does that matter?

[Otis Grapsas]

It's not easy for someone to see the light, but a beautiful movie as seen above cannot be denied by anyone. This piece as we all know was shot on a Canon 5d + 7D.

Run all the test you want, their is a reason these HDSLRs are being used. If Canon themselves would have know people were going to be using them for movies they would have sold them more expensive. We stumbled upon this technology and I can never see myself going back.

I'm Done Here

Please enlighten me Cieloent!

"Canon, who provided two pre-production 7Ds and a host of L-series lenses; Tiffen, who provided all of our lighting; and Cinevate, who sent us much needed gear and sponsored an incredible behind-the-scenes campaign viewable at cinevate.com"

It's good location, large cast, good lighting, nice costumes etc. But it is only designed to sell dSLRs and accessories. All I see is a high compression 720p web version which is useless for checking image quality. I have used 7D, 5DII and GH1 and I wasn't impressed. The image quality is very low compared to a modern camcorder.

In the world I live in, a tool is judged by its capacity to shoot good images and take post processing without artifacting. No camera manufacturer that produces real filmmaking tools bothers with this type of marketing. It's pure consumer oriented marketing. They just produce their tools and let cinematographers discover them. Shoot with them, go throught post, do plates, key, color correct and see how the image holds up for bluray and film output. That's the only real test and you don't need to go to exotic locations with tons of sponsors on your side. A dSLR fails in these tests and no amount of extra money or accessories can fix that. The problem is in the sensor readout method and the codec implementation.

 

[Otis Grapsas]

Otis, while I respect what you are saying, I think it's something of a misnomer to state that the T2i is grossly inferior to the 5D. Clearly the 5D is better in a number of respects, and it's capabilities in poor light are vastly better, but the T2i on a cost bases might be the best bargain out there. Yes, if money is taken off the table, the 5D is certainly a better camera, but if money is taken off the table then these DLSRs are essentially 1-trick ponies.

The image quality is measurably inferior. I don't think the T2i makes a good purchase because the price difference from the 5DII is very small when you complete a camera package (lenses, accessories etc). We invest so much time in shooting, at this price level it doesn't make sense to save a little bit of cash.

For someone that just needs a dSLR with a single lens (kit lens perhaps), the entry level dSLRs do make sense.

 

[PerroneFord]

I have used 7D, 5DII and GH1 and I wasn't impressed. The image quality is very low compared to a modern camcorder.

What kind of "modern camcorder" do you mean? A RED? An Alexa? An HPX170, and EX1, an HMC150, and FlipHD? There are many kinds and levels of modern camcorder. The HDSLRSs are better than some and worse than others. Just like any camera.

In the world I live in, a tool is judged by its capacity to shoot good images and take post processing without artifacting.

Really? That's an interesting world you live in. Because even on movies like Avatar, The Dark Knight, etc., the post process involved artifacting and loss of information. Even with the 8k scans of the IMAX footage. Must be a wonderful world you live in!

No camera manufacturer that produces real filmmaking tools bothers with this type of marketing. It's pure consumer oriented marketing.

Really? Does Canon produce real filmmaking tools? Arri? Is Sony's F35 a real filmmaking tool? Because I've certainly seen advertising from all of those companies aimed at consumers (people who buy $100k camera are consumers too) that are very similar to what we see here.

They just produce their tools and let cinematographers discover them. Shoot with them, go throught post, do plates, key, color correct and see how the image holds up for bluray and film output. That's the only real test and you don't need to go to exotic locations with tons of sponsors on your side. A dSLR fails in these tests and no amount of extra money or accessories can fix that. The problem is in the sensor readout method and the codec implementation.

So it's your impression that real filmmaking cinematographers would feel that HDSLRs fail in their ability to go through post, and certainly could not compare to high end camcorders and most CERTAINLY not film? Right? A DSLR couldn't POSSIBLY impress the likes of people who do this in Hollywood on big budget films? And no one would DARE say they they would use one of these cheap cameras in their real films.. yes?

