From: Ray Fischer on 18 Jul 2008 23:29
John Sheehy <JPS(a)no.komm> wrote:
>Steve <steve(a)example.com> wrote in
>> On Thu, 17 Jul 2008 22:15:06 GMT, John Sheehy <JPS(a)no.komm> wrote:
>>>That makes no sense at all. I am not comparing cameras; I am comparing
>>>the effects of pixel density. I am fighting the myth that pixel noise
>> Actually, you're not doing that at all. What you are doing is
>> comparing the effects of resolution at the focal plane. No one can
>> argue that higher pixel density gives higher resolution at the focal
>> plane. And that's the only thing your test proves ... the obvious.
>> Only you can argue that this somehow translates to higher pixel
>> density gives better noise and DR performance in any real world image.
>> It doesn't, and the rest of the world knows that to be true no matter
>> how much you profess otherwise.
>I don't know what to say to you; you are being totally illogical. Can
>you find *any* way in which the 400D crop is less noisy than the FZ50
Oh, you're trying to justify your inane test.
Here's what really counts: Noise per pixel.
>I don't know what it is, but I can not for the life of me figure out how
>people can hold such unrealistic models in their heads.
Such as your daft notion of thinking that the size of the image on the
sensor is something that people care about?
From: Roger N. Clark (change username to rnclark) on 18 Jul 2008 23:55
John Sheehy wrote:
> none <nobodyhome(a)yahoo.com> wrote in
>> As a real, scientific study, this effort fails horribly, to show what
>> you were setting out to demonstrate.
> No one has ever proved the opposite! Does that mean nothing to you?
But they have and you ignore those studies like you ignore me in this
thread. And you ignore the math.
> The basic problem, I think, is that people have big problems, in general,
> with the differences between relatives and absolutes, and have their
> thoughts warped by using the wrong one in context.
> This is a demonstration of the difference between the sensitivity and to
> a lesser extent, the resolution of pixels of extreme density difference,
> to address the assumption held by most (though their number is
> decreasing) that higher pixel density means more noise.
Once again, let's work a simple example.
Sensor A has 100 pixels for every pixel in sensor B.
Both have pixels with the same read noise = 4 electrons.
Assume 100% fill factors and same QE.
Compute dynamic range for the area of pixel in sensor B.
sensor A: X = Signal A = Signal B if we sum the signal from each pixel
in A (sum(100* signal in one pixel).
read noise: sensor A = 4*sqrt(100) = 40 sensor B= 4 electrons.
Dynamic range, DR: Sensor A DR= X/40, Sensor B: X/4
The large pixel has higher DR.
In your computation of DR, you summed the signal and then
divided the noise in the small pixel sensor by the number of pixels.
In the above example, you would come up with:
Sensor A read noise = 4*(sqrt(100)/100 = 0.4. That is wrong.
In real world sensors, fill factor decreases as pixel size becomes
small, and so does full well. Then there are other factors that
influence performance, including effects on QE with very small pixels.
For example look at the absorption length of photons in silicon.
A red photon of 6000 angstroms has an absorption length of 5 microns.
What happens to red light in a sensor with 2-micron pixels, when
only 63% of the photons are absorbed in the first 5 microns,
86% after 10 microns and so on? So what do you think happens to image
quality when it takes 5, 10, 15 microns for a red light photon to be
absorbed, far away from the 2-micron pixel?
There are multiple reasons why larger pixels perform better
and all these factors affect image quality. And these multiple
factors combined with the desire for spatial resolution makes
an optimum in image quality.
Of course, you could team up with Ilya who also advocates insanely small
pixels (I think it was 1/2 micron pixels at one pint) and make
a camera that is better than everything else out there, make lots
of money and prove the rest of the world wrong.
From: Roger N. Clark (change username to rnclark) on 18 Jul 2008 23:55
John Sheehy wrote:
> The direct comparison of what? I am getting the sense that most of the
> critics here have no viable working model in their heads of what a sensor
> is, and what it does, yet they feel obliged to make all kinds of comments
> assuming some kind of magic.
Of course you have yet to respond to any of my posts, including
where I point out errors including the equations you use and
the reasons you come up with the erroneous conclusions you do.
> And where is your counter-evidence? There is none. There is no evidence
> that higher pixel density lowers DR or increases noise.
I already proved you were using incorrectly different equations
for high and low end to come up with the erroneous conclusions.
