Latest News - Nikon D5000

The LV sensor gets the exact same amount of light that the main
sensor gets.
No, it doesn't. There are three mirrors and a focus screen sitting
between the light leaving the lens and the LV sensor.
I am sure the mirrors suck up a lot of the light, as does the focusing screen.

I just took a light meter to a 30 year old focusing screan for a hassy (not known to be the brightest - it isn't an accumat) and lost a stop thru the it. So, allow me to revise my comment:

The LV sensor gets 50% of light that the main sensor gets, which can be compensated thru specialized design and pixel pitch.

--
It is ALL a conspiracy!
 
The LV sensor gets the exact same amount of light that the main
sensor gets.
No, it doesn't. There are three mirrors and a focus screen sitting
between the light leaving the lens and the LV sensor.
I am sure the mirrors suck up a lot of the light, as does the
focusing screen.

I just took a light meter to a 30 year old focusing screan for a
hassy (not known to be the brightest - it isn't an accumat) and lost
a stop thru the it. So, allow me to revise my comment:

The LV sensor gets 50% of light that the main sensor gets, which can
be compensated thru specialized design and pixel pitch.

--
It is ALL a conspiracy!
--

That's a possibility, although losing a stop of light is pretty critical. One of the advantages of the video of the Nikon DLSRs (and more so with the ff 5Dii) is that you get a relatively large sensor with a large enough pixel pitch to facilitate good low light performance. Using a much smaller secondary sensor, with half the light hitting it to begin with, would really put Sony at a disadvantage in this regard. Maybe Sony should name their main sensor live view "Slow View," and that way they can emphasis that it's for slow focusing, and it would characterize all of the other DSLR maker's live view as slow?? lol.
 
That's a possibility, although losing a stop of light is pretty
critical. One of the advantages of the video of the Nikon DLSRs (and
more so with the ff 5Dii) is that you get a relatively large sensor
with a large enough pixel pitch to facilitate good low light
performance. Using a much smaller secondary sensor, with half the
light hitting it to begin with, would really put Sony at a
disadvantage in this regard.
Not only that - they would need to use a focus screen without AF sensor markings (Pentax K2000 anyone?) and with minimal coarseness (which together with the small and dim viewfinder would make successful MF effectively impossible).

Thus you would end up with a camera which for the sake of a video function with some (limited) AF functionality (let's face it - DSLR lenses and AF modules haven't exactly been designed for video) has its still photo capability significantly impaired. I'm not sure whether that would be a good direction for Sony to take.

But then again I'm not holding my breath for Sony to introduce video with their next DSLR generation after that recent somewhat sobering interview with the Sony rep ( http://translate.google.co.uk/translate?u=http%3A%2F%2Fwww.focus-numerique.com%2Fnews_id-1283.html&sl=fr&tl=en&hl=en&ie=UTF-8 ).
 
Canon 500D - Sensor size - 22 x 15mm
Pixels / mm - 216

Estimated A350 LV sensor width - 10mm wide (determined by measuring the width of a viewfinder - 12mm)

1080p width - 1980 pixels
pixels/mm - 198

It has better pitch, ad gets more intense light (viewfinder funnel effect)

Also, I never see AF sensors when I look thru the view finder of my D300 until it locks on, so I am willing to be that Sony, even with all of their shortcomings, can find a way to make the AF marks invisible when video was on.

--
It is ALL a conspiracy!
 
Canon 500D - Sensor size - 22 x 15mm
Pixels / mm - 216

Estimated A350 LV sensor width - 10mm wide (determined by measuring
the width of a viewfinder - 12mm)

1080p width - 1980 pixels
pixels/mm - 198

It has better pitch, ad gets more intense light (viewfinder funnel
effect)

Also, I never see AF sensors when I look thru the view finder of my
D300 until it locks on, so I am willing to be that Sony, even with
all of their shortcomings, can find a way to make the AF marks
invisible when video was on.
Ok, I'm with you so far. I'd guess the biggest hurdle would be the amount of light that the focusing screen absorbs, and the development cost of putting in the LCD overlay like Nikon uses. Maybe by using the new EXMOR R in the viewfinder, Sony would negate much of the light lost through the viewfinder mirrors and focusing screen? If so, the biggest lost would be the extra DOF, which could actually be a positive to some. All of this being said, why not just use the main sensor? Because of AF?
 
All of this being said, why not just use the
main sensor? Because of AF?
AF - yes. And, I am sure they do not want to have to back track from the position that the 2 sensor LV system is better. If they could make it work, I think it would be another positive thing to set them apart from the rest of the crowd.

--
It is ALL a conspiracy!
 
