Hi all, just looking for some advice here. I have a Canon Ixus 200 IS (PowerShot SD980 IS in the States). Can anyone tell me what the best settings are for the pixels? I have the options for:

12M, 8M, 5M, 2M, 0.3M.

The sensor size is only 1/2.3" (6.16x4.62 mm, 0.28 cm²) and I have it set to 8M pixels but should I up to to 12M or lower it? I read conflicting advice online

Secondly, the auto ISO keeps ISO quite high (250 for daytime indoor shots with flash, and 400 for night shots indoors). Can I simply force it to do 100 with no ill effects?

Photo 1. Indoors, daytime (dull), auto ISO (250)
http://www.discopanda.com/gallery2/d/12637-1/IMG_0035.jpg

Photo 2. Indoors, daytime (dull), ISO 100
http://www.discopanda.com/gallery2/d/12643-1/IMG_0037.jpg

Photo 3. Indoors, daytime (dull), ISO 80
http://www.discopanda.com/gallery2/d/12640-1/IMG_0036.jpg

Photo 4. Indoors, night, Auto ISO
http://www.discopanda.com/gallery2/d/12618-2/IMG_0173.JPG

Photo 5. Indoors, night, Auto ISO
http://www.discopanda.com/gallery2/d/12414-2/IMG_0006.JPG

They seem a bit 'rubbish'. Any help?

Thanks all!

In theory reducing the number of pixels will result in less noise in the images and also reduce the amount of blurring you get from camera shake. But that does depend somewhat on how the camera steps down the number of pixels generated.

The ISO is purely the sensitivity of the sensor, when you up the ISO the camera tries harder to extra data from the sensor. This allows you to use slower shutter speeds without camera shake but at the same time increases noise in the picture.

"Can I simply force it to do 100 with no ill effects?"

No. Locking it at 100 when the camera wants to increase it will reduce the noise in your photos, but the trade off is that the shutter speeds will be longer and therefore more camera shake.

If you want to keep the noise down but avoid camera shake in low light on a compact camera then you have three options:

- put it on a tripod (or rested on something) and lower the ISO
- use the flash
- buy a proper camera

Photography is always a balance of compromises, turning something up always means turning something else down.

My brother-in-law got me a Gorillapod for Christmas. Works pretty well with our HD DVcam as well.

Thanks Roger, I think I've found my dad's birthday present.

I would go with Andy's advice and buy a proper camera

Even with a tripod or Gorillapod you may still need to use a higher ISO to get your subject somewhat sharp, but your camera looks pretty noisy even at 250ISO.

Does it have exposure compensation? If it does add 1/3 or 2/3 exposure compensation. Over exposing a shot slightly will reduce the visible noise in the picture as most noise will be in the shadows, you can then use post production to bring the black levels back to where they should be for a cleaner looking picture. This will allow you to use a higher ISO, a better shutter speed but keep the noise to a minimum in the shadows. Give it a try, it;s a trick used by many pros.

Cheers

Cris.

'Proper cameras' are all well and good, but not when you want to travel light. Compacts have their place.

Yes Cris, I thought the images are overly noisy and don't know why this should be the case. My older Ixus 500 produced better images I'm sure.

Will turning the pixels down to 5M help?

You are expecting quite a lot from any camera to have pictures look half decent in room lighting with no flash. Most room lights have a horrible Color Rendering Index- so yes they are always going to look dull no matter what camera you use. Digital Cameras can do a lot but they can't really get over physical limitations like light or lack of it.

Shoot at the maximum resolution of the camera unless for some reason you're trying to save space on your memory card. Lowering this resolution will not and can not improve your images. It won't affect exposure or the camera's ability to resolve detail. The sensor always captures the full resolution, the setting only alters what gets recorded to your memory card in the final stage - your camera will do a size reduction.

If you intend to have some choice in what to do with the images at some point in the future, you want them with as high a resolution as possible. That will let you apply different reduction or filtering options on a computer to achieve the best results possible for the intended application. You can batch automate a reduction as a second set of images for instance if you want to give them to someone for email or other purposes where highest quality doesn't matter.

I took a look at a couple of the images posted. They look fine. As expected anyway. You ay benefit from some white balance tweaks as well as focus and reducing camera shake. You might also benefit from pushing the exposure on some of them. In the images I saw, the white envelopes in the foreground were blown out, the shot included elements at a very wide exposure which your camera just can't capture. You'll need to supplement the lighting of the darker parts of the image if you want to equalize the exposure over the whole frame.

