Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > Digital Photography > Re: Going from 3 to 4 MP good for 8x10 prints?

Reply
Thread Tools

Re: Going from 3 to 4 MP good for 8x10 prints?

 
 
CNT
Guest
Posts: n/a
 
      09-16-2003
That was how I looked at this way, but found out from few trustworthy names
in here saying the 5MP makes it worse than the 4MP, so better off going with
4MP for picture quality. If numbers were more important, then go for the
5MP. BUT, for solely 8x10, maybe the 5MP has it's propose? I am no where
near camera expert, so you know.

Chuck

> If you're going to leave your 3 MP camera behind, why not jump up to a
> 5 MP instead of just 4 MP? If the extra resolution isn't needed, you
> can always use the extra detail to have more flexibility in croping.
> The Canon S50 is well priced for it's features and image quality.



 
Reply With Quote
 
 
 
 
gr
Guest
Posts: n/a
 
      09-16-2003
"CNT" <(E-Mail Removed)> wrote
> That was how I looked at this way, but found out from few trustworthy

names
> in here saying the 5MP makes it worse than the 4MP, so better off going

with
> 4MP for picture quality.


?!?

All else being equal, 5MP beats 4MP for quality. Perhaps some 5MP camera are
of poorer quality and design than some 4MP cameras, but in general a 5MP
camera will give you higher quality results. Check out sample pictures from
the various camera review sites and see for yourself.

I think it's not really worth upgrading from 3MP to 4MP. If you're going to
upgrade, you may as well go for 5MP or higher.


 
Reply With Quote
 
 
 
 
Rafe B.
Guest
Posts: n/a
 
      09-16-2003
On Tue, 16 Sep 2003 07:10:09 -0400, "gr"
<(E-Mail Removed)> wrote:

>"CNT" <(E-Mail Removed)> wrote
>> That was how I looked at this way, but found out from few trustworthy

>names
>> in here saying the 5MP makes it worse than the 4MP, so better off going

>with
>> 4MP for picture quality.

>
>?!?
>
>All else being equal, 5MP beats 4MP for quality. Perhaps some 5MP camera are
>of poorer quality and design than some 4MP cameras, but in general a 5MP
>camera will give you higher quality results. Check out sample pictures from
>the various camera review sites and see for yourself.
>
>I think it's not really worth upgrading from 3MP to 4MP. If you're going to
>upgrade, you may as well go for 5MP or higher.
>



"All else being equal" is the problem.

One of the main issues with consumer or even prosumer
digicams is noise. Sensor areas are still very small,
around 10mm on a side for the chip itself.

If you cram 25% more sensors into the same area, you
end up with noisier pixels.

So the extra resolution comes at the cost of extra noise.

That's why dSLRs (e.g. the 10d) are significant; it's the
first big jump in sensor size. And even then, the sensor
is still around 1/3 the area of a 35 mm frame.


rafe b.
http://www.terrapinphoto.com
 
Reply With Quote
 
Ken Alverson
Guest
Posts: n/a
 
      09-16-2003
"Rafe B." <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
>
> One of the main issues with consumer or even prosumer
> digicams is noise. Sensor areas are still very small,
> around 10mm on a side for the chip itself.
>
> If you cram 25% more sensors into the same area, you
> end up with noisier pixels.
>
> So the extra resolution comes at the cost of extra noise.


Certainly, unless something else improved, the noise per pixel is going to go
up with smaller pixels. However, has anyone done a comparison of the noise
across the same area of the image between 4 and 5 MP cameras? A piece of the
image that is 200x200 on a 4 MP camera will be 224x224 on a 5 MP camera. I
suspect that the 224x224 piece, despite having higher per-pixel noise, would
have less noise when resampled to 200x200 than the 4 MP image would.

