Is video raising cost of bodies? Is it wasted for many shooters?

[unfocused]

This is ridiculous. Video doesn't increases the cost of cameras, it reduces it.

Cameras are a mass produced object. It's not like your 5DIII is being handcrafted and each feature must be individually installed. It's a mass production item.

Let me explain. Manufacturer "A" makes $100 profit on each unit they sell of Product "B." They have a goal of $10,000 profit. They sell 100 units, they make $10,000. But, if they add feature "C" to the product, it will reduce their per unit profits by $10.

But at the same time, they know that adding feature "C" will boost their sales by 30%, so when they sell Product "B" at $90 profit, they actually net $11,700, exceeding their sales goal. So, instead of selling Product "B" at $90 profit, because they have increased their sales volume, they decide to cut the price by $3 and sell it at $87 profit. The company still surpasses their sales goals and consumers get a product at less cost, but with more features.

As long as the ratio between the increased sales due to video exceeds the marginal cost of including the feature, the per unit cost (which is what we consumers should be concerned with) is actually lower with video than without.  This is what is called the "industrial revolution" and it happened about 150 years ago.

[paul13walnut5]

Ah but the cost of the plastic switch could have made the 5D3 36MP instead. 

Or some such absolute bo11ocks.

I've stopped banging my head against the wall.  Nikon trumped Canon for a change, so it must be the fault of the video (which the D800 also has)

[Orangutan]

I was asking the simple question - has the 5DIII design been compromised as a stills camera by incorporating video?

Probably not.  We can never know for certain.   Therefore it's not worth fretting about.

Not all simple questions have simple answers.

[preppyak]

Really?  How do you come to that conclusion?  On a chip that is neither natively 5:4, 4:3 or 16:9 aspect into the bargain?

I'll let another user answer that for you

a slight increase to 22mp is intended for EOS-Movie, horizontal pixels 5760 is exactly 3 times of 1920, this indicates that 5D Mark III will do a full-sensor readout and 3x3 pixel binning to produce clean, sharp and almost-no-moire HD video.

Now, that's hypothetical for the reasoning, though it makes sense to make a sensor that can be used among many cameras (including the C cameras) that works well with video capture. But, they might also have found 22mp to be compromise between low-light and resolution that they liked best.

That moire is basically non-existent in video would lead me to believe that the sensor size isn't accidentally a solution for that, but, I wouldn't say its the only reason. Canon settled on 21mp for the 5dII with video as an after-thought; it might just be the magic number for them

Are there any components where compromises have to be made which will detrimentally impact the design for a still camera? - for example 22mp is a video size, perhaps it might have been as cheap and easy to make a 28mp sensor which was optimised for still image

It's an interesting question. I'd have to imagine that there was a slight compromise there...though, a true video person would be even happier with 4x4 binning, which would mean a sensor in the 35mp range. Canon have made sensors MUCH larger than that, so, they are clearly capable of such a thing. I imagine they tested it and decided that for the target market (wedding, event, indie movie set), low-light handling was more important than that added resolution, and their sensors only allowed them to go so far.

So, sure, it could have been 28mp, but, do you really need that extra resolution if it cost camera a stop of light, or didn't solve issues from the older 5dII (moire, etc)

[AvTvM]

This is ridiculous. Video doesn't increases the cost of cameras, it reduces it.

wrong.

As long as a KB-sensored "dedicated videocam" with video-performance similar to a 5D3 costs in excess of 10k, Canon figures, they can charge a much higher price than they could for a video-less stills camera. Obviously Video-users will happily pay anywhere up to 9999 - thus driving up the price of the cam for all stills-only users who are forced to cross-subsidize the video crowd, as Canon does not offer a video-less version of the camera (say a a reasonable 2500) but prefers to demand a whopping 3.500 for a video-enabled camera that only a minority of purchasers will ever use to regularly capture video.   

The only one "spoiling the game" for Canon is Nikon with its D800, which offers the whole video shebang plus better stills performance for 600 less. This is why Nikon will sell a boatload of their new DSLRs and is gaining market share against Canon in 2012.

[preppyak]

This is why Nikon will sell a boatload of their new DSLRs and is gaining market share against Canon in 2012.

Got a link for that? As I imagine it'd be pretty impossible for the D800 to have gained Nikon market-share when its barely even available in stores

[dilbert]

I was asking the simple question - has the 5DIII design been compromised as a stills camera by incorporating video?

No.

[Orangutan]

Apparently we've got three lines of reasoning:

1. Basic video features increase the cost to produce the camera enough to affect end-user price.  This one is likely false, but ultimately unknowable.  (marginal cost to add basic video is low, and overall sales would suffer without video)

2. High-quality video features alter design decisions, resulting in compromise of still features.  Maybe, but it's Canon's game and they can do what they want.  The market will sort this out.

3. High-quality video features drive up consumer demand, resulting in higher end-user prices.  If true, this is music to Canon's ears, and this would be a great business decision.  Like it or hate it, we're stuck with it.

For high-end bodies (5 and up) it's possible that high-quality video increases price through consumer demand.  If so, there's nothing to be done but vote with your wallet.

For consumer-level bodies (7D, XXd and below) video is a requirement of this market segment, but the quality is not high enough to pull in serious video enthusiasts, so body price is not affected.

These are just opinions and speculations, I have no hard data.