Then take an hour of your day and watch these three short pieces:

http://www.zacuto.com/shootout


Because here we have an oppurtunity to see what happens when several DSLRs are stacked against Kodak and Fuji film, and some of the more prolific Cinematographers working in the world today get a chance to sit down in a real theater and screen them against each other. They had full workflows, including grading on what looked like a DaVinci but may have been a Quantel, and everything else you'd do for a feature film with major release. The failures of the cameras are there for all to see. And the success are there as well. And in large part, these test was responsible for me buying my first HDSLR to accompany my "professional camcorder".

 

[PerroneFord]

The image quality is measurably inferior. I don't think the T2i makes a good purchase because the price difference from the 5DII is very small when you complete a camera package (lenses, accessories etc). We invest so much time in shooting, at this price level it doesn't make sense to save a little bit of cash.

For someone that just needs a dSLR with a single lens (kit lens perhaps), the entry level dSLRs do make sense.

If we are comparing length and size matters, 999mm is measurably inferior to 1000mm. But if the level of tolerance is +- 10mm and the cost of 999mm is 1/4 the cost of 1000mm then we have a dilemma. And this is where we are with these DSLRs. Yes the 5D is better. And I have stated that it is. In nearly every measurable way. However, the vast majority of people purchasing these cameras will never release anything beyond a DVD or BluRay, and mostly will be producing 720p web videos.

It is my opinion that the T2i more than exceeds the standard for production at that level, and in fact, is quite capable of producing video suitable for the movie theater. Though it may be measurably inferior to the 5D or other cameras, it is MORE than enough for the great majority of people spending money at this level. In much the same way IMAX is measurably better than 35mm but movies continue to be made and distributed on 35mm. And largely for the same reasons.

 

[Otis Grapsas]

Otis,

I take note that you are not in a native English speaking country, and I can only guess that my last post was not clear. I did not say that movies with $30M budgets were being shot on DSLRs. I said movies with those kinds of budgets have USED DSLRs in filming.

Among those are Ironman 2, and the upcoming George Lucas film "Red Tails". The implication that the Ironman franchise needs to "save money in the camera department" is beyond silly. And I would also doubt very much that their exaggerated their budget by 20x, lest the movie would not have gotten financed. I won't even speak to the idea that Canon was involved in funding a movie at that level to get their camera in it.

I am not someone who took up the idea of the HDSLR easily. I failed to see them as much more than a toy when they started, and to be fair, that's about what they were when first released. It wasn't until Canon started to pay some real attention to these cameras, mostly by prodding of people at ASC level, that they statted to become viable tools. To ignore this, is simply burying one's head in the sand. No, they are not the best cameras on the market. Nor are they trying to be. They ARE a viable solution for independent filmmakers who have a camera budget of less than US$10k. In that market, they are likely the best tool available right now for narrative filmmaking. With a $50k camera budget there are better options available. And certainly at the 100k level cameras like the Alexa can be considered.

Sony and Panasonic may be developing a more friendly layout around a large sensor and kudos to them for that. But right now, today, they are not available, and the 5D, 7D, T2i, D3s, GH1, etc. are available and getting the job done for quite a number of people. Read into that what you will.

Frankly, I don't see the new Panasonic bringing much more to the table for it's $6k in terms of narrative filmmaking. I can see how it will be a better fit for event work, or some other uses. But by the time you add a set of rails, external monitor, follow focus, mattebox, external power, and all the other things required for filmmaking, the form factor of the DSLR is just as viable as the larger form factor of the camcorder. .

Many cameras are used in many films for many reasons. Cost, size or any other factor. What is important is when an independently funded project (not accessories companies, sponsors etc) with a high budget uses a dSLR as the main camera for a commercial 1080p bluray or film output. Then we can say that film A WAS SHOT with that camera. We haven't seen that yet because the current crop of dSLRs are not good enough. Every colorist knows they are a pain. The dSLR projects that get good results, need no post processing, no keying or effects, and they manage to nail the exposure on set.

