foveon chip

I think it's a question of whether color or detail is what you're after. If color is more important, the Foveon has 10.5 million 'spots' of color, but only 3.5 million 'spots' of light; if it's detail, the new Nikon, Canon, and Fuji DSLRs have 6+ million 'spots' of light, but only that many 'spots' of color, too.

Apples and oranges, in a way. If you have to recognize a face in the crowd across a football field, the 6MP sensors give you a better chance to do it; but I'd really like to see colors that aren't a mottled blue/green/red in close-ups and tabloid prints.

Dale Bickley said:

I think it's a question of whether color or detail is what you're
after. If color is more important, the Foveon has 10.5 million
'spots' of color, but only 3.5 million 'spots' of light; if it's
detail, the new Nikon, Canon, and Fuji DSLRs have 6+ million
'spots' of light, but only that many 'spots' of color, too.

Apples and oranges, in a way. If you have to recognize a face in
the crowd across a football field, the 6MP sensors give you a
better chance to do it; but I'd really like to see colors that
aren't a mottled blue/green/red in close-ups and tabloid prints.

Click to expand...

The Bayer pattern sensor inteprolates 2 out of 3 of the RGB values for every pixel. Errors in any one of these contribute to errors in luminance or detail.

The projection of the RGB signal into the luminance dimension is 0.299R 0.587G and 0.144B. So, in a Bayer pattern sensor, each G pixel gets 58.7% of the luminance information hitting that pixel, each R pixels gets 29.9% and each B pixels gets 11.4%

With a 6MP Bayer interpolated sensor, you might think that you're getting 6MP worth of "detail" information, but you're not.

--Ron ParrFAQ: http://www.cs.duke.edu/~parr/photography/faq.htmlGallery: http://www.pbase.com/parr/

Ron Parr said:

Dale Bickley said:

I think it's a question of whether color or detail is what you're
after. If color is more important, the Foveon has 10.5 million
'spots' of color, but only 3.5 million 'spots' of light; if it's
detail, the new Nikon, Canon, and Fuji DSLRs have 6+ million
'spots' of light, but only that many 'spots' of color, too.

Apples and oranges, in a way. If you have to recognize a face in
the crowd across a football field, the 6MP sensors give you a
better chance to do it; but I'd really like to see colors that
aren't a mottled blue/green/red in close-ups and tabloid prints.

Click to expand...

The Bayer pattern sensor inteprolates 2 out of 3 of the RGB values
for every pixel. Errors in any one of these contribute to errors
in luminance or detail.

The projection of the RGB signal into the luminance dimension is
0.299R 0.587G and 0.144B. So, in a Bayer pattern sensor, each G
pixel gets 58.7% of the luminance information hitting that pixel,
each R pixels gets 29.9% and each B pixels gets 11.4%

With a 6MP Bayer interpolated sensor, you might think that you're
getting 6MP worth of "detail" information, but you're not.

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/

Click to expand...

Hi Ron,

Your mathematics are too much for me and probably for others as well. Either way, you are not saying which is a better system the typical Bayer pattern CCD or the Foveon that requires none of that mathematical interpolation.
Pleas clarify.
Rinus

Ron, I am convinced the Foveon is the future. I started
looking at the Sigma used camera prices on the net, and
was bidding on a SA-5 on EBAY, to get used to their equipment
layout, and to start a collection of their lenses. I hit a
seemingly strange fact tho that has me backed off until I
get some more info. At all of the sites, the SA-5 sells for more
than the SA-7 or even the 9 in some cases. The SA-7 can
be purchased for 179$, and the SA-9 for around 300$. Most
of the SA-5's are running in the 400$ range, even tho the
AF is supposed to be better on the 7 and the 9, and the frame
rate is faster. I would like to know why the 5 is so much
more expensive before going on, as the SD-9 is based on the
SA-9.

Dale Bickley said:

I think it's a question of whether color or detail is what you're
after. If color is more important, the Foveon has 10.5 million
'spots' of color, but only 3.5 million 'spots' of light; if it's
detail, the new Nikon, Canon, and Fuji DSLRs have 6+ million
'spots' of light, but only that many 'spots' of color, too.

Apples and oranges, in a way. If you have to recognize a face in
the crowd across a football field, the 6MP sensors give you a
better chance to do it; but I'd really like to see colors that
aren't a mottled blue/green/red in close-ups and tabloid prints.

Click to expand...

In all seriousness, you should go with whatever looks better. There are many, many things that happen between an idea and a product and the final execuation as well as the price will influence the outcome. We can't know for sure until we see the tests.

