Why no prosumer with APS-sized CCD ?
Alfred Molon
DSLR CCDs in a much more compact package (and don't need interchangeable
lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
such a compact camera ? Given that DLRs now cost less than $1000, it
can't be the cost alone.
--
Alfred Molon
------------------------------
Olympus 4040, 5050, 5060, 7070, 8080, E300 forum
at http://groups.yahoo.com/group/MyOlympus/
Olympus 8080 resource - http://myolympus.org/8080/
Roland Karlsson
news:MPG.1c4b414a3...@news.supernews.com:
> I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ? Given that DLRs now cost less than $1000, it
> can't be the cost alone.
Me to.
My guess is that the manufacturers, faulty or not,
believes that there exist no market.
The prosumer cameras have rather hefty zoom lenses with
rather good aperture value. Most are 3x zoom, many are
5x and some are even 10x.
If you use a larger sensor, you have to make compromises
there. Either less zoom or worse max aperture, if you
don't want large cameras that are very expensive. DSLR
normal zoom lenses are normally F/3.5 where prosumer
are F/2.0.
Moreover, the APS sized sensors used in SLR do not
allow for live preview. So, either you have to make
special sensors for those cameras or you have to
abandon live preview.
/Roland
Timo Autiokari
>I guess there are lots of people who would like to have the low
>noise of DSLR CCDs in a much more compact package,
Me too!
>so why has no manufacturer yet launched such a compact camera ?
>Given that DLRs now cost less than $1000, it can't be the cost alone.
If there was such a high quality point&shoot digicam available then
there would be a high risk that consumers would finally realize the
*extremely* poor quality of the current digicams. This could results
some law suits etc.
AAlso why to use a 20 USD imager chip in the P&S cameras when a 2 USD
imager sells very well already.
Timo Autiokari
Roland Karlsson
news:pfb2u09gh6gndd3i3...@4ax.com:
> If there was such a high quality point&shoot digicam available then
> there would be a high risk that consumers would finally realize the
> *extremely* poor quality of the current digicams. This could results
> some law suits etc.
I think my reply was more balanced :)
> AAlso why to use a 20 USD imager chip in the P&S cameras when a 2 USD
> imager sells very well already.
Hmmm ... that is a valid point .. although I would be surprised
if the chip in a DSLR only costs $20.
/Roland
Conrad Weiler
<< it can't be the cost alone >>
Don't bet on that. The beancounters add pennies.
I'd like to see actual size comparisons between CCD and CMOS sensors used in
digital cameras.
Best,
Conrad
Conrad Weiler
Camp Sherman, Oregon
Mark B.
news:MPG.1c4b414a3...@news.supernews.com...
>I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ? Given that DLRs now cost less than $1000, it
> can't be the cost alone.
> --
I believe it is mostly cost. A larger imager will require a larger lens
than what is going in the current crop of small-sensor compacts. Not only
that, they require better electronics to handle the larger files.
Mark
David J. Littleboy
"Conrad Weiler" <wei...@aol.com> wrote:
> << it can't be the cost alone >>
>
> Don't bet on that. The beancounters add pennies.
>
> I'd like to see actual size comparisons between CCD and CMOS sensors used
in
> digital cameras.
http://www.dpreview.com/learn/?/Glossary/Camera_System/Sensor_Sizes_01.htm
The 2/3" sensors are 96.8 sq mm, whereas the 20D is 345 sq mm. So you'd
expect the 20D to be about three to four times as sensitive for the same
noise, and, surprise, ISO 1600 on the 20D looks about the same as ISO 400 on
the 8MP P&S cameras.
David J. Littleboy
Tokyo, Japan
Alfred Molon
says...
> Moreover, the APS sized sensors used in SLR do not
> allow for live preview. So, either you have to make
> special sensors for those cameras or you have to
> abandon live preview.
Should be no problem making an APS-sized CCD with live preview. If the
tiny ones have this feature, there is no reason why the bigger CCDs
can't have it.
Alfred Molon
> I believe it is mostly cost. A larger imager will require a larger lens
> than what is going in the current crop of small-sensor compacts. Not only
> that, they require better electronics to handle the larger files.
The filesize should be the same - it only depends on the resolution, not
on the CCD size (unless of course the larger CCD camera generates RAW
files with more bit depth, but even then the increase would be moderate
- 16 bits vs 12 bits for instance).
Joseph Meehan
> I guess there are lots of people who would like to have the low noise
> of DSLR CCDs in a much more compact package (and don't need
> interchangeable lenses, an TTL viewfinder etc.), so why has no
> manufacturer yet launched such a compact camera ? Given that DLRs now
> cost less than $1000, it can't be the cost alone.
Maybe not a direct answer to your question, but keep in mind that the
way CCD's are manufactured and the limits placed on them by the
manufacturing process makes for certain "sweet" sizes. Changing the shape
or size can greatly change the cost of each unit. I suspect they are made a
number at a time on a circular blank. Making them say just a 1/16" larger
may reduce the yield per blank by 20% or more, thereby increasing the price
by 25%.
--
Joseph Meehan
26 + 6 = 1 It's Irish Math
Timo Autiokari
>I think my reply was more balanced :)
Well ... yes, considering that I forgot to mention that most of the
manufacturers do not want to provide such a high quality P&S digicam
simply because doing so would inevitably eat the market share of their
dSLRs.Those manufacturers want to sell the high price tag dSLR
glassware and other add-on too.
You did mention two very important issues, the lens (aperture and zoom
range) and the live preview feature.
I'm not sure what is the typical speed of the 3x lenses of the P&S
film cameras but I'd guess that f/2.8 can be quite easily made for APS
size, e.g. digicams do not require a lot of resolving power. I'd
gladly take a high quality digital P&S with "only" 3x zoom but I'd not
want a slow lens.
The live preview has effect on image quality versus imager size, a
sensor that does not provide the live preview is using all of its
active photon sensing area for capturing the image. Sensors with live
preview only use half of the active area for image capture and the
other half for the video stream. The larger the active area is the
higher is image quality so the live preview is very costly in terms of
image quality.
For the high quality P&S digicam I would gladly take the similar range
finder that the film P&S cameras have.
Timo Autiokari
Andrew Koenig
news:MPG.1c4b414a3...@news.supernews.com...
>I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ?
Because it wouldn't be particularly compact.
Roland Karlsson
> Should be no problem making an APS-sized CCD with live preview. If the
> tiny ones have this feature, there is no reason why the bigger CCDs
> can't have it.
Nope - but the point was that there are no such CCD.
You cannot just build such a camera.
/Roland
Roland Karlsson
> For the high quality P&S digicam I would gladly take the similar range
> finder that the film P&S cameras have.
The main problem is that "the market" does not understand that.
It is only a matter of comparing features. All cameras give
fantastic and outstanding pictures in the ads, so its all
about features. So - "live preview" wins over "higher quality".
You cannot tick off "higher qualiy" in the fact sheet,
only "high quality" - and that is always ticked off.
/Roland
Roland Karlsson
news.ops.worldnet.att.net:
> Because it wouldn't be particularly compact.
Compact 35 mm film cameras are compact.
Compact APS film cameras are even compacter.
Digital prosumer cameras can be even compacter,
but usually are not. On the contrary.
But you are right - if the APS camera shall have a 5x F/2.0
zoom, it will not be in particular compact. But, an APS
size sensor and a 30 mm F/1.4 lens would be super compact.
