Go Back   Steve's Digicams Forums > Digicam Help > General Discussion

Reply
 
Thread Tools Search this Thread
Old Dec 3, 2002, 7:41 AM   #1
Junior Member
 
Join Date: Dec 2002
Posts: 8
Default How many megapixels do I need?

I really like the Canon G-3. I'm a little (maybe too much) hung up on the 4 vs 5 megapixel specification. There are several 5MP cameras in the same price range. I'm new to digital, but a long time photographer with a Canon EOS film system. I always have tried to use the finest grain films for the situation I am photographing. I use my slides frequently projected very large in lecture halls and auditoriums. I also enlarge and mount some of my better photos.

1. What is the theoretical size limit for an enlargement (print) of a 4MP camera 8x10, 11x14?

2. How much difference will there be on a projected image from a 4MP vs. 5MP camera in an auditorium where the images are typically 30 feet or even larger. Certainly the resolution of the projector and size of file will enter into the quality of the projected image as well.

If anyone could help me understand this better I would greatly appreciate it. Thanks in advance.

Frank
Frank Chang is offline   Reply With Quote
Sponsored Links
Old Dec 3, 2002, 9:17 AM   #2
Senior Member
 
BillDrew's Avatar
 
Join Date: Jun 2002
Location: Hay River Township, WI
Posts: 2,512
Default

The max print size from a specified MPixel size depends on issues like how well the camera is in focus, amount of detail in the subject, amount of sharpening, viewing distance, phase of the moon, ... and most importantly - the eye of the beholder.

All else equal (it never is), the maximum output size increases as the square root of the megapixel ratio, e.g., a 5 MP camera can produce a print 11.8% larger in linear dimension than a 4 MP camera. Nice to have, but not likely to the the crucial deciding factor between two cameras. The 5MP camera is likely to be better than a 4 MP because it is likely to be a newer and improved design - not simply because of the increased pixels.
BillDrew is offline   Reply With Quote
Old Dec 3, 2002, 10:52 AM   #3
Senior Member
 
Join Date: Sep 2002
Posts: 3,585
Default

To the 2d point of your question: The reproduction of an image will depend more on the projector than the image. I doubt you could tell any difference in an image projected in a lecture hall.
gibsonpd3620 is offline   Reply With Quote
Old Dec 3, 2002, 2:28 PM   #4
Junior Member
 
Join Date: Dec 2002
Posts: 8
Default

Thanks,
1. Both Cameras are brand new. They are the Canon G-3 (4MP) and the Olympus 5050 (5MP). My impression taking and comparing a few shots in the store was that the overall exposure of the Canon G-3 was more pleasing, the Olympus flash seemed to overpower the subject, underexposing all the ambient light in the backround (too dark) and washing out the subtle existing light on the subject itself. The Canon exposures were better balanced (IMHO). Unfortunately that's about as much testing as the sales person would tolerate.
2. Does anyone have experience with these two cameras side by side outside comparing clarity of optics and overall exposure and focusing accuracy? The sales person claims the Canon G-3 is better than the Olympus 5050 in these comparisons.
3. 11.8%? Are there rules of thumb or recommendations regarding # of megapixels and recommended print size before the megapixels become the limiting factor in resolution?
Thanks again for the responses, any one else please chime in.
Frank Chang is offline   Reply With Quote
Old Dec 3, 2002, 4:37 PM   #5
Junior Member
 
Join Date: Dec 2002
Posts: 7
Default

I think you should also keep in mind that just because a camera advertizes 5MP, doesn't necessarily mean there will be more DETAIL than a 4MP camera.
This is because some manufacturers interpolate up to their advertized MP number, from a lower resolution sensor. For example, the sensor actually captures 3.5MP worth of data, but the camera processes/interpolates it up to 4MP. It's not creating any detail in this process, but often, upsampling to higher resolutions helps reduce aliasing artifacts.
However, Photoshop can do this just as well (and probably better) due to its more advanced algorithms (bicubic sampling for example).

-Ryan
RyanDinan is offline   Reply With Quote
Old Dec 3, 2002, 5:17 PM   #6
Senior Member
 
Join Date: Aug 2002
Posts: 2,162
Default

There are some tests you can do without a cam if res. is your only problem. But as already said, newer cams may have better sensitivity/lower noise so MPix is not everything.

Anyway, take a large very high quality photograph of similar scene content to your intended projection images, or search the NASA site for a very high quality uncompressed image (large file). 5 and 4 Mpix will be equivalent to about 40Mb and 32 Mb file sizes.

Scan your photo at a res. which gives you file sizes of say 50-60Mb. Then downsample in a photo editor to the horizontal and vertical res of the 2 cams. Find out what res. and compression ratio modes are used in both cams (web site handbooks) and save as compressed JPEG to get a similar file size.