I just don't see
Sony's current implementation of secondary sensor live view as a
logical way to get video into their DSLRs. Smaller sensor, less
light hitting it, AF points and various focus screen issues.
Maybe not in 2008. But, have you checked your calendar lately, buddy? It's 2009! ;)

--
It is ALL a conspiracy!
 
Canon 500D - Sensor size - 22 x 15mm
Pixels / mm - 216

Estimated A350 LV sensor width - 10mm wide (determined by measuring
the width of a viewfinder - 12mm)
10mm - hardly!! After all sensors cost money - and the bigger they are the more they cost. 10mm would be significantly larger than the sensors used in todays digicams. While I haven't found the specs for the LV sensor size, I guess it's safe to assume that they chose a fairly small sensor with a lens in the front to adjust the image circle.
1080p width - 1980 pixels
pixels/mm - 198

It has better pitch
(extremely doubtful (s.a.)),
and gets more intense light (viewfinder funnel
effect)
what exactly is the "viewfinder funnel effect" ? We had just established that the LV sensor gets less light, hadn't we?
Also, I never see AF sensors when I look thru the view finder of my
D300 until it locks on, so I am willing to be that Sony, even with
all of their shortcomings, can find a way to make the AF marks
invisible when video was on.
I wouldn't be too sure of that. IIRC Nikon holds some pretty effective patents on viewfinder overlays (which is why no other manufacturer offers something similar).
 
I congratulate you on getting it all figured out.

I certainly hope that Sony has people heading up their design teams that are creative enough to get around "big rocks in the road". Or, as some apparently call them "insurmountable roadblocks".

--
It is ALL a conspiracy!
 
There is a lot for us to glean from this move at Nikon and the recent low end Canon announcement in terms of its implication to the dSLR photographic industry in general and our Sony dSLRs in particular. Up until this point in time, as a general rule, same-generation higher end cameras received higher end sensors and produced technically superior pictures. For all intents that is no longer true or at least much less true. The Canon prosumer 50D and its much less expensive brethren the 500D have similar sensor and output. The pro/prosumer D300 body, the enthusiast D90 and the just announced high-end consumer D5000 all have the same or similar sensor. That’s essentially the same sensor as in the Sony A700 and they must all be made by Sony. I suspect this means that what we are looking at in the new unannounced lower end Sony’s will be this same sensor. That’s not a bad thing of course. There will be very modest IQ gains in the a700 replacement; the technology just isn’t there. And IMO increasing pixels in an APS camera is counter productive right now and is more a marketing ploy. We are back to the film-era camera days where camera body innovations are slow in coming…especially innovations that really enhance the photographic experience. IQ improvements will be even slower in coming.

Like it or not, live view and its variants will lead to a blurring of video and still photography. It will enhance some people’s experience, probably not mine. It’s coming however. What bothers me is that we seem to be getting sidetracked in our vision of what we need to take photographs; what I view as REAL improvements for us as photographers. We can always use improvements in auto focus, we can always use improvements in dynamic range, we can always use improvements to micro adjust individual lenses as in the a900, we can always use a better viewfinder and we can always use improvements in in-camera processing. Look at how little press and forum bandwidth is given to the a900’s wonderful film-like curve in the highlights. I don’t understand why it isn’t universally applauded. Instead I read slams on high ISO noise (I suspect they are intrinsically linked to the way these images are processed). I don’t need an in-camera ‘web sized B&W psychedelic posterization filter’; please. I have Photoshop. I don’t need an in-camera television either but we’ll probably see one before long.

With all that said there are big unknowns to me, other than Sony’s introduction timeline. Mainly what is their worldview of the positioning and economic viability of 12-15MP full-frame sensor, lower-noise camera. Clearly they have the wherewithal to produce one. I’d personally love one at a decent price-point. But are there enough of ‘ME’ to warrant making one? For that matter is there enough of me to warrant making an a700 replacement that isn’t a knee jerk reaction to the wants of their marketing department.

Time will tell, not my idle speculation. LOL

Bruce

--
http://www.pbase.com/misterpixel
 
There is a lot for us to glean from this move at Nikon and the recent
low end Canon announcement in terms of its implication to the dSLR
photographic industry in general and our Sony dSLRs in particular.
Up until this point in time, as a general rule, same-generation
higher end cameras received higher end sensors and produced
technically superior pictures. For all intents that is no longer
true or at least much less true. The Canon prosumer 50D and its much
less expensive brethren the 500D have similar sensor and output. The
pro/prosumer D300 body, the enthusiast D90 and the just announced
high-end consumer D5000 all have the same or similar sensor. That’s
essentially the same sensor as in the Sony A700 and they must all be
made by Sony. I suspect this means that what we are looking at in
the new unannounced lower end Sony’s will be this same sensor.
That’s not a bad thing of course. There will be very modest IQ gains
in the a700 replacement; the technology just isn’t there. And IMO
increasing pixels in an APS camera is counter productive right now
and is more a marketing ploy. We are back to the film-era camera
days where camera body innovations are slow in coming…especially
innovations that really enhance the photographic experience. IQ
improvements will be even slower in coming.