/Roland
Roland Karlsson
> I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ? Given that DLRs now cost less than $1000, it
> can't be the cost alone.
I have a question for you Alfred.
Why do you ask this question again and again?
This is at least the fourth time you ask it.
And you get the same answer every time.
Do you hope to get better answers?
/Roland
trad...@yahoo.com
get more features and a better product. Why should Nikon or Canon,
etc. put all the goodies in a P&S when they can allure you with their
higher-end products. What may happen is that a company that doesn't
have a professional series may come out with a prosumer camera with the
larger sensor, etc. and this will force the hand of the other
companies. I own a Nikon 8800 and it has a tack sharp lens from 35-380
with ED galss and VR. It has certain limitations as does every camera
in this category, however, if it was as fast as the D70 the 8800 would
steal away market share from semi-pros. With the D70 the consumer will
probably buy several lenses and that means more dollars for the
companies who make these cameras.
Charlie Self
>In article <Xns95D98CC12...@130.133.1.4>, Roland Karlsson
>says...
>
>> Moreover, the APS sized sensors used in SLR do not
>> allow for live preview. So, either you have to make
>> special sensors for those cameras or you have to
>> abandon live preview.
>
>Should be no problem making an APS-sized CCD with live preview. If the
>tiny ones have this feature, there is no reason why the bigger CCDs
>can't have it.
Sure. And putting wings on a battleship and flying it should be no problem. But
will it do anything useful?
Charlie Self
"One of the common denominators I have found is that expectations rise above
that which is expected." George W. Bush
Alfred Molon
says...
> The live preview has effect on image quality versus imager size, a
> sensor that does not provide the live preview is using all of its
> active photon sensing area for capturing the image. Sensors with live
> preview only use half of the active area for image capture and the
> other half for the video stream.
Not sure if I understand this, because you have to read out the image in
some way, even with a CCD with no live preview.
Alfred Molon
says...
> > I guess there are lots of people who would like to have the low noise of
> > DSLR CCDs in a much more compact package (and don't need interchangeable
> > lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> > such a compact camera ? Given that DLRs now cost less than $1000, it
> > can't be the cost alone.
>
> I have a question for you Alfred.
>
> Why do you ask this question again and again?
> This is at least the fourth time you ask it.
When was the last time I put this question (about the prosumer with the
APS-sized CCD) ?
Dutch Flyer
"te...@yahoo.com" <trad...@yahoo.com> wrote in message
news:1105285310.9...@c13g2000cwb.googlegroups.com...
A 35-380mm lens certainly isn't tack sharp, and would shows it's
ugly head on a APS-sized sensot very fast.
Alfred Molon
says...
> The live preview has effect on image quality versus imager size, a
> sensor that does not provide the live preview is using all of its
> active photon sensing area for capturing the image. Sensors with live
> preview only use half of the active area for image capture and the
> other half for the video stream. The larger the active area is the
> higher is image quality so the live preview is very costly in terms of
> image quality.
By the way, even if what you say is true, video streaming is just
320x240, so it's just 240 lines out of 2000+ (in a typical 6 or 8MP CCD)
which would have their area halved. One line out of eight, which would
translate into a 12% lower average ISO.
David J Taylor
[]
> A 35-380mm lens certainly isn't tack sharp, and would shows it's
> ugly head on a APS-sized sensot very fast.
It depends how much you pay for it!
andr...@littlepinkcloud.invalid
> I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ? Given that DLRs now cost less than $1000, it
> can't be the cost alone.
Such a camera would be very much like a D70, but with a fixed zoom
lens. I can't imagine why anybody would want such a thing.
Andrew.
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>I guess there are lots of people who would like to have the low noise of
>DSLR CCDs in a much more compact package (and don't need interchangeable
>lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
>such a compact camera ?
You can buy an Epson RD-1, tape over the lens-release button and pretend the
lens is nonremovable, I guess. It satisfies your other requirements - no TTL
viewing, compact, APS-C sensor.
Bit expensive though. Give it a couple of years, see if Zeiss and Cosina do
similar cameras.
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>In article <Xns95D9A8886...@130.133.1.4>, Roland Karlsson
>says...
>
>> > I guess there are lots of people who would like to have the low noise of
>> > DSLR CCDs in a much more compact package (and don't need interchangeable
>> > lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
>> > such a compact camera ? Given that DLRs now cost less than $1000, it
>> > can't be the cost alone.
>>
>> I have a question for you Alfred.
>>
>> Why do you ask this question again and again?
>> This is at least the fourth time you ask it.
>
>When was the last time I put this question (about the prosumer with the
>APS-sized CCD) ?
June, according to Google.
OK, back then, you said you didn't care if it was an SLR or not, as long as
it's compact, but this seems to be a rehash of that thread.
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>In article <gkf2u0tspeh8voo07...@4ax.com>, Timo Autiokari
>says...
>
>> The live preview has effect on image quality versus imager size, a
>> sensor that does not provide the live preview is using all of its
>> active photon sensing area for capturing the image. Sensors with live
>> preview only use half of the active area for image capture and the
>> other half for the video stream.
>
>Not sure if I understand this, because you have to read out the image in
>some way, even with a CCD with no live preview.
Such a sensor is predicated on the principle that you can read it out in the
dark, thus you don't need a duplicate charge store for each photosite which
isn't light-sensitive.
Once you introduce the need for continuous exposure, you can't take
advantage of that any more.
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>
>Should be no problem making an APS-sized CCD with live preview.
Your main reason given for wanting such a camera was low-noise. These
sensors are low noise due to their large capacity storage wells. Adding a
decent live readout capability reqauires a second cell of the same capacity
for each photosite. Doing this halves the capacity of the wells. This gives
you noise.
Roland Karlsson
news:10u2rud...@news.supernews.com:
> Such a camera would be very much like a D70, but with a fixed zoom
> lens. I can't imagine why anybody would want such a thing.
No, it would not. It could be much compacter.
/Roland
Frode P. Bergsager
> Every company has a line of products and as you go up in that line you
> get more features and a better product. Why should Nikon or Canon,
> etc. put all the goodies in a P&S when they can allure you with their
> higher-end products. What may happen is that a company that doesn't
> have a professional series may come out with a prosumer camera with
> the larger sensor, etc. and this will force the hand of the other
> companies.
Then why don't Sony produce this camera then? They have no SLR-series to
worry about, and they produce their own (and most others) sensors also.
Use your money to vote. This product is not technically impossible, but
there is no market for it unless people stop to buy prosumer cameras and buy
SLR instead. Then some companies may try "impossible" things to get their
customers back.
--
Frode P. Bergsager
Norway
Dave Martindale
>Should be no problem making an APS-sized CCD with live preview. If the
>tiny ones have this feature, there is no reason why the bigger CCDs
>can't have it.
Sure they *could* have it. But the feature eats some silicon area, so
the active light-sensing area of each photosite would be reduced
compared to a same-size sensor without live preview. And thus you lose
some of the noise advantage of the large sensor - which is why you
wanted a larger sensor in the first place.
Dave
Roland Karlsson
news:B0fEd.8929$HA5.1...@juliett.dax.net:
> Use your money to vote. This product is not technically impossible,
> but there is no market for it unless people stop to buy prosumer
> cameras and buy SLR instead. Then some companies may try "impossible"
> things to get their customers back.
