How is this for A2 performance?

[RedDawg]

This is fairly a fairly typical small sensor noise pattern for a
deep blue sky. If shot in JPEG, or when converting from RAW, I'd
suggest turning sharpening to low, and then when processing in your
image editor,

--
David

David, I realize you're talking about a sky here where you have to
deal with a large, mostly textureless area. But I have obtained my
best results when setting my A1 in-camera sharpening to "hard". The
image below was shot in RAW at ISO100 (A1's lowest ISO), 1/60,
f5.6. I then processed with PS CS using the Adobe Camera RAW
Plug-In. After all adjustments were made, I then set the sharpness
to the max just prior to saving to JPG.

I don't understand the point of setting the camera to hard sharpening if you are going to shoot RAW. RAW just provides you with the RAW data off the sensor. It's then up to you to set sharpness, contrast, etc in Photoshop.

I realize that this image is under controlled lighting, but I think
the sharpness speaks for itself. I would suggest to mikejac that he
tinker with the in-camera sharpness to see which result is the best.

If there is no improvement in the noise issue after this, I would
definitely return the camera for service.



--
Rommaker
http://www.pbase.com/rommaker
http://www.momentsnoticed.com

 

[Rommaker]

I don't understand the point of setting the camera to hard
sharpening if you are going to shoot RAW. RAW just provides you
with the RAW data off the sensor. It's then up to you to set
sharpness, contrast, etc in Photoshop.

Maybe I'll try all of the settings to see if there's any difference ... all I know is this setup works for me.

--
Rommaker
http://www.pbase.com/rommaker
http://www.momentsnoticed.com

 

[mikejac]

Can you put the original RAW file for download somewhere, so I can
play with it a little bit? If you don't have any place to put, I
can provide provide you with a temporary username/password for my
FTP server.

E.Z.

Thanks. Could you give me access to your server and I'll download the RAW original for you to look at in more detail.

 

[RedDawg]

Every camera is a compromise, an attempt to balance conflicting
needs: size, weight, speed, image quality, features, reliability,
build quality, cost (and probably a few more).

I agree.

There is no such thing as "best" and there's hardly even "better".
It's all a matter of which compromise, which balance, each of us
prefers.

There's no doubt that the bigger sensor in DSLRs can produce better
RAW images and as a consequence better out of camera images.
However, there's a lot of technology OUTSIDE the camera that can
take lower quality data (eg. noise) and improve it considerably. So
an 8MP digicam can produce results that are close, and sometimes
even identical, to those produced by a DSLR.

Well, I guess it depends. Depends on final output, print vs. onscreen, etc. But it's just nice to know that I can print at some very large sizes, or do some very heavy cropping, and still get excellent results. I do believe that my 10 gives significantly better image quality than my A2. And it frustrated me terribly that I couldn't get good bokeh. It seems to me that so much of good photography depends on manipulating depth-of-field. I found the A2 to be very limited it this area.

Regarding the color accuracy that you mention - technical tests
show that the 10D and the A2 are both very good but neither is
perfect. Every camera has unique color characterstics, that depend
on the sensor, the RGB filter, the signal processing, white
balance, interpolation, etc. Sensor size does make some difference
for color accuracy, but all the other factors are by far more
determining. In some cases the 10D fares better than the A2 and in
other cases it's the other way around.

Yes, but it's nice to get nicely saturated, accurate colors righ out of the camera. I'm not talking Fujifilm colors, just rich color. I think the A2 could stand improvement in this area.

I had, actually still have, a Canon EOS 3 (film), Canon 550EX
flash, and several lenses. I considered going for the 10D or 300D,
but then I realized that in the last 3 years I hardly take photos
anymore. I avoid taking my camera and lenses because I'm tired of
carrying around big and heavy gear. For me an A2 can - and does -
produce far superior results compared to what a seldom used 10D can
produce.

For me the A2 offers a better compromise than any DSLR, because it
gives me a lot of creative and technical freedom, and also because
I want to take it with me and use it. Show me an SLR that
provides a better compromise (ie. great handling, anti-shake, as
compact as the A2, big sensor, and priced at $1,000 including a
quality 28-200 lens) - and I'll switch.