Best noise results will be achieved when shooting at the camera's native sensitivity setting. So set the native ISO for best results in that regard. However, the best picture is a picture taken/captured. If you don't push the sensitivity in some instances, you might not be able to get an image at all. So you decide if you want a less than stellar image or no image at all in those situations.

You probably shouldn't expect to be able to handhold the camera for steady images at a shutter speed lower than 1/60th of a second. Even at that speed you'll probably still shake a few of them. It's inevitable with a small camera and the inability to properly support it like you would with a larger and heavier camera body. If you're using flash in a room dark enough to otherwise be severely underexposed (or not exposed at all) at the current shutter speed, then you shouldn't need to worry about shake. As long as the aperture opens enough to expose the scene with the flash, the shutter speed is going to be somewhat irrelevant (a longer shutter will only expose the ambient lit scene).

If you do want to replace your camera with something small, though not as small, and with better quality and controls, take a look at the Canon G11 or similar.

I, too, wish I could find a camera that fit all of these requirements at the same time:

- Very compact size.
- Can take pictures in low light that are not blurry because of a long exposure.
- Can take pictures in low light that are not grainy because of too-high ISO.

Unfortunately, there is not yet any such thing. Compact cameras need small image sensors which get noisy at high ISO.

If you want grain-free and blur-free images at high ISO settings, you need a bigger, more expensive DSLR camera. You crank open the aperture and manually focus on your subject. You get a great picture, albeit with a very narrow focal depth.

One thing I might settle for is a compact camera with a flash that'll reach more than five feet. But I haven't found one yet.

Here's an example of the problem I want the flash to solve: If I'm in the fifth row at an indoor musical performance, the flash only illuminates rows 4, 3, and 2, but never the performer themselves. Why on earth is that? It's like the shutter is opening too soon... the autofocus sensor should know the distance to the target in the center of my viewfinder, and thus open the shutter when the light from the flash has had a chance to reach that spot and bounce back to me. Instead, the shutter gets all done with its job before the flash has even had a chance to do any good.

There are other valid reasons to lower the camera's resolution.

If you're not a serious photographer, i.e., your pictures are just happy-snaps for posting on the web, and you're sure you won't need to print them out at high resolution, then you can save a lot of hassle and time by just shooting at the target resolution.

Above and beyond the memory card space savings, shooting at a lower resolution means that you'll waste less time transferring files, waste less time waiting for image processing tools to work on the image (not just resizing, but things like color correction and sharpening too), and your overall storage space on the hard disk of your computer will be less over time.

There's another reason that you might want to shoot at a lower resolution. Some cameras allow the lower resolution to increase your zoom, without having to use "digital zoom" features. My camera does this and it calls the feature "EZ" zoom. It works like this:

Let's say your camera has a lens that will go to a 3X zoom. At full resolution, that's the most you'll get before you have to push into digital zoom (resampling). But if you're shooting at a lower resolution, it can, instead of resampling the image to the lower rez, simply use the pixels in a smaller rectangle in the center of the image sensor instead of the entire image sensor. You'll get a higher zoom level but still at a 1:1 pixel sample. In other words, the lens might be at 3X, but since only the middle 2/3 or so of the image sensor is being used, the photo ends up at something like 4x actual zoom.

It's the same as if you'd taken the high resolution version of the picture and cropped it down in photoshop after the fact. But you get the benefit of being able to zoom and aim at the higher zoom level on your camera's LCD display, and you don't have to do the post-processing.

Yes it can. If the camera just does a simple average of pixels when reducing the image size then it would have the effect of reducing the noise in the image. The camera could go a step further than that and do some more clever processing along the way, though I'm not sure if any of them do.

Of course, any noise reduction that the camera can do by downsampling the image, you can also do better in a photo manipulation program after the fact.

But yeah, if the camera's downsampling does a decent job of noise reduction, and you don't need the high rez, then it could save you the postprocessing step.

Agreed. But I suspect the number of people who do any form of post process is absolutely tiny. And the number of people with compact camera who post process is every smaller.

What I think the next step will be in compact digital cameras:

When you press the shutter button, it's not a single exposure. The camera samples the entire full-rez CCD at a very high frame rate, much like a tiny little movie, then, internally, combines the images in an intelligent way which uses the temporal sampling to remove the grainy noise and also improves the sharpness of objects which are not moving within that frame. For objects which are moving within a frame, it locates the least blurry sample and uses that.

