Optical vs. interpolated resolutions, illustrated

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

george_preddy@yahoo.com wrote:

> http://www.pbase.com/optical_vs_interpolated

LOL, the foveon shot is a 571K file and the bayer is compressed down to
127K with loads of nasty jpeg artifacts. I'm sure there isn't any Hokey
pokey going on with the edits either..

Hmm I wonder if the person who put this up has any bias?
--

Stacey

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

Yes, the Bayer ones are clearly much better, in spite of the smaller
file sizes and the attempt to cheat.

And what's with the awful color on the Sigma? I didn't think it would
be *that* bad.

Thanks for the warning, George, but I think we already *know* that the
Sigma/Foveon thing was a disaster.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

In article <1117956367.533480.218200@f14g2000cwb.googlegroups.com>,
george_preddy@yahoo.com says...
> http://www.pbase.com/optical_vs_interpolated
>
>

Why didn't you take the same shot with both cameras so that people could
make direct comparisons? I was able to make some comparisons with the
shots provided.

1) The Foveon image apears to have muddied colors, while the Bayer
images have bright, vidid colors.

2) the Foveon image appears out of focus, while the Bayer images appear
sharp throughout the image.

From the image samples provided it appears Foveon images are inferior to
Bayer images. Perhaps it would be best to take the same shots with each
camera to make direct one on one comparison, because these images
clearly favor Bayer imaging.

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

On 5 Jun 2005 00:26:07 -0700, george_preddy@yahoo.com wrote:

>http://www.pbase.com/optical_vs_interpolated

What's with the B.S. grew print on a black background???
-Rich

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

dylan wrote:
> I see it now, take some poor pictures, reduce the size, enlarge them again
> and they aren't as good as the originals.

With Foveon images, that is correct. With Bayer the image is almost
completely unchanged by downsizing to quarter-size, as is shown so
nicely.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

KennyJr wrote:
> In article <3gfpngFc5qpoU1@individual.net>, fotocord@yahoo.com says...
> > george_preddy@yahoo.com wrote:
> >
> > > http://www.pbase.com/optical_vs_interpolated
> >
> > LOL, the foveon shot is a 571K file and the bayer is compressed down to
> > 127K with loads of nasty jpeg artifacts. I'm sure there isn't any Hokey
> > pokey going on with the edits either..
> >
> > Hmm I wonder if the person who put this up has any bias?
> >
>
> They seem biased towards Bayer images.

You both need to learn how to use a web browser. There are these
things called thumbnails, you click them to open the originals.

No wonder Bayer users never post images bigger than 640x480, they don't
know these sites automatically create thumbnails for you.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

<george_preddy@yahoo.com> wrote in message
news:1117979230.592887.61690@g47g2000cwa.googlegroups.com...
>
>
> dylan wrote:
>> I see it now, take some poor pictures, reduce the size, enlarge them
>> again
>> and they aren't as good as the originals.
>
> With Foveon images, that is correct. With Bayer the image is almost
> completely unchanged by downsizing to quarter-size, as is shown so
> nicely.
>

Choose some decent examples to try this on, not this collection of poor
shots. You can even see the bayers are degraded even on these poor examples.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

<george_preddy@yahoo.com> wrote in message
news:1117979882.992319.125360@g43g2000cwa.googlegroups.com...
>
>
> RichA wrote:
>> On 5 Jun 2005 00:26:07 -0700, george_preddy@yahoo.com wrote:
>>
>> >http://www.pbase.com/optical_vs_interpolated
>>
>> What's with the B.S. grew print on a black background???
>
> Some people are bound to have a nasty reaction when they are shown that
> the digital camera they bought only delivers about 25% of its
> advertised, full color resolution.

Even if this was true at least they would be the correct colors.
>

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

In article <1117979882.992319.125360@g43g2000cwa.googlegroups.com>,
<george_preddy@yahoo.com> wrote:

> Some people are bound to have a nasty reaction when they are shown that
> the digital camera they bought only delivers about 25% of its
> advertised, full color resolution.

Doesn't it make you sad to think of the outrageous amount of money you
spent and all you got was a 3.24MP kiddie toy?

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

In article <1117979665.765878.173480@z14g2000cwz.googlegroups.com>,
george_preddy@yahoo.com says...
>
>
> KennyJr wrote:
> > In article <3gfpngFc5qpoU1@individual.net>, fotocord@yahoo.com says...
> > > george_preddy@yahoo.com wrote:
> > >
> > > > http://www.pbase.com/optical_vs_interpolated
> > >
> > > LOL, the foveon shot is a 571K file and the bayer is compressed down to
> > > 127K with loads of nasty jpeg artifacts. I'm sure there isn't any Hokey
> > > pokey going on with the edits either..
> > >
> > > Hmm I wonder if the person who put this up has any bias?
> > >
> >
> > They seem biased towards Bayer images.
>
> You both need to learn how to use a web browser. There are these
> things called thumbnails, you click them to open the originals.
>
> No wonder Bayer users never post images bigger than 640x480, they don't
> know these sites automatically create thumbnails for you.
>
>

I had three browser windows open; one for the thumb nails, one for the
Foveon pitures, and one for the Bayer pictures.