That's what I kept telling myself, that I didn't want to carry around a bag of lenses. And even after I ordered my 10D, I still wondered whether I was doing the right thing. I can be a lazy photographer sometime, just wanting to carry the essentials. But then I tried the 10D, and I was astounded. By the sharpness, the color accuracy, the smoothness. The speed. WIth the 10D I don't have to carry another flash, as the on-board flash is really useful, not weak like on the A2. And, by bumping up the ISO to 400, I can light a large room, and still get very very little noise. And I just bought a Sigma zoom that is a 28-200 equiv. just like on the A2. Now it's not as fast as the GT lens, but it is sharp, and it will be perfect for everyday casual photography. And guess what, the pair still fits in the camera bag that I was using for the A2.

I understand that your needs and motivations are different than
mine, and you strive for clean image off the sensor, better color
and faster operation. But in that case the 10D is not really a good
camera - if you compare it to the 1Ds, 1D MarkII, Kodak DCS/c, or a
Hasselblad H1 + Phase One P25 sensor.

But, you'll say, those systems are much more expensive, bigger,
heavier, require a lot of storage, etc. etc. - that's right but if
you want the ultimate quality, why stop at the 10D.

Well, that's a little bit of a slippery slope argument, but I do understand your point.

Maybe because it's the one compromise that lets you take the
photos you want? ;-)

That is true.

 

[JusGene]

A2 pics are pretty good for a small-sensor, 'P&S on steroids', SLR-like camera, compared to ALL the pro-quality DSLRs. Also compared to great film scanned on hi-rez scanners.

BUT, the A2 isn't either of those, and thus is only capable of what it is capable. Next some 1Ds/Mark II owner will say they have less shutter lag, or better dynamic range.

I could more understand comparing the 30D and the 10D (which I'm sure you'll find in the Canon forums).

COLOR could be pretty good on the A2 as well, with perfect lighting conditions, etc. I hope we have a seperate DSLR forum as soon as the KM DSLRs hit, so we don't see people asking why the A2 8MP doesn't exceed the 7D 6MP....

--
JusGene
http://www.pbase.com

 

[Rommaker]

A2 pics are pretty good for a small-sensor, 'P&S on steroids',
SLR-like camera, compared to ALL the pro-quality DSLRs. Also
compared to great film scanned on hi-rez scanners.

BUT, the A2 isn't either of those, and thus is only capable of what
it is capable. Next some 1Ds/Mark II owner will say they have less
shutter lag, or better dynamic range.
--
JusGene
http://www.pbase.com

Well said, JG ... and right on the bullseye.

--
Rommaker
http://www.pbase.com/rommaker
http://www.momentsnoticed.com

 

[David Goolsby]

I have to agree with what you said.

The thing that amazes me is that people have been told on numerous occasions that the image quality of an Ax will not match that of a DSLR, and yet they are surprised at the difference when they upgrade to a DSLR. I can't count how many times it has been stated that megapixels don't tell the whole story.

The Ax line is a very fine all-in-one camera and the images it produces can be quite amazing. But it has been stated before that each system (point-and-shoot, SLR-like, and DSLR) has its positives and negatives. It is up to the consumer to do their homework and decide which system will best meet their needs.

A2 pics are pretty good for a small-sensor, 'P&S on steroids',
SLR-like camera, compared to ALL the pro-quality DSLRs. Also
compared to great film scanned on hi-rez scanners.

BUT, the A2 isn't either of those, and thus is only capable of what
it is capable. Next some 1Ds/Mark II owner will say they have less
shutter lag, or better dynamic range.

I could more understand comparing the 30D and the 10D (which I'm
sure you'll find in the Canon forums).

COLOR could be pretty good on the A2 as well, with perfect lighting
conditions, etc. I hope we have a seperate DSLR forum as soon as
the KM DSLRs hit, so we don't see people asking why the A2 8MP
doesn't exceed the 7D 6MP....

--
JusGene
http://www.pbase.com

--
David

I shoot people all the time–but I use cameras, not guns!
http://homepage.mac.com/david_g_force

 

[Dennis]

Every camera is a compromise, an attempt to balance conflicting
needs: size, weight, speed, image quality, features, reliability,
build quality, cost (and probably a few more).

There is no such thing as "best" and there's hardly even "better".
It's all a matter of which compromise, which balance, each of us
prefers.

Wow - EZ - my entire philosophy on gear all summed up right there ! Too few people get that every piece of gear, every different film, comes with compromises.