This is a bit synthetic and not precise science (but it is cam-less). Now zoom the images to either the output print size you're interested in at the resolution of your printer, or projection resolution and size - but remember to stand back from your monitor at an equivalent auditorium viewing distance, or scale the magnification, or borrow a projector and feed it your test files.

There are lots of caveats, However, I think you'll find that resolution will be more limited by the projector (already said) and what you really need to know is how much compression can be tolerated for your output image sizes, before JPEG artefacts become annoying.

I haven't tried this so don't blame me if I've missed something out, but you should get the idea of cam-less testing for res. only, without going to the shop! Don't forget, if you're presenting high resolution SVGA images, you need to check the time it might take to render them in a pc and pace your presentation accordingly.

You might have already guessed that you could shoot film and scan a large quality print , if you don't like the results from your tests!
voxmagna is offline   Reply With Quote
Old Dec 4, 2002, 2:07 AM   #7
Junior Member
 
Join Date: Dec 2002
Posts: 8
Default Does size really matter?

Wow,
I'm just a dumb surgeon, I did not really understand answers I received.

OK I do understand the resolution of the projector will be the limiting factor on the projected images and have experienced laptop purgatory when the image files were too large to load and the presentation stalls.

My first question was what size can I enlarge a photgraph to with a 4MP (Canon G3) camera compared to a 5MP (Olympus C-5050) camera. For a 4MP camera there must be a size where the resolution of 4MP becomes a limiting factor in clarity and a 5MP will be able to clearly produce a slightly larger print (11.8% larger according to BillDrew). It seems to me that the difference should be closer to 20% between 4MP and 5MP. I'm not doubting BillDrew, but asking for a dimension such as 11x14, 13x17, 16x20?

I'm not trying to be disrepectful and really appreciate the efforts of you all to try and educate this simple new to the digital world doctor.

The more difficult corrollary question is, since not all camera images are equal, how much does size (MP) matter, and how does one measure quality, clarity, of images between different cameras to determine which camera to purchase? I'm obviously trying to separate the image quality of the Canon G3 and the Olympus C5050 and how the one megapixel difference will affect my enlargements.

Thanks for any more comments anyone would care to make regarding my dilemma!

Frank Chang
Frank Chang is offline   Reply With Quote
Old Dec 4, 2002, 5:14 AM   #8
Senior Member
 
Join Date: Aug 2002
Posts: 2,162
Default

Quote:
The more difficult corrollary question is, since not all camera images are equal, how much does size (MP) matter, and how does one measure quality, clarity, of images between different cameras to determine which camera to purchase?
The answer is probably with great difficulty between 4 & 5 Mpix, since there is probably no clear marginal difference based on resolution alone (which I and others have mentioned).

A digicam is a multi-functional product with compromised design to work in as many situations as users might want. BUT they don't do all these things equally well. So, some models might be better in lower light, have bigger lenses, more zoom and features, but are big and cumbersome for some users. Other cams might be smaller more compact, claim to have the same 'features' but don't perform as well at the extremes.
voxmagna is offline   Reply With Quote
Old Dec 4, 2002, 8:00 AM   #9
Senior Member
 
BillDrew's Avatar
 
Join Date: Jun 2002
Location: Hay River Township, WI
Posts: 2,512
Default

There is no standard for how large a print can be made based only on the number of pixels in the camera. None. Never has been, never will be.

Some folks will claim they can make a perfect 8x10" print from a 0.2 MPixel Barbie Cam - and they are correct if the subject lacks detail and the print is viewed across a large room while drinking the third gin-and-tonic.

Others will claim that an 8x10" print made from the output of a 200 MPixel camera is not good enough because they cannot make out the species of a moth at the top of a tree in the background using a 20x loupe.
Quote:
".... the Olympus flash seemed to overpower the subject, underexposing all the ambient light in the backround (too dark) and washing out the subtle existing light on the subject itself."
IMHO, the built in flash in all cameras (digital and chemical) are junk. If flash is important to you, you want a camera that can use an external flash. That is an example of the kind of issue which can be more important than the difference between 4 and 5 MP.
BillDrew is offline   Reply With Quote
Old Dec 5, 2002, 1:30 AM   #10
Junior Member
 
Join Date: Dec 2002
Posts: 8
Default Thanks

Thanks for everyone trying to educate me. I really appreciate the feedback and time you have spent helping me start to understand the more important variables between cameras. I think I'm starting to get it. Correctly prioritize features to match specific shooting situations the camera will be used for.
Frank C
Frank Chang is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

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 Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 2:16 AM.