I'm not aware of Red Tails material being actually used on the output. I have heard of a rumour. I believe it is not very probable because there are smaller and better solutions.

I find it hard to believe that Iron Man 2 uses a dSLR on the final output either. If they needed 20 crash cams perhaps, but there could be smaller, cheaper and better solutions for that also.

Are these rumours actually verified? I have heard about Harry Potter "being shot" on dSLRs also

So, to put this straight, none of these films are "being shot on dSLR".

My comment on budget was for a high budget feature film that claimed to be using a dSLR as a main camera. I share your view, that accessories and dSLR manufacturers can only fund very low budgets. That was the point of the 20x figure.

You can spot the marketing most of the times. They are usually using the low end dSLRs in these shoots even though that can get 5dIIs for an extra 15 euro or so per day of rental. Does this make sense? It does, if the project is being used to push the low end dSLR! 7D is a lot hotter in sales due to its low cost, so they always have a 7D in those shoots besides the 5DII.

I agree with everything you say on camera choice vs budget. It's how real world production works.

The new Panasonic will do a little on the codec department and treat the alias problem which is important. It will also make the camera a better tool, closer to what a modern camcorder is. It should do well on the market. Many people avoid dSLRs simply because they are not fully featured camcorders. I don't know how large that segment is but it could be important.

 

[Otis Grapsas]

By modern camcorder I mean an EX1 or something like that. The codec is better, the image quality is higher. And it doesn't cost that much more considering it comes with a lot of extras (vs a dSLR).

The level of artifacting in big budget post production is unimportant for 1080p output. We can admit it works quite well Avatar didn't look very good in the "video" shots IMHO, but the graphics made up for that.

Sony and Panasonic do not do viral marketing of this type. They just attend trade shows, do their press releases and pay for magazine ads.

Yes, my opinion is that dSLR are not capable of modern film production due to the codec and sensor readout limitations.

I have seen the Zacuto tests. Zacuto know the value of the dSLR market. It's a huge market. Their comparisons mean nothing to me, they are bogus. They are engineered to make the dSLRs look good and I could write a 10 page article on how they do it and why it is nonsense. Even if you don't understand how reality can be skewed from an engineering point of view, I don't understand how you take something like this seriously coming from the leading dSLR accessory manufacturer. It only takes common sense to be sceptical against this type of test coming from this type of source.

 

[PerroneFord]

Many cameras are used in many films for many reasons. Cost, size or any other factor. What is important is when an independently funded project (not accessories companies, sponsors etc) with a high budget uses a dSLR as the main camera for a commercial 1080p bluray or film output.

Given this, if the standard of measure is whether a large budget feature is using a camera as it's main camera, then most "modern camcorders" by your definition fail to meet the standard. And thus I would be forced to view my EX1 in the same way as my T2i, or even the 5D. None has been used for a large budget feature so none are "good enough".

Then we can say that film A WAS SHOT with that camera. We haven't seen that yet because the current crop of dSLRs are not good enough. Every colorist knows they are a pain. The dSLR projects that get good results, need no post processing, no keying or effects, and they manage to nail the exposure on set.

I certainly agree that the current crop of DSLRs is NOT good enough for primary acquisition for a feature film. But short of a RED, none of the other low cost cameras are either.

I'm not aware of Red Tails material being actually used on the output. I have heard of a rumour. I believe it is not very probable because there are smaller and better solutions.

Correct, it is a rumor. Though I have no reason to doubt the source.

I find it hard to believe that Iron Man 2 uses a dSLR on the final output either. If they needed 20 crash cams perhaps, but there could be smaller, cheaper and better solutions for that also.