But, you'll say, those systems are much more expensive, bigger,
heavier, require a lot of storage, etc. etc. - that's right but if
you want the ultimate quality, why stop at the 10D.

Maybe because it's the one compromise that lets you take the
photos you want? ;-)

Again, just got done with a debate on the Minolta forum on Yahoo groups where someone was espousing the greatness of "full frame" as if it were the holy grail. My take on it is that it's no more "ideal" than any other sensor size - the Maxxum lens owner has obviously already chosen that some arbitrary stopping point on the quality scale (35mm film versus medium format or large format) is "good enough" and "better" isn't worth spending more money or carrying heavier gear. Yet too often all the people who have chosen 35mm in the film world as a compromise suddenly start preaching "bigger is better" when it comes to digital sensors.

Enjoying your posts !

• Dennis

 

[E.Z]

Maybe because it's the one compromise that lets you take the
photos you want? ;-)

EZ.

--
Hmmmm... If I may be permitted to say......

I think your responses on these forums are models of reasoned, and
most diplomatically expressed(!) thinking -- quite a humbling
experience for the rest of us!

Please stick around ... we can use people like you to keep us in
order. ;-)

With highest regards,

Baz

Wow, thanks!

However I must share the credit with many people - let me explain:

When I finally decided to go digital, I discovered that although I have good knowledge of photogrpahy and computers, I really know nothing about digital photography. So I studied. I read dozens of free articles on various web sites, and I read thousands of forums posts here on DPreview. I learned a great deal. Then when I narrowed down my choice to "prosumer" (I hate this term) diigcams, I went on another quest to know and understand each and every detail about them, so I can choose the best one for me.

In the end, I know a quite a bit about digital photgrpahy, from sensors and pixel counts to post processing techniques. But, there's still a lot more to learn and no doubt that technology will bring about a lot of new "stuff" that we'll have to figure out and deal with.

Anyway all this knowledge came from others, and most of the time I "repackage" it in my own words. Of course I add my own original thoughts and opinions, but as the saying goes, I'm standing on the shoulders of giants. So I'll take this opportunity to thank, and point other to, the greatest sources of information:

• http://dpreview.com - for the camera reviews, forums and as the gateway to all the rest.

• http://luminous-landscapes.com - for priceless reviews, articles, tips and techniques.

• http://dpfwiw.com - for eye opening articles about human vision and photography.

• The good people on Sony and Minolta forums here on DPR, for a lot of valuable information, opinions and of course some really great photos.

• And many other web sites that contribute free knowledge for the benefits of everyone.

Thanks again for the kind words. Of course I'll stick around and contribute as my time permits.

Eyal.

 

[JusGene]

And I don't even attempt to deny all the opinions that come across these threads. Heck, I have one too!

Just the reference of full-frame makes me cringe. Shoot - most cameras, like the Ax, are full-frame. Just a different 'frame'. I know, most people talke about 35mm (forgeting to refer to it) but think about this.

A common querie now is whether the upcoming 7D can even do full-frame with A/S? Guess what; the Ax sensor is full-frame - "to it's lens"! So if the A/S doesn't give it a problem, why would there be anything different to the 7D with a sensor which matches IT'S frame/lens!?!

Ciao

A2 pics are pretty good for a small-sensor, 'P&S on steroids',
SLR-like camera, compared to ALL the pro-quality DSLRs. Also
compared to great film scanned on hi-rez scanners.

BUT, the A2 isn't either of those, and thus is only capable of what
it is capable. Next some 1Ds/Mark II owner will say they have less
shutter lag, or better dynamic range.

I could more understand comparing the 30D and the 10D (which I'm
sure you'll find in the Canon forums).

COLOR could be pretty good on the A2 as well, with perfect lighting
conditions, etc. I hope we have a seperate DSLR forum as soon as
the KM DSLRs hit, so we don't see people asking why the A2 8MP
doesn't exceed the 7D 6MP....

--
JusGene
http://www.pbase.com

--
David

I shoot people all the time–but I use cameras, not guns!
http://homepage.mac.com/david_g_force

--
JusGene
http://www.pbase.com

 

[E.Z]

Well, I guess it depends. Depends on final output, print vs.
onscreen, etc. But it's just nice to know that I can print at some
very large sizes, or do some very heavy cropping, and still get
excellent results. I do believe that my 10 gives significantly
better image quality than my A2. And it frustrated me terribly that
I couldn't get good bokeh. It seems to me that so much of good
photography depends on manipulating depth-of-field. I found the A2
to be very limited it this area.

