overnight flight next to Canon engineer...

[TravelShooter]

...recently, who opened up just a bit:

1. prototypes constructed constantly regardless of price
2. all are built around the sensor
3. sensors are most expensive camera piece
4. video has taken over as driving force despite still advances, too
5. most prototypes do not go to market because it cannot be figured
out how to bring down costs to sell enough units

Summarizing:  your "dream" cameras exist as prototypes
but are too expensive to become market models

[ontarian]

Really interesting stuff, thanks for sharing the good insight.  The video part makes me sad, the sensor part happy.

[jseliger]

The video comment does make sense: for non-pros, the image quality difference among major dSLRs is so low as to be negligible. Occasionally I point friends to this article: http://www.kenrockwell.com/tech/mpmyth.htm on the subject, although I think Rockwell understates the value of cropping. Nonetheless, for most people, cameras became "good enough" a long time ago. Hence the popularity of cell phone cameras among so many people.

But video is still the "developing" phase, and it can be used to differentiate between dSLRs. Video drove me to sell an XTi and buy a t2i. That, and high Craigslist prices on the former and relatively low eBay prices on the latter, but still.

[Fleetie]

I know I've said this before, but I really dislike - resent, even - the idea that a significant part of the cost of the 5D3 that I'll buy will have gone into developing video functionality that I will never, EVER, use.

Just sell me a 5D3 without any video functionality, commensurately cheaper, please!

Or, spend the same money enhancing the stills functionality.

If I wanted a camcorder, guess what: I'd go out to a camcorder shop, and buy one.

Sorry to repeat the same old rant.

[motorhead]

Fleetie,

Well said, I could not agree more. Video cameras are a completely different animal and should stay that way.   

[Gothmoth]

The video comment does make sense: for non-pros, the image quality difference among major dSLRs is so low as to be negligible. Occasionally I point friends to this article:

quoting rockwell as a source .... lol.. good joke.   
that guy says tomorrow the opposit of what he said today.
as long as he makes a few bucks out of his website.

he will tell you that 12 MP are enough for most people (something i can agree too... and better photographer then him said that before) and on the next page he will tell you how great film is and how he can scan ~40MP out of analog film and how superior film is over digital.

[kubelik]

I know I've said this before, but I really dislike - resent, even - the idea that a significant part of the cost of the 5D3 that I'll buy will have gone into developing video functionality that I will never, EVER, use.

Just sell me a 5D3 without any video functionality, commensurately cheaper, please!

Or, spend the same money enhancing the stills functionality.

If I wanted a camcorder, guess what: I'd go out to a camcorder shop, and buy one.

Sorry to repeat the same old rant.

do you still resent the video aspect when you realize that, without video functionality, there would be no 5D3 forthcoming?

a lot of the success of the 5D2 was canon's successful ability to integrate new functionality in multiple vectors while reducing the cost of the body from its predecessor.  without video, products simply would not be coming to market at all, at any price, because they would not be competitive in today's market.

this common rant to me is something like if someone were to complain about anti-lock brakes in cars.  how often do you truly use the anti-lock brake function in your car?  how often do you really need your seat belts?  or airbags for that matter?  or high-beam lamps?  or the spare tire?  heck, cars worked fine before variable-valve timing and direct fuel injection, why should we as consumers carry the cost of research into fuel-efficient technologies?

[bwhitz]

I know I've said this before, but I really dislike - resent, even - the idea that a significant part of the cost of the 5D3 that I'll buy will have gone into developing video functionality that I will never, EVER, use.

Just sell me a 5D3 without any video functionality, commensurately cheaper, please!

Or, spend the same money enhancing the stills functionality.

If I wanted a camcorder, guess what: I'd go out to a camcorder shop, and buy one.

Sorry to repeat the same old rant.

This just shows a sever lack of understanding about why the 5D became popular for video. The video you can capture is MILES ahead of any "camcorder" in terms of aesthetics. The 5D is one of only TWO digital camera systems in the world that records FF35 video.

As far as to why people would want that for video over a regular camcorder at the camcorder store... well because it DOESN'T LOOK LIKE A CAMCORDER. If you think that going from FF35 to APS-C is bad... try FF35 compared to 1/3" video camera chip... then come back and talk about video on the 5DIII being nonsense. It opened up a completely new look for cinematographers and directors that was, no more than 5 years ago, only available on $250,000 camera systems.

[torger]

I know I've said this before, but I really dislike - resent, even - the idea that a significant part of the cost of the 5D3 that I'll buy will have gone into developing video functionality that I will never, EVER, use.

