Megapixel's / Resolution Qualty

I have a Canon SD400 5 megapixel point and shoot camera and I love it

for it's quality and compact design.

 

I have several questions.

 

It has 3 sizes that I can choose from

L = 2592x1944

M1 = 2040x1536

M2 = 1600x1200

S = 640x480

 

My question is the following. If the camera is a 5 megapixel camera

does it matter what resolution/size I use?

 

OR does the megapixels get smaller as i choose a smaller resolution.

I.E. L = 5megapixels, M1 = 4megapixels, M2 = 3Megapixels and S =

2megapixels?

 

Because I thought it the camera is 5mp then no matter what size i

choose it will have 5mp per square inch or some unit of measure.

 

Thank you

It is only 5 mp at its highest setting, IE 2592x1944 = 5.04 mp.

 

I generally always shoot at the highest setting and reduce the size as needed in PS.

L=5, M1=3, M2=2, S=0.3

As Jean-Baptiste says, always shoot at the highest.

Ooops, I mean as Troy said.

Multiply those figures together and you have the Mp total at each setting.

 

The other option you have is to do with compression

Super-fine, Fine, and Normal .... don't take that normal to mean that is the normal way to use the camera :-) Normal is the lowest setting to be used only when away from home and running out of space on the card for more shots ........ so for quality the largest resolution and superfine .... :-)

 

Although Fine may be suitable for your needs .. check the jpg file size as they are copied out of the camera and listed in Windows Explorer. If they range between 1.4 and 2Mb that is very similar to my 5Mp Nikon with it's 4:1 compression and that is a good compromise between top quality and speed in downloading to the storage card. I use it all the time. File sizes vary with the tonal complexity of the image when they are using the jpg system.

In general I agree with shooting at the lowest jpg compression, eg. SHQ. However, much depends on your camera.

 

I have an Oly 5060 and E-300 and have done a lot of comparisons between SHQ and HQ (as well as with TIFF) and, frankly, I am very hardpressed to tell the difference. Wrotniak and other reviewers have pointed out that Oly has superb compression algorithms which in practical terms mean that for virtually all practical purposes it's difficult to tell the difference between SHQ and HQ. What does this mean to me? When out for a day or so with a 1 gig card I shoot at SHQ, but when off on a very long trip I often step down to HQ because I prefer to keep my images both on the card and in a backup. Now and then I triage up to SHQ or Raw. Most commercial work I do is RAW. I don't do a lot of printing, and never any at more than 8x10 and I get stunning results with very little post-process tweaking. I might add that my daughter uses Canon and Fuji cameras and the differences between SHQ and HQ seem much greater. So, get to know your camera.

Thanks for the info guys. I feel a little silly now that someone pointed out that you multiply the resolution and you get your megapixel rate.

 

Awesome advice

Nobody is silly asking a question that they do not know the answer to which ignorance of can lead them into strife ... good luck :-)

On top of what others have said, I think it's important to think in terms of the number of independent pixels one has. For example, I have a 5Mp P&S, but the pixels are only 2¼ microns across. This means the lens needs to be able to resolve over 400 lines/mm for each pixel to be seeing something different from the pixel(s) right next to it. Since this is a virtual impossibility -- a lens that could do 150-200 lines/mm would be superb -- the pixels are not independent of each other. Another way of saying this is that the image is oversampled by the detector.<P>

 

Thus it's not surprising that my "fine" level files are about 2½ MB max. That's all the information (almost) that's there. (¾-1 byte per pixel is a very high quality JPEG imo). Though cameras differ in their internal details, generally when you choose the "lower" quality modes you get fewer pixels because the sensor's pixels are binned (say 2x2) as they're read out. If the image is oversampled to begin with, this doesn't result in the loss of very much real resolution, and it can lower the noise so that the resulting image looks better.<P>

 

Some simple experiments can show you how much you can go down in the number of pixels before you start to trade actual stored image quality for file size.<P>