Just remember that the greatest photogrpahers in history used equipment that doesn't even get close to the A2. And still their work is admired. What counts most is not the technical quality of a photo, but the contents of the photo - what it expresses and what it does to the viewer.

A camera has to be just good enough as to let the photogrpaher capture that special photo, and just provide good enough image quality that doesn't interfere with the photo. All the rest is nice to have but doesn't make better photos. Most of us tend to forget it, myself not an exception!

It's always easier to discuss all kinds of measurable indicators, because the information is available and easy to express. But how can we explain what makes a better photo? We can't, so we sometimes forget that this is the important 90% of the issue.

Yes, but it's nice to get nicely saturated, accurate colors righ
out of the camera. I'm not talking Fujifilm colors, just rich
color. I think the A2 could stand improvement in this area.

While I haven't taken my A2 for any serious excursion, I did shoot a few hundred test shots around where I live. I get very good colors, and I rather that the camera don't saturate them but leave them as close to the natural color as possible. When saturation is needed, then I think it's better to do it in post processing with full control over what and how much.

then I tried the 10D, and I was astounded. By the sharpness, the
color accuracy, the smoothness. The speed. WIth the 10D I don't
have to carry another flash, as the on-board flash is really
useful, not weak like on the A2. And, by bumping up the ISO to 400,
I can light a large room, and still get very very little noise. And
I just bought a Sigma zoom that is a 28-200 equiv. just like on the
A2. Now it's not as fast as the GT lens, but it is sharp, and it
will be perfect for everyday casual photography. And guess what,
the pair still fits in the camera bag that I was using for the A2.

No doubt the 10D is a wonderful camera and certainly it's a great camera for you , which (for you) is the most important consideration. The question is, does it make you a better photogrpaher compared to when you used the A2? Can you now capture that great photo that you couldn't with the A2? That you'll answer to youself, with time.

But, you'll say, those systems are much more expensive, bigger,
heavier, require a lot of storage, etc. etc. - that's right but if
you want the ultimate quality, why stop at the 10D.

Well, that's a little bit of a slippery slope argument, but I do
understand your point.

For me the slope is going the DSLR way, because once down that road I'm not sure I can resist getting the best (told ya - I also fall into that "best quality" trap). I'm also not sure how I can afford the best, but that's another story...

Maybe because it's the one compromise that lets you take the
photos you want? ;-)

That is true.

You're happy with your compromise and I'm happy with mine, what more can we ask for (hmmm... maybe a 1D Mark II + 14mmL + 28-200/2.8L + ...)

EZ.

 

[E.Z]

Every camera is a compromise, an attempt to balance conflicting
needs: size, weight, speed, image quality, features, reliability,
build quality, cost (and probably a few more).

There is no such thing as "best" and there's hardly even "better".
It's all a matter of which compromise, which balance, each of us
prefers.

Wow - EZ - my entire philosophy on gear all summed up right there !
Too few people get that every piece of gear, every different film,
comes with compromises.

Well, it cost me quite a bit to know what I know now. When I was younger I always wanted "the best" so either I didn't have anything or I paid way too much for something that I couldn't use to its fullest.

One example is my decision to buy the Canon EOS 3. I had an EOS 5 that was really a great camera for me and did everything I wanted or needed. But when it wore out I decided it's time to "upgrade" to a real pro machine - the EOS 3. Was that a mistake! Except for better dust protection and slighlty faster AF, the EOS 3 didn't provide me anything that I didn't have enough of before, but it cost twice the price of the EOS 5. It's a great camera no doubt, but photo-journalists can tell the difference, not me.

So now I know...

Again, just got done with a debate on the Minolta forum on Yahoo
groups where someone was espousing the greatness of "full frame" as
if it were the holy grail. My take on it is that it's no more
"ideal" than any other sensor size - the Maxxum lens owner has
obviously already chosen that some arbitrary stopping point on the
quality scale (35mm film versus medium format or large format) is
"good enough" and "better" isn't worth spending more money or
carrying heavier gear. Yet too often all the people who have
chosen 35mm in the film world as a compromise suddenly start
preaching "bigger is better" when it comes to digital sensors.

