Point & shoots, no improvement as long as sensors stay SMALL

David J Taylor wrote:
> Doug McDonald wrote:
>> David J Taylor wrote:
>>
>>> Sensor quantum efficiency has along way to go in both compact and
>>> DSLR cameras, but today the cost is a limitation.
>>>
>> Huh? The sensor QE is pushing the limit of 100%, at least
>> at the best wavelengths. OF course, lots and lots of
>> light is lost in the Bayer filter. That's why pro video
>> cameras use three CCDs and dichroic prisms to split the
>> colors without losing light.
>>
>> Doug McDonald
>
> Oh, OK, so why do the professional applications demand back-illumination?
>

The last little bit. But first things first: the Bayer filter is THE
big light-eater! You want more photoelectrons, use a prism-dichroic
system and three CCDs. Of course, this means a completely redesigned
optical system with a very long back focus, or a relay lens.

Doug McDonald
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

Doug McDonald wrote:

> The last little bit. But first things first: the Bayer filter is THE
> big light-eater! You want more photoelectrons, use a prism-dichroic
> system and three CCDs.

The original Foveon studio camera worked like that. Too bad the silicon
color separation didn't work as well as was hoped, but I read somewhere
that Canon was working on something similar.
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

SMS 斯蒂文• å¤� wrote:

> Doug McDonald wrote:
>
>> The last little bit. But first things first: the Bayer filter is THE
>> big light-eater! You want more photoelectrons, use a prism-dichroic
>> system and three CCDs.
>
>
> The original Foveon studio camera worked like that. Too bad the silicon
> color separation didn't work as well as was hoped, but I read somewhere
> that Canon was working on something similar.

There has been at least one hand-held digital camer which used dichroic
mirrors to split the image into different colours for capture on
seperate sensors.
It was made by Minolta and was hellishly expensive, due to having
multiple image sensors and a complicated mirror/prism setup inside of it.
Every extra part adds to the cost of building an item.
With a 3ccd digicam with an acceptable resolution for still photography,
I would imagine that the big cost would be aligning the whole system, as
even a small miss-alignment with a mirror or sensor could produce rather
hideous artifacts in the final image.
Not to mention that you'd also be paying for three high-resolution image
sensors instead of just one.
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

On Wed, 14 Nov 2007 12:00:22 +1100, dj_nme <dj_nme.RemoveThis@hotmail.com> wrote:

>SMS ???• ? wrote:
>
>> Doug McDonald wrote:
>>
>>> The last little bit. But first things first: the Bayer filter is THE
>>> big light-eater! You want more photoelectrons, use a prism-dichroic
>>> system and three CCDs.
>>
>>
>> The original Foveon studio camera worked like that. Too bad the silicon
>> color separation didn't work as well as was hoped, but I read somewhere
>> that Canon was working on something similar.
>
>There has been at least one hand-held digital camer which used dichroic
>mirrors to split the image into different colours for capture on
>seperate sensors.
>It was made by Minolta and was hellishly expensive, due to having
>multiple image sensors and a complicated mirror/prism setup inside of it.
>Every extra part adds to the cost of building an item.
>With a 3ccd digicam with an acceptable resolution for still photography,
>I would imagine that the big cost would be aligning the whole system, as
>even a small miss-alignment with a mirror or sensor could produce rather
>hideous artifacts in the final image.
>Not to mention that you'd also be paying for three high-resolution image
>sensors instead of just one.

Maybe someone else has the link, I no longer remember the URL, but there's some
new sensor-array development coming down the line where this very routine is
going to be applied at the pixel level. A larger micro-lens is going to focus
the light into a series of two 45-degree dichroic mirrors beneath to send 2
colors off to the side and the 3rd continue to the base of the sensor array. The
extra lateral space needed for the side-sensors will be compensated for by the
larger micro-lens above the stack. Giving 100% of the sensor area usable at full
intensity for all colors. So instead of splitting the colors with dichroic
mirrors before hitting 3 different sensor arrays, working out the difficulties
of alignment, etc., just one sensor-array will accomplish it all without any
extra space needed in front of it so all pre-existing optics and camera body
configurations will be useable with it.
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

In article <2eukj3l8e3qck6ml0f7ud4jhvg20cfrtal DeleteThis @4ax.com>, Charlie Danke
<cdanke DeleteThis @mailinator.org> wrote:

> Maybe someone else has the link, I no longer remember the URL, but there's
> some
> new sensor-array development coming down the line where this very routine is
> going to be applied at the pixel level. A larger micro-lens is going to focus
> the light into a series of two 45-degree dichroic mirrors beneath to send 2
> colors off to the side and the 3rd continue to the base of the sensor array.

that's the nikon sensor that has a patent but nothing more than that
(yet). it is also not clear that this will actually be better than
current sensors, or able to be manufactured for a reasonable price.