If you look back att film cameras, then you can see that
pro and semipro SLR sold very well. Compact cameras with
no particular image quality at all also sold very well.
But - better compact cameras sold only to very special
people. Those cameras were also rather expensive.
Now - the camera makers does the same with digital cameras.
/Roland
Roland Karlsson
@mughi.cs.ubc.ca:
> Sure they *could* have it. But the feature eats some silicon area, so
> the active light-sensing area of each photosite would be reduced
> compared to a same-size sensor without live preview. And thus you lose
> some of the noise advantage of the large sensor - which is why you
> wanted a larger sensor in the first place.
Yepp - and this is probably the most important reason
why they don't make any compact cameras with large sensors.
They have to give up live preview to get all the noise
advantage. And taking away features is a "no no".
/Roland
paul
> I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ? Given that DLRs now cost less than $1000, it
> can't be the cost alone.
You would need an enormous lense to come close to the zoom range of P&S.
There doesn't even exist any such lense for 35mm with that wide of a
zoom range. Plus the DSLR's are not really sharper, just that they have
more dynamic range which is kind of subtle really for the P&S crowd it's
not something that jumps out at you like megapixels and super telephoto
ability.
Dave Martindale
>Not sure if I understand this, because you have to read out the image in
>some way, even with a CCD with no live preview.
The difference is whether the image needs to be read out with the
shutter open or closed. (This applies to CCD; I don't know if CMOS
sensors have similar tradeoffs).
In a DSLR-type sensor, you have nothing but the light-sensing photosites
themselves to hold and shift the charges around on the CCD. The
exposure sequence is something like:
0. (Note that the shutter is normally closed on a DSLR)
1. Read the CCD rapidly to flush all charge from it
2. Open the shutter for the exposure time, accumulating charge
proportional to light intensity
3. Close the shutter
4. Start reading out the CCD, sweeping charges to the corner(s) where
a charge to voltage conversion is done
The fundamental limitation of this sort of CCD is that it is *always
sensitive to light*. If you try to read out the image while the chip is
exposed to light, you'll get a smeared mess, as the charge packets pick
up additional electrons while passing through cells that are both
light-sensitive and being used as shift registers. Thus the readout
step (#4 above) is always done when the CCD is in darkness.
To avoid this smearing, chips capable of video output (which means that
they must be read out while still exposed to light) have separate
sensing and readout registers. Once per frame, the entire image is
transferred rapidly from the sensing cells to the readout cells (which
are not affected by light). Then the next exposure takes place in the
sensing cells in parallel with the previous exposure being read from the
readout cells.
In video CCDs, a common way to do this is simply to make the chip twice
as tall as it would otherwise be and cover the lower half with black
paint. During vertical retrace, the entire image from the upper
(light-sensing) portion of the chip is rapidly clocked into the lower
(black painted) portion. Then the lower half is read out during the
image time. But this makes the CCD twice the size of its sensing area,
which is not practical for large still camera CCDs.
Another method is to put the storage/transfer registers interleaved with
the rows of sensing cells. This takes silicon area away from the
sensors. It also means you're clocking the transfer registers while
hoping the signals don't affect the sensing cells a few microns away.
It's just easier to make everything work well if the exposure and
readout phases are separated in time, and the readout takes place in the
dark.
Dave
Dave Martindale
>By the way, even if what you say is true, video streaming is just
>320x240, so it's just 240 lines out of 2000+ (in a typical 6 or 8MP CCD)
>which would have their area halved. One line out of eight, which would
>translate into a 12% lower average ISO.
I can't see that ever working. If 1/8 of the sensor rows are
manufactured differently than the other 7/8 of the rows, it's nearly
impossible to make their response the same as the others. It's not a
matter of 12% lower ISO, it's a matter of having every eighth scanline
not match its neighbours, and thus getting streaks in the image.
If a CCD is set up for video output, *every* row and column will provide
video output. It's entirely possible that the electronics uses
"binning" (adding the charge from NxN cells together in the analog
electronics) for faster A/D conversion, but you've still got to pay the
price of the extra transfer registers for every pixel.
Dave
Alfred Molon
> Your main reason given for wanting such a camera was low-noise. These
> sensors are low noise due to their large capacity storage wells. Adding a
> decent live readout capability reqauires a second cell of the same capacity
> for each photosite.
Not for each photosite. Video streaming is only 320x240, i.e. only 240
lines out of a total of 2000+ would need to be modified.
Alfred Molon
> June, according to Google.
>
> OK, back then, you said you didn't care if it was an SLR or not, as long as
> it's compact, but this seems to be a rehash of that thread.
What thread was that ? Anyway, I can't remember all posts I make.
Alfred Molon
> Alfred Molon <alfred_mo...@yahoo.com> writes:
>
> >By the way, even if what you say is true, video streaming is just
> >320x240, so it's just 240 lines out of 2000+ (in a typical 6 or 8MP CCD)
> >which would have their area halved. One line out of eight, which would
> >translate into a 12% lower average ISO.
>
> I can't see that ever working. If 1/8 of the sensor rows are
> manufactured differently than the other 7/8 of the rows, it's nearly
> impossible to make their response the same as the others. It's not a
> matter of 12% lower ISO, it's a matter of having every eighth scanline
> not match its neighbours, and thus getting streaks in the image.
You would multiply the different row with a constant factor (or use a
different gain for the row amplifier). Noise will be a bit higher, but
only for that specific row (and not for the other 7 rows).
Roland Karlsson
> What thread was that ? Anyway, I can't remember all posts I make.
I do :)
No - I don't. But - I remembered this one.
/Roland
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>In article <c888b2-...@narcissus.dyndns.org>, Chris Brown says...
>
>> June, according to Google.
>>
>> OK, back then, you said you didn't care if it was an SLR or not, as long as
>> it's compact, but this seems to be a rehash of that thread.
>
>What thread was that ? Anyway, I can't remember all posts I make.
"Why go dSLR?" was the thread title.
Whilst looking this up, I note that you also started a thread in September,
"Why separate AF sensors in DSLRs ?", where the subject of DSLR sensors and
how they differ from the sensors in compacts was also done to death.
Asking almost the same question every 3 months isn't going to change the
answers, you know. If you want a big sensor, just buy a camera with one in.
There's even one that's not an SLR these days.
...or shoot film and scan it. You can use really huge image capture areas
then.
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>
>You would multiply the different row with a constant factor (or use a
>different gain for the row amplifier). Noise will be a bit higher, but
>only for that specific row (and not for the other 7 rows).
I'm sure you'll forgive the marketing departments of the various camera
manufacturers for not yet coming up with the appropriate marketing spin to
allow them to viably produce a sensor that overlays your image with bars of
noise every 8 lines.
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>In article <gj78b2-...@narcissus.dyndns.org>, Chris Brown says...
>
>> Your main reason given for wanting such a camera was low-noise. These
>> sensors are low noise due to their large capacity storage wells. Adding a
>> decent live readout capability reqauires a second cell of the same capacity
>> for each photosite.
>
>Not for each photosite. Video streaming is only 320x240, i.e. only 240
>lines out of a total of 2000+ would need to be modified.
But at that resolution, your image will be so piss-poor that you won't have
a clue about critical focusing. This is important because a larger sensor
and the focal lengths that accompany it give you a much smaller depth of
field, and stuff that would be in focus on a zoom compact, when focused
using a low-res preview, will be horribly blurred when taken on a larger
sensor.
