It plainly says it was shot with pre-production firmware. It doesn't count. I'm not saying the 10D doesn't produce colour moire, but it is very rare. I have posted shots with colour moire on before. I have only a handful of shots with any colour sliasing at all, never mind full moire.
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Foveon/Bayer comparison
- Thread starter J. Michael
- Start date
R
Ron Parr
Guest
What are you talking about? I checked the exif and it indicates firmware 1.0.0, which is the same as in Phil's review.
--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
Here is a an actual set of unretouched samples, unlike the ones you showed, where the not-3.4MP Bayer crop had obviously been oversharpened til it screamed.
Below are two out of camera samples, one from a "5.24MP-rated" Sony F717, the second is unretouched from a "3.43MP-rated" Foveon. The Sony has an in-camera sharpness setting of +1, the Foveon's sharpness level is set via SPP to 0.0. Both are 1024 x 768 pixels, 100% crops.
Can you guess which one is the Bayer with a 50% higher MP rating? I have taken thousands upon thousands of Bayer images (over 10K all told), most at MPs rated even higher the 717's. Foveon's claims are understated...
Ron. I acknowledge it is possible but it is of a VERY low level on the 10D and presumable the 300D. Anyway I am out of this thread and the sigma forum. If I stay here I waste my time conversing with loons like SG...
http://www.trytel.com/~pguidry/vacation.html
--
http://www.trytel.com/~pguidry/vacation.html
http://www.mwscomp.com/movies/grail/grail-04.htmMy main complaint is using the extremely biased Foveon Marketing
images, which don't paint an anywhere near fair view of how Bayer
cameras perform. For example:
Here is what they say bayer and X3 look like on text:
Here is the difference in a review:
Hardly what Foveon tells us we will see.
I feel like I am talking to the Black Knight guarding the bridge where after having hacked off his arms and legs, he says: "Just a flesh wound"... Of course SG would instead say: "He's pining for the Fjords". Not only does he not acknowledge the situation at hand, he spouts something completely irrellevant.
SG I will only comment on the title of your message. Foveon claims being understated?? The Above is Actual foveon claims right off their web page compared to actual reality of the same situation. Either A: You are lying through your teeth about what you yourself believe. or B: You have inhaled too many foreign substances and actually believe that.
Given A or B, there is little point exchanging words with liars or drug abusers. Since you are the forums main booster. I shall quit this forum to you...
Here is a quote for Black night skit....
BLACK KNIGHT:
I'm invincible!
ARTHUR:
You're a looney.
BLACK KNIGHT:
The Black Knight always triumphs! Have at you! Come on, then.
[whop]
[ARTHUR chops the BLACK KNIGHT's last leg off]
BLACK KNIGHT:
Oh? All right, we'll call it a draw.
ARTHUR:
Come, Patsy.
BLACK KNIGHT: Oh. Oh, I see. Running away, eh? You yellow bast@rds! Come back here and take what's coming to you. I'll bite your legs off!
I will leave this loon to guard the Sigma bridge but I am out of here...
It said prototype at the top of the page, and to be aware that it is not necessarily representitive of final production output. But if that what the EXIF says then fair enough.
http://www.mwscomp.com/movies/grail/grail-04.htmMy main complaint is using the extremely biased Foveon Marketing
images, which don't paint an anywhere near fair view of how Bayer
cameras perform. For example:
Here is what they say bayer and X3 look like on text:
Here is the difference in a review:
Hardly what Foveon tells us we will see.
I feel like I am talking to the Black Knight guarding the bridge
where after having hacked off his arms and legs, he says: "Just a
flesh wound"... Of course SG would instead say: "He's pining for
the Fjords". Not only does he not acknowledge the situation at
hand, he spouts something completely irrellevant.
SG I will only comment on the title of your message. Foveon claims
being understated?? The Above is Actual foveon claims right off
their web page compared to actual reality of the same situation.
Either A: You are lying through your teeth about what you yourself
believe. or B: You have inhaled too many foreign substances and
actually believe that.
Given A or B, there is little point exchanging words with liars or
drug abusers. Since you are the forums main booster. I shall quit
this forum to you...
Here is a quote for Black night skit....
BLACK KNIGHT:
I'm invincible!
ARTHUR:
You're a looney.
BLACK KNIGHT:
The Black Knight always triumphs! Have at you! Come on, then.
[whop]
[ARTHUR chops the BLACK KNIGHT's last leg off]
BLACK KNIGHT:
Oh? All right, we'll call it a draw.
ARTHUR:
Come, Patsy.
BLACK KNIGHT: Oh. Oh, I see. Running away, eh? You yellow bast@rds!
Come back here and take what's coming to you. I'll bite your legs
off!
I will leave this loon to guard the Sigma bridge but I am out of
here...
Chris Lewis167065
Leading Member
- Messages
- 500
- Reaction score
- 0
That's a BAD Monty Python sketch? Hmm, so SG10 is the Black Knight? I guess you would be...help me out...I can't think of any insignificant 3 year old little boys in the movie.