Like it or not, live view and its variants will lead to a blurring of
video and still photography. It will enhance some people’s
experience, probably not mine. It’s coming however. What bothers me
is that we seem to be getting sidetracked in our vision of what we
need to take photographs; what I view as REAL improvements for us as
photographers. We can always use improvements in auto focus, we can
always use improvements in dynamic range, we can always use
improvements to micro adjust individual lenses as in the a900, we can
always use a better viewfinder and we can always use improvements in
in-camera processing. Look at how little press and forum bandwidth
is given to the a900’s wonderful film-like curve in the highlights. I
don’t understand why it isn’t universally applauded. Instead I read
slams on high ISO noise (I suspect they are intrinsically linked to
the way these images are processed). I don’t need an in-camera ‘web
sized B&W psychedelic posterization filter’; please. I have
Photoshop. I don’t need an in-camera television either but we’ll
probably see one before long.

With all that said there are big unknowns to me, other than Sony’s
introduction timeline. Mainly what is their worldview of the
positioning and economic viability of 12-15MP full-frame sensor,
lower-noise camera. Clearly they have the wherewithal to produce
one. I’d personally love one at a decent price-point. But are there
enough of ‘ME’ to warrant making one? For that matter is there
enough of me to warrant making an a700 replacement that isn’t a knee
jerk reaction to the wants of their marketing department.

Time will tell, not my idle speculation. LOL

Bruce

--
http://www.pbase.com/misterpixel
--

Well said, but I disagree with the last paragraph. The Sony Exmor chip has the lowest read noise around at low ISO, and Canon's CMOS has lower read noise at high ISO. It has nothing to do with pixel numbers, and everything to do with sensor design. So far, Sony is unable to get high quality enough, small gain amps into the in-column chip architecture, so the price is paid at higher ISO. That is part of the reason why some have found that keeping the A900 at ISO 320-400 or less and boosting in the RAW converter provides lower noise than shooting at ISO 800+ in camera. At low ISO, the EXMOR-chipped cameras are the best around. Producing a lower MP EXMOR FF chip would save in hard drive space and be faster for high speed shooting, but it would not be lower in high ISO noise than a 24MP version downsampled, unless Sony drastically changed their CMOS chip design (assuming the same AA and CFA were used.) If Sony plans a 10fps sport shooter, then I could imagine a 15MP FF chip, but that won't exactly be cheap :) If Sony plans a lower price FF chip, I'd bet they'd stick with 24MP in order to amortize the cost with the A900.
 
Even though I am sure sony knows more about what is and is not a dead end, and the pitfalls associated with each road, I am sure Sony will appreciate your enlightened conclusion and thank you for saving them the time.

--
Nobody is paying either of us for brand management
 
Sony is in a better position to take advantage of movie mode than any
of the other makers using main sensor LV.
That is certainly debatable but others can haggle over that.

They are at a disadvantage in other ways. Shifting to doing high quality video involves more than the camera body. Think about it, video is shooting movement, and shooting while moving, or making apparent motion.

First way to move, use a zoom lens to apparently move closer or farther away. Doing this with the current manual zooms on DSLR lenses is not going to be very smooth. So a new series of zoom lenses will have to be added to the DSLR lens set. These will have smooth power zoom, and will maintain perfect focus while zooming. Why will this be a problem for Sony? Well, Canon and Nikon have a pretty complete set of DSLR lenses, it won't hurt them much to shift their lens designers to designing the new system of lenses. But Sony still has a good ways to go to have a complete set of lenses. Their designers will have to be switched from completing that set to designing the new video set, actually probably redesigning every zoom they have. This will slow way down the production of new still photography lenses. Or if they don't do that their conversion of DSLRs to video cameras will fall behind the others. Sony cannot just bring across the lenses from their current video cameras as those are designed for a different sensor system. Though they should at least know what a good video lens is like.

Second way to move is to pan. And pan is not just horizontal but in virtually any kind of direction. Again it must be smooth and is often combined with zoom. This is more a problem for end users than Sony. All the folks with their ball head tripods will find those not very good for this. Video heads are a different category, mostly damped three way heads. So either folks will carry two heads along with their tripod or give up their ball heads. The video heads can be used for stills.