Indulge me for a moment and let me try to explain what I was saying above: Some people have the impression that Bayer sensors get the "detail" right, and just do interpolation for the color. I'm intepreting "detail" to mean that the B&W part of the image will be correct with a Bayer pattern sensor. (I'm going to use brightness interchangeably with luminance here even though it's not strictly correct because the usage is more intuitive.)

My reply was an attempt to dispel the myth that Bayer interpolated sensors always get the right brightness information. Here's why: Our perception of brightness comes from a combination of red, green and blue. Suppose our eyes are hit by red at intensity 100, blue at intensity 100 and green at intensity 100. How bright would this look to us? Well, we might just add the levels up to together and say that it has a brightness of 300. This would suggest that if we took 100 from the blue and added it to the green (R=100, G=200, B=0) we'd see a different color, but our overall perception of brightness would be the same.

Now, it turns out we can't trade off one color for the other in exactly this way. Doubling green makes things look much brighter to us, but doubling blue doesn't have much effect because our eyes are much more sensitive to green. The numbers in my previous message came from scores that people assign to the different components of color to determine how much they contribute to our perception of brightness. However, the exact scores aren't really needed to understand the point and it just complicates the math, so let's assume that the total brightness level comes from adding together the R, G, and B values.

With a Bayer sensor each photocell sees only one of R, G, or B. Let's suppose some light with values R=100, G=100, B=100 hits a blue sensor. The total brightness for this light should be 300. What the blue pixel sees is B=100. Now the blue pixel needs to guess its red and green values based upon its neighbors. Since green and red influence our perception of brightness so much, a small mistake in these can cause a big mistake in the brightness we perceive for the blue site. Suppose we incorrectly guess R=50 and G=50, then the brightness for R=50, G=50, B=100 is 200, not 300.

So, what has happened is that by guessing the G and R values incorrectly, the Bayer interpolation algorithm has caused a serious mistake in the brightness of a pixel (off by 1/3), and this can cause a mistake in our perception of the detail in the image. Bayer interpolation algorithms can and do make incorrect guesses.

Is this making more sense?

--Ron ParrFAQ: http://www.cs.duke.edu/~parr/photography/faq.htmlGallery: http://www.pbase.com/parr/

I think you're brave.

I think that Foveon approach is a great idea that has a good chance of ultimately succeeding, but I'll be watching the reviews of their first generation products very carefully before I make any decisions.

--Ron ParrFAQ: http://www.cs.duke.edu/~parr/photography/faq.htmlGallery: http://www.pbase.com/parr/

Nothing brave about me. If you read the user reviews
of these film cameras, the feedback is very good, even for
service issues. In the 300 and the 5, most of the complaints
are about focus speed, much like the feedback you read
about the D30. The feedback on the 7 says it is better, and
the 9 is supposed to be better than that as far as the AF.
I am just confused as to why the 5 would be priced higher.
It may be a body materials issue, but I don't know yet. I
want to play with the toys, but i don't have lots of cash to
spend. In this area the Sigma cameras and lenses seem a
good compromise, and I am a disabled person that plays with
the cameras, not a pro. I am also betting the Sigma DSLR
will have to come in lower than the Nikon, Canon, and the
Fuji to make any real market penetration, making it a good
choice for my first DSLR also. If not, then the reconditioned
D30 prices should fall like a rock, and that will be the first, and
I have a film SLR that I lost on, but not a lot.

Ron Parr said:

I think you're brave.

I think that Foveon approach is a great idea that has a good chance
of ultimately succeeding, but I'll be watching the reviews of their
first generation products very carefully before I make any
decisions.

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/

Click to expand...

Ron,

I admire your persistance...if it was me I would have given up a long time ago.

Honestly, I agree with almost everything you've presented so far. The theory, examples, explanations etc. The points have been more than proved.

The problem here is like many things in the IT world (digital cameras being simply a part of that), is that unfortunately too many people find it either too complex, are informed by inaccurate information/marketing, or for want of a better word, form 'illogical' conclusions based upon an almost passionate embrace on a particular technology.

Personally I've made use of digital photography (for one reason or another), since the first comercially available cameras were released, then providing a whopping 320x240 B&W image.

With the release of the first range of 'megapixel' color cameras I started to think that I might no longer need 35mm film. Unfortunately after purchasing or trying various models out I noticed various problems with image quality, smoothing of colors, chromatic aberration, diagonal pixellation etc. Problems I've still found with every single camera since including current models such as the Canon D30/Sony 707 etc (heck maybe I'm just fussy!)