That's right, the first one is the "5MP" Bayer image.
Can you guess which one is the Bayer with a 50% higher MP rating?
I have taken thousands upon thousands of Bayer images (over 10K all
told), most at MPs rated even higher the 717's. Foveon's claims
are understated...
Have you looked at the original in 36-bit color space? JPEG ruins you test, because it cuts the RAW-output SD-9 colorspace to 1/4096th the original size, then cuts the relatively number of unique colors in that tiny colorspace in half and dithers them during compression.I seldom see moire in my SD9 images, but here is an example (100%
crop):
Other than the obvious luminance moire, there is also a hint of
chromatic moire if your eye is very sensitive
Typical SD-9 numbers (average-ish) for the same RAW file output in different ways...
16-bit TIF - 2-3M unique colors present
8-bit TIF - 600K unique colors present
JPEG(12) - 250K unique colors present
JPEG of a JPEG - more than 250K colors present (addition is statistical noise)
R
Ron Parr
Guest
Compression doesn't cause moire.
jpeg compression is a deterministic process. It cannot introduce "statistical noise."
--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
It causes worse.
It doesn't matter what you personally want to label it, its unacceptably low quality for serious work, and "analyzing moire" in an image that's had its color palatte reduced to less than 1/8000th the original, then dithered, is, well, a joke. Sorry to be the one to let you in on it.
Yi Ding
Veteran Member
Our fearless warrior sg10:
Will you please spare a fraction of the time you spend on the forum to read the article http://www.wfu.edu/~matthews/misc/DigPhotog/alias/ ?
Here is the 16-bit TIFF version of the example I posted, cropped from the 16-bit TIFF output in SPP: http://www.physics.ohio-state.edu/~ding/sd9/moire_16bit.tif ; Feel free to play with it and/or print it out. BTW, the X3 sensor's dynamic range is 61 dB, which corresponds 10 stops in EV or 10 bits in the digitized data. 16 bits are way redundant, and the true data loss in an 8-bit image file format is about a factor of 4 for one channel.
You may want to take some pictures of similar fabric at different distances/focal lengths and see how the moire pattern changes. If you can convince yourself that moire does exist in some SD9 pictures due to the reasons explained in Rick Matthews' article, please let us know. We are humans and humans have errors. It is not a shame for a man to acknowledge his errors, and it is not a shame for a camera to have some limitations, but it is a shame if a man lacks the courage to acknowledge he has got errors.
You are a good photographer, but the ability to take good photos isn't equal to sufficient knowledge in digital photography. Your degrees also don't guarantee you have the instinct to judge right and wrong without further learnings and careful considerations. Even a real expert has to be open-minded; let along us ameteurs...
Also, it is not a good idea to use absolute language. For example, you said "That's because it's a JPEG", and by saying this, you meant that you are 100% sure the cause of moire is the file format. It would have been way better if you instead stated a speculation like "could that be due to the JPEG compression?" You see, for this instance, although you felt very very confident that the moire pattern was caused by the compression, it turned out not to be the case. If you raised many of your points in the form of speculations and leave yourself some room for correction, there would have been less flame wars...
Yi
Will you please spare a fraction of the time you spend on the forum to read the article http://www.wfu.edu/~matthews/misc/DigPhotog/alias/ ?
Here is the 16-bit TIFF version of the example I posted, cropped from the 16-bit TIFF output in SPP: http://www.physics.ohio-state.edu/~ding/sd9/moire_16bit.tif ; Feel free to play with it and/or print it out. BTW, the X3 sensor's dynamic range is 61 dB, which corresponds 10 stops in EV or 10 bits in the digitized data. 16 bits are way redundant, and the true data loss in an 8-bit image file format is about a factor of 4 for one channel.
You may want to take some pictures of similar fabric at different distances/focal lengths and see how the moire pattern changes. If you can convince yourself that moire does exist in some SD9 pictures due to the reasons explained in Rick Matthews' article, please let us know. We are humans and humans have errors. It is not a shame for a man to acknowledge his errors, and it is not a shame for a camera to have some limitations, but it is a shame if a man lacks the courage to acknowledge he has got errors.
You are a good photographer, but the ability to take good photos isn't equal to sufficient knowledge in digital photography. Your degrees also don't guarantee you have the instinct to judge right and wrong without further learnings and careful considerations. Even a real expert has to be open-minded; let along us ameteurs...
Also, it is not a good idea to use absolute language. For example, you said "That's because it's a JPEG", and by saying this, you meant that you are 100% sure the cause of moire is the file format. It would have been way better if you instead stated a speculation like "could that be due to the JPEG compression?" You see, for this instance, although you felt very very confident that the moire pattern was caused by the compression, it turned out not to be the case. If you raised many of your points in the form of speculations and leave yourself some room for correction, there would have been less flame wars...
Yi
Dominic Groß
Veteran Member
[DISCLAIMER]
!Any resemblance to real persons, living or dead is pureley coincidental!