One does not have to pan with a tripod, one can buy steadycams or shoulder mounts. And the camera will work better for this if it's designed to balance on the shoulder, which will require redesign of both camera body and lenses. With that the current optical viewfinder will have little reason to exist as it's designed for the current camera position close in front of the face. In any case the transition to video will probably mean the end of the optical viewfinder as it would be blacked out for shooting anyway.

Another area of some interest is how SSS will interact with video. Stabilization in video has generally been in lens or post processing of the sensor output. if SSS does not work well with video then Sony will be at a disadvantage against companies that use in lens systems. They will have to start over on stabilization. probably by importing their stabilization from their current video cameras. At least that will be designed for the full framing rate of video.

The only advantage Sony really has in this conversion of DSLRs to video cameras is that they are already highly experienced in building high quality video gear.

Of course if this video thing is just to produce P&S style video then only some of the changes will be necessary. But the argument seems to be that there must be video in DSLR's to do high quality video movies.

Walt
 
All of this being said, why not just use the
main sensor? Because of AF?
AF - yes. And, I am sure they do not want to have to back track from
the position that the 2 sensor LV system is better. If they could
make it work, I think it would be another positive thing to set them
apart from the rest of the crowd.

--
It is ALL a conspiracy
Let's all hope that Sony can offer a camera with the "fast" LV like the A3XX series but with movie mode. This would give them a decided advantage over the competition IMO.

How could they do it? I'm not sure. I don't design cameras for a living. That's why Sony has an engineering force.

I do hope however that they would not be petty and short sighted to the point that they would do a course correction if they to. Sometimes the other guy has the best idea. If you can beat it, join it (but I'm not suggesting Canon/Nikon have the best movie solution).

--
Rick
 
Like Walt mentioned, it'll be interesting to see how Sony deals with video stabilization, since Canon/Nikon have it in the lens. If sensor based stabilization works with video, then that's another issue with having movie mode off of the secondary sensor.
 
Even though I am sure sony knows more about what is and is not a dead
end, and the pitfalls associated with each road,
They sure should by now, after they hit so many of them in the last years (dead-ends, that is).
I am sure Sony will
appreciate your enlightened conclusion and thank you for saving them
the time.
I won't get my hopes up. So far they never did. They usually just send a check "for services related to brand management". Didn't you ever get one of those? Frankly, I could do with a bottle of wine or some pounds of kobe beef and a thank you note from Sir Howard (Stringer) - money is so impersonal, but what can you do?
--
Nobody is paying either of us for brand management
Oops, so I assume you don't get those checks then...
 
What bothers me
is that we seem to be getting sidetracked in our vision of what we
need to take photographs; what I view as REAL improvements for us as
photographers. We can always use improvements in auto focus, we can
always use improvements in dynamic range, we can always use
improvements to micro adjust individual lenses as in the a900, we can
always use a better viewfinder and we can always use improvements in
in-camera processing.
This is the exact problem of the current thrust. None of it improves image quality a bit. The development effort involved in these things is not insignificant.

If the word is truly that image quality does not count, as it is now, then we know what's going to happen. it will not be pretty. And many, myself included will be highly disappointed.

If film were at all economically viable and still being supported then that's where i'd be looking. But it is not, and with every brand of camera just becoming a display horse for bells and whistles it's very sad. I no longer expect to see development effort put into improving photographic quality. It no longer helps you with reviewers, end users don't care. So why spend money on it.

I was considering today if I had the money i should switch to the top end Canon FF cameras. Not because I think they are better, I don't. I do not even think FF is better. But because they have the greatest proportion of folks who make a living off high quality DSLR still photography. The death of the DSLR and image quality will reach there last. But it's a hugely overpriced area to go into. I don't have the money for such things.

Sony has stated that they have no plans to put video into their DSLRs, but that does not mean we will necessarily see improvements in image quality in the next DSLRs they introduce. It could just mean that they want to do the video better and have all their development effort going into video for farther down the road. They are at a point they can choose, continue to improve image quality of DSLRs or go to re-inventing them to hold bells and whistles. I do think it's only a matter of time until the bells and whistles are all we hear about from them.

Walt
 
Like Walt mentioned, it'll be interesting to see how Sony deals with
video stabilization, since Canon/Nikon have it in the lens. If
sensor based stabilization works with video, then that's another
issue with having movie mode off of the secondary sensor.
Stabilization is an excellent point. How does Sony do it in their current camcorders? I thought they did it electronically but I don't know the details (I think the sensor is bigger than the frame and they change the framing on the sensor ). Will we possibly see a hybrid camera? And is it in Sony's best interest to offer video in the DSLRs if it eats into the camcorder market? I imagine if DSLR video becomes something more than a novelty then I think Sony will have no choice but to add it in some form to remain competitive.

--
Rick
 

Keyboard shortcuts

Back
Top