As I just wrote in another post, people tend to focus on technicalities, because the information is easy to grasp and readily available. So yes, a full frame sensor is technically better than a small digicam sensor. And medium format sensor is better than both, and the sensors used on those huge astronomical telescopes are better still.

So, ask that fellow who likes full frame, can he prove his point by taking a photos with a 1x1 meter sensor and get it to show in an exhibition? Probably not. And with a digicam? Already proven by many.

Enjoying your posts !

Thanks!

EZ.

 

[mikejac]

Can you put the original RAW file for download somewhere, so I can
play with it a little bit? If you don't have any place to put, I
can provide provide you with a temporary username/password for my
FTP server.

E.Z.

Thanks. Could you give me access to your server and I'll download the RAW original for you to look at in more detail.

 

[E.Z]

Thanks. Could you give me access to your server and I'll download
the RAW original for you to look at in more detail.

Check your email @fsmail.net !

EZ.

 

[José B]

SLR-like cameras like the 7xx and Axx. Just revisiting the landscape/macro pics I have taken from my 7Hi and they are as sharp as any I've taken with my Maxxum cameras and shot on Elite 5400.

However, I shoot portraitures/candids and sports/action a lot more and by a long shot, I favour my Maxxum equipment because of faster AF-performance and of course-----the wonderful bokeh my Maxxum and Sigma optics have produced on my images.

Cheers,

José

Just the reference of full-frame makes me cringe. Shoot - most
cameras, like the Ax, are full-frame. Just a different 'frame'. I
know, most people talke about 35mm (forgeting to refer to it) but
think about this.

A common querie now is whether the upcoming 7D can even do
full-frame with A/S? Guess what; the Ax sensor is full-frame - "to
it's lens"! So if the A/S doesn't give it a problem, why would
there be anything different to the 7D with a sensor which matches
IT'S frame/lens!?!

Ciao

A2 pics are pretty good for a small-sensor, 'P&S on steroids',
SLR-like camera, compared to ALL the pro-quality DSLRs. Also
compared to great film scanned on hi-rez scanners.

BUT, the A2 isn't either of those, and thus is only capable of what
it is capable. Next some 1Ds/Mark II owner will say they have less
shutter lag, or better dynamic range.

I could more understand comparing the 30D and the 10D (which I'm
sure you'll find in the Canon forums).

COLOR could be pretty good on the A2 as well, with perfect lighting
conditions, etc. I hope we have a seperate DSLR forum as soon as
the KM DSLRs hit, so we don't see people asking why the A2 8MP
doesn't exceed the 7D 6MP....

--
JusGene
http://www.pbase.com

--
David

I shoot people all the time–but I use cameras, not guns!
http://homepage.mac.com/david_g_force

--
JusGene
http://www.pbase.com

--
Come and visit me at:
http://www.pbase.com/jmb_56

 

[José B]

as any I've taken with my Maxxum cameras and shot on Elite 5400.

should be scanned on Elite 5400.

--
Come and visit me at:
http://www.pbase.com/jmb_56

 

[E.Z]

A common querie now is whether the upcoming 7D can even do
full-frame with A/S? Guess what; the Ax sensor is full-frame - "to
it's lens"! So if the A/S doesn't give it a problem, why would
there be anything different to the 7D with a sensor which matches
IT'S frame/lens!?!

There's a reason to question the feasability of AS with bigger sensors. I don't know if those who raise the doubt think about the same reason I describe below, or maybe they have other concerns in mind.

The anti-shake system is basically a combination of a motion detector and electrical (servo) motors. When the motion detector signals that the camera is moving in a certain direction, the motors move the sensor assembly to counter the movement, in order to keep the sensor in a constant position and orientation.

The motion detector can be a gyroscope, or an array of piezo-electric sensors (maybe some other technology?). I think that gyroscopes are either too power consuming (mechanical types) or too expensive (ring laser types) to be used in a camera.

Piezo-electric sensors can be tiny and consume no power. Actually a piezo-electric sensor produces electric current in a proportional response to pressure that is applied upon it. Full 3D motion detection requires at least 6 sensors, 2 for each plane of motion (eg. left-right, up-down, and forward-backward, or to be more precise we're dealing with spherical movement so it's pitch, yaw and tilt).

Ok, so there are 6 electrical currents that "describe" exactly how the camera moves (accelerates) on each axis. Those current inputs are connected to an electrical circuit that measures them constantly. The other side of this circuit is connected to 3 motors, each moving the sensor on a different axis. The circuit takes the input currents, does some calculations and generates output currents that move the motors in exactly the right amount to counter the "input movement".

