Reading mode: Light Dark

Sigma SD16 - How many pixels?

Go to last post
Tim Binns said:
I agree that the camera isn't that important, but I believe that
Sigma needs to put out a 20ish mp or full frame camera soon or there
wont be any thinking of the future for Sigma camera. There just wont
be a Sigma camera anymore.
Click to expand...
If you want so many pixels - save double-size from SPP.

That way you get an interpolated many-pixel image manually. Just like what the competition makes - except they automated this upsizing step in-camera.

Sigma doesn't bother, because it doesn't really improve the image.

Helge Hafting
 
Helge Hafting said:
If you want so many pixels - save double-size from SPP.
That way you get an interpolated many-pixel image manually. Just like
what the competition makes - except they automated this upsizing step
in-camera.

Helge Hafting
Click to expand...
helge,

proof please ?

that is not true, if they talk of NATIVE size, interpolated statments are not allowed.
 
Helge Hafting said:
If you want so many pixels - save double-size from SPP.
That way you get an interpolated many-pixel image manually. Just like
what the competition makes - except they automated this upsizing step
in-camera.

Sigma doesn't bother, because it doesn't really improve the image.

Helge Hafting
Click to expand...
I think you missed my point. I doesnt matter to me how may pixels I have as long as I have a good image, but to the rest of the world when they look for a new camera usually the first concern is how many pixels.

I was saying that if Sigma wants to continue to sell cameras they are going to have to get into the pixel wars, not keep saying that even though we have less pixels our quality is just as good. In the real world "nobody listens to that".
--
Tim Binns

Just doing the best I can with what God gave me.

http://www.pbase.com/tobinns
 
--
There are more important features Sigma should think about than pixels.

What about a new sensor size? An 24x16mm APS-C sensor size instead of 20,7x13,8mm.
LCD screen with RGB colors, 3" screen. OLED?
Increase the buffer size and data processing speed, alot!
Better high ISO performance
A better way to mount the dust/IR filter. Some sort of screw in type.
and so on....

Sigma shouldn't say that SD-14/15 is a 14Mp camera, it isn't. It's a 4,6Mp X 3 or 4,6Mp camera with 3 layers.
 
NorthPentax said:
Sigma shouldn't say that SD-14/15 is a 14Mp camera, it isn't. It's a
4,6Mp X 3 or 4,6Mp camera with 3 layers.
Click to expand...
And Bayer cameras shouldn't be described as 10Mp without the caveat that this is their luminance resolution and the colour resolution is 2.5Mp for red and blue and 5Mp for green (typically) clearly stated somewhere...

The technical specifications do indeed state the output file resolution "x3" for the camera resolution, except when describing the in-camera upsized JPEG from the SD14.

FWIW, the Advertising Standards Authority in the UK were satisfied that Foveon's definition of a 14Mp sensor was acceptable.

--
GeekGoth, Writer
http://www.geextreme.com/
Sigma, Nikon, Fuji and Olympus DSLRs
Music, cars (Citroën mostly), computing, media
 
Gotoh,

please stop telling hyped marketing stories.
If You reduce a 12MP bayer picture and a Sigma picture to BW, or LAB,
You end with a picture having 12MP bayer spatial resolution, and
4,6MP on the SIGMA side.

I´m not talking about color quality rendition, picture impression or else, simply
the ability to resove lines per mm.
The bayer is unfortunately still ahead here, and the gap unfortunately widens.

Sigma has to react, another and faster CPU is a pleasant side effect, but not the
elementary goal.
 
After seeing the samples from the DP2 I do not expect very much from the SD15. Hopefully it will not have the lock-up green cast issues with the new processor and much better battery life. But it seems the same issues with high ISO noise and red/blue channel nasites lives on.

Since the sensor is a carry over for all practical purposes I may be what goes to the wood shed next revision.

A 6MP DX 1.5 crop sensor would up the resolution without tweaking the pixel pitch much. But other revisions to the design may be needed like improved on chip pixel amplifiers/ADC circuits. Maybe some Isolation layers between the P-N wells to eliminate cross talk. Another big option would be to switch to backlighting the "stack" Making the blue diode the widest and the easy to over saturate red the narrowest. Sony has just started doing this with their exmour CMOS sensors, although this has been used already for years with some CCD designs.

A revised DX would also be suitable for the DP series that Sigma seems to have as a niche all to itself still.
DX would also keep all the DC lens owner on the reservation too.

Body handling could be a real challenge, as nice, solid and straight forward as the SD14 is its not even close to the experience of shooting with my 5 year old D70 design. Endless battery life. Instant previews on the screen. Reliable metering and flash control. Also independent control wheels for shutter speed and aperture when shooting in manual.
Paul
--
A bad day of train chasing is better than a good day at work.
http://peterzpicts.smugmug.com/
 
Gene Hack said:
please stop telling hyped marketing stories.
If You reduce a 12MP bayer picture and a Sigma picture to BW, or LAB,
You end with a picture having 12MP bayer spatial resolution
Click to expand...
Except the maximum sharpness that could be achieved has been hampered by the intermediate stage of processing the raw, where the file you get has 12M RGB pixels with values calculated from adjacent sites. You get a desaturated interpolated file, rather than a true monochrome capture.