<http://www.dpreview.com/news/0708/07080901nikonimagesensor.asp>
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

In article <fhcj18$qoq$1@news.ks.uiuc.edu>, Doug McDonald says...

> Huh? The sensor QE is pushing the limit of 100%, at least
> at the best wavelengths. OF course, lots and lots of
> light is lost in the Bayer filter. That's why pro video
> cameras use three CCDs and dichroic prisms to split the
> colors without losing light.

Well no, QE is in the 20-30% range. It's close to 100% only for back-
illuminated CCDs. Ask Roger if you don't believe me.
--

Alfred Molon
------------------------------
Olympus 50X0, 7070, 8080, E3X0, E4X0 and E5X0 forum at
http://tech.groups.yahoo.com/group/MyOlympus/
http://myolympus.org/ photo sharing site
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

Alfred Molon wrote:
> In article <fhcj18$qoq$1@news.ks.uiuc.edu>, Doug McDonald says...
>
>> Huh? The sensor QE is pushing the limit of 100%, at least
>> at the best wavelengths. OF course, lots and lots of
>> light is lost in the Bayer filter. That's why pro video
>> cameras use three CCDs and dichroic prisms to split the
>> colors without losing light.
>
> Well no, QE is in the 20-30% range. It's close to 100% only for back-
> illuminated CCDs. Ask Roger if you don't believe me.

I'll trust you. But I'm surprised. Most modern design CCDs are
in the 80% range at 800 nm even if not back illuminated. 30% at
400 nm I could believe as typical.

Doug McDonald
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

Scott W biphoto DeleteThis @hotmail.com posted to rec.photo.digital:

> JosephKK wrote:
>> nospam nospam DeleteThis @nospam.invalid posted to rec.photo.digital:
>>
>>> In article <4734d22d$0$79918$742ec2ed@news.sonic.net>, SMS
>>> ÊñØËíÇÊñáÂ’Ä¢ §è <scharf.steven DeleteThis @geemail.com> wrote:
>>>
>>>>> So do none of the other Live View DSLR models also offer
>>>>> auto-focus in
>>>>> Live View? If it's Live View, doesn't it to have to be contrast
>>>>> detection?
>>>> I was only listing non-D-SLRs with hybrid auto-focus. Yes, of
>>>> course you're correct, the live view D-SLRs will also use
>>>> contrast detection when using live view.
>>> not all slrs with live view have contrast detection. canon, for
>>> example, does not. to obtain focus in live view, they drop the
>>> mirror and focus.
>>
>> Moving parts (other than in the lens) in a DSLR? There is no
>> design reason to do such a thing.
>>
>
> You really don't know how a DSLR works, do you?
> The mirror has to swing out of the way for each shot.
>
> Scott

That is how _film_ SLR cameras work, i know, i have some. With
digital there is no need for such mechanical contrivances, thus they
are eliminated to keep costs down, i know, i have some. There is
another twit here that thinks that they still use mechanical shutters
in DSLRs. What part of "electronic shutters are faster and more
reliable" can't the twit understand? As for camera sounds they were
included in cheaper cameras via sounders to relax the Dale six-pack
types by sounding like all that mechanical ruckus were actually still
going on.
Can you say marketing?
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

-hh recscuba_google.RemoveThis@huntzinger.com posted to rec.photo.digital:

> Scott W <biph....RemoveThis@hotmail.com> wrote:
>> JosephKK wrote:
>> >
>> > Focal-plane shutters are from the film
>> > SLR technology. Electronic sensors do not need any such thing.
>>
>> Well yes he may be an idiot, but DSLR do have focal plane shutters.
>
> Unless they were to re-introduce the pellicle mirror, such as what
> was in the 1995 vintage Canon EOS-1N RS.
>
>> This is why there is a limit on the flash sync speed.
>
> It depends. Mirror movement is one factor, but so too are the speed
> of the shutter's blades: typically at the higher speeds, because
> the closing blade is trailing behind the opening blade, the frame is
> never fully open, so if you were to use a strobe at this time, you
> would only expose part of the total frame.
>
> If you want to see this, I recall that on some older cameras, you
> can
> force this condition to occur. One way to try to circumvent this
> shutter speed limitation is to modulate the strobe in pulses, IIRC,
> timed for illuminating each partial frame.
>
>
> -hh

