So it appears there are still advances to be made, improvements yet to
come...
Ok, it's only a 5Mp sensor at this point, but if this design (which
makes me think, why haven't they done this before? (or have they? -
this isn't really my field)...) can be produced relatively cheaply and
for higher res sensors as well, it will put some good competitive
pressure back into the market.
Mark Thomas wrote:
> (Just beating Rich or Focus to it.. don't go here if a Sony hater..(O
>
> http://www.imaging-resource.com/NEWS/1213308645.html
>
> So it appears there are still advances to be made, improvements yet to
> come...
>
> Ok, it's only a 5Mp sensor at this point, but if this design (which
> makes me think, why haven't they done this before? (or have they? -
> this isn't really my field)...) can be produced relatively cheaply and
> for higher res sensors as well, it will put some good competitive
> pressure back into the market.
I didn't see where it was cheap but that'd be great. I would get a Nikon
D30 Fx format even at 4MP if it was cheap & small.
"Conventionally, consumer digital video camcorders and digital still
cameras have been required to combine high resolution capable of
capturing every detail of the subject matter, and miniaturization
oriented to portability. In order to meet these needs, image sensor
development has focused on miniaturizing pixel size, while maintaining
imaging characteristics. However, in addition to these ongoing
requirements, in recent years demand for improved levels of minimum
subject illuminance and rapid image capture has also increased. This has
led to the requirement of image sensors with improved signal-to-noise
ratio and other features capable of realizing improved overall picture
quality."
It seems I've heard of back illuminated in astro & scientific work...
Mark Thomas wrote:
> (Just beating Rich or Focus to it.. don't go here if a Sony
> hater..(O
> http://www.imaging-resource.com/NEWS/1213308645.html
>
> So it appears there are still advances to be made, improvements yet to
> come...
>
> Ok, it's only a 5Mp sensor at this point, but if this design (which
> makes me think, why haven't they done this before? (or have they? -
> this isn't really my field)...) can be produced relatively cheaply and
> for higher res sensors as well, it will put some good competitive
> pressure back into the market.
I just wish that Fuji had presented a document like this, instead of the
mystique that they have allowed to surround their supposed high-ISO
performance.
Mark Thomas wrote:
> (Just beating Rich or Focus to it.. don't go here if a Sony hater..(O
>
> http://www.imaging-resource.com/NEWS/1213308645.html
>
> So it appears there are still advances to be made, improvements yet to
> come...
>
> Ok, it's only a 5Mp sensor at this point, but if this design (which
> makes me think, why haven't they done this before? (or have they? -
> this isn't really my field)...) can be produced relatively cheaply and
> for higher res sensors as well, it will put some good competitive
> pressure back into the market.
Thanks Mark,
Interesting.
As Sony are definitely not noise champions, this could help them catch
up to Canon. As I said recently, it would be good if they could reduce
noise coming in before trying to filter it with resultant resolution loss.
I wonder if they intend to ramp this up to a FF sensor at 16 Mpix or
higher...
(They'd probably sell it to Nikon beofore putting it in their own body.)
Mark Thomas wrote:
> (Just beating Rich or Focus to it.. don't go here if a Sony hater..(O
>
> http://www.imaging-resource.com/NEWS/1213308645.html
>
> So it appears there are still advances to be made, improvements yet to
> come...
>
> Ok, it's only a 5Mp sensor at this point, but if this design (which
> makes me think, why haven't they done this before?
<snip>
From the site: " However, compared to conventional
front-illuminated structures, back-illuminated structures
commonly causes problems such as noise, dark current,
defective pixels and color mixture that lead to image
degradation and also cause a decrease in the signal-to-noise
ratio." There's your answer.
In article <BAx4k.4229$Mu.1821@trndny07>, Marvin says...
> From the site: " However, compared to conventional
> front-illuminated structures, back-illuminated structures
> commonly causes problems such as noise, dark current,
> defective pixels and color mixture that lead to image
> degradation and also cause a decrease in the signal-to-noise
> ratio." There's your answer.