And here lies the problem. The bigger the sensor, the bigger the motors must be, and of course more power is needed to drive them. This in turn has impact on the design of the camera in terms of battery life, heat generation and physical size. Of those issues, I think heat generation is the most important, because if the heat "leaks" to the sensor plate, it will increase the noise in the image.

All this doesn't mean that AS is impossible with bigger sensors. My guess is that Minolta will have anti-shake in their DSLRs. It's a great competitive advantage so they won't let it slip away. However, it's still a technical challenge, so we can't just assume that "it works".

EZ.

PS - I'm not really familiar with the internals of Minolta's anti-shake, but I do know a bit (just a bit) about missile navigation systems, and the idea is about the same...

 

[SteveYC]

I was comparing the noise from the Minolta A2 vs. the Olympus C5050, shooting at ISO64. It was a street scene, handheld, daylight - I let the cameras pick the exposure. Both cameras picked roughly identical exposures.

I took 2 photos with the A2, one at 28mm and another ~ 40mm equivalent. I compared them to a picture I took with the Oly at 35mm equivalent. All photos were shot within an interval of a few minutes.

The 28mm shot is on the upper right, the 40mm equiv is on the lower left. The Olympus is in the upper left. You'll see that the 28mm shot is at 100% and matches up well against the Oly C5050 at 100%. While the 40mm shot is at 66% to match the Oly shot.

The C5050 has a smaller sensor than the A2, I did the math, and it looks like the pixel size is just about the same as the A2, but since it is a smaller sensor, the total pixel count is lower.

So, you'd think that at 100% image, they would have comparable noise. Not so, the Minolta is in fact much, much noisier and the resolution and microcontrast are unimpressive as well.



To me, it is pretty clear that pixel for pixel at 100%, the Minolta A2 is really no match for my old Olympus. While if you shoot at a comparable focal length and downrez, the sharpness may be superior, but it doesn't entirely compensate for the additional noise. I had another series of test shots that included patches of the sky - the A2 was very clearly noisier than the C5050.

The Minolta A2 is as good as it gets in terms of digicam handling, but even with RAW, the noise of the sensor and low contrast of the lens are a let down. The Luminous Landscapes site says that the A2 has agood resolution, but I suspect that the contrast and microcontrast ratings would be low.

I don't know that any of the other 8mp cameras would be better in terms of noise, since Phil Askey's test seem to show the Minolta as among the best for noise among the 8mp cameras. If it weren't for the improved EVF on the A2, I'd be sorely tempted to save a lot of money and pick up an A1.

As others have noted, photography isn't about "pixel peeping". The A2 has incredibly good handling, but people should be clear about the trade-offs they are making. I'm not at all advocating getting the C5050 - it has lag times and write times that are measured in geological epochs, and the handling is nowhere near the A2 - but the image quality seems better, even taking into account the extra pixels and sensor size of the A2.

I have produced a crop of the 100% image. Trouble is that jpeg
artefacts are now showing up at this magnification. However, I
think you can see from this what I mean about noise and lack of
sharpness (image was sharpened before cropping)

 

[betlog]

And I don't even attempt to deny all the opinions that come across
these threads. Heck, I have one too!

Just the reference of full-frame makes me cringe. Shoot - most
cameras, like the Ax, are full-frame. Just a different 'frame'. I
know, most people talke about 35mm (forgeting to refer to it) but
think about this.

A common querie now is whether the upcoming 7D can even do
full-frame with A/S? Guess what; the Ax sensor is full-frame - "to
it's lens"! So if the A/S doesn't give it a problem, why would
there be anything different to the 7D with a sensor which matches
IT'S frame/lens!?!

I wouldn't say it's full-frame, full frame is generally used for the 35mm film size, Sure AS could work on a full frame camera but wouldn't you need new lenses that projected image circles larger than 35mm to use it? It was easy to amke a lens for an anti-shake sensor, it's different when your starting with the lens and crafting the anti-shake around it, either use a smaller sensor or produce a new lens.

 

[E.Z]

Mike, who started this thread, let me have his original RAw file for some post processing, to check how the photo can be improved in terms of (less) noise and (more) sharpness. I decided to use only common tools and techniques, apply any action to the whole image (ie. no selections), and limit the job to about 10 minutes. As for the tools those are Photoshop and its built in tools, and Neat Image for noise reduction (there's no effective tool built into Photoshop for NR).