Mark B.
news:MPG.1c4b6215f...@news.supernews.com...
> In article <Xns95D98CC12...@130.133.1.4>, Roland Karlsson
> says...
>
>> Moreover, the APS sized sensors used in SLR do not
>> allow for live preview. So, either you have to make
>> special sensors for those cameras or you have to
>> abandon live preview.
>
> Should be no problem making an APS-sized CCD with live preview. If the
> tiny ones have this feature, there is no reason why the bigger CCDs
> can't have it.
>
The much larger CCD & CMOS sensors used in DSLRs aren't capable of doing
live preview.
Mark
Roland Karlsson
news:4ll8b2-...@narcissus.dyndns.org:
> "Why go dSLR?" was the thread title.
>
> Whilst looking this up, I note that you also started a thread in
> September, "Why separate AF sensors in DSLRs ?", where the subject of
> DSLR sensors and how they differ from the sensors in compacts was also
> done to death.
>
> Asking almost the same question every 3 months isn't going to change
> the answers, you know. If you want a big sensor, just buy a camera
> with one in. There's even one that's not an SLR these days.
Hmmm - I thought it was 4 times. But that does not matter,
three times is at least one time too many.
But - I have also asked this question way back then.
And I am quite sure that Alfred was in that thread also.
That might be the fourth.
/Roland
Alfred Molon
> >> June, according to Google.
> >>
> >> OK, back then, you said you didn't care if it was an SLR or not, as long as
> >> it's compact, but this seems to be a rehash of that thread.
> >
> >What thread was that ? Anyway, I can't remember all posts I make.
>
> "Why go dSLR?" was the thread title.
>
> Whilst looking this up, I note that you also started a thread in September,
> "Why separate AF sensors in DSLRs ?", where the subject of DSLR sensors and
> how they differ from the sensors in compacts was also done to death.
>
> Asking almost the same question every 3 months isn't going to change the
> answers, you know.
1. I didn't start the "Why go dSLR?" thread, only made a couple of
posts.
2. "Why separate AF sensors in DSLRs ?" is a very different question
from "Why no prosumer with APS-sized CCD ?"
Justin Thyme
>I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ? Given that DLRs now cost less than $1000, it
> can't be the cost alone.
also the reason why people buy prosumer compacts in the first place. They
buy prosumer compacts instead of SLR's because they are more compact,
smaller, and offer a good range of zoom to cover most needs so there is no
need to change lenses.
Take for example the Pana FZ10. This camera has a lens that is the
equivalent to 420mm at F2.8. Because of it's small sensor, this lens can be
made fairly small and light. If the sensor was APS sized, the lens would
need to be 280mm to give the same equivalence. If you have ever held a 300mm
F2.8 lens, you would know that they are anything other than small and light.
Would you buy a camera with a lens like that? Ok, so if we want to make the
camera reasonably small and light then we have to reduce the lens size. To
give something about the same size as the current crop of prosumers, the
lens would fall to be something close to the equivalence of 28-80 F3.5-F5.6.
By coincidence this is the size lens that ships with DSLR's, because it is
small and compact. The prosumer camera of course would be pretty much the
same size as a DSLR if it had the same lens, so then we ask - if two cameras
are the same size, similar price, and have pretty much the same lens, would
you buy the one with the replaceable lens or the one with the fixed lens?
The answer is pretty obvious, because while 28-80 will cover 80% of needs
pretty well, most people will at some stage want the ability to go to a
longer lens.
But they made compact 35mm cameras and they have a 35mm sized sensor I hear
you say. Yes they did, but lenses longer than 150mm were pretty rare, the
very odd camera would go to 200mm. The Nikon Zoom 150ED for example has a
lens that zooms to 150mm, but at 150mm it is F12!!! even at it's widest
angle it is a pretty pedestrian F5.6.
So that is why there is no APS sized prosumer camera - it wouldn't be able
to offer any advantages. It would have to have either a very short lens,
thus losing one of it's big features (versatility), or it's lens would have
to become very slow.
If you really want a prosumer with an APS sized sensor, buy a
D70/EOS300/istDS/7D and superglue the standard lens to it. Except for live
preview, this would pretty much match what you would be getting if they made
an APS prosumer
paul
> In article <MPG.1c4ba8bc8...@news.supernews.com>,
> Alfred Molon <alfred_mo...@yahoo.com> wrote:
>
>> In article <gj78b2-...@narcissus.dyndns.org>, Chris Brown says...
>>
>> Video streaming is only 320x240, i.e. only 240
>> lines out of a total of 2000+ would need to be modified.
>
>
> But at that resolution, your image will be so...
I know this is sort of a different issue but the D70 sensor is about
420x280 and can zoom in to about 65% actual pixel size, just for reference.
Conrad Weiler
<< By the way, even if what you say is true, video streaming is just
>320x240, so it's just 240 lines out of 2000+ (in a typical 6 or 8MP CCD) >>
I have a 6MP Fuji S7000 that has 640x480 video.
Best,
Conrad
Conrad Weiler
Camp Sherman, Oregon
Darrell
"Chris Brown" <cpb...@ntlworld.no_uce_please.com> wrote in message
news:ud78b2-...@narcissus.dyndns.org...
The Epson is a Cosina...
FoulDragon
with a trick lens so it stays somewhat in sync with the main lens, and then the
real 35mm-sized sensor?
An interesting alternative: Use a SLR like rig, where the light's sent to a
second low-res sensor for the preview befoe the press.
A third approach: Develop a sensor which can either get scanned 30 times a
second and produce crap, yet ample for preview, images, or work like a normal
sensor.
I'm sort of imagining (if it's possible), assemble the guts of a nice 6MP
sensor on top of a 640x480 sensor of similar size.
--
Marada Shra'drakaii
Chris Brown
The problem with zooming in in Alfred's scenario is that he's not capturing
the data which you'd want to zoom into in the first place, because he's only
modified sufficient cells to drive his low-res display.
Having a zoom abaility requires modifying more, which means you get more
noise.
Chris Brown
Roland Karlsson <roland_do...@bonetmail.com> wrote:
>Chris Brown <cpb...@ntlworld.no_uce_please.com> wrote in
>news:4ll8b2-...@narcissus.dyndns.org:
>> Asking almost the same question every 3 months isn't going to change
>> the answers, you know. If you want a big sensor, just buy a camera
>> with one in. There's even one that's not an SLR these days.
>
>Hmmm - I thought it was 4 times. But that does not matter,
>three times is at least one time too many.
I said every 3 months, not just 3 times. ;-)
And you're right. It does get a bit silly.
Chris Brown
Alfred Molon <alfred_mo...@yahoo.com> wrote:
>
>1. I didn't start the "Why go dSLR?" thread, only made a couple of
>posts.
Ah, so a few months later, you decide the issue needs to be rehashed in a
thread of its own?
>2. "Why separate AF sensors in DSLRs ?" is a very different question
>from "Why no prosumer with APS-sized CCD ?"
Go through the thread in Google. It ended up covering pretty much exactly
the same gound that is being covered here again.
It's not clear what you're expecting to hear that's any different from the
last time. The sensor technology hasn't really changed, apart from Fuji's
little oddity, which seems to be more like the apochraphal talking dog than
anything else, in that the interest is not in doing it well, but in doing it
at all.