Since 5Dmk3 will be a mass market product I don't think the video development costs will lead to that much extra cost in the product itself. Similar type of hardware would still be required for a stills-only product. And if I've understood correctly, live view exists partly thanks to video development, and I surely love my live view for still life photography.

I'll start complain when they start optimizing EF optics for video rather than stills though... :-)

[Fleetie]

I know I've said this before, but I really dislike - resent, even - the idea that a significant part of the cost of the 5D3 that I'll buy will have gone into developing video functionality that I will never, EVER, use.

Just sell me a 5D3 without any video functionality, commensurately cheaper, please!

Or, spend the same money enhancing the stills functionality.

If I wanted a camcorder, guess what: I'd go out to a camcorder shop, and buy one.

Sorry to repeat the same old rant.

This just shows a sever lack of understanding about why the 5D became popular for video. The video you can capture is MILES ahead of any "camcorder" in terms of aesthetics. The 5D is one of only TWO digital camera systems in the world that records FF35 video.

As far as to why people would want that for video over a regular camcorder at the camcorder store... well because it DOESN'T LOOK LIKE A CAMCORDER. If you think that going from FF35 to APS-C is bad... try FF35 compared to 1/3" video camera chip... then come back and talk about video on the 5DIII being nonsense. It opened up a completely new look for cinematographers and directors that was, no more than 5 years ago, only available on $250,000 camera systems.

If the 5D2 is so great at video, then why do I keep seeing people on here whining about serious video problems in the 5D2 that Canon "really should have fixed by now, and had BETTER fix in the 5D3!"?

Based on that, I will not accept that the 5D2 is such a dream for video.

So yes, I do have a "serious lack of understanding" of why video is such a good thing. Video in a DSLR is something I do not want, and that I think just adds useless and unwanted purchase cost to the many of us who buy them for stills only. And I bet that the majority (I do dare say "majority") of 5D2 owners do not use the video functionality at all, or to any significant extent.

[epsiloneri]

I know I've said this before, but I really dislike - resent, even - the idea that a significant part of the cost of the 5D3 that I'll buy will have gone into developing video functionality that I will never, EVER, use.

And as I've also said it before, it is not evident that including video features would make the 5D3 more expensive. If video features contribute to sell more units, they may actually make the camera less expensive.

[Chewy734]

And I bet that the majority (I do dare say "majority") of 5D2 owners do not use the video functionality at all, or to any significant extent.

I would have to agree with that statement as well. Although I know this cannot be considered a good, randomized, worldwide sample size, but most people I know with a 5D2 use its video function rarely (if at all).

[neuroanatomist]

I'll start complain when they start optimizing EF optics for video rather than stills though... :-)

Oh, you mean like the new Power Focusing mode on the new MkII supertele primes, intended specifically for pulling focus during video shooting?

[papa-razzi]

And as I've also said it before, it is not evident that including video features would make the 5D3 more expensive. If video features contribute to sell more units, they may actually make the camera less expensive.

This really is true.  I have been product manager for several consumer electronics products - volume is the largest factor in driving down manufacturing costs.  Fixed costs get allocated to larger numbers of units and the per/unit allocation becomes smaller; parts are much cheaper when purchased in large volumes, the factory can be more efficient, etc.  If video features leverage existing parts and most of the work/cost is firmware, then yes, video will drive the cost of your still camera down - to the extent it drives volume up.

The concept of video R&D driving up the cost of your still camera, and not wanting to pay for the extra video features may seem intuitive at first, but it is missing the bigger picture with all the other things that impact the cost of the camera.

Personally, I rarely (as in I've used it twice in a year) use the video on my 7D, but it doesn't get in the way of anything I use the camera for, so I don't care about it.

[Bob Howland]

And as I've also said it before, it is not evident that including video features would make the 5D3 more expensive. If video features contribute to sell more units, they may actually make the camera less expensive.

This really is true.  I have been product manager for several consumer electronics products - volume is the largest factor in driving down manufacturing costs.  Fixed costs get allocated to larger numbers of units and the per/unit allocation becomes smaller; parts are much cheaper when purchased in large volumes, the factory can be more efficient, etc.  If video features leverage existing parts and most of the work/cost is firmware, then yes, video will drive the cost of your still camera down - to the extent it drives volume up.

The concept of video R&D driving up the cost of your still camera, and not wanting to pay for the extra video features may seem intuitive at first, but it is missing the bigger picture with all the other things that impact the cost of the camera.

Excellent points, especially about the firmware. In my experience, there are constant negotiations between Marketing who wants to add features to drive volume up and Engineering Management who wants to keep NRE (Non-Recurring Expenses) down. But I'm talking about things that cost $100,000 with volumes in the hundreds and low thousands.