Just some background about my post processing capability (or lack thereof): I've only recently entered the world of digital photogrpahy. I did occasionally used Paint Shop Pro in the past, but only during the last few months I started using Photoshop. So far I've worked on a few dozens of test photos and that's about it. Whatever I can do in 10 minutes, someone more experienced can probably do better and in half the time. My computer is based on Athlon 2600 with 1GB RAM, and my monitor is uncalibrated 17" LCD.

I've made 2 versions of the processed image: one with only minimal processing to remove noise and improve sharpness, and another with some additional work on color and contrast (also minimal). Each photo sample is preceded by the exact steps I have taken in order to produce it. At the bottom of each photo there's a link to the full frame photo, that you can copy into a new browser window.

Both images were saved as JPEG with a medium-high quality setting (8 in photoshop, which is equivalent to 66%). This of course is necessary to avoid too big file sizes, but the JPEG compression re-introduced some of the noise in the sky. The same file saved with lossless compression, or even higher JPEG quality, doesn't show any such noise.

First version: Noise Reduction and Sharpening only

Steps taken
----------------------

RAW Import:

• 16 bits / Adobe color space
• White Balance - as determined by camera.
• Tint/Brightness/Shadow/etc. - defaults.
• Noise Reduction - None.
• Sharpening - None.
• Lens/Chromatic corrections - None.

Noise Reduction with NeatImage:

• Didn't use special calibration target.
• Auto calibration on image.
• Auto fine tune.
• Noise Levels:
• High Freq = 0%
• Mid Freq = 0%
• Low Freq = 0%
• Y Channel = 0%
• Cr Channel = +25%
• Cb Channel = +25%
• Noise Reduction amounts:
• High Freq = 75%
• Mid Freq = 100%
• Low Freq = 100%
• Y Channel = 70%
• Cr Channel = 100%
• Cb Channel = 100%
• Other setttings:
• Very Low Freq = Off
• Smooth Edges = Off
• High Quality = On
• High Resolution = On

Sharpening with Unsharp Mask:

• Parameters:
• Amount = 500%
• Radius = 1
• Threshold = 5

Save result:

• Convert to 8 bits.
• Jpeg quality 8/12 (ie. 66%).

Link to full frame: zvi.org/photo/pict0023.jpg

Second version: Noise Reduction, Sharpening, some Color & Contrast enhancements

Steps taken
----------------------

RAW Import:

• 16 bits / Adobe color space
• White Balance - first Auto, then changed color temp to 6500.
• Tint/Brightness/Shadow/etc. - defaults.
• Noise Reduction - None.
• Sharpening - None.
• Lens/Chromatic corrections - None.

Noise Reduction with NeatImage:

• Didn't use special calibration target.
• Auto calibration on image.
• Auto fine tune.
• Noise Levels:
• High Freq = 0%
• Mid Freq = 0%
• Low Freq = 0%
• Y Channel = 0%
• Cr Channel = +25%
• Cb Channel = +25%
• Noise Reduction amounts:
• High Freq = 80%
• Mid Freq = 100%
• Low Freq = 100%
• Y Channel = 80%
• Cr Channel = 100%
• Cb Channel = 100%
• Other setttings:
• Very Low Freq = On
• Smooth Edges = Off
• High Quality = On
• High Resolution = On

Further processing:

• Convert to Lab color space.
• Local Contrast Enhancement using Unsharp Mask:
• Amount = 25%
• Radius = 50
• Threshold = 0
• Sharpening using Unsharp Mask:
• Select L channel only.
• Amount = 300%
• Radius = 1
• Threshold = 5

Save result:

• Convert to RGB.
• Convert to 8 bits.
• Jpeg quality 8/12 (eg 66%).

Link to full frame: zvi.org/photo/pict0023_better.jpg

Just in case I have understated the message above: a lot more can be done with some more time and experience. For example noise reduction and sharpening can be applied seperately to skies, dark areas, and detailed subjects, to achieve much better results. Also color balance and contrast can be tweaked better and applied differentially to various parts of the image. However that's outside the scope of this test (and my capabilities).

Judge for yourself if A2 images can turn out clean and sharp, with just a bit of post processing.

EZ.