Ken


 
Reply With Quote
 
VT
Guest
Posts: n/a
 
      09-16-2003
On Tue, 16 Sep 2003 12:16:27 GMT, Rafe B. <(E-Mail Removed)>
wrote:


>One of the main issues with consumer or even prosumer
>digicams is noise. Sensor areas are still very small,
>around 10mm on a side for the chip itself.
>
>If you cram 25% more sensors into the same area, you
>end up with noisier pixels.
>
>So the extra resolution comes at the cost of extra noise.
>


Agreed - and the more comprehensive tests show this -
at lower light levels and at higher ISO speeds.

Other tests seem not to being this up as a significant issue.

So which is right?

Perhaps neither,
perhaps both......


It is significant (in favor of the 4Mp)
if one does low light (high ISO) photography.

It is also significant (in favor of the 5Mp)
if one does general/normal light level photography.

I would say for most people not specializing in low light photography
the 5Mp makes a better choice, because of the added crispness of the
overall photo (although not easily seen until enlargements of about
10x8 or larger).

In fact the most vocal person about the "shortcomings" of the 5Mp
Canon S50 - was from Phil Askey - who also actually said this in his
S50 review:

http://www.dpreview.com/reviews/canons50/page12.asp

QUOTE:
When the 1/1.8" (7.2 x 5.3 mm) five megapixel CCD was first announced
there was universal concern about levels of noise which would be
produced from such small pixel cell sizes. While not quite as clean as
some newer three megapixel digital cameras levels of noise exhibited
certainly aren't higher than from previous four or five megapixel
digital cameras. Taking into account the S50's higher than indicated
sensitivity we can see from the graph below that it matches the Sony
DSC-V1 all the way up the sensitivity range.
UNQUOTE

So is noise "significant"? or not

The best way to determine this is to print out the test sample images
from both the intended 4Mp and 5Mp digicams (of which there are plenty
on the web) to the typical largest size one would normally want (use a
significant cropped area to save paper/ink) and see if one can see any
difference - if not, go for the cheaper camera - if one can - then
determine if any price differential is worth the added image
quality....... I guess only you can be the judge - as they say YMMV.

Having done my own ad-hoc printing "tests" to 10x8 I found -

3Mp was satisfactory - but I could detect some image breakdown - but
only with a 2.5x magnifying glass.

I could not see image breakdown on the 4Mp, and for me there was
noticable improvement over the 3Mp.

5Mp did add more crispness/snap to the print - although it was hard
for me to quantify.

That thread was:
Resolution - when is Enough ENOUGH? (a personal view)
date: 9/9/2003
http://groups.google.com/groups?hl=e...4ax.com&rnum=1


For me someshere between the 4Mp and 5Mp - ie: 227ppi to 260ppi
printing was "good enough".
--
Vincent
remove CLOTHES for e-mail

http://UnknownVincent.cjb.net/
 
Reply With Quote
 
gr
Guest
Posts: n/a
 
      09-16-2003
"Rafe B." <(E-Mail Removed)> wrote
>
> One of the main issues with consumer or even prosumer
> digicams is noise. Sensor areas are still very small,
> around 10mm on a side for the chip itself.


Read these:
http://www.wrotniak.net/photo/c5050/c5050-dark.html
http://www.wrotniak.net/photo/c5050/c5050-noise.html


 
Reply With Quote
 
Rafe B.
Guest
Posts: n/a
 
      09-17-2003
On Tue, 16 Sep 2003 21:38:25 -0400, "gr"
<(E-Mail Removed)> wrote:

>"Rafe B." <(E-Mail Removed)> wrote
>> On Tue, 16 Sep 2003 16:53:13 -0400, "gr"
>> <(E-Mail Removed)> wrote:
>>
>> >"Rafe B." <(E-Mail Removed)> wrote
>> >>
>> >> One of the main issues with consumer or even prosumer
>> >> digicams is noise. Sensor areas are still very small,
>> >> around 10mm on a side for the chip itself.
>> >
>> >Read these:
>> >http://www.wrotniak.net/photo/c5050/c5050-dark.html
>> >http://www.wrotniak.net/photo/c5050/c5050-noise.html

>>
>>
>> And then what?
>>
>> Both my digicams show noise in shadow detail,
>> and in very low light situations. The G2 is worse
>> than the 10D, not surprsingly.