You need alternative processing software to get the full resolution - without having previously interpolated the colour data - from a CFA sensor.

The ASA agreement is a verifiable fact.

There's a difference between "hyped marketing stories" and "technical data". Remind me again why you use a Sigma camera if there's no benefit?

--
GeekGoth, Writer
http://www.geextreme.com/
Sigma, Nikon, Fuji and Olympus DSLRs
Music, cars (Citroën mostly), computing, media
 
Interesting name.

Let's see now. We are discussing pixels and I assume you have a great idea of what a pixel is. Otherwise, here is a little lesson:

http://video.google.com/videoplay?docid=-5655850487750051532

Sorry that it is long, but it seems the learning process around here is also long when it comes to understanding what pixels are. So make some popcorn and enjoy the show. If you have an SP9, hook it up using the USB cable and download ten images to SPP.

Of course, **** Lyon could be considered biased, but you are welcome to write him if you dispute his research. Good luck!

So, let's do the math: 4.7 million x3 = 14.1 million pixels. Sounds pretty accurate.

Note that I was not yet talking about spatial locations or megapixels (the latter is a pretty abused word in its own right), but I will now. If you want to be precise, there are 4.7 million spatial locations with 14.1 million pixels arrayed over three layers.

One might fall for the nonsense written by Uli (Gene the Hack to keep him current with right-wing American politics) that the 14.1 million count means nothing to the 4.7 million count. He would be correct if the CFA camp were not claiming that their spatial location count has everything to do with resolution.

The fact is that spatial location is sometimes the best way to approach the question. The Foveon imager is the only one that can resolve detail to Nyquist. The CFA imagers with their NIP feature don't come close.
NorthPentax said:
--
There are more important features Sigma should think about than pixels.

What about a new sensor size? An 24x16mm APS-C sensor size instead of
20,7x13,8mm.
LCD screen with RGB colors, 3" screen. OLED?
Increase the buffer size and data processing speed, alot!
Better high ISO performance
A better way to mount the dust/IR filter. Some sort of screw in type.
and so on....

Sigma shouldn't say that SD-14/15 is a 14Mp camera, it isn't. It's a
4,6Mp X 3 or 4,6Mp camera with 3 layers.
Click to expand...
--
Laurence

Never look down to test the ground before taking your
next step; only he who keeps his eye fixed on the far
horizon will find the right road.

Dag Hammarskjold

http://www.pbase.com/lmatson/root
http://www.pbase.com/sigmadslr/root
http://www.pbase.com/cameras/sigma/dp1
http://www.pbase.com/cameras/sigma/sd14
http://www.pbase.com/cameras/sigma/sd10
http://www.pbase.com/cameras/sigma/sd9
 
Gene Hack said:
Gotoh,

please stop telling hyped marketing stories.
If You reduce a 12MP bayer picture and a Sigma picture to BW, or LAB,
You end with a picture having 12MP bayer spatial resolution, and
4,6MP on the SIGMA side.
I´m not talking about color quality rendition, picture impression or
else, simply
the ability to resove lines per mm.
The bayer is unfortunately still ahead here, and the gap
unfortunately widens.
Sigma has to react, another and faster CPU is a pleasant side effect,
but not the
elementary goal.
Click to expand...
--
Laurence

Never look down to test the ground before taking your
next step; only he who keeps his eye fixed on the far
horizon will find the right road.

Dag Hammarskjold

http://www.pbase.com/lmatson/root
http://www.pbase.com/sigmadslr/root
http://www.pbase.com/cameras/sigma/dp1
http://www.pbase.com/cameras/sigma/sd14
http://www.pbase.com/cameras/sigma/sd10
http://www.pbase.com/cameras/sigma/sd9
 
I respect you very much and I agree and disagree with the pixel concept.

**** Lyon's research has a lot of merit on defining a pixel. I don't think he can claim sensors as a pixel when Foveon sensors are stacked by 3 to define a pixel on a xy axis of a 2 dimensional surface. The same claim by mosaic sensors are also false since it has to interpolate the actual output of the picture elements.

I feel I have a 14MsensorElement camera which produces approximately 4.2MP of usable picture elements. I think the definition needs a bit more refining to define what a pixel truly is these days instead of using shop-talk words of yesterday to define it, and our forefathers never really had an idea what tomorrow would bring. No matter what the true definition is my Foveon rocks.

Regards,
Larry
Laurence Matson said:
Let's see now. We are discussing pixels and I assume you have a great
idea of what a pixel is. Otherwise, here is a little lesson:

http://video.google.com/videoplay?docid=-5655850487750051532

Sorry that it is long, but it seems the learning process around here
is also long when it comes to understanding what pixels are. So make
some popcorn and enjoy the show. If you have an SP9, hook it up using
the USB cable and download ten images to SPP.