You can't tell that to the Black Knight, he will neither believe it nor stop attacking people. The Black Knight is a fighter, a ridicoulos one indeed but still a fighter who will never stop.
Since he is such a fighter he will never ever admit that he he lost and is wrong. Simply forget about him, since he is quite handicaped because of the choped of parts of his body (or should I say ruined reputation by lots of dumb posts).
I know that it is hard to see people believing in what the Black Knight writes and even thanking him for some of his BS pictures, but after all it is not morally correct to fight a guy who has no legs and arms (and another necessary part of his head was missing all the time).
In some ways the Black Knight is comparable to the guy below, he might realize his lost position or not but never admits that he might be wrong. In his opinion he and his points are still strong.
And according to the latest conspiracy theories the person above is the one who invented the Supa dupa Fuji Super CCD SR and also created the marketing fuzz around it....
[DISCLAIMER]
!Any resemblance to real persons, living or dead is pureley coincidental!
Regards from Old Europe,
Dominic
http://www.pbase.com/sigmasd9/dominic_gross
!Any resemblance to real persons, living or dead is pureley coincidental!
You can't tell that to the Black Knight, he will neither believe it nor stop attacking people. The Black Knight is a fighter, a ridicoulos one indeed but still a fighter who will never stop.
Since he is such a fighter he will never ever admit that he he lost and is wrong. Simply forget about him, since he is quite handicaped because of the choped of parts of his body (or should I say ruined reputation by lots of dumb posts).
I know that it is hard to see people believing in what the Black Knight writes and even thanking him for some of his BS pictures, but after all it is not morally correct to fight a guy who has no legs and arms (and another necessary part of his head was missing all the time).
In some ways the Black Knight is comparable to the guy below, he might realize his lost position or not but never admits that he might be wrong. In his opinion he and his points are still strong.
And according to the latest conspiracy theories the person above is the one who invented the Supa dupa Fuji Super CCD SR and also created the marketing fuzz around it....
[DISCLAIMER]
!Any resemblance to real persons, living or dead is pureley coincidental!
--
Regards from Old Europe,
Dominic
http://www.pbase.com/sigmasd9/dominic_gross
I prefer "common sense" warrior, but ok.
Doesn't help. I don't know anyone with 36-bit display capabilty--its big bucks. Same for true 36-bit printers, you need to send out to pro service. That doesn't mean an SD-9 image with 2 to 3 million colors embedded in it should be analyzed at the 200K unique color level--that makes no sense.
I already listed the typical unique color counts for all. Re-rendering an image with a few million unique colors in it, watered down to a few hundred thousand, then dithered, is poor venue for claiming the SD-9 exhibts moire. See what I'm saying?
I've done it a lot, I've never seen moire, but thats in 24-bit colorspace which is 1/4096th of the original. You can't really blame the camera until you view or print in 36-bit space.
I'm just pointing out that these tests are clearly inconclusive, becasue they've all failed to realize a dominant factor is influencing their results, that's all.
Thank you, I don't consider that remotely true but I appreciate the kind words.
I am 100% sure JPEG'ing induces dominant artifacts much like the ones you are discussing. JPEG artifacts alone (cutting the colorspace to 1/8192nd of the original, then dithering to compensate) are sufficient to invalidate any potential artifacts that may or may not be present--the only thing we know for sure is that PEG'ing artifacts are present.
If a "flame war" arrises because I pointed out that one cannot hope to accurately scrutinize what is left after one pours 8000 gallons of raw data into a 1 gallon bucket, then I'm sorry to hear that. :^(
But that is what these reviewers are doing, whether they realize it or not.
R
Ron Parr
Guest
Compression does many things.
Moire is caused by an interaction between the frequency of the pattern photographed and the arrangement of the pixels. There's a clear and consistent explanation for what's going on. Why try to explain it away by attributing it to forces you obviously don't understand?
It's not a question of what I personally call it. It's a matter of fact.
A sensor without an AA filter will produce moire. That's just a matter of fact too. What's laughable is your effort to find other explanations for something that anybody with a lick of knowledge understands to be unavoidable.
Even if by some miraculous string of coincidences you found an example where moire could be attributed to compression, that wouldn't change the basic fact that sensors without AA filters will produce moire under the right conditions.
--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
R
Ron Parr
Guest
The problem has essentially nothing to do with the number of colors that can be represented in the output medium. If you think otherwise, provide a technical explanation on how subtler gray rendering than is possible with a typical monitor would reduce moire in a B&W image.
You have no technical basis for your claim that there is any other factor involved.
He has already proved that the artifacts are not coming from jpeg compression by providing a tiff. BTW, compression does not dither.
jpeg compression was designed so that at low compression levels, the information discarded is not visible to humans. You talk about 36 bit displays, but I doubt you really understand what the limits of your own eyes are.
--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/
JPEG shrinks the colorspace by 8000x, then introduces artifacts and dithering that eliminate any hope of seeing and/or quantifying any subtle detail in any image. It works fine for identifying Bayer moire, because that is anything but subtle.
Keyboard shortcuts
- f
- Forum