Chris Brown
Indeed it is. I meant Cosina releasing one themselves under the Voigtlander
brand, rather than Epson building a digicam around their body. The current
situation is a bit like what Kodak used to do with Canon and Nikon film
bodies, ripping their guts out and turning them into DSLRs.
Zeiss Ikon is the one to watch. Their recent reentry into the 35mm
rangefinder market looks suspiciously like they're actually developing a
lens line for a digital M mount camera, and the current film bodies are
testing the water.
Chris Brown
Justin Thyme <pleasedo...@nowhere.com> wrote:
>
>If you really want a prosumer with an APS sized sensor, buy a
>D70/EOS300/istDS/7D and superglue the standard lens to it. Except for live
>preview, this would pretty much match what you would be getting if they made
>an APS prosumer
Even then, he could just pretend the viewfinder was a really high
resolution, wide gamut EVF.
David J. Littleboy
"Chris Brown" <cpb...@ntlworld.no_uce_please.com> wrote:
>
> Zeiss Ikon is the one to watch. Their recent reentry into the 35mm
> rangefinder market looks suspiciously like they're actually developing a
> lens line for a digital M mount camera, and the current film bodies are
> testing the water.
Hmm. I think it's more for people who realize that the Contax G lenses are
better than Leica lenses but don't like AF.
David J. Littleboy
Tokyo, Japan
andr...@littlepinkcloud.invalid
> andr...@littlepinkcloud.invalid wrote in
> news:10u2rud...@news.supernews.com:
>> Such a camera would be very much like a D70, but with a fixed zoom
>> lens. I can't imagine why anybody would want such a thing.
> No, it would not. It could be much compacter.
The lenses would be just as big, because they scale to the size of the
sensor. OK, you wouldn't need the lens mount, but you'd still need
everything else. You'd need the viewfinder and the associated
pentaprism -- LCD screens are no substitute for a viewfinder. So
what's lost to make it more compact?
Andrew.
Chris Brown
*peers about*
Phew, you didn't post in in r.p.e.35mm, but be careful, *they* might hear
you...
Anyway, I dunno. I very much got the impression from reading between the
lines when they made their announcement that they definitely had their eyes
on digital capture. ISTR they were making a big fuss about the retrofocal
nature of their lenses, and their FAQs specifically mentioned digital, and
essentially said "watch this space, nudge nudge, wink wink".
I do hope there's more development in this area anyway. I quite like
rangefinders, and a digital one would make a good alternative to an SLR.
David J. Littleboy
"Chris Brown" <cpb...@ntlworld.no_uce_please.com> wrote:
> David J. Littleboy <dav...@gol.com> wrote:
> >
> >Hmm. I think it's more for people who realize that the Contax G lenses
are
> >better than Leica lenses but don't like AF.
>
> lines when they made their announcement that they definitely had their
eyes
> on digital capture. ISTR they were making a big fuss about the retrofocal
> nature of their lenses, and their FAQs specifically mentioned digital, and
> essentially said "watch this space, nudge nudge, wink wink".
How about a link?
The 25, 28, and 35mm lenses are all Biogons. They are all nearly symmetric
designs. The corresponding Leica lenses are all far more asymetric and the
Leica 24mm uses aspheric elements. The (Japanese) review at hand comments to
the effect that the Zeiss 25mm Biogon completely disproves the myth that
aspheric surfaces are required for wide angle performance<g>. They really
liked the Zeiss 35/2.0 lens, and thought it was much nicer than the Leica
lens, although the Leica was a tad sharper wide open. (The Contax G 35mm
lens is a tad iffy wide open too.)
The article didn't cover anything 21mm or wider or longer than 50mm.
Oh, yes. The Zeiss lenses all project at least as far if not further into
the camera than the corresponding Leica lens.
> I do hope there's more development in this area anyway. I quite like
> rangefinders, and a digital one would make a good alternative to an SLR.
Yes. Given that there's unhappiness with Canon wide angle lenses for the
1Ds2, a rangefinder might be fun. But Canon's the only player with an
acceptable full-frame sensor, and it's unlikely that they'd sell to Zeiss at
a reasonable price. So I'm expecting a full-frame 'tweener in the US$3,000
range from Canon long before a usable digital rangefinder. Sigh. I wish the
Kodak sensor were not such a dog*. Sigh. It would be nice if someone gave
Canon something even vaguely resembling competition.
*: No microlenses, no AA filter, high noise. No thanks.
Bart van der Wolf
"David J. Littleboy" <dav...@gol.com> wrote in message
news:crtvbc$th3$1...@nnrp.gol.com...
SNIP
> Given that there's unhappiness with Canon wide angle lenses
> for the 1Ds2, a rangefinder might be fun.
For subjects like architecture or landscape photography, the Contax
21mm f/2.8 Zeiss Distagon T* with an adapter for the EOS range looks
extremely attractive:
http://www.pbase.com/smoody/image/37834176
and the 35-70mm zoom from Zeiss looks pretty good as well:
http://www.dslrexchange.com/image_dir/jorge/3570at70f8Ship.jpg
http://www.dslrexchange.com/image_dir/jorge/3570at70f8Ship.jpg
But then Canon's 24-70mm L is no slouch either..., but not really
*very* wide angle, even for a full frame sensor.
Bart
paul
>
> I know this is sort of a different issue but the D70 SENSOR [sic] is about
Obviously I meant to say LCD, not SENSOR.
Mitch...@aol.com
(drumm roll) larger sensor cells. Larger sensor cells can capture more
photons befor they overflow (ne. saturate). So, why is this important?
Noise! There is a noise associated with capturing a signal in quantized
states (photon -> electron) called Possion noise. It ends up that
Possion noise is equal to the SQRT(signal electrons). So sensors with
larger cells have lower Possion noise. And when coupled with low noise
amplification and large bit-count A/D converters the resulting images
have (drum roll) lower noise.
Modern dSLR camera have sensors that saturate around 65Ke- while the
P&S camera have sensor sthat saturate around 16Ke-. So the dSLR have
about one half the Possion noise of the P&S cameras.
So why not put large sensors in the P&S cameras? Cost! silicon sensors
have a cost structure dominated by area and the relationship between
area and cost is above cubic and below exponential. So when on compares
the 100mm**2 P&S sensors to the 400mm**2 dSLR sensors, the dSLR sensors
cost around 60X as much. The full frame sensors cost around another 8X
as much.
Sorry, high volume cannot overcome physics.
Dutch Flyer
faggot.
"Alfred Molon" <alfred_mo...@yahoo.com> wrote in message
news:MPG.1c4b414a3...@news.supernews.com...
>I guess there are lots of people who would like to have the low noise of
> DSLR CCDs in a much more compact package (and don't need interchangeable
> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
> such a compact camera ? Given that DLRs now cost less than $1000, it
> can't be the cost alone.
Roland Karlsson
@news.supernews.com:
> The lenses would be just as big, because they scale to the size of the
> sensor. OK, you wouldn't need the lens mount, but you'd still need
> everything else. You'd need the viewfinder and the associated
> pentaprism -- LCD screens are no substitute for a viewfinder. So
> what's lost to make it more compact?
We are talking about a compact camera with maximal
picture quality. We are not talking about a ZLR with
large sensor.
No penta prism.
No flapping mirror.
No 100% upscaled lens.