It's not the fact that Bayer may do a good job and fool people in 70/80/90% of cases. To me the fundamental problem is the way the Bayer works. It's a good 'hack', but at the end of the day it's like driving a car with square wheels and using some fancy form of active suspension to even out the bumps. Basically it's a technological dead end. I simply cannot see an argument for it's continuation. Who cares if the first 'round' wheels are only just as good as a square wheel with lots of fancy bit's on it. Do people really want to argue that the square wheel is better?...for what?

Basically that's why I gave up on purchasing a digital camera 3-4 years, used film ever since. Now the Foveon is here maybe I'll change my mind.

Ron Parr said:

In all seriousness, you should go with whatever looks better.
There are many, many things that happen between an idea and a
product and the final execuation as well as the price will
influence the outcome. We can't know for sure until we see the
tests.

Indulge me for a moment and let me try to explain what I was saying
above: Some people have the impression that Bayer sensors get the
"detail" right, and just do interpolation for the color. I'm
intepreting "detail" to mean that the B&W part of the image will be
correct with a Bayer pattern sensor. (I'm going to use brightness
interchangeably with luminance here even though it's not strictly
correct because the usage is more intuitive.)

My reply was an attempt to dispel the myth that Bayer interpolated
sensors always get the right brightness information. Here's why:
Our perception of brightness comes from a combination of red, green
and blue. Suppose our eyes are hit by red at intensity 100, blue
at intensity 100 and green at intensity 100. How bright would this
look to us? Well, we might just add the levels up to together and
say that it has a brightness of 300. This would suggest that if we
took 100 from the blue and added it to the green (R=100, G=200,
B=0) we'd see a different color, but our overall perception of
brightness would be the same.

Now, it turns out we can't trade off one color for the other in
exactly this way. Doubling green makes things look much brighter
to us, but doubling blue doesn't have much effect because our
eyes are much more sensitive to green. The numbers in my previous
message came from scores that people assign to the different
components of color to determine how much they contribute to our
perception of brightness. However, the exact scores aren't really
needed to understand the point and it just complicates the math, so
let's assume that the total brightness level comes from adding
together the R, G, and B values.

With a Bayer sensor each photocell sees only one of R, G, or B.
Let's suppose some light with values R=100, G=100, B=100 hits a
blue sensor. The total brightness for this light should be 300.
What the blue pixel sees is B=100. Now the blue pixel needs to
guess its red and green values based upon its neighbors. Since
green and red influence our perception of brightness so much, a
small mistake in these can cause a big mistake in the brightness we
perceive for the blue site. Suppose we incorrectly guess R=50 and
G=50, then the brightness for R=50, G=50, B=100 is 200, not 300.

So, what has happened is that by guessing the G and R values
incorrectly, the Bayer interpolation algorithm has caused a serious
mistake in the brightness of a pixel (off by 1/3), and this can
cause a mistake in our perception of the detail in the image.
Bayer interpolation algorithms can and do make incorrect guesses.

Is this making more sense?

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/

Click to expand...

Just by looking at the physical layout of the sensors what we will see when comparing it to the 6mp sensors will be as follows (IMO):

1. more resolution and texture with blue, red and anything in between and around those colors.
2. equal, maybe little less resolution at and around green
3. quite a bit less resolution with white-gray-black spectrum

The above will translate into better looking (texture-rich) colorful highlights and midtones but slighly less in the lower midtowns and shadows. The 6mp Bayer will produce better black and white images (if someone wanted that). The "softness" of Bayer images is over. What is bothering me with those images (from d1x and d30 etc.) the sudden drop of resolution (clipping) when red or blue objects are highlighted.--Best wishes,Zoli

I love the square wheels analogy!

--Ron ParrFAQ: http://www.cs.duke.edu/~parr/photography/faq.htmlGallery: http://www.pbase.com/parr/

Thanks for the good explanation Ron.It only seems to reinforce that a Foveon type CCD would be the preferred device. I am sorry to say that the Foveon chip seems to be wasted in this Sigma camera but it is a good start to show what the Foveon type chip can do. The market will tell.

Never say never is what I told a Pro dealer in my hometown when he explained to me why the CMOS will never fly (a few years ago). The first satelite receiving dishes were large and the amplifiers to go with it were massive and expensive. Now we have mini receivers and dishes and they will be replaced to even better and smaller.

With all the problems associated with the Bayer pattern and interpolation of same, I believe a CCD that acts like film holds a more promising future.

