10 vs 8 MP

I am trying to decide if 10 MP is significantly better than 8MP. ( just want
to decide whether to buy Canon XTi or 30d )
The very respectfule sources say - no difference. While the image is bigger
actually Canon's 8MP CMOS is ever sharper than 10 Mop sensor.
Read here for example at the bottom of the page
http://www.dpreview.com/reviews/canoneos400d/page24.asp
Many other similar sources agree ( for example cnet.com)

Now compare the following two images from the same source.

http://www.dpreview.com/reviews/Cano...so0100_std.JPG
http://www.dpreview.com/reviews/Niko...0d_iso0100.JPG

Magnify the images to 400% and try to read the vertical text on the middle
bottle.
The text produced by Canon XTi's is readable the one from Canon 30D (or
350D) is not.

To me this is a very big difference. Am I missing something ?

VC skrev:

Now compare the following two images from the same source.

http://www.dpreview.com/reviews/Cano...so0100_std.JPG
http://www.dpreview.com/reviews/Niko...0d_iso0100.JPG

Magnify the images to 400% and try to read the vertical text on the middle
bottle.
The text produced by Canon XTi's is readable the one from Canon 30D (or
350D) is not.

To me this is a very big difference. Am I missing something ?

First off, I thought both texts were readable - and I downloaded the
30D image first. But yes, the text on the 400D reads slightly easier.

Please note, though, that the 10MPsensors, whether from Canon, Sony or
anyone else, all seem to employ some degree of on-chip noise reduction.
On-chip noise reduction presumably increases with ISO sensitivity. And
small detail with extremely high contrast - such as white text on
black, or vice versa - is less blurred by the effects of noise
reduction than small detail with low contrast.

Thus, the 400D may well be a tad better at capturing the detail of
high-contrast text than the 30D at ISO100. But it is still a very open
question whether this performance can be generalised to low-contrast
detail such as fur or foliage, or whether it can be generalised at all
to, say, ISO800.

Jan Böhme

VC wrote:
I am trying to decide if 10 MP is significantly better than 8MP. ( just want
to decide whether to buy Canon XTi or 30d )
The very respectfule sources say - no difference. While the image is bigger
actually Canon's 8MP CMOS is ever sharper than 10 Mop sensor.
Read here for example at the bottom of the page
http://www.dpreview.com/reviews/canoneos400d/page24.asp
Many other similar sources agree ( for example cnet.com)

Now compare the following two images from the same source.

http://www.dpreview.com/reviews/Cano...so0100_std.JPG
http://www.dpreview.com/reviews/Niko...0d_iso0100.JPG

Magnify the images to 400% and try to read the vertical text on the middle
bottle.
The text produced by Canon XTi's is readable the one from Canon 30D (or
350D) is not.

Assuming you mean the white on black text.
The 400D is a bit better in those images, yes. The CAPS lettering is
readable and the lower case, almost so.
The 30D the CAPS lettering is somewhat readable, the lower case not at all.

10 Mpix is only 11% better resolution than 8 Mpix ... but when you hit a
borderline case (like this) the text goes from 'just readable' to 'not
readable' over that small difference. Threshold case.

They should both render 8x12's very cleanly and double that if you don't
examine them up too close.

Cheers,
Alan

--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.

Splitting digital hairs.
You must live a charmed life if you have nothing better to agonize over than
this piffle.

VC wrote:
I am trying to decide if 10 MP is significantly better than 8MP. (
just want to decide whether to buy Canon XTi or 30d )
The very respectfule sources say - no difference. While the image is
bigger actually Canon's 8MP CMOS is ever sharper than 10 Mop sensor.
Read here for example at the bottom of the page
http://www.dpreview.com/reviews/canoneos400d/page24.asp
Many other similar sources agree ( for example cnet.com)

Now compare the following two images from the same source.

http://www.dpreview.com/reviews/Cano...so0100_std.JPG
http://www.dpreview.com/reviews/Niko...0d_iso0100.JPG

Magnify the images to 400% and try to read the vertical text on the
middle bottle.
The text produced by Canon XTi's is readable the one from Canon 30D
(or 350D) is not.

To me this is a very big difference. Am I missing something ?