No live preview.
Just a compact camera with extraordinary picture quality.
It will be smaller than a DSLR by a large factor.
It probably weights below 200 grams and you can have it in your
little pocketses.
Unfortunately we also are talking $3000.
The latter will make this a rare beast.
NOTE - this price is not motivated from a technical view.
But the strange beings at the marketing dept always
will try to get as much as possible from odd ball
solutions.
/Roland
Chris Brown
David J. Littleboy <dav...@gol.com> wrote:
>
>"Chris Brown" <cpb...@ntlworld.no_uce_please.com> wrote:
>> David J. Littleboy <dav...@gol.com> wrote:
>> >
>> >Hmm. I think it's more for people who realize that the Contax G lenses
>are
>> >better than Leica lenses but don't like AF.
>>
>> Anyway, I dunno. I very much got the impression from reading between the
>> lines when they made their announcement that they definitely had their
>eyes
>> on digital capture. ISTR they were making a big fuss about the retrofocal
>> nature of their lenses, and their FAQs specifically mentioned digital, and
>> essentially said "watch this space, nudge nudge, wink wink".
>
>How about a link?
>
>The 25, 28, and 35mm lenses are all Biogons.
A quick Google confirms that they are Biogons, so not retrofocal after all.
Clearly my memory is playing tricks on me. However, Zeiss do claim that:
"The lenses have already been developed taking into account the special
requirements of a foreseeable digital camera"
http://www.zeiss.com/C1256A770030BCE0/WebViewAllE/1A30F92F41D05AB4C1256F1C002E544A
(scroll down to "Focused on the Future")
Perhaps that's what I remembered, and assumed they must have been talking
about retrofocal designs. Clearly not, but one does have to wonder what
they're getting at - one would have thought that the Biogon design was the
last sort of lens you'd be using if you were designing a
Digitally-future-proof system.
>Yes. Given that there's unhappiness with Canon wide angle lenses for the
>1Ds2, a rangefinder might be fun. But Canon's the only player with an
>acceptable full-frame sensor,
I don't see them doing a digital version any time soon, and I suspect it
won't be full-frame, so something like Cosina/Voigtlander's 12mm may be in
order. Having said that, it'll be interesting to see what Leica does with
the digital M.
Timo Autiokari
>We are talking about a compact camera with maximal picture quality.
Yes.
I would be very happy with 2MP imager with say 10um x 10um sensors, in
3:2 arrangement.
There could be a prism or surface mirror to bend the light path
similarly like some of the ultracompacts have.
Timo Autiokari
Dave Martindale
>You would multiply the different row with a constant factor (or use a
>different gain for the row amplifier). Noise will be a bit higher, but
>only for that specific row (and not for the other 7 rows).
You'll never get an exact constant ratio in gain between the video and
non-video rows, so it will never match exactly in brightness. And don't
you think someone will notice when an otherwise-uniform area like sky
has more noise every 8th line?
Dave
Alfred Molon
> >1. I didn't start the "Why go dSLR?" thread, only made a couple of
> >posts.
>
> Ah, so a few months later, you decide the issue needs to be rehashed in a
> thread of its own?
>
> >2. "Why separate AF sensors in DSLRs ?" is a very different question
> >from "Why no prosumer with APS-sized CCD ?"
>
> Go through the thread in Google. It ended up covering pretty much exactly
> the same gound that is being covered here again.
>
> It's not clear what you're expecting to hear that's any different from the
> last time. The sensor technology hasn't really changed, apart from Fuji's
> little oddity, which seems to be more like the apochraphal talking dog than
> anything else, in that the interest is not in doing it well, but in doing it
> at all.
yawn
Alfred Molon
says...
> I would be very happy with 2MP imager with say 10um x 10um sensors, in
> 3:2 arrangement.
>
> There could be a prism or surface mirror to bend the light path
> similarly like some of the ultracompacts have.
What would you use such a camera for (just curious) ?
Alfred Molon
Mitch...@aol.com says...
> Modern dSLR camera have sensors that saturate around 65Ke- while the
> P&S camera have sensor sthat saturate around 16Ke-.
Where do you get these numbers from ?
Alfred Molon
> You'll never get an exact constant ratio in gain between the video and
> non-video rows, so it will never match exactly in brightness.
I'm not so sure about that. There could be a calibration step during the
production phase, to measure the gains and ensure that the brightness
levels match.
> And don't
> you think someone will notice when an otherwise-uniform area like sky
> has more noise every 8th line?
Maybe, but that's the RAW data and to get RGB data you interpolate using
data from at least +- 2 neighbouring lines.
So basically the extra noise of the 8th line spreads to the 6th, 7th and
9th and 10th line, getting "smoothed" away a bit in the process - and
that's assuming that RGB conversion uses the smallest possible mask.
It's entirely possible that noise lines won't be visible in the RGB
image.
But anyway, I'm still not convinced that you have to cut some lines into
half and that quality _must_ be compromised to obtain live preview.
There is always a technical solution which solves the problem.
Roland Karlsson
> But anyway, I'm still not convinced that you have to cut some lines into
> half and that quality _must_ be compromised to obtain live preview.
> There is always a technical solution which solves the problem.
>
You could try to synchronize some kind of shutter.
You could use a rotating shutter - like those in movie cameras.
It would be large though.
You could use an electro-optical shutter. That might compromise
picture quality and steal some light.
Mayby you could flap a slit shutter back and forth.
Hmmm ... it would not survive long.
How do shutters in large formet movie cameras look? Are they
mega things?
/Roland
Roland Karlsson
> yawn
Yes - sort of.
Sorry for starting this branch.
I was genuinely curious as you (and I and Timo
and others) alreday had discussed those things several
times before. In other threads - not neccesarily
started by you or me and not neccessarily on this
main topic.
/Roland
J...@no.komm
Roland Karlsson <roland_do...@bonetmail.com> wrote:
>Yepp - and this is probably the most important reason
>why they don't make any compact cameras with large sensors.
>They have to give up live preview to get all the noise
>advantage. And taking away features is a "no no".
But who needs "all" the advantage? Just scaling up the tiny sensors
with live preview to APS size will double or triple the signal-to-noise
ratio.
--
<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <J...@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
J...@no.komm
"Mark B." <mbohnt...@comcast.net> wrote:
>The much larger CCD & CMOS sensors used in DSLRs aren't capable of doing
>live preview.
He knows that. He's stating a perceived need.
Do you define tomorrow by yesterday's technology?
Mark B.
news:4q96u0dbc2kscgdt5...@4ax.com...
> In message <E-SdnXNO6rW...@comcast.com>,
> "Mark B." <mbohnt...@comcast.net> wrote:
>
>>The much larger CCD & CMOS sensors used in DSLRs aren't capable of doing
>>live preview.
>
> He knows that. He's stating a perceived need.
>
> Do you define tomorrow by yesterday's technology?
He stated it as if it could be done by flipping a switch on the sensor.
Obviously, it isn't that easy or it would have been done already.
Mark
timo.au...@aim-dtp.net
> What would you use such a camera for (just curious) ?
Alfred, I'd carry it in my pocket all the time and every now and then
I'd take high quality pictures with it. What else there is to do with
an ultracompact ultra high quality camera?
Timo Autiokari
Graham Holden
>It is Physics! Larger sensors with the same number of pixels have
>(drumm roll) larger sensor cells. Larger sensor cells can capture more
>photons befor they overflow (ne. saturate). So, why is this important?