A post camera interpolation based on maintaining edge sharpness like Genuine Fractals could enlarge these images easier and better than the problem loaded Bayer files. Foveon could create smaller files and create some real speed for the Journalist Shooter. In Portraits I can always tell digital from film by just looking at the rainbow effects in the hair due to the bayer CCD problems.
Why did they have to make that Foveon chip so small.
Rinus
http://www.scienceandart.org/photography/rinus/showcase.html

Not a bad approach (to get a Sigma film camera now, and accumulate lenses). You would then have both a film version and a Foveon version, for whatever advantages each may have, plus the ability to directly compare the digital with the film.

And post the results for us...

Gary Eickmeier

Bob,

I wouldn't worry about why the SA9 is less than the SA5. Sigma no longer makes it any more anyway. I own both the SA9 and SA5, and ergonomically the SA9 is a much more pleasurable camera to work with. Its layout and position of controls make it one of the most responsive cameras out there, in its class. I can see why the SD9 is based on its design. That is why I happen to be attracted to this site. When I heard about the SD9 I wanted to know more about it. Anyway, test out the SA9 and you'll see for yourself.

Bob Williams said:

Dale Bickley said:

I think it's a question of whether color or detail is what you're
after. If color is more important, the Foveon has 10.5 million
'spots' of color, but only 3.5 million 'spots' of light; if it's
detail, the new Nikon, Canon, and Fuji DSLRs have 6+ million
'spots' of light, but only that many 'spots' of color, too.

Apples and oranges, in a way. If you have to recognize a face in
the crowd across a football field, the 6MP sensors give you a
better chance to do it; but I'd really like to see colors that
aren't a mottled blue/green/red in close-ups and tabloid prints.

Click to expand...

Click to expand...

... or is it really a minus? More comparison stuff probably will degrade Bob's quality of life

Gary Eickmeier said:

plus the ability to
directly compare the digital with the film.

And post the results for us...

Gary Eickmeier

Click to expand...

--Michael

Gary, by the time they reach my pricepoint, anything I
would post would be already known and reported info.
As far as I am concerned, the three remaining questions
about the Foveon Bayer do not apply to me today. I am
a hobbiest with different needs than the pro's. I am
quite sure even the older Sigma SA-300 would be capable
of doing what I need. If the G-1 had a usable ISO 400 and
I could switch lenses, it would be ok for most of what I
do. ISO 50 with 100mm limit is not. What I do, and
want to do is best summed up by this story. I did not
take this shot, but I watched it being taken. Second
day of the beginning of the elk rut, and the herd bull
is standing under a large oak with a hillside and dense
foliage behind him. He has his head thrown back screaming
a challenge, while his antlers are still tangled in the over
head branch he is busy destroying. Steam is coming out
of his nostrils, even tho it is a nice day. He has his back
legs together, urinating over the scent glands, while he
screams his challenges. This shot was taken from a car,
with the engine shut off to stop the vibration, and the
camera was mounted on a window mounted support.
It was taken at about 8 feet, but had to have a large
DOF to get the entire animal. It was taken with Canon
Pro gear I think. The younger bulls were across the
clearing watching to make sure the big bull did not attack
them. When the photographer left, and as I was driving
out, the big bull lowered his head and charged straight
at my truck. I hit the gas, thinking he was going for my
truck, but he had just lowered his head and was using
me as camoflage to get a running start towards the
young bulls without them seeing him. He chased them
around for about an hour, and never got farther than
about 100 feet from my truck. If I had had a good
film camera, and lots of film, I would still be trying to
afford getting it all developed. This was not a once in
a lifetime chance, as I can get pretty much the same
shots this fall from the same location, depending only on
me getting up to be there on the correct days at 7 in
the morning.
George, what is the big difference that is causing the
price difference. I just let a used SA-5 go by on EBAY
for 132$ because of the 179$ price for the 7 from
B&H. How good or bad is the auto focus on the 7's
and 9's? Are they as bad about eating batteries as
some reports mention, and can AA nihm's be used
instead of the crxx camera batteries, like they can on
the SD-9? Thanks very much for the feedback. It
is hard to get good info on these cameras, as a lot of
the reviews are done by pro's and start with statements
like, This is a good camera with a lot of high end features,
but no pro would be caught dead with one. As you read
the rest of the review, you can see the obvious bias
against the camera, while at the same time they praise it
as a great camera, but for consumers or beginners only.

george wilson said:

Bob Williams said:

Dale Bickley said:

I think it's a question of whether color or detail is what you're
after. If color is more important, the Foveon has 10.5 million
'spots' of color, but only 3.5 million 'spots' of light; if it's
detail, the new Nikon, Canon, and Fuji DSLRs have 6+ million
'spots' of light, but only that many 'spots' of color, too.