[paul13walnut5]

For consumer-level bodies (7D, XXd and below) video is a requirement of this market segment, but the quality is not high enough to pull in serious video enthusiasts, so body price is not affected.

Good enough for the Avengers and Black Swan though @Orangutan

I found the 5D2 impossible to shoot on without all the bells and whistles, and the bells and whistles defeated the point of the small form, also depth of field was so unforgiving, and the rolling shutter is much worse on the 5D2 than anything thats came since, including the 7D, 550D 600D.

The 7D also has live HDMI output at 1080, which for a long time made it the best option for folk wanting hidef live monitoring.   In fact when I bought my 7D it was the only DSLR with a dedicated PAL /25p mode.

And I'm not a serious enthusiast.  I actually get paid.  No DSLR is going to replace a conventional ENG camera for many types of job, but for some they offer awesome bang per buck.

And whichever method of pixel binning (the 3x3 would worry me, as this then takes no account of the bayer pattern, which would actually cause more false colouration) if the AA filter isn't right (and it isn't on a camera optimised for 18, 21 or 22MP) then you still need to do a little work in post, which along with selective use of MarvelsDSLR plug in (free) irons out most moire problems.

[Orangutan]

Good enough for the Avengers and Black Swan though @Orangutan

Granted, there are exceptions, but it's probably not enough to affect the price.

[preppyak]

For high-end bodies (5 and up) it's possible that high-quality video increases price through consumer demand.  If so, there's nothing to be done but vote with your wallet.

For consumer-level bodies (7D, XXd and below) video is a requirement of this market segment, but the quality is not high enough to pull in serious video enthusiasts, so body price is not affected.

I'd say you nailed it...though, I think the 7D might fall into category 1 simply because it hits a market that wasn't really covered in any way (action sports). To get a weather-proofed body that is very portable with great IQ; it basically didn't exist. Now they have 7D's (and GoPro's), and its proliferated that market a good deal.

But that is less because of the IQ, it's more for the build quality in such an affordable camera.

[unfocused]

Apparently we've got three lines of reasoning:

1. Basic video features increase the cost to produce the camera enough to affect end-user price.  This one is likely false, but ultimately unknowable.  (marginal cost to add basic video is low, and overall sales would suffer without video)

2. High-quality video features alter design decisions, resulting in compromise of still features.  Maybe, but it's Canon's game and they can do what they want.  The market will sort this out.

3. High-quality video features drive up consumer demand, resulting in higher end-user prices.  If true, this is music to Canon's ears, and this would be a great business decision.  Like it or hate it, we're stuck with it.

For high-end bodies (5 and up) it's possible that high-quality video increases price through consumer demand.  If so, there's nothing to be done but vote with your wallet.

For consumer-level bodies (7D, XXd and below) video is a requirement of this market segment, but the quality is not high enough to pull in serious video enthusiasts, so body price is not affected.

These are just opinions and speculations, I have no hard data.

You are missing the most important and obvious option:

High-quality video features drive up consumer demand, resulting in more sales and reducing the per unit price to still photographers.

[AvTvM]

Granted, there are exceptions, but it's probably not enough to affect the price.

Ibelieve demand for video-use 7D / 5D2s though a minority of purchasers has still been high enough to impact  the price.

We would know for sure, had Canon sold these cams in "still photography version" and a "Video-enabled" version for twice the price. These DSLRs would have still been dirt cheap for professional video use. In that model, Canon would have made those users pay, who are actually using the video crap, rather than the majority of purchasers who hardly ever use their DSLRs for video.

In the current model stills photographers are actually subsidizing commercial video use and get cameras that are not optimized for still photography but for video use. 

[preppyak]

In the current model stills photographers are actually subsidizing commercial video use and get cameras that are not optimized for still photography but for video use.

Consider all the upgrades that differentiate the 5dII from the 5dIII and tell me which were critical to videographers...and which were critical to a stills shooter. You'd really have to be insane to say its not optimized for stills, since the only video specific upgrade is the choice of video codecs.

6fps, 61pt AF, a processor dedicated to AF, hell, even the MP aren't what a videographer would demand normally (more or less would be more ideal). It has no XLR inputs, it has no ND filters, no focus peaking, no zebras, I could go on and on. If the 5DIII was an 8MP camera that did 4k video (like Sony's FS-700), you might have a point.

If you want Canon to charge double for a video specific camera...they did. Look at all the cine cameras Canon just put out. The 1D-C literally costs double the 1D-X

[Orangutan]

You are missing the most important and obvious option:

High-quality video features drive up consumer demand, resulting in more sales and reducing the per unit price to still photographers.

Unfocused, I'm basically on your side, and I did not miss that argument: I was just re-iterating my understanding of the viewpoints of the "yes" camp.  One one hand, more sales drive down per-unit cost by spreading fixed costs across a larger number of units.  On the other hand, end-user price *could* hypothetically go up if demand exceeds supply, resulting in higher per-unit profit.  E.g. iPads.  Does it actually do this?  I don't know, and I doubt we have access to data to prove one way or another.  We only know that per-unit production cost goes down with increased sales.

My main argument against the "yes it does" camp is that there is no reason to believe that video adds significantly to the R&D and manufacturing costs.  Any price increase would be due to market forces, and you can't expect any company to forgo extra profit to accommodate a minority of purists.

Personally, I believe the 5D3 was designed as a wedding/event camera, and it hits that spot pretty nicely. 

In the end, we just don't know the answer, so it's not worth fretting over.