>Noise! There is a noise associated with capturing a signal in quantized
>states (photon -> electron) called Possion noise. It ends up that
>Possion noise is equal to the SQRT(signal electrons). So sensors with
>larger cells have lower Possion noise.
Being over-pedantic (and to make sure I understand things), surely the
larger sensors have higher [absolute] noise (since you're SQRTing a larger
number of collected photons), but lower RELATIVE noise (in other words, a
higher signal-to-noise ratio)?
Regards,
Graham Holden (g-holden AT dircon DOT co DOT uk)
--
There are 10 types of people in the world;
those that understand binary and those that don't.
andr...@littlepinkcloud.invalid
> andr...@littlepinkcloud.invalid wrote in news:10u4phoqqghct57
> @news.supernews.com:
>> The lenses would be just as big, because they scale to the size of the
>> sensor. OK, you wouldn't need the lens mount, but you'd still need
>> everything else. You'd need the viewfinder and the associated
>> pentaprism -- LCD screens are no substitute for a viewfinder. So
>> what's lost to make it more compact?
> We are talking about a compact camera with maximal
> picture quality. We are not talking about a ZLR with
> large sensor.
> No penta prism.
> No flapping mirror.
I don't get it. How would you view the image with no viewfinder and
no live preview?
> No 100% upscaled lens.
The lens has to be at least as large as the current DX lenses.
> No live preview.
> Just a compact camera with extraordinary picture quality.
> It will be smaller than a DSLR by a large factor.
Well, it seems that all you've lost is the viewfinder.
Andrew.
Chris Brown
<andr...@littlepinkcloud.invalid> wrote:
>Roland Karlsson <roland_do...@bonetmail.com> wrote:
>
>> No penta prism.
>> No flapping mirror.
>
>I don't get it. How would you view the image with no viewfinder and
>no live preview?
Who said anything about not having a viewfinder?
Besides, viewfinders are overrated - have to get the camera out to look
through them.
Which reminds me, I really need to make up a couple of 56*56mm frames, with
a piece of string attached to the corner of them, and a knot tied 80mm from
the end. Very useful for hiking when all your gear is in your backpack and
you don't want to get the old Rolleiflex out just to see if a potential shot
will even frame decently.
This technique is a bit fiddlier with a digicam though, as the sensors are a
bit small. Of course you can scale the frame up and use a correspondingly
longer piece of string...
Alfred Molon
timo.au...@aim-dtp.net says...
> Alfred, I'd carry it in my pocket all the time and every now and then
> I'd take high quality pictures with it. What else there is to do with
> an ultracompact ultra high quality camera?
If you only need a pocket camera, then any digital pocket camera
currently on the market would be fine (unless you want to shoot at ISO
6400). I was thinking you were into astrophotography and needed an extra
low noise camera.
Alfred Molon
says...
> Being over-pedantic (and to make sure I understand things), surely the
> larger sensors have higher [absolute] noise (since you're SQRTing a larger
> number of collected photons), but lower RELATIVE noise (in other words, a
> higher signal-to-noise ratio)?
Exactly - what matters is the signal-to-noise ratio.
Dave Martindale
>How do shutters in large formet movie cameras look? Are they
>mega things?
They are generally rotating discs, and they are large because they have
to block the light from a lens that may have a large rear element.
The simplest variant is just a disc with a fixed-angle sector cut out of
it to give a fixed shutter angle. There are also more complex cameras
with a variable-angle shutter. This it two discs rotating on a common
axis with a differential gear mechanism used to maintain a constant
(but adjustable) angle between the shutter blades as they rotate.
All of the above are for non-reflex viewing cameras. To get reflex
viewing, you still use a rotating disc but tilt it at 45 degrees to the
lens axis (instead of perpendicular) and mount a front-surface mirror to
the front of the disc. This way, the image light either goes straight
to the film, or is reflected into the viewfinder optics.
Yes, all this is quite large compared to the image frame size.
Dave
andr...@littlepinkcloud.invalid
> In article <10u7mjo...@news.supernews.com>,
> <andr...@littlepinkcloud.invalid> wrote:
>>Roland Karlsson <roland_do...@bonetmail.com> wrote:
>>
>>> No penta prism.
>>> No flapping mirror.
>>
>>I don't get it. How would you view the image with no viewfinder and
>>no live preview?
> Who said anything about not having a viewfinder?
Well, I'm assuming that for the camera to be reasonably versatile it'd
need a zoom lens. okay, I suppose a rangefinder might be an
alternative.
Let's say you had something like a digital Leica M; it still wouldn't
be that small. People have been talking about an "ultracompact
camera", which it wouldn't be.
Andrew.
Roland Karlsson
> But who needs "all" the advantage? Just scaling up the tiny sensors
> with live preview to APS size will double or triple the signal-to-noise
> ratio.
Yepp - technically correct. But - in practice I assume that
the only ones that want to buy that camera want optimal
image quality. Only full-frame-fill sensors wil deliver that.
/Roland
Roland Karlsson
news:5uk5u09agd4veelr7...@4ax.com:
> I would be very happy with 2MP imager with say 10um x 10um sensors, in
> 3:2 arrangement.
Ill make a prediction Timo :)
You will never see that camera.
A compact camera with 2 million high quality pixels is not really all
that easy to sell. As I have said before - all cameras have super
high quality according to the manufacturers and the retailers. You
cannot market the camera as super - duper - extra - fantastic quality.
It will not float.
You - and 15 more people on this planet are waiting for this camera.
Maybe they can convince 85 more to buy it - and now they have sold 100.
An - thats it. It will be a mega-lose product.
/Roland
Darrell
"Roland Karlsson" <roland_do...@bonetmail.com> wrote in message
news:Xns95DBE6E6E...@130.133.1.4...
short of "optimal image quality"? It's not the size it's what you do with
it!!!
;)
Roland Karlsson
> Didn't the Kodak DSC-14N have a full-frame (24x35mm) sensor that fell
> well short of "optimal image quality"? It's not the size it's what you
> do with it!!!
>
I did not say full-frame, I said full-fill-frame, i.e that there are
(almost) no parts of the chip that is not used as a sensor.
/Roland
J...@no.komm
andr...@littlepinkcloud.invalid wrote:
>Alfred Molon <alfred_mo...@yahoo.com> wrote:
>> I guess there are lots of people who would like to have the low noise of
>> DSLR CCDs in a much more compact package (and don't need interchangeable
>> lenses, an TTL viewfinder etc.), so why has no manufacturer yet launched
>> such a compact camera ? Given that DLRs now cost less than $1000, it
>> can't be the cost alone.
>Such a camera would be very much like a D70, but with a fixed zoom
>lens. I can't imagine why anybody would want such a thing.
The D70 doesn't have live preview.
I think that Alfred's concern is that live preview is restricted to
small sensors with high noise. There is a gap; live preview and
low-to-medium noise, in the current market.
Another gap is the lack of tiny-sensor cameras with small pixel pitches
that take SLR lenses, or even larger sensor with many tiny pixels,
intended of course for sunlit subjects or strong artificial light. The
Canon 500mm f4L IS lens can resolve at least 2x to 3x as well as any of
the sensors you can currently hook it up to can deliver. A sensor like
the ones in the best current 5MP "prosumer" cameras attached to a 500mm
Canon could do wonders, in bright light. We're talking a FOV equivalent
to a 2000 to 2500mm 35mm lens, with reasonable pixel-to-pixel sharpness.