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

dylan wrote:
> <george_preddy@yahoo.com> wrote in message
> news:1117979882.992319.125360@g43g2000cwa.googlegroups.com...
> >
> >
> > RichA wrote:
> >> On 5 Jun 2005 00:26:07 -0700, george_preddy@yahoo.com wrote:
> >>
> >> >http://www.pbase.com/optical_vs_interpolated
> >>
> >> What's with the B.S. grew print on a black background???
> >
> > Some people are bound to have a nasty reaction when they are shown that
> > the digital camera they bought only delivers about 25% of its
> > advertised, full color resolution.
>
> Even if this was true at least they would be the correct colors.

There's no way to say "even if" with a straight face, when the images
are right in front of you. Amyone can reproduce the reults themselves.

When I bought my 1DMkII, of course I knew it was a 2MP camera with nice
color. Nice color is worth a lot, but when intelligent people compare
a Bayer DSLR to a Foveon DSLR they understand that dividing the Bayer's
sensor's monochrome pixel real estate into 3, very small, RGB sensors,
results in similarly very low full color resolution.

And isn't it interesting that the "pro" review web sites are more than
happy to participate in the scam to rip you off as spectacularly as
possible, by testing only using B&W resolution targets?

What a shame it is not to have a real camera review site available,
only paid-for sales-front sham rip off artists, pretending to be unpaid
by the same Bayer makers they "test" (in B&W).

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

<george_preddy@yahoo.com> wrote in message
news:1117988841.569923.109310@z14g2000cwz.googlegroups.com...
>
>
> Hannah wrote:
>> <george_preddy@yahoo.com> wrote in message
>> news:1117982525.921793.137480@g14g2000cwa.googlegroups.com...
>> >
>> > > You twerp. No pros use Canon?
>> >
>> > Lot's of pros use Canon, but few understand what they have: a
>> > monochrome sensor sub-divded into 3 very small RGB sensors.
>> >
>>
>> You idiot. 99.8% of pros use Canon. Certainly, none use Sigma. I know one
>> or
>> two loyal to Nikon.
>
> I was able to verify your stats, you are right, in the years 1997-2005
> only 2 Nikon DSLRs were sold.
>
>> >The 1DMkII is a good camera, but it's only 2MP in RGB, as shown.
>>
>> Twerp.
>
> It is tough when you realize you were ripped off to the tune of 400%.
> But realize that Canon DLSRs, even though the Bayer design sub-divides
> their avertised monochrome MPs in to 3 very small RGB exposures,
> resulting in similarly low full color resolution, are decenty made
> camera bodies once you break the $4000 barrier.
>
> Canon DSLRs generally produce nice colors too, which is why I bought my
> 2MP 1DMkII. Even if the full color resolution of a $4000-5000 2MP
> 1DMkII isn't quite the equal of a $425 3.4MP SD9 pro body, it still
> takes good pics most of the time.
>

Golly George, since you like the SDx, why did you waste your money on the
canon?

perhaps because you're lying!

Oh, we knew it. Just wanted to make sure you know we knew it

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

Tom Scales wrote:
> > Canon DSLRs generally produce nice colors too, which is why I bought my
> > 2MP 1DMkII. Even if the full color resolution of a $4000-5000 2MP
> > 1DMkII isn't quite the equal of a $425 3.4MP SD9 pro body, it still
> > takes good pics most of the time.
>
> Golly George, since you like the SDx, why did you waste your money on the
> canon?

Did you read the above? Canon DSLRs are nice cameras. But they do
have very low color MPs for the money. I'm sure its nice to be a Bayer
maker, able to get away with advertising monochrome MPs, but it forces
prices awfully high for the MPs you get.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

On 5 Jun 2005 07:42:05 -0700, george_preddy@yahoo.com wrote:

>> You twerp. No pros use Canon?
>
>Lot's of pros use Canon, but few understand what they have: a
>monochrome sensor sub-divded into 3 very small RGB sensors.

They understand results.
They also understand that your misunderstanding of how cameras work
doesn't affect the results.

--
Big Bill
Replace "g" with "a"

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

KennyJr wrote:
> The thing I get out of your comparison images is that the Bayer images
> look much better right out of the camera regardless what you did in post
> processing.

I like both of the Bayer images, the 2MP Canon 20D image is a nice shot
and so is the 1MP Kodak image.