Why would a gain in sensitivity by a factor of 3 produce higher noise
levels? Doesn't make sense.
Spectral efficiency of standard front-illuminated sensors is less than
30%, while it's close to 100% for back-illuminated sensors.
--
? "Paul Furman" <paul-@-edgehill.net> ?????? ??? ??????
news:IEn4k.7947$jI5.7657@flpi148.ffdc.sbc.com...
> Mark Thomas wrote:
>> (Just beating Rich or Focus to it.. don't go here if a Sony hater..(O
>>
>> http://www.imaging-resource.com/NEWS/1213308645.html
>>
>> So it appears there are still advances to be made, improvements yet to
>> come...
>>
>> Ok, it's only a 5Mp sensor at this point, but if this design (which makes
>> me think, why haven't they done this before? (or have they? - this isn't
>> really my field)...) can be produced relatively cheaply and for higher
>> res sensors as well, it will put some good competitive pressure back into
>> the market.
>
> I didn't see where it was cheap but that'd be great. I would get a Nikon
> D30 Fx format even at 4MP if it was cheap & small.
>
> "Conventionally, consumer digital video camcorders and digital still
> cameras have been required to combine high resolution capable of capturing
> every detail of the subject matter, and miniaturization oriented to
> portability. In order to meet these needs, image sensor development has
> focused on miniaturizing pixel size, while maintaining imaging
> characteristics. However, in addition to these ongoing requirements, in
> recent years demand for improved levels of minimum subject illuminance and
> rapid image capture has also increased. This has led to the requirement of
> image sensors with improved signal-to-noise ratio and other features
> capable of realizing improved overall picture quality."
>
> It seems I've heard of back illuminated in astro & scientific work...
>
> Large-Format Science-Grade CMOS Active Pixel Sensors for Extreme ...
> http://www.ercim.org/publication/Erc...5/waltham.html
>
> 515 Full Frames per Second
> Ideal for highly dynamic, low light experiments.
Yes, but have you heard of the F frame cat sensor? It can be found in any
adult cat, and because it has a tapetus lucidum has excellent night vision
(low light), also very good MP rating in bright light, the problem is
meowing which can distract from picture taking.
--
Tzortzakakis Dimitrios
major in electrical engineering
mechanized infantry reservist
hordad AT otenet DOT gr
Tzortzakakis Dimitrios wrote:
> Paul Furman wrote:
>> Mark Thomas wrote:
>>
>>> (Just beating Rich or Focus to it.. don't go here if a Sony hater..(O
>>> http://www.imaging-resource.com/NEWS/1213308645.html
>>> So it appears there are still advances to be made, improvements yet to
>>> come...
>>>
>>> Ok, it's only a 5Mp sensor at this point, but if this design (which makes
>>> me think, why haven't they done this before? (or have they? - this isn't
>>> really my field)...) can be produced relatively cheaply and for higher
>>> res sensors as well, it will put some good competitive pressure back into
>>> the market.
>>
>> I didn't see where it was cheap but that'd be great. I would get a Nikon
>> D30 Fx format even at 4MP if it was cheap & small.
I meant 5MP as stated. 6MP would be better.
>> "Conventionally, consumer digital video camcorders and digital still
>> cameras have been required to combine high resolution capable of capturing
>> every detail of the subject matter, and miniaturization oriented to
>> portability. In order to meet these needs, image sensor development has
>> focused on miniaturizing pixel size, while maintaining imaging
>> characteristics. However, in addition to these ongoing requirements, in
>> recent years demand for improved levels of minimum subject illuminance and
>> rapid image capture has also increased. This has led to the requirement of
>> image sensors with improved signal-to-noise ratio and other features
>> capable of realizing improved overall picture quality."
>>
>> It seems I've heard of back illuminated in astro & scientific work...
>>
>> Large-Format Science-Grade CMOS Active Pixel Sensors for Extreme ...
>> http://www.ercim.org/publication/Erc...5/waltham.html
>>
>> 515 Full Frames per Second
>> Ideal for highly dynamic, low light experiments.