> Every shred of
> so-called evidence assumes a 100% pixel view, uses JPEGs, or unequal
> conversions (No commercial RAW converter converts different cameras the
> same way; Brands that have a history of high RAW noise get extra NR by
> default, just like the manufacturer's JPEGs do).
No it does not. It is fundamental physics, which you seem to ignore.
I've also put out what is probably to most complete model of
sensor performance one can find on the net, which counters what
you say. You have no actual model. You incorrectly process images
to make your case. I've pointed that out several times and you
will not even acknowledge it.
> Guess you're out of the loop, but there's a big showdown going on at
> DPReview right now, because the intelligesia there have challenged Phil
> on his pixel-centric, noise-reduced, unequal conversion comparisons, and
> it has become clear to him that his most knowledgeable forum members
> almost unanimously believe that the conclusions are invalid. He was
> engaged in such a discussion, and when asked for proof of his viewpoint,
> he became mute. Hopefully, he's doing a lot of thinking about this.
Do you see the irony in your statement. I show you your errors
and you become mute.
> Many things that people find clear are actually false, under the hood.
> People like simplicity, more than they are interested in truth.
> There are three layers to this onion. There is the hypothetical blind
> consumer who thinks that more MPs are always better in every way. Then
> below that there are people who think that more MPsruins images in a
> given sensor size.
And your 4th layer seems to think that the more pixel density is better.
It seems most people here realize that there is an optimum pixel
density that trades focal plane resolution with S/N, dynamic range,
and real resolution. That is also what the scientific models
From: ARChie G. on 19 Jul 2008 00:12
On Sat, 19 Jul 2008 03:25:40 GMT, John Sheehy <JPS(a)no.komm> wrote:
>I've given far more than just a hint of what I am demonstrating here, yet
>almost no one who has responded seems to understand. My thoughts are
>going way over your head, apparently.
The reason being that you are making complete sense, so those that understand
have no reason to support your viewpoint. The usual useless mass of resident
trolls who are desperate for attention of any kind are making fools of
themselves. As usual. I see no need to support what you are saying because it's
100% true. Only an idiot would take their outdated urban legends parroted from
the beginnings of their misunderstandings and try to apply them as an argument
to your proven results. This newsgroup is filled to the brim with them
(monkey-see-monkey-do idiots). That's why it's never worth posting here. You get
the same idiots espousing the same nonsense for years now. Idiots supporting
idiots. The lowest common denominator has amassed here in great numbers. They
prove it daily and never realize they are the joke of the photography world. I
too have proven Phil at DPReview wrong on too many counts to mention. Yet they
keep taking Phil's tests as some sort of gospel. They are taking gospel from a
complete and total idiot. They're just too stupid to realize that.
You have stated your proof. It is clear you are right and they don't know what
the hell they are talking about. Leave it at that. The rest of us already see
them for what they are--for years.
From: John Sheehy on 19 Jul 2008 00:25
ASAAR <caught(a)22.com> wrote in
> How about Nikon's 6mp D40 vs. the 10mp D40x? Or Canon's A610 vs.
> A620? Announced at the same time, they used the same size 1/1.8"
> sensors, where the former had 5mp vs. the latter's 7mp. There are
> more such pairings, but I'd be surprised if you'd accept any of
> them, as they wouldn't aid your agenda.
Why don't you show us, with homogenous RAW conversions, how they
compare, both upsampled, and downsampled?
If I did it, you would say I'm lying or rigged something if you didn't
like the results.
I don't know about the Canons, but other people have done the D40 vs
D40x thing, and have reported the D40X downsampled to 6 MP to be better,
at least at the lower ISOs. They are too close in pixel density, IMO,
to make a useful comparison for *general* purposes, as the pixel pitch
only decreases by about 22% in the D40X. It would just be for those two
I remember looking at the RAW data from the Nikons a while back, and I
remember that the two cameras had very different QEs, with the D40X
being more sensitive, photon-wise, but the D40 having lower read noise
at higher ISOs. So, I would expect the D40 to be best at shadows at
high ISOs, and the D40X to be better at everything else.
Well, can you get someone to provide the RAWs (not RAW conversions!)
from the two Nikons, same scene and lens, and make sure the lighting and
the exposure is the same? That would compare them for RAW SNR at the
ISOs provided. DR is a bit tougher, as you really need to know what
you're doing, as you have to figure out the RAW sensitivity, RAW
clipping points, and alter the exposure time to make the cameras match
in tones relative to RAW saturation.
<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS(a)no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><