The resolving power of the 500mm lens is going to waste on current
DSLRs. You can only utilize it with teleconverters, and that loses a
lot of light and slighty compromises the optics.
Roland Karlsson
@news.supernews.com:
> Let's say you had something like a digital Leica M; it still wouldn't
> be that small. People have been talking about an "ultracompact
> camera", which it wouldn't be.
Yes - it could.
There are lots of compact 35 mm cameras with high quality optics,
and an APS camera would be smaller.
Have you seen the 35 mm Minox? Thats small. Have you seen
the 35 mm Contax thingies - thats reasonable small and
very high quality. Now - decrease the volume of the Contax with
a factor 3 and you have a very compact APS sensor sized
camera of high quality.
/Roland
Roland Karlsson
@mughi.cs.ubc.ca:
> Yes, all this is quite large compared to the image frame size.
So - Alfred - if you want a thingie with a large
rotating disc - you have your camera :)
/Roland
Wayne J.
> If you only need a pocket camera, then any digital pocket camera
> currently on the market would be fine (unless you want to shoot at ISO
> 6400). I was thinking you were into astrophotography and needed an
> extra low noise camera.
For the majority of consumers compactness is the most important atribute
when they choose a camera. It's quite amazing the sacrfices in picture
quality consumers a willing to make in order to get a smaller camera. So
I don't think it's such a stretch that there should be a demand for a
camera that delivers both compactness and low noise high ISO pictures.
A fast high quality lens is going to reduce compactness and increase
camera cost more than simply going with a larger sensor. It seems clear
to me that the cheapest way to get a good compact camera is to invest in
a larger sensor. So even though I think it a great idea, I too wonder
why such a camera doesn't exist.
I don't think it is just a matter of APS sized sensors being too
expensive. The fact that large sensors cost several times more than the
small sensors is often used to justify the cost of DSLR cameras but It's
not the main reason. It does't make sense that a digital camera costs
$500 more than it's film equivilent just because it happens to have a
$50 sensor instead of a $10 sensor. The main reason for the cost of
those cameras has to do with market forces. They are new and highly
desirable products and as such command a premium price. Canon also
practically has a monopoly on CMOS sensors. As long as people are
willing to pay the high prices and as long as Canon can maintain it's
technological advantage, the price will stay high.
I also don't think it is a technological limitation of being unable to
make a live preview APS sensor. As others have suggested, there is no
reason why a small sensor can't just be scaled up in size.
I think the problem lies in the fact that DSLR cameras command premium
prices and are highly profitable to make. A compact version with an APS
sensor could not command a high enough price in the market to justify
undermining DSLR sales and just as importantly lens sales. Certain
camera companies would probably be just about willing to give away a
camera body just to get you hooked on their own line of lenses. The fact
that they are atually making very large coin on the camera body alone is
just icing on the cake.
Eventually though, like all electronic devices the price of DSLRs will
drop like a stone when the market becomes saturated and competition
increases. When the price drops to the point that it is no longer
necessary to pretend that APS sensors are hugely expensive then I think
you will see a compact large sensor camera intended for the mass market.
Wayne
Roland Karlsson
news:Xns95DBACD568274...@207.35.177.134:
> I also don't think it is a technological limitation of being unable to
> make a live preview APS sensor. As others have suggested, there is no
> reason why a small sensor can't just be scaled up in size.
Yes it can - but it will not be "high quality". It will only
be "higher quality".
/Roland
J...@no.komm
Roland Karlsson <roland_do...@bonetmail.com> wrote:
>I did not say full-frame, I said full-fill-frame, i.e that there are
>(almost) no parts of the chip that is not used as a sensor.
I thought we were talking about live preview cameras.
J...@no.komm
"Mark B." <mbohnt...@comcast.net> wrote:
><J...@no.komm> wrote in message
>news:4q96u0dbc2kscgdt5...@4ax.com...
>> In message <E-SdnXNO6rW...@comcast.com>,
>> "Mark B." <mbohnt...@comcast.net> wrote:
>>>The much larger CCD & CMOS sensors used in DSLRs aren't capable of doing
>>>live preview.
>> He knows that. He's stating a perceived need.
>> Do you define tomorrow by yesterday's technology?
>He stated it as if it could be done by flipping a switch on the sensor.
>Obviously, it isn't that easy or it would have been done already.
I don't know; I think it has more to do with assumptions by
manufacturers about what people will buy. Many people are very
conservative about technology, and don't want new things until after
they already exist.
I bet many of the things we take for granted in current digital cameras
were things that people were saying they didn't need a few years ago.
Chris Brown
<andr...@littlepinkcloud.invalid> wrote:
>Chris Brown <cpb...@ntlworld.no_uce_please.com> wrote:
>> Who said anything about not having a viewfinder?
>
>Well, I'm assuming that for the camera to be reasonably versatile it'd
>need a zoom lens. okay, I suppose a rangefinder might be an
>alternative.
Have you genuinely never seen a compact camera with an optical, non-ttl
zooming viewfinder? There are loads.
Justin Thyme
> and an APS camera would be smaller.
them. One I like to compare against is the Panasonic FZ10 - 420mm Equiv at
F2.8. To get this out of a camera with an APS sized sensor would require a
280mm F2.8 (close enough to 300mm). Have you ever held a 300mm F2.8 lens? I
have and it is not small, and not light.
35mm compacts do one of two things - they either have fixed lenses at
moderate speeds (eg 35mm F4), or they have moderate zoom lenses at very
pedestrian speeds (eg 150mm F12). APS film camera's also faced the same
problems, although not quite as severe as 35mm. An APS sized camera would
have to make the same compromises, so for a fairly minor improvement in
image quality, you would either have to sacrifice the big zoom lens, or
sacrifice lens speed (or both).
David J. Littleboy
<J...@no.komm> wrote in message
news:kjm8u01t0jcel2r2o...@4ax.com...
> In message <Xns95DBE9B65...@130.133.1.4>,
> Roland Karlsson <roland_do...@bonetmail.com> wrote:
>
> >I did not say full-frame, I said full-fill-frame, i.e that there are
> >(almost) no parts of the chip that is not used as a sensor.
>
> I thought we were talking about live preview cameras.
Not necessarily. If the object is to design a _sensible_ P&S camera that
uses an APS-C sensor, then the obvious thing is a rangefinder/squintfinder
with a separate AF unit, e.g. a slightly scaled-down Contax G or Contax T.
But then the silliness starts: folks wanting video and AF off the sensor and
all sorts of uselessness, and the possibilities for a good sensible design
go away.
David J. Littleboy
Tokyo, Japan
David J Taylor
[]
> I think the problem lies in the fact that DSLR cameras command premium
> prices and are highly profitable to make. A compact version with an
> APS sensor could not command a high enough price in the market to
> justify undermining DSLR sales and just as importantly lens sales.
> Certain camera companies would probably be just about willing to give
> away a camera body just to get you hooked on their own line of
> lenses. The fact that they are atually making very large coin on the
> camera body alone is just icing on the cake.
Got it in one!
Mind you, I think that the high-end point and shoot - the ZLRs - also make
a mint for the manufacturers.....
Cheers,
David