Apples and oranges, in a way. If you have to recognize a face in
the crowd across a football field, the 6MP sensors give you a
better chance to do it; but I'd really like to see colors that
aren't a mottled blue/green/red in close-ups and tabloid prints.

Click to expand...

Click to expand...

Click to expand...

Hi Bob,

Honestly I cannot explain why the SA5 is still selling in some cases higher than the SA9 except to say that it doesn't really matter since it is no longer made. However...

Trust me when I say that the SA9 is a much better engineered camera. They both have single central-only crosshair autofocus points, sensitive to both horizontal and vertical lines, but it is much more sensitive in the SA9, which is why there is no infra red focus beam on the SA9 like there is on the SA5. As a result, the SA9 is a smaller camera but ironically much more comfortable to manage and operate. Also there was some funky control positions on the SA5 that made it a bit frustrating to use convieniently. On other cameras controls that are usually hidden behind a menu or put on opposite sides of the camera body are placed right next to one another on the SA9, making its use very user friendly and quick. For example, going from single to continuous fps, to mirror lock up, to 2 or 10 sec. delay, to multi exposure, to rapid rewind, etc. is easily reached by a simple turn of a dial, as opposed to pushing a button first then turning a dial while holding down the button as in the SA5. Changing ISO settings, flash settings, wireless remote settings, metering settings is also very rapid in the SA9 because of its simplified, practical layout (just a push of a button, turn of a dial). All of what would practically be used most in real world settings is right there for you. The SA5, although a nice camera, isn't as convienient or practically laid out.

They use the same batteries, no aa alternative, and I haven't personally had any rapid battery burn out with it. It works incredibly well with the amazing EF500 Super Flash (really versatile flash!) although Sigma corp. told me that they will have a new flash coming out with the same feature set as the current one except made specifically for the SD9. The new flash will be backward compatible with the SA9.

In my opinion it is as useful to use in the same settings as the Canon Elan 7 although without the eye control and some custom funtion controls that in my opinion are just decorative add ons (IMHO only ). It is certainly in the same class as the Nikon N80 IMHO. Sigma designed it to compete with those two cameras, and I think feature for feature, it does very well.

Why do reviews damn the camera with faint praise? I think any pro who would shoot with the Canon or Nikon Elan/N80 would find as much pleasure with this one. Part of the reason it is spoken of as only an amateur camera is because: 1- it is not bulletproof, although it has a nice, solid body 2- lacking in trendy features or a zillion "custom funtionless" funtions 3-there is not a lot of accessories to go with it, even though what there is is very versatile for many real world settings 4- (the biggest) LENS SELECTION! However there is an abundance of high quality EX Series lenses that in some cases outguns or matches or comes pretty darn close to major brand name lenses, 5- Very small market share, low resale value 6- Built in bias toward the major brands. 7- Although their 3rd body and their best, in terms of history with bodies ( not lenses ) its not been very long.

Phil seems to indicate that the SD9 has some unique features that (in line with Sigmas philosophy) are practical and very useful. Whether it is a hit with buyers or not, I don't think will break Sigma. Nontheless, if you consider the camera for what it offers on its own terms without predjudice, even with the mount and EX lenses with FOVEON chip, Sigma may definitely have a monster of a camera on its hands that may change the course of its own history. Personally, I think it will sell for around $1800.

It will sell primarily if not exclusively to those first time digital SLR users, to those who own the SA9, and to those who haven't heavily invested in other brand names that's it. I cannot see Nikon or Canon people with a history with those brands buying into it, FOVEON chip or not. If the FOVEON chip is dynamite, than maybe Minolta and Pentax will follow, possibly Nikon or Canon ( I doubt it ) and then things will get really interesting!
Hope this was helpful, if too long!