>
>The point is that it's no worse than the 35mm film we all used to love and
>enjoy. The whole issue with noise in digital cameras is way over-hyped. It's
>not a big deal, unless you choose to shoot at high ISO (in which case, it's
>as noisy as high ISO film). What's the big deal about noise? A 5MP camera
>won't even show it in an 8x10" print, unless you examine the shadows with a
>magnifying glass!



I'll be happy to post some "comparative" samples of digicam
noise vs. noise from scanned film.

Both are ugly, though (IMHO) the digicam noise is more
detrimental to the print.

It's not so much an issue of high-ISO as low-light. The
noise and artifacts are absent in other parts of the same
image(s) that are not in shadow.


rafe b.
http://www.terrapinphoto.com
 
Reply With Quote
 
Michael Meissner
Guest
Posts: n/a
 
      09-17-2003
"CNT" <(E-Mail Removed)> writes:

> That was how I looked at this way, but found out from few trustworthy names
> in here saying the 5MP makes it worse than the 4MP, so better off going with
> 4MP for picture quality. If numbers were more important, then go for the
> 5MP. BUT, for solely 8x10, maybe the 5MP has it's propose? I am no where
> near camera expert, so you know.


In general the 5MP cameras are noiser than 4MP cameras, since the sensor is
generally the same physical size, but packed closer together. Quality of the
lens, compression algorithms, and in-camera processing also play a part in the
ultimate picture quality. FWIW, I get great 8x10's from my 2 megapixel Olympus
C-2100UZ (but then the lens and minimal compression were real good).

--
Michael Meissner
email: http://www.velocityreviews.com/forums/(E-Mail Removed)
http://www.the-meissners.org
 
Reply With Quote
 
Dr. Bob
Guest
Posts: n/a
 
      09-22-2003

>
>

lens, compression algorithms, and in-camera processing also play a part in
the
> ultimate picture quality. FWIW, I get great 8x10's from my 2 megapixel

Olympus
> C-2100UZ (but then the lens and minimal compression were real good).


I have gotten what I consider fine 8x10's from my HP618 (2mp). I have just
ordered a Canon G-3 (4 mp), and I'll be interested to see how much diff.
there is. BTW, I usually upped the HP file size to 3mp, using Genuine
fractals, before printing.


 
Reply With Quote
 
Michael Meissner
Guest
Posts: n/a
 
      09-24-2003
"Dr. Bob" <(E-Mail Removed)> writes:

> >
> >

> lens, compression algorithms, and in-camera processing also play a part in
> the
> > ultimate picture quality. FWIW, I get great 8x10's from my 2 megapixel

> Olympus
> > C-2100UZ (but then the lens and minimal compression were real good).

>
> I have gotten what I consider fine 8x10's from my HP618 (2mp). I have just
> ordered a Canon G-3 (4 mp), and I'll be interested to see how much diff.
> there is. BTW, I usually upped the HP file size to 3mp, using Genuine
> fractals, before printing.


Most of my printing is on commercial Fuji Frontier machines, and I don't bother
upscaling the image, since the printer does continous tone printing.

--
Michael Meissner
email: (E-Mail Removed)
http://www.the-meissners.org
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
color 8x10 repros, NYC would be nice Ray Normandeau Digital Photography 0 10-17-2003 08:37 PM
How do I print a portrait in Photoshop 8x10 Standard Size with constrained proportions? rbaker3 Digital Photography 5 10-15-2003 05:22 AM
Resolution for an 8x10 from Wal-Mart? bballxdotcom Digital Photography 7 09-07-2003 07:25 AM
FA: 8x10 ART COLOR NATURE PHOTO 11x14 FRAMED NEW Airline Pedestal Digital Photography 8 08-12-2003 06:18 PM
Question Re: 8X10 Printers Doug Yarnold Digital Photography 4 07-29-2003 01:56 PM



Advertisments