> I have a Kodak DX6490 that lets me choose between saturated and nuetral

If you like photography, you really should be shooting RAW so you can
adjust everything.


> colors. I prefer the nuetral setting. Of course that doesn't change the
> fact the colors are muddy in the Foveon image.
>
> > > 2) the Foveon image appears out of focus, while the Bayer images appear
> > > sharp throughout the image.
> >
> > Funny one.
> >
>
> You may think it's funny, but I like a camera that produces sharp clean
> images.

I do think it's funny that you think a 1MP color camera with a pinky
finger nail sized sensor is competitive with a pro 3.4-color MP DSLR.
3.43 color MPs is the highest resolution pro DSLR sensor you can buy
for under $8000.

Not to mention the $8000 4MP-color 1Ds, although is has 20% more more
MPs ( = about 5% higher color resolution in theory), also the need for
a blur filter, plus it always has to use the VREY low quality outer
frame of the lens. Those limitations combined, give the 3.43MP-color
SD9 higher achieved full color resolution in most cases at a little
over 1/20th the price.

It is minor drawback that the SD9 body is so much larger than the 1Ds
body. I like large camera bodies but the 1Ds body is big enough
without adding another inch or so to the height. OTOH, the Canon's
dR/20D family are way too small.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

Big Bill wrote:
> On 5 Jun 2005 07:42:05 -0700, george_preddy@yahoo.com wrote:
>
> >OBVIOUSLY when you overlay 3 monochromatic R+G+B exposures, the
> >resulting composite image's full color resolution is equal in dimension
> >to any one of the 3 RGB originals.
>
> Possibly; that's obviously open to discussion

No it isn't.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

On 5 Jun 2005 10:02:17 -0700, george_preddy@yahoo.com wrote:

>
>
>Big Bill wrote:
>> On 5 Jun 2005 07:42:05 -0700, george_preddy@yahoo.com wrote:
>>
>> >> You twerp. No pros use Canon?
>> >
>> >Lot's of pros use Canon, but few understand what they have: a
>> >monochrome sensor sub-divded into 3 very small RGB sensors.
>>
>> They understand results.
>> They also understand that your misunderstanding of how cameras work
>> doesn't affect the results.
>
>I never said my 2MP 1DMkII can't produce results. The 2MP results are
>great.


There's that misunderstanding again.
If you applied yourself to understanding instead of simply insisting
you're right when the facts prove otherwise, you'd be much better off.

--
Big Bill
Replace "g" with "a"

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

On 5 Jun 2005 10:00:10 -0700, george_preddy@yahoo.com wrote:

>I do think it's funny that you think a 1MP color camera with a pinky
>finger nail sized sensor is competitive with a pro 3.4-color MP DSLR.
>3.43 color MPs is the highest resolution pro DSLR sensor you can buy
>for under $8000.

When you get to define things, you canmake all sorts of claims.
What, exactly, is a "pro 3.4-color MP DSLR"? I can't fnd such a
description anywhere except from you.
And you certainly aren't in a class to be able to make such
definitions, since you still don't understand the definition of a
"pixel".

--
Big Bill
Replace "g" with "a"

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

Big Bill wrote:
> On 5 Jun 2005 10:02:17 -0700, george_preddy@yahoo.com wrote:
>
> >I never said my 2MP 1DMkII can't produce results. The 2MP results are
> >great.
>
> There's that misunderstanding again.
> If you applied yourself to understanding instead of simply insisting
> you're right when the facts prove otherwise, you'd be much better off.

Ok then it doesn't produce great results, have it your way.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

Big Bill wrote:
> When you get to define things, you canmake all sorts of claims.
> What, exactly, is a "pro 3.4-color MP DSLR"? I can't fnd such a
> description anywhere except from you.

It's easy, all manufacturers except Foveon list monochrome MPs. SO if
you see an MP rating it is the monochrome reoslution. To take a color
image, the sensor surface is subdivided to take 3 very small
simultaneous RGB exposures which are then overlayed to make color MPs.
The size of the tiny exposures is the color MP rating of the camera.

That tiny image is then interpolatively upscaled (placeholders
inserted) to bring it back upto the monochrome sensor's image
dimensions. Thus, fooling customers to the tune of claiming 400%
higher MPs than the camera can optically support. Everyone in the
industry obviously understands this, or the standard "pro" review sites
wouldn't take money from Bayer manufactures to show only monochrome
resolution results.

dpreview.com's paid monochrome sales pitch...

SD9:
http://www.dpreview.com/reviews/Si [...] m__sd9.jpg

6MP Canon D60:
http://www.dpreview.com/reviews/Si [...] eosd60.jpg

Now, color resolution tests from an obscure scientific article...