That is film camera properties. Not digital camera properties.
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

Scott W biphoto DeleteThis @hotmail.com posted to rec.photo.digital:

> IDIOT ALERT! wrote:
>> On Sun, 11 Nov 2007 13:41:57 -1000, Scott W <biphoto DeleteThis @hotmail.com>
>> wrote:
>>
>>> -hh wrote:
>>>> Scott W <biph... DeleteThis @hotmail.com> wrote:
>>>>> JosephKK wrote:
>>>>>> Focal-plane shutters are from the film
>>>>>> SLR technology. Electronic sensors do not need any such thing.
>>>>> Well yes he may be an idiot, but DSLR do have focal plane
>>>>> shutters.
>>>> Unless they were to re-introduce the pellicle mirror, such as
>>>> what was in the 1995 vintage Canon EOS-1N RS.
>>>>
>>>>> This is why there is a limit on the flash sync speed.
>>>> It depends. Mirror movement is one factor, but so too are the
>>>> speed
>>>> of the shutter's blades: typically at the higher speeds, because
>>>> the closing blade is trailing behind the opening blade, the frame
>>>> is never fully open, so if you were to use a strobe at this time,
>>>> you would only expose part of the total frame.
>>>>
>>>> If you want to see this, I recall that on some older cameras, you
>>>> can
>>>> force this condition to occur. One way to try to circumvent this
>>>> shutter speed limitation is to modulate the strobe in pulses,
>>>> IIRC, timed for illuminating each partial frame.
>>> I was not clear in what I wrote.
>>> You are right, at fast shutter times it is a slit that moves
>>> across the film/sensor. This is what limits the shutter time when
>>> using the flash, unless you put the flash in high speed mode, but
>>> what a waste of power.
>>>
>>> I have never seen distortion from a slit moving across the sensor,
>>> just does not seem to be problem, as much as one person would want
>>> us to believe it is.
>>>
>>
>> Yes, let's all just put the blinders on (how else can they justify
>> buying a DSLR?), and ignore this "just does not seem to be problem"
>> that only happens with a focal-plane shutter.
>>
>> http://farm1.static.flickr.com/8/8798593_5cebc86c21.jpg
>>
>> LOL
>>
>> No wonder none of them can ever get any photos out of their cameras
>> that are worth a plug-nickel. They can't even see something that is
>> this obviously wrong with their chosen cameras.
>>
>> "Totally blind" is too kind of a description for them. But it does
>> explain so many things about the blatant misinformation that they
>> relentlessly projectile-vomit about different cameras and
>> photography as a whole. I bet you were hoping that you'd still have
>> some credibility after this. Kiss any last ounce of it, that you
>> ever imagined having, completely good-bye.
>>
>> Let's add "psychotic" and "fool" to the "total idiot" requirement
>> needed to be a DSLR owner, just to be more accurate.
>>
>>
> I have taken many a photo of helicopters in flight with a DSLR and
> never gotten a curved blade.
>
> I shoot both point and shoots and DSLRs and have a pretty good idea
> of what each can do, it seems you have never shoot a DSLR.
>
> Scott

A history lesson for anybody that may care, toss "copal square shutter
history" sans quote marks to any search engine. Konica (equivalent)
from the 1960s. Apparently first sold in Nikkormat cameras.
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

Ron Hunter rphunter RemoveThis @charter.net posted to rec.photo.digital:

> SMS 斯蒂文• å¤� wrote:
>> Ron Hunter wrote:
>>
>>> While sensor size, at whatever level of advancement plays a
>>> significant role in image quality, it is inaccurate to say that
>>> 'no improvement' in image quality is possible without a larger
>>> sensor. Other factors come into play, and the improvements ARE
>>> substantial, and they continue.
>>
>> The only "improvement" is noise reduction software which is always
>> a compromise. Unfortunately the physics are what they are, and
>> noise is directly related to pixel size. It will continue to get
>> worse until the pixel size stops shrinking (at least on Bayer
>> sensors).
>>
>> Until there's an alternative to the Bayer sensor, we'll continue to
>> see degradation in high-ISO noise if the sensor size stays the same
>> but the resolution goes up. Look at Nikon, which for years derided
>> full frame sensors as being unnecessary, but finally capitulated
>> when they just could not compete with Canon any longer at the high
>> end.
>
> Digital signal processors get better all the time. Other
> improvements, just from my first camera to the current one are:
> Better display, better lens, more resolution, multiple scene modes,
> faster focus, better focusing options, viewfinder with diopter
> adjustment, burst mode, better view/review, closeup mode, closer
> focus, red-eye reduction (not worth much), and better movie mode,
> and last, but not least, much shorter shutter lag.
> Now if you want to call those 'no improvements' we just will have to
> agree to disagree, because they were quite significant to me.