In that kind of a situation, the small amounts of hand shake that you get from a handheld camera would actually improve the image instead of messing it up. It's possible to get sub-pixel resolution and come out with a sharper image if you can intelligently postprocess multiple frames taken from ever-so-slightly different times and positions. It's a technique sometimes used to improve the resolutions of planetary surface photographs from space probes, for example.

Of course this would only work for certain kinds of photos, you'd want to turn off that mode for action shots (or have the camera intelligently decide when to turn off the mode).

Exactly. Which is why the "more megapixels" trend in compact cameras seems silly to me. I'd rather see those extra pixels being put to use improving the image quality rather than making the file sizes unnecessarily larger.

Tony, your camera is just digitally zooming the image and then reducing its size. There's no way for your #x lens to project an image on the sensor any bigger than it normally does.

Andy, reducing the resolution in camera doesn't leave you with a better picture ever. it just leaves you a down sampled picture. Everyone down samples whether they know it or not. When you put your images on flickr, facebook or wherever, they're going to scale and process the image. When you print, the image again gets processed, you don't have to do it manually.

If you're not going to do anything at all but select a lower megapixel image size from your camera's settings, you're only saving space. You will never have anything better than you would by leaving it at its max.

All the noise reduction and other enhancement features are applied to the native image if those options are available and turned on. Again, the only thing you're getting is a very quick bicubic (hopefully) down sample. Most software is going to do this on a PC or server without flinching and with better results. Further, many one-stop enhancements will feature a lot more than a simple bicubic reduction for even better results - that you still won't have to wait around for terribly long. Most photo cataloging software has batch abilities for stuff like this. But we don't need to be concerned with "most people" anyway, since we're making recommendations only to one person here.

And the recommendations were about getting better pictures. Shooting at native resolution and then processing in desktop software is what's going to get you the best pictures. Shooting in RAW even better still if you have that option available (some older Canon P&S do with a firmware hack)

If you're going keep a camera set at a lower resolution, I suggest saving money as well by buying a camera with a less dense sensor. At least then you have physics working for you with larger photosites. Pick up a 5MP or 7MP Canon and save a ton of cash, time and space.

Nikon Coolpix P100

...when combined with Nikon's Best Shot Selector you have a better than average chance of getting a sharp shot of a still subject while holding the camera. BSS is a high-speed shooting setting that takes up to 10 shots while the shutter release is pressed and saves only the sharpest shot.

tanstaafl.

Incorrect. You're misunderstanding the way the system works. Let's see if I can explain it again less confusingly.

I will preface this by saying that this is a feature only available on some brands and models of camera, not all of them. Conceptually, though, it's quite simple once you understand it.

The 3x zoom lens is not projecting a bigger image onto the sensor. That's the point: It's projecting the same 3x size image onto the full width and height of the sensor.

Let's say the sensor is, I dunno, 1000 pixels wide, just to make the math simple.

Then, to grab the image off the sensor, let's say I only use the middle 640 of those pixels.

Instead of taking the whole image off the whole sensor and downsampling it, I'm still using the 1:1 pixels of the sensor as they already exist.

But because I'm only looking at the cropped middle 640 of those pixels, I'm effectively getting a farther zoom than 3x.

This of course only works if you're deliberately using a lower rez than the sensor is capable of. And of course, you'd get the same result if you'd simply shot at the full rez and then cropped the picture after the fact.

Make sense?

Although that's a great feature that I would LOVE TO HAVE (that's essentially what I do, only I have to manually cull out the sharpest shot myself, a painstaking process that I hate), it's not what I'm talking about.

I'm talking about advanced image processing that combines multiple images, not just selecting the best one. But rather, selecting the best features of multiple images and combining them together.

Very true. Nothing that can be done in-camera can match what's possible with desktop post-processing. Assuming you don't mind the hassle of doing it.

Overall, nothing is going to be as good as getting a big, proper camera with a big image sensor and a big lens. No matter how much postprocessing you do, you'll never get a photo out of a happysnap camera that matches it.

I wish very much that there could be a huge leap in CCD technology that allowed us to get DSLR-quality images out of a happysnap.

A number of relatively recent sensors do pixel binning, combining the outputs of multiple sensor elements to make one bigger pixel. You could argue that it's just interpolating those pixels and "downsampling", but it seems to produce better images than what that would imply.

Ok, so it's even more simple. You're talking about an auto-crop. How the FX Nikon cameras can pull DX-sized images from the middle of the FX sensor. It's a reach to call that reach however. Because when displayed at the same pixel scale as the full sensor image, the magnification is the same for either image.