If I manually reduce the resolution at which my camera takes pictures using the menus, does that reduce pixel density and so does it lead less noise?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Lowering Res reduces Noise?
- Thread starter dingdongbubble
- Start date
- Sort by reaction score
Not really. Noise is a function of the sensor. How the camera processes that sensor data can reduce the noise to some degree, but essentially only by blurring the photo. I think that effectively, the picture might look a little less noisy if you use a lower resolution setting, but you'd get better results by shooting in RAW and doing noise reduction with software on your computer--which would take a lot more time, obviously.
Digital Rebel XT
The Canon 350d is known to have almost no noise even at high ISOs. So if I get some other camera that isnt so good at noise, and shoot in RAW and process the image on my computer, is it possible to get similar, if not better, results?
The Canon 350d is known to have almost no noise even at high ISOs. So if I get some other camera that isnt so good at noise, and shoot in RAW and process the image on my computer, is it possible to get similar, if not better, results?
The Canon 350D has very good noise performance; but it sure sounds like you've gotten a sales job from a Canon fanboy.
Most all current generation (or one back) digital SLRs also have very good noise performance - but read some objective reviews to learn about specific cameras.Post-processing noise reduction almost always has some trade-off against picture quality - it may be acceptably small in terms of how much quality is impacted, but it's there.
in camera processing.
So I should just stick to Canon's built in processing right? And what about when I am shooting in RAW? Do I have to do my own processing on my computer?
So I should just stick to Canon's built in processing right? And what about when I am shooting in RAW? Do I have to do my own processing on my computer?
It depends on your camera.
On a Sigma SD-9, SD-10, or SD-14 it indeed does.
For most (if not all) Bayer sensor cameras it does not. So the answer is no - unless you are a lucky Sigma owner
When you are shooting in RAW, you have the option of doing your own processing (as you would with JPEG for that matter). But you will probably find that, for the vast majority of the shots you take, you will not have to do any noise reduction at all - what comes out of the camera is pretty darn good already.
it depends on the setting for the Canon 350d - I use this in work... always shoot in RAW and do the process in CS3 camera RAW 4.0
it's a good camera, but I've seen results on the default settings - pretty terrible, depends on what your shooting
What you're talking about is called binning. Basically it's a post processing "feature" (if you want to call it that) that allows you to reduce noise by grouping a bunch of pixels into one. A 2 by 2 bin means that you take 4 pixels and average them to produce one. While it resuces the visible chroma noise, it produces a horrible image that most would deem worthy of a 19th century impressionist painter. You essentially lose all detail in the image and make a nice smeary mess. The Lumix FZ-50 does this for ISO 3200 images as shown here.
The best thing you can do to ensure that you have less noise in images that are taken with cameras that have noise issues at certain levels is to make sure that they are properly exposed (especially in the shadows), and use slower shutter speed when the light calls for it (this may mean using a tripod!!! )
The best thing you can do to ensure that you have less noise in images that are taken with cameras that have noise issues at certain levels is to make sure that they are properly exposed (especially in the shadows), and use slower shutter speed when the light calls for it (this may mean using a tripod!!!
)In theory yes. consumer camera don't actually "bin" pixels like a scientific camera can. Binning does the addition in analog mode before the DAC so you loose some sampling and quantization noise. These camera can electrically change the size of the pixels on the sensor.
Consumer camera does a full resolution readout of the sensor and then down samples using software. How much better the noise is depends in the exact way the down sampling is done. (Hence the advice to do the down sample in Photoshop rather then the camera) These are several sources of noise in a camera but if you are talking about low light and high ISO the noise or rather the signal to noise ratio should improve by a factor of the square root of the number of pixels that are combined, In the real world I think the improvement is a little less than this do to not-perfect down sample algorithm, quantization and so on. So reducing the size from 1000 x 1000 to 500 x 500 is a 4X reduction and should improve signal to noise ration by almost 2x, If you shoot raw and use a good photoedit software then it can get closer to 2X.
What does this mean visually. I think you need a 2x reduction to be noticable to the eye
no way. noise removal always involves a tradeoff with detail. starting with the lowest noise possible is the best option. besides, that sounds like a lot of time spent to get back to square one
The best way to reduce noise on almost any camera (except a lucky Sigma owner) is to lower the in camera sharpening to its lowest setting, and post process, unless you can shoot RAW.