ISO & Quality
- Thread starter z101
- Start date
Barks
Suspended / Banned
- Messages
- 326
- Name
- Paul
- Edit My Images
- Yes
There are lots of things that have an impact on noise, not just the pixel count. If it's 18mp on a full frame and 12mp on a copped sensor (aps-c etc) then there's a chance that the 18mp full frame would be less noisy. But this also takes in to account things like each camera's image processing chip and how it handles noise reduction.
But generally speaking on the same sized sensor then the 12mp will be less noisy than the 18mp. Again generally speaking the more pixels/photosites crammed on to the sensor the noisier the images get.
But generally speaking on the same sized sensor then the 12mp will be less noisy than the 18mp. Again generally speaking the more pixels/photosites crammed on to the sensor the noisier the images get.
- Messages
- 43,206
- Name
- Alan
- Edit My Images
- No
This is probably one of the most debated and argued about topics 
Some think that more mega pixies makes the sensor noisier and some think that the noise will be less noticeable because the pixies are smaller...
Technology marches on and you'd hope that each new sensor would be better at controlling noise than the sensor it replaces but each new sensor seems to have a higher mp count than the one it replaces so it's pretty much impossible to get two cameras with two sensors of different mp counts that use the same sensor technology so that you can do a fair comparison.
Personally I look at shots taken with Canon 18mp cameras and feel less than impressed, especially when compared to those from Sony sensors but is that because of lower mp counts or just better Sony technology? Who knows...
Some think that more mega pixies makes the sensor noisier and some think that the noise will be less noticeable because the pixies are smaller...
Technology marches on and you'd hope that each new sensor would be better at controlling noise than the sensor it replaces but each new sensor seems to have a higher mp count than the one it replaces so it's pretty much impossible to get two cameras with two sensors of different mp counts that use the same sensor technology so that you can do a fair comparison.
Personally I look at shots taken with Canon 18mp cameras and feel less than impressed, especially when compared to those from Sony sensors but is that because of lower mp counts or just better Sony technology? Who knows...
450D or 400D vs 60D or 50D
Hmm.. I would like to see the same image from each.
Thanks to everyone that replied so far.
Barks
Suspended / Banned
- Messages
- 326
- Name
- Paul
- Edit My Images
- Yes
Technology marches on and you'd hope that each new sensor would be better at controlling noise than the sensor it replaces but each new sensor seems to have a higher mp count than the one it replaces so it's pretty much impossible to get two cameras with two sensors of different mp counts that use the same sensor technology so that you can do a fair comparison.
That about sums things up nicely.
Last edited:
Barks
Suspended / Banned
- Messages
- 326
- Name
- Paul
- Edit My Images
- Yes
The July 2011 edition of Digital Camera magazine had a "shoot out" with Canon v Nikon cameras in the various levels of DSLR's (entry, intermediate etc) and they do a good breakdown of things like this, even though it's canon v nikon they're comparing you can at least compare the Canon's or Nikon's of various levels against each other if you want to.
I think the new edition of that magazine is out now but i think you can back issues from their website if you're interested.
HoppyUK
Suspended / Banned
- Messages
- 23,200
- Name
- Richard
- Edit My Images
- No
The biggest factor affecting noise is format/size, eg full frame vs crop format. A physically larger sensor area collects more light, pretty much regardless of how it is divided up in terms of pixels, so the signal to noise ratio is higher, though pixel density is a factor.
The main difference between those cameras, which are all the same size, is sensor development technology and newer processing engines. There are several generations between a 400D and 60D.
petersmart
Suspended / Banned
- Messages
- 5,000
- Edit My Images
- Yes
It also depends on the final picture size.
I can take a pic at 3200 ISO on my camera - Canon 1DsII - and reduce it to say 1024 px on the longest side and noise will be pretty unnoticeable.
But pixel peep at 100% and the noise will be obvious.
It also depends on the exposure - under expose and noise again becomes more obvious - so the ETTR technique also helps reduce noise in a pic.