"Believing the technology that powers the Iron Man armor was stolen from his family, Vanko constructs his own “Whiplash” armor and confronts Stark in a spectacular action sequence at the Monaco Grand Prix. Filming the sequence required the second unit, shot and directed by Jonathan Taylor, ASC, to shoot background plates and shots of a speeding Rolls Royce along the actual Monaco circuit. (See sidebar) Taylor recalls, “The Monaco police force and auto club were very cooperative, but because it’s a working city, we had only an hour or two for each section of track. It was just frantic, and we literally had only one hit at it. Second-unit first AD Michael Moore and I planned the whole thing at Manhattan Beach. It was a question of getting all the equipment set up and having a well-oiled machine.” Taylor’s equipment list included a high-speed camera-tracking vehicle, a Porsche 928 rented from Propulsion in Paris, driven by Jean-François Dubut and mounted with VistaVision cameras from Geo and Procam; an insert car rigged with VistaVision cameras, Arri 435s and Canon EOS 5D Mark IIs; and a Mercedes SUV-mounted Russian Arm rented from Bickers Action. Additionally, ground cameras were positioned to grab shots of the Rolls speeding by, and even a helicopter, rented from Flying Pictures, was employed for aerial shots. “We had all the toys, and we ran a whole convoy around the track,” Taylor continues. “It was quite a trick to pull the whole thing off.” "


"Taylor also incorporated Canon EOS 5D Mark II DSLRs, fitted with Canon lenses, as crash cameras. “One of the problems with doing action stuff is finding interesting places to put the camera,” says Taylor, adding that he could “actually put the 5Ds on the cars we were going to crash. We cut holes for the lenses in small Pelican cases that we painted to match the cars. We got some amazing shots. Of course, it’s not film quality, but for a 12-frame cut in an action piece, it holds up very well.”


http://www.theasc.com/ac_magazine/May2010/IronMan2/page3.php


Certainly, they are not saying the 5D was good enough to stand on it's own. But for numerous reasons it was the correct choice here. There was more emphasis placed on the use of the camera in this movie in HDVideoPro magazine in May, but that is not available online, and my copy is at home so I cannot quote it accurately for you. But they were used and they do appear in the final film cut.

Are these rumours actually verified? I have heard about Harry Potter "being shot" on dSLRs also

So, to put this straight, none of these films are "being shot on dSLR".

By your definition, I certainly agree.

The new Panasonic will do a little on the codec department and treat the alias problem which is important. It will also make the camera a better tool, closer to what a modern camcorder is. It should do well on the market. Many people avoid dSLRs simply because they are not fully featured camcorders. I don't know how large that segment is but it could be important.

I think the primary advantage to a full sized camcorder will be the inputs and outputs. XLR inputs on the camera is of nearly no consequence for narrative filmmaking. It's highly doubtful the Panasonic will have timecode in or out, so that negates a large part of need for it's size. It might have and SDI port, and if so, that would be worth the cost alone. I wonder whether we will see an HDMI port and I would expect that we will. If we do, then that will certainly add a revolutionary step.

I agree that people are avoiding DSLRs because they lack some camcorder features. But the biggest of these seem to be the lack of useful monitoring options (a killer on the GH1, and pretty inconvenient on the Canon's), and the lack of XLR inputs which is more important to the solo shooters out there and event folks.

There is certainly room for improvement in this market, but the BIGGEST is when we get RAW off these sensors in 24p/25p or 30p. That, more than anything, will change the game.

 

[PerroneFord]

I don't understand how you take something like this seriously coming from the leading dSLR accessory manufacturer. It only takes common sense to be sceptical against this type of test coming from this type of source.

I can take them seriously, because the tests I saw mirror my own experience. I've been lucky enough to take 5D footage through post and put it on the big screen. And I am about to do it again with our current production. In my opinion, for our budget (which was less than US$10k total), the footage held up nicely and it has been well received. Our short has played across the country, will be playing in Scotland this fall, and has been requested for an Italian film festival.

I had access to a digital projector when I was grading the film and was able to see how it held up during post. For what I needed from it, it was fine. Certainly, you may feel differently, and I can respect that.