System usability improvements, yes they are, sensitivity vs noise
improvements, no they are not. SMS is right on this one.
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

JosephKK wrote:
> Ron Hunter rphunter.TakeThisOut@charter.net posted to rec.photo.digital:
>
>> SMS 斯蒂文• å¤� wrote:
>>> Ron Hunter wrote:
>>>
>>>> While sensor size, at whatever level of advancement plays a
>>>> significant role in image quality, it is inaccurate to say that
>>>> 'no improvement' in image quality is possible without a larger
>>>> sensor. Other factors come into play, and the improvements ARE
>>>> substantial, and they continue.
>>> The only "improvement" is noise reduction software which is always
>>> a compromise. Unfortunately the physics are what they are, and
>>> noise is directly related to pixel size. It will continue to get
>>> worse until the pixel size stops shrinking (at least on Bayer
>>> sensors).
>>>
>>> Until there's an alternative to the Bayer sensor, we'll continue to
>>> see degradation in high-ISO noise if the sensor size stays the same
>>> but the resolution goes up. Look at Nikon, which for years derided
>>> full frame sensors as being unnecessary, but finally capitulated
>>> when they just could not compete with Canon any longer at the high
>>> end.
>> Digital signal processors get better all the time. Other
>> improvements, just from my first camera to the current one are:
>> Better display, better lens, more resolution, multiple scene modes,
>> faster focus, better focusing options, viewfinder with diopter
>> adjustment, burst mode, better view/review, closeup mode, closer
>> focus, red-eye reduction (not worth much), and better movie mode,
>> and last, but not least, much shorter shutter lag.
>> Now if you want to call those 'no improvements' we just will have to
>> agree to disagree, because they were quite significant to me.
>
> System usability improvements, yes they are, sensitivity vs noise
> improvements, no they are not. SMS is right on this one.
>

Read the subject line. Is there anything about sensitivity vs noise
there? BTW, Kodak has made a change to the way the Bayer sensor works
that CLAIMS to improve the light sensitivity, but I haven't see proof
that it will make any difference. Will have to wait and see, but I have
a lot of confidence in the ability of technology to overcome obstacles
to progress.
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL 

"JosephKK" <joseph_barrett RemoveThis @sbcglobal.net> wrote in message
news:W1Q_i.6892$TR5.2745@nlpi061.nbdc.sbc.com...
> Scott W biphoto RemoveThis @hotmail.com posted to rec.photo.digital:
>
>> JosephKK wrote:
>>> nospam nospam RemoveThis @nospam.invalid posted to rec.photo.digital:
>>>
>>>> In article <4734d22d$0$79918$742ec2ed@news.sonic.net>, SMS
>>>> ÊñØËíÇÊñá’Ä¢ §è <scharf.steven RemoveThis @geemail.com> wrote:
>>>>
>>>>>> So do none of the other Live View DSLR models also offer
>>>>>> auto-focus in
>>>>>> Live View? If it's Live View, doesn't it to have to be contrast
>>>>>> detection?
>>>>> I was only listing non-D-SLRs with hybrid auto-focus. Yes, of
>>>>> course you're correct, the live view D-SLRs will also use
>>>>> contrast detection when using live view.
>>>> not all slrs with live view have contrast detection. canon, for
>>>> example, does not. to obtain focus in live view, they drop the
>>>> mirror and focus.
>>>
>>> Moving parts (other than in the lens) in a DSLR? There is no
>>> design reason to do such a thing.
>>>
>>
>> You really don't know how a DSLR works, do you?
>> The mirror has to swing out of the way for each shot.
>>
>> Scott
>
> That is how _film_ SLR cameras work, i know, i have some. With
> digital there is no need for such mechanical contrivances, thus they

This is a joke, right?
You've removed the lens of your dSLR and verified that there is no mirror?
Or are you doing this all on the experience of your vivid imagination?
 >> Stay informed about: Point & shoots, no improvement as long as sensors stay SMALL