SD9:
http://www.outbackphoto.com/artofr [...] _00200.jpg

Canon 10D:
http://www.outbackphoto.com/artofr [...] 71_RT8.jpg

Obviously the SD9 wins in both cases, but when shooting monochrome
targets the results are much closer.

The "pro" review sites are more than happy to fool people in exchange
for money, that's what makes them "pro" by their standard--the ability
to collect. In reality, to be a "pro" you need to conform to
professional ethics, which requires putting the good those you serve
above your bottomline. So while these sites are certainly profitable,
they are certainly not professional.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

"dylan" <no@mowhere.com> wrote in message
news7ud9b$hl1$1@newsg3.svr.pol.co.uk...
>
> <george_preddy@yahoo.com> wrote in message
> news:1117956367.533480.218200@f14g2000cwb.googlegroups.com...
>> http://www.pbase.com/optical_vs_interpolated
>>
>
> What's it trying to show / say ?

What has this *downsizing* bollocks got to do with real life photography
anyway..? Really I not being facetious, I don't get it...

I admit I was jumping up and down with excitement when I first saw the
Foveon exhibition stand some years ago. With the progress of CMOS image
sensor technology I sincerely hope that it will be a viable option in the
future. Meanwhile I have to be realistic and use a camera that produces
superior results by successfully "fudging" the information together: it
works though! The proof is in the print.
Foveon might make it eventually, as compared to CCDs their sensor is still
in its infancy. They dared to take a completely different approach and it
might pay off in the future. Didn't Canon suffer initially when they decided
to build their own electronics for their digital cameras? Nikon follow in
their shadow now with other people STILL making some of their innards. (I'm
a professional Nikon user by the way!)

Regards,

Craig.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

<snip>
> It's easy, all manufacturers except Foveon list monochrome MPs. SO if
> you see an MP rating it is the monochrome reoslution. To take a color
> image, the sensor surface is subdivided to take 3 very small
> simultaneous RGB exposures which are then overlayed to make color MPs.
>
It doesn't work like that with CCDs. They're analogue devices and you can't
address the elements in such a manner. Plus the complex algorithms use
information from neighbouring elements during the "fudging" process. It's
not as cut and dry as you're making it out, although I see the point that
you're trying to make. It's like we're all cheating, but it's not hurting
anyone and I like the smiles on my customers' faces!
It's a little like the argument I used when the Nikon D1 was first in use
for film versus digital. (My mate has a mobile phone that has a camera
producing better looking results than the D1)

Craig.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

In article <66b6a1pr6qrbscus1aofguubdnoe737p7r@4ax.com>, Big Bill
says...

> Bayer-filtered cameras are truly whatever resolution they claim to be
> (with notable exceptions); color is interpolated, but luminance is
> unique for each pixel.

Well, no. While the green channel is a good enough approximation of
luminance, the red and blue channels are not.
--

Alfred Molon
------------------------------
Olympus 4040, 5050, 5060, 7070, 8080, E300 forum at
http://groups.yahoo.com/group/MyOlympus/
Olympus 8080 resource - http://myolympus.org/8080/

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

<george_preddy@yahoo.com> wrote in message
news:1118004051.205850.243800@g43g2000cwa.googlegroups.com...
>
>
> Bart van der Wolf wrote:
>> <george_preddy@yahoo.com> wrote in message
>> news:1117997855.375671.63640@g44g2000cwa.googlegroups.com...
>> > No it's not. You absolutley must have a R, G, and B sensor to
>> > produce
>> > a true full color pixel.
>>
>> Double nonsense, again, as usual.
>> The result is a trichromatic color (which fails with metameric color),
>
> It doesn't matter if the 3 primary color model is imperfect, it's the
> same model used for both technologies. Obviously.
>
> The difference is that a Foveon sensor captures 3 huge RGB exposures,
> while a Bayer sensor divides a monochromatic device so that it may
> capture 3 very small RGB exposures--until you start paying car prices.
>

Too bad it doesn't work worth a damn.

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital (More info?)

On 5 Jun 2005 11:18:00 -0700, george_preddy@yahoo.com wrote:

>
>
>Big Bill wrote:
>> On 5 Jun 2005 10:02:17 -0700, george_preddy@yahoo.com wrote:
>>
>> >I never said my 2MP 1DMkII can't produce results. The 2MP results are
>> >great.
>>
>> There's that misunderstanding again.
>> If you applied yourself to understanding instead of simply insisting
>> you're right when the facts prove otherwise, you'd be much better off.
>
>Ok then it doesn't produce great results, have it your way.

It's the 2MP part that shows your misunderstanding.

Since you say you own Canon, why not post a link to a few of these
great photos?

--
Big Bill
Replace "g" with "a"