Many of the filmmakers on this site are hoping to get their movies screened at the kinds of festivals we've had success with, and to them I can say that the 5D and the T2i are most CERTAINLY good enough. But be prepared to reshoot if your movie actually gets picked up.

 

[Otis Grapsas]

Crash cam use makes sense. It will be perhaps 1 second on the final edit for each cut? Nobody has the time to notice image quality as long as color is matched.

I'm not comparing dSLRs to a high end digital cinema camcorder, different price levels. I just believe that for most people a $5,000 euro camcorder will make a better tool. When people ask me I always tell them to rent the dSLR before investing large amounts of money in building a kit. They cost virtually nothing to rent for a day.

I watched the Zacuto dynamic range test for a second time and took some notes. The numbers they mention come out of nowhere.

24:00 on first video.

Film clearly has the full 12stops the chart contains (13distinct exposure points). We know it has even more. The gradation is excellent.

5D (they say 10.5 stops) has only 8 stops (3 white points, 3 black points, 8 distinctive non white/black points). The 8 distinct steps contain a range of 7 stops, the next step to the dark is practically the same with the previous one but lets add 0.5 to be generous, and add 0.5 stop on the highlights. 8 stops. No relation to the 8stops coming from film though. You can push film shadows a lot because they are not blocky artifacts.

7D (they say 11 stops) is 8 stops also. The exposure level is just slightly darker to the 5D level.

The 1D (they say 11) is about 9.

The D3s (they say 11) is 8.5 stops.

The GH1 (they say 9) is actually 9.

That's based on their own chart.

These cameras cannot do 11 stops, not even in RAW 12/14bit full resolution images in 16bit output. They can do something over 10 on their lowest sensitivity but only 8 stops in a realistic ISO. They 2 lower stops will be very compromised. Because of the compression there is virtually no texture there, it's just blocking.

 

[Gary Huff]

Otis,

Here's the deal in simple terms:

"I can get shallow DOF with the HDSLRs! Then it must be the best tool for every job!"

"But the resolution is only 600-700 lines, you have to fight jello/skew/moire, and the implementation of AVC is the consumer side of the codec and only has a color space of 4:2:0!"

"Shallow DOF, shallow DOF, I CAN'T HEAR YOU!"

EDIT: And you're wasting your time with Cieloent...I'm sure you can tell by now that professionalism is not one of his strong suits.

 

[Otis Grapsas]

I can take them seriously, because the tests I saw mirror my own experience. I've been lucky enough to take 5D footage through post and put it on the big screen. And I am about to do it again with our current production. In my opinion, for our budget (which was less than US$10k total), the footage held up nicely and it has been well received. Our short has played across the country, will be playing in Scotland this fall, and has been requested for an Italian film festival.

I had access to a digital projector when I was grading the film and was able to see how it held up during post. For what I needed from it, it was fine. Certainly, you may feel differently, and I can respect that.

Many of the filmmakers on this site are hoping to get their movies screened at the kinds of festivals we've had success with, and to them I can say that the 5D and the T2i are most CERTAINLY good enough. But be prepared to reshoot if your movie actually gets picked up.

At a $10k budget, nobody would complain. It's excellent for the money.

The viewers are not engineers. If you can work around the limitations and nail the exposure, a dSLR will work.

I only object to the idea that a dSLR can compare with high end equipment.

 

[PerroneFord]

I only object to the idea that a dSLR can compare with high end equipment.

I don't think anyone has put that forth, certainly no one who understand what is going on here. And I certainly would not put that forth. I've shot on RED once, have an EX1, and now have a T2i. They are all VERY different tools aimed at very different users.

I would say that for much of what I have seen from folks here and at other forums, putting together a short with $10k-$20k would be a luxury. I know we are doing our current film with less than that, but we didn't have to put the 5D and glass in the budget this time.

If you asked me if I'd rather be doing this on the EX1, I would say no. I had it on set, and never used it. If I had a RED on set I would have gladly put the 5D away, re-lit, and shot with it.