>
> Yes, but have you heard of the F frame cat sensor? It can be found in any
> adult cat, and because it has a tapetus lucidum has excellent night vision
> (low light), also very good MP rating in bright light, the problem is
> meowing which can distract from picture taking.
Meowing wouldn't bother me if the cat could produce a raw file <g>.
? "Paul Furman" <paul-@-edgehill.net> ?????? ??? ??????
news:8nI4k.5452$89.1757@nlpi069.nbdc.sbc.com...
> Tzortzakakis Dimitrios wrote:
>> Paul Furman wrote:
>>> Mark Thomas wrote:
>>>
>>>> (Just beating Rich or Focus to it.. don't go here if a Sony hater..(O
>>>> http://www.imaging-resource.com/NEWS/1213308645.html
>>>> So it appears there are still advances to be made, improvements yet to
>>>> come...
>>>>
>>>> Ok, it's only a 5Mp sensor at this point, but if this design (which
>>>> makes me think, why haven't they done this before? (or have they? -
>>>> this isn't really my field)...) can be produced relatively cheaply and
>>>> for higher res sensors as well, it will put some good competitive
>>>> pressure back into the market.
>>>
>>> I didn't see where it was cheap but that'd be great. I would get a Nikon
>>> D30 Fx format even at 4MP if it was cheap & small.
>
> I meant 5MP as stated. 6MP would be better.
>
>>> "Conventionally, consumer digital video camcorders and digital still
>>> cameras have been required to combine high resolution capable of
>>> capturing every detail of the subject matter, and miniaturization
>>> oriented to portability. In order to meet these needs, image sensor
>>> development has focused on miniaturizing pixel size, while maintaining
>>> imaging characteristics. However, in addition to these ongoing
>>> requirements, in recent years demand for improved levels of minimum
>>> subject illuminance and rapid image capture has also increased. This has
>>> led to the requirement of image sensors with improved signal-to-noise
>>> ratio and other features capable of realizing improved overall picture
>>> quality."
>>>
>>> It seems I've heard of back illuminated in astro & scientific work...
>>>
>>> Large-Format Science-Grade CMOS Active Pixel Sensors for Extreme ...
>>> http://www.ercim.org/publication/Erc...5/waltham.html
>>>
>>> 515 Full Frames per Second
>>> Ideal for highly dynamic, low light experiments.
>>
>> Yes, but have you heard of the F frame cat sensor? It can be found in any
>> adult cat, and because it has a tapetus lucidum has excellent night
>> vision (low light), also very good MP rating in bright light, the problem
>> is meowing which can distract from picture taking.
>
> Meowing wouldn't bother me if the cat could produce a raw file <g>.
>
Of course, it will eat a raw fillet;-)
--
Tzortzakakis Dimitrios
major in electrical engineering
mechanized infantry reservist
hordad AT otenet DOT gr
[A complimentary Cc of this posting was sent to
Alfred Molon
<alfred_molon@yahoo.com>], who wrote in article <MPG.22bccd0e26085cb898bcb3@news.supernews.com>:
> > From the site: " However, compared to conventional
> > front-illuminated structures, back-illuminated structures
> > commonly causes problems such as noise, dark current,
> > defective pixels and color mixture that lead to image
> > degradation and also cause a decrease in the signal-to-noise
> > ratio." There's your answer.
>
> Why would a gain in sensitivity by a factor of 3 produce higher noise
> levels? Doesn't make sense.
??? A radically different manufacturing process could produce radically
different noise level...
> Spectral efficiency of standard front-illuminated sensors is less than
> 30%, while it's close to 100% for back-illuminated sensors.
Do not think so. Front-illuminated sensor have the apex QE (in their
spectral-response curve) closer to 45% (it is the Bayer filter which
reduces the QE 3x, to the current value of about 15%; but Bayer filter
remain on most of these new designes). I should not expect more than
90% back-illuminated performance from a mass-produced first-generation
chip (of course, in 3 years it may become much better than
scientific-grade equipment - due to major R&D money flow).
8dB snr increase? 
Linear Mode