Bob Williams said:

Member said:

George, what is the big difference that is causing the

Click to expand...

price difference. I just let a used SA-5 go by on EBAY
for 132$ because of the 179$ price for the 7 from
B&H. How good or bad is the auto focus on the 7's
and 9's? Are they as bad about eating batteries as
some reports mention, and can AA nihm's be used
instead of the crxx camera batteries, like they can on
the SD-9? Thanks very much for the feedback. It
is hard to get good info on these cameras, as a lot of
the reviews are done by pro's and start with statements
like, This is a good camera with a lot of high end features,
but no pro would be caught dead with one. As you read
the rest of the review, you can see the obvious bias
against the camera, while at the same time they praise it
as a great camera, but for consumers or beginners only.

george wilson said:

Bob,
I wouldn't worry about why the SA9 is less than the SA5. Sigma no
longer makes it any more anyway. I own both the SA9 and SA5, and
ergonomically the SA9 is a much more pleasurable camera to work
with. Its layout and position of controls make it one of the most
responsive cameras out there, in its class. I can see why the SD9
is based on its design. That is why I happen to be attracted to
this site. When I heard about the SD9 I wanted to know more about
it. Anyway, test out the SA9 and you'll see for yourself.

Click to expand...

Click to expand...

SA-7 for 179, plus a starter lens, or the SA-9 for 300+?
Recommendation for a good cheap starter lens in the
200mm range on the long end? Never worry about long
winded when answering me!!!! I am one of the people
most challenged by the 6000 character limit here!
george wilson, jr. wrote:

Hi George, many thanks for all the useful information about the SA9. I have never owned a Sigma camera (only lenses), but your comments agree with the reviews of the SA9 that I have read in magazines etc. It sounds like a nice body comparable with the Canon EOS 30 and the Nikon F/N80 (here in the UK the names of many cameras are very slightly different compared to their US equivalents).

I am seriously considering buying a Sigma SD9 (hence my interest in your posting) and will be comparing it closely with the Canon D60 and the Nikon D100. I have one very simple question that you can probably answer: the SA7, SA9 and SD9 all have a recess on the right hand side of the body (as you face the front of the camera) immediately below the label "SA-7" etc. It looks like a sensor of some sort -- at first I assumed that it was an infra red assist light or similar to aid focussing in low-light, but this is clearly not the case. So what exactly is that little recess?

Terry.

Meanwhile, back at the ranch . . .

I downloaded the sample images from the Foveon site and have been messing around with them to see how they hold up to manipulation. For comparison, I've been doing the same manipulations to a shot taken by a DX1 that a customer brought in last fall. Both photos were high-quality JPEG when I started.

The Foveon photo did well in resampling and resizing -- JPEG artifacts came up before pixelization became noticable, and the colors stayed pretty true, which is what you'd expect. What was a bit of a surprise was that when I got to the level of over-sharpening the pictures, the Foveon sample was very predictable, and produced less of a halo effect than I'm used to. It broke up into a mosaic pattern eventually, but it really held up well.

Hi, Bob,
Hi, Terry,
To Bob:

For me, the better value is the SA9. The SA7 has a top shutter speed of 1/2000 sec. and only 1/90 sec. top sync speed. The SA9 has 1/8000 sec. shutter speed and 1/180 sec. top sync speed, and a larger hand grip which adds to its comfort. Plus, longer battery life. As far as a lens is concerned, I have heard/read some really good things about their 28-135mm f/3.8-5.6 macro lens. However, their 28-200mm got generally decent reviews, some rating it higher than the Tamron version. I tend to stick with their EX series lenses.
To Terry,

That small recess is the sensor for their wireless remote control, which releases the shutter from about 19 feet away, in front of the camera. It can also be used in conjuntion with the mirror lock up function, and there are three seperate channels on it. I don't have one yet, but one reviewer says that it is so sensitive that you can fire it from behind the camera as well, therefore a wired shutter release is not needed.

Now that the Canon is officially about 2100/2200 US dollars, I'll bet the Sigma will be closer to about 1800 dollars, since Sigma seems to model itself closely with Canon mid range products and always comes in with a lower price.

Terry Sessford said:

Hi George, many thanks for all the useful information about the
SA9. I have never owned a Sigma camera (only lenses), but your
comments agree with the reviews of the SA9 that I have read in
magazines etc. It sounds like a nice body comparable with the
Canon EOS 30 and the Nikon F/N80 (here in the UK the names of many
cameras are very slightly different compared to their US
equivalents).

I am seriously considering buying a Sigma SD9 (hence my interest in
your posting) and will be comparing it closely with the Canon D60
and the Nikon D100. I have one very simple question that you can
probably answer: the SA7, SA9 and SD9 all have a recess on the
right hand side of the body (as you face the front of the camera)
immediately below the label "SA-7" etc. It looks like a sensor of
some sort -- at first I assumed that it was an infra red assist
light or similar to aid focussing in low-light, but this is clearly
not the case. So what exactly is that little recess?

Terry.

Click to expand...