 

[Gary Huff]

Would you use the EX1 if it had the same shallow DOF look as the 5D?

If not, why not?

 

[PerroneFord]

Otis,

Here's the deal in simple terms:

"I can get shallow DOF with the HDSLRs! Then it must be the best tool for every job!"

"But the resolution is only 600-700 lines, you have to fight jello/skew/moire, and the implementation of AVC is the consumer side of the codec and only has a color space of 4:2:0!"

"Shallow DOF, shallow DOF, I CAN'T HEAR YOU!"

EDIT: And you're wasting your time with Cieloent...I'm sure you can tell by now that professionalism is not one of his strong suits.

While I understand the implications here, and sypmathize, I think it's an oversimplification.

There are numerous tools to get screen depth in a film shoot. One is by having physical space. However, this is often not available to people on very limited budgets. Big spaces cost more money.

Another is by careful and selective lighting. Large lighting packages cost large money. And hiring someone who understands HOW to light with large lighting packages costs even more money. Again, this is generally something that small filmmakers cannot afford.

So shallow DOF becomes a primary tool for obtaining separation between our primary and secondary elements. It's a crutch. It's less than ideal. But, it may be the only tool you have to work with given the budget. Thus it gets overused.

I would LOVE to have a Nanoflash hooked to my EX1 with a Letus Ultimate on the front and PL glass. But I don't. I can't afford PL glass and the Letus is out of my reach for the moment as well. And if I use it, I have to increase my lighting package. And I can't afford HMIs, so I need to get a 5K or 10k generator to run my lighting package.

Or I can do what I did thsi weekend. I can use compact fluorescent practicals with strong bulbs, fluorescent tube practicals, and use them to actually light the set. And I can take a sensitive camera like the 5D, set it for ISO400 or ISO640 and get the shots I need. The actors stay cool, I don't need a generator, and we get the picture done on time and under budget. The producer is happy, the director is happy, and I am happy. Will we win an Oscar? No. Does it matter? No.

I don't know a single DSLR filmmaker whould wouldn't be THRILLED if their camera was shooting 4:2:2 100mbps or RAW with no Jello. But they don't. So like filmmakers have always done, you use the best tools you can to tell the story you want to tell. What more can you do?

 

[Otis Grapsas]

So far, I hear about 2 reasons: DOF and color. The colorimetry on a dSLR is quite nice.

 

[PerroneFord]

Would you use the EX1 if it had the same shallow DOF look as the 5D?

If not, why not?

No, I would not. At least not of the shoot I am on now. I used the EX1 for my last shoot but it was a different kind of film. For me, it's not the shallow DOF. For me it's more about the lighting.

On my last shoot, I ran camera, lights, and audio. I had no help. So I didn't use sync sound, and I had to set every light. The EX1 allowed me to work faster and more efficiently. So that was the tool of choice. On this shoot we have a camera op, director, first AC, script sup., AD, sound mixer/boom op, craft services, etc. Different kind of movie, different camera package.

 

[Otis Grapsas]

No, I would not. At least not of the shoot I am on now. I used the EX1 for my last shoot but it was a different kind of film. For me, it's not the shallow DOF. For me it's more about the lighting.

On my last shoot, I ran camera, lights, and audio. I had no help. So I didn't use sync sound, and I had to set every light. The EX1 allowed me to work faster and more efficiently. So that was the tool of choice. On this shoot we have a camera op, director, first AC, script sup., AD, sound mixer/boom op, craft services, etc. Different kind of movie, different camera package.

If you could improve one element on the dSLR, what would you prefer?

- Global shutter CMOS/CCD
- 150mbit intraframe codec
- Proper OLPF filter
- Usability features such as audio, long recording times, etc

Also, would you trade shallow DOF for any of these?

 

[PerroneFord]

If you could improve one element on the dSLR, what would you prefer?