Regarding enlargement/printing, the difference is pretty insignificant.

If I were you, I'd wait about 4 weeks to see Canon's nearly-certain
replacement of the 30D (February should bring significant
announcements)...and then decide to either wait a month or two for thenew
model...or pick up the outgoing model for less, since prices usually drop at
that point.

--
Images (Plus Snaps & Grabs) by Mark² at:
www.pbase.com/markuson

babaloo wrote:
Splitting digital hairs.
You must live a charmed life if you have nothing better to agonize
over than this piffle.

Oh I don't know. Camera makers have put in quite an effort to make these
small (+2MP) upgrades which causes understandable confusion in the buying
public with just such questions. It's not a dumb question at all.

--
Images (Plus Snaps & Grabs) by Mark² at:
www.pbase.com/markuson

babaloo wrote:
Splitting digital hairs.
You must live a charmed life if you have nothing better to agonize over than
this piffle.

Since there is a discernable difference in the text readability (despite
the CW that 8 to 10 Mpix is a trifle) then it is not splitting hairs nor
is it agonizing.

--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.

In article , VC
writes
I am trying to decide if 10 MP is significantly better than 8MP. ( just want
to decide whether to buy Canon XTi or 30d )
The very respectfule sources say - no difference. While the image is bigger
actually Canon's 8MP CMOS is ever sharper than 10 Mop sensor.
Read here for example at the bottom of the page
http://www.dpreview.com/reviews/canoneos400d/page24.asp
Many other similar sources agree ( for example cnet.com)

Now compare the following two images from the same source.

http://www.dpreview.com/reviews/Cano...ed/Studio/30d_
iso0100_std.JPG
http://www.dpreview.com/reviews/Niko...Studio/400d_is
o0100.JPG

Magnify the images to 400% and try to read the vertical text on the middle
bottle.
The text produced by Canon XTi's is readable the one from Canon 30D (or
350D) is not.

To me this is a very big difference. Am I missing something ?

Probably the fact that your links are to cameras with different lenses?

More important than any effect near the limit of resolution is what
lenses your are going to use. I purchased the 350D in preference to the
20D or 30D because the difference in price did not make it twice as good
optically, and was enough to purchase an extra lens.
--
Ian G8ILZ
There are always two people in every pictu the photographer and the viewer.
~Ansel Adams

VC wrote:
I am trying to decide if 10 MP is significantly better than 8MP. ( just want
to decide whether to buy Canon XTi or 30d )
The very respectfule sources say - no difference. While the image is bigger
actually Canon's 8MP CMOS is ever sharper than 10 Mop sensor.
Read here for example at the bottom of the page
http://www.dpreview.com/reviews/canoneos400d/page24.asp


Broadly correct. The 400D (XTi) is the softest-imaging of all the
current 10 megapixel DSLRs - I've tested all of them now. It is probably
designed this way to reduce noise levels. The 30D is a pretty sharp 8
megapixels, because they made changes to lower the level of
anti-aliasing and recover more fine image detail.

If you can afford the 30D, which is a far superior object physically,
there's no contest. I find the 400D colours, contrast and sharpness
quite disappointing but my daughter, who uses it permanently and shoots
stock landscapes and stuff for our library files, is under instructions
never, ever to shoot anything except RAW and it's amazing what Adobe
Camera Raw can do to pep up the results.

David

Prometheus wrote:

Now compare the following two images from the same source.

http://www.dpreview.com/reviews/Cano...ed/Studio/30d_
iso0100_std.JPG
http://www.dpreview.com/reviews/Niko...Studio/400d_is
o0100.JPG

To me this is a very big difference. Am I missing something ?


Probably the fact that your links are to cameras with different lenses?

Nope. The images above are both for Canon and both tests were with the
Canon 50 mm f/1.4.

IAC, the centre (where the text occurs) of a cropped sensor 50 f/1.8
Nikkor and 50 f/1.4 Canon are almost the same sharpness.

http://www.photodo.com/product_44_p4.html Canon 50 f/1.4
http://www.photodo.com/product_257_p4.html Nikkor 50 f/1.8
(x axis is mm from center of the sensor)

Cheers,
Alan

--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.