Of course, **** Lyon could be considered biased, but you are welcome
to write him if you dispute his research. Good luck!

So, let's do the math: 4.7 million x3 = 14.1 million pixels. Sounds
pretty accurate.

Note that I was not yet talking about spatial locations or megapixels
(the latter is a pretty abused word in its own right), but I will
now. If you want to be precise, there are 4.7 million spatial
locations with 14.1 million pixels arrayed over three layers.

One might fall for the nonsense written by Uli (Gene the Hack to keep
him current with right-wing American politics) that the 14.1 million
count means nothing to the 4.7 million count. He would be correct if
the CFA camp were not claiming that their spatial location count has
everything to do with resolution.

The fact is that spatial location is sometimes the best way to
approach the question. The Foveon imager is the only one that can
resolve detail to Nyquist. The CFA imagers with their NIP feature
don't come close.
NorthPentax said:
--
There are more important features Sigma should think about than pixels.

What about a new sensor size? An 24x16mm APS-C sensor size instead of
20,7x13,8mm.
LCD screen with RGB colors, 3" screen. OLED?
Increase the buffer size and data processing speed, alot!
Better high ISO performance
A better way to mount the dust/IR filter. Some sort of screw in type.
and so on....

Sigma shouldn't say that SD-14/15 is a 14Mp camera, it isn't. It's a
4,6Mp X 3 or 4,6Mp camera with 3 layers.
Click to expand...
--
Laurence

Never look down to test the ground before taking your
next step; only he who keeps his eye fixed on the far
horizon will find the right road.

Dag Hammarskjold

http://www.pbase.com/lmatson/root
http://www.pbase.com/sigmadslr/root
http://www.pbase.com/cameras/sigma/dp1
http://www.pbase.com/cameras/sigma/sd14
http://www.pbase.com/cameras/sigma/sd10
http://www.pbase.com/cameras/sigma/sd9
Click to expand...
 
Well since the body is not that small anyway .... and since all of Sigma's EX-DC lens work on APS-C 1.5X ( Sony, Pentax, Nikon ) also ... Why not just made the sensor larger to facilitate better photosites and more photosites ...

FF would be great but that's luxury though
--
  • Franka -
 
Franka T.L. said:
Well since the body is not that small anyway .... and since all of
Sigma's EX-DC lens work on APS-C 1.5X ( Sony, Pentax, Nikon ) also
... Why not just made the sensor larger to facilitate better
photosites and more photosites ...

FF would be great but that's luxury though
--
  • Franka -
Click to expand...
Fully agree. And there's very simply way of doing that. Just change the aspect ratio to 4:3, while maintaining 2640 pixels side to side. This will increase the area by 12.5% and pixel count by same proportion to 5.2m without reducing the pitch size. That brings the sensor size to a converting factor of almost 1.5X

A small step further by slightly increasing the horizontal pixels to 3000 at the 4:3 aspect ratio will give us a 6.75m pixel sensor. That’s a significant 45% increase in pixel counts and should easily better any APS sensor camera in terms of IQ. equiped with a new (or old by then) TRUE II engine, it should deliver no worse ISO performance than SD14. SD16 with such a sensor should have great appeal to many for landscape and other daylight photography, and as OS lenses become more readily available, it can even be a great general purpose camera.

Allow me to indulge further. Put this new sensor into a DP Zoom with a 15-40mm 2.8 lens, it will be my camera for a long time to come.

--
Maple
 
...the actual crop factor of the current Foveon
chip should be 1.75x. (At least if you use their
published pixel pitch of 7.8 micrometers for
computing the sensor size). If you take the
same photosite size and build a full frame
chip the resulting resolution would be

3080 x 4620 or 14 Megapixels

(if you want width and height to be jpg friendly,
i.e divideable by 8, it should be 3072 x 4608).

If they improved the electronics to better deal
with low-light shots, this would be a phenomenal
sensor.

Greetings
--
Robert F. Tobler
http://ray.cg.tuwien.ac.at/rft/Photography/
 
Guys,
evangelists,
hat wearers,
DV bunch ;-)

thank You for the flowers and I love the north Pentax and Right Wing labeling ;-)
especially if it´s formulated by a swiss.You make it too easy.

Of course the Sigma has 14Megapixels, of course the colours are, let´s say more interesting than any other brand out there, of course the dynamics are in the highlights much better, that´s why I love to use the SDs for a hobby (yet),
but
You must not mistake
picture quality with enlargability, due to RESOLUTION.
The Sigma only has a BW resolution of 4.6 MP and You know that.Of course it has

more possibilities to interpretate natural colours, but in the end its a combination of

image quality and being able to print to big sizes WITHOUT having to use any interpolation routine.

And no, I´m not Jörg Preddimann, fo sho.
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top