- Global shutter CMOS/CCD
- 150mbit intraframe codec
- Proper OLPF filter
- Usability features such as audio, long recording times, etc

Also, would you trade shallow DOF for any of these?

Now THIS gets to the crux of the matter.

I'll address each:

Global Shutter - This is important for reducing skew. In the films I shoot, this is not an issue. So the HDSLRs and the EX1 work great for me.

Codec - To me this is the largest limiting factor. The DSLRs offer no workaround for this. The EX1 does, which is why I bought it.

OLPF filter - I have control of wardrobe and background on set. So this one doesn't bother me too much.

Usability features - I shoot sync sound so I could care less about audio inputs or outputs. I've never had a take go beyond 4 minutes, so the 12 minute limit on the Canon's is immaterial to me. External monitor solves the issue of the small LCD. No different than the EX1 or other camcorders. But the ONLY usability feature I'd really like to have is full HD output on both the LCD and over the HDMI output with a CLEAN output. I'd like to have a monitor for the first AC along with the camera op. That is just so fundamental to the process, it's a glaring omission.


And would I trade shallow DOF for any of these. No. Because I already have. That is what I have in the EX1. We've had that for quite some time with traditional camcorders. From the F900, the Varicam, the HVX200, the EX1, etc. All of them offer the same excellent resolution, ports, codecs, etc. None have the shallow DOF. Adding shallow DOF to them was a $2500+ proposition and often a loss of light from 1-3 stops.

At a cost of $800, the T2i's compromises become commensurate with it's cost. And for me, that made it worth it. The 7D was close. For me PERSONALLY, the 5D's compromises were not worth it's cost. I came to the EX1 from the same perspective. The rolling shutter, poor handholding ergonomics etc, were more than outweighed by it's performance.

Again, I think shallow DOF is nice, but often overused. And when we say shallow DOF, what do we mean? Personally, I am lighting our current movie to F4.0 most of the time. Occasionally at 5.6 and even more rarely F2.0. It's hard to pull focus at F2 on a moving camera and with a moving subject. I like the look of F4.

 

[Otis Grapsas]

Thank you for the analysis.

A high quality codec is indeed the most expensive feature on a camcorder today.

When you mention clean output, what do you mean? High resolution? Most 7"/8" on camera monitors display a 800x450 pixel image. I use a 850x450 display and I find it quite sharp.

If you mostly use f2.8 and smaller apertures, you do not really need a s35 sensor, you could use a 2/3" with fast lenses at 4x lower ISO. There are no 2/3" cameras in that price range yet though.

 

[PerroneFord]

A high quality codec is indeed the most expensive feature on a camcorder today.

Yes, it is. And it's a good reason that external recording is so sought after. So you can use a NanoFlash, a Wafian, a Codex, etc. If the camera can output it's RAW information, or can give an SDI or HDMI output, then the person shooting can determine the codec.

When you mention clean output, what do you mean? High resolution? Most 7"/8" on camera monitors display a 800x450 pixel image. I use a 850x450 display and I find it quite sharp.

When I say a clean output, I mean a 1080p output that has no graphics, letters, or anything else on it. Suitable for recording.

If you mostly use f2.8 and smaller apertures, you do not really need a s35 sensor, you could use a 2/3" with fast lenses at 4x lower ISO. There are no 2/3" cameras in that price range yet though.

Correct. Assess the economics. The S35 sensor in the T2i cost me $799. Buying reasonable glass for it might cost another $2k. So for less than $3k I have a camera setup that can record fairly clean images in 10-15FC of light. What would it cost me to do the same with a 2/3" camera?

And with the 2/3" camera I am going to NEED fast glass. How much extra will I pay for F1.4 or F1.8 glass over the F2.8 or F4.0 glass that I can use with that S35 sensor? And once I am into that 2/3" sensor, I no longer have that shallow DOF that I might want again. It ceases to be an option. I ALWAYS have the option on the HDSLR even if I don't use it. So then I am back to putting a DOF adapter in front of the camera again.