And you can of course use NR programs such as Neat Image and use higher ISO levels with reduced noise.
.
I can take a pic at 3200 ISO on my camera - Canon 1DsII - and reduce it to say 1024 px on the longest side and noise will be pretty unnoticeable.
But pixel peep at 100% and the noise will be obvious.
It also depends on the exposure - under expose and noise again becomes more obvious - so the ETTR technique also helps reduce noise in a pic.
And you can of course use NR programs such as Neat Image and use higher ISO levels with reduced noise.
.
- Messages
- 43,206
- Name
- Alan
- Edit My Images
- No
Yet again we disagree
The format size in itself does not decide the number of photons falling on a particular pixel, the design of the chip, the processing or the resultant signal to noise ratio.
If format size in itself is the deciding factor I wonder why a FF sensor will have more noise in one area of the shot than in another, simple, it's to do with the number of photons and signal to noise.
scottthehat
Suspended / Banned
- Messages
- 6,290
- Name
- scott
- Edit My Images
- Yes
this might help, its an image comparison site its not perfect but has alot of cameras, click on all at the top.
http://www.imaging-resource.com/IMCOMP/COMPS01.HTM
http://www.imaging-resource.com/IMCOMP/COMPS01.HTM
Barks
Suspended / Banned
- Messages
- 326
- Name
- Paul
- Edit My Images
- Yes
It's my understanding that it's more to do with photosite (pixel) density on a sensor than the sensor size itself. If you have a 35mm (FF) and a 22mm (APS-C) sensor both with say 15mp then the photosites on the 35mm sensor will be slightly bigger and therefore have more light hitting them, allowing them to get a more accurate reading and will give you a better signal to noise ratio.
But as we've already discussed here image noise has so much more to it than simply this one aspect.
But as we've already discussed here image noise has so much more to it than simply this one aspect.
- Messages
- 43,206
- Name
- Alan
- Edit My Images
- No
Most people don't realise the effect that a connection, a length of track or a length of conductor or even proximity to other devices and signals has on the signal you want to work with.
This is why it's probably impossible for end users and even labs to do a fair comparison as the goal posts are continually moving and you'll probably never see two sensors that can be truly and fairly compared so that you can answer such a basic question like will more pixels mean more noticeable noise in my final image.
This is why it's probably impossible for end users and even labs to do a fair comparison as the goal posts are continually moving and you'll probably never see two sensors that can be truly and fairly compared so that you can answer such a basic question like will more pixels mean more noticeable noise in my final image.
HoppyUK
Suspended / Banned
- Messages
- 23,200
- Name
- Richard
- Edit My Images
- No
Yet again we disagree
The format size in itself does not decide the number of photons falling on a particular pixel, the design of the chip, the processing or the resultant signal to noise ratio.
If format size in itself is the deciding factor I wonder why a FF sensor will have more noise in one area of the shot than in another, simple, it's to do with the number of photons and signal to noise.
I disagree that we disagree
squishy
Suspended / Banned
- Messages
- 1,902
- Name
- Adam
- Edit My Images
- No
It's somewhat limited though, all the comparison images are jpgs using default in camera settings.
So different and unknown levels of sharpening and noise reduction means that the images are pretty useless for side by side comparisons.
Johnnybugs
Suspended / Banned
- Messages
- 53
- Name
- John
- Edit My Images
- Yes
does anyone actually shoot at 1600 ?
All the photos I have seen at 1600 were shot at that iso to give a grainy noise to the composition ....... surely then the best camera is the one with the worst noise ratio
All the photos I have seen at 1600 were shot at that iso to give a grainy noise to the composition ....... surely then the best camera is the one with the worst noise ratio
Nawty
Suspended / Banned
- Messages
- 6,473
- Name
- Ned
- Edit My Images
- Yes
You are missing the point - there is SOOOOOOOOO much more to it than just ISO setting and that's why comparisons on anything other than RAW's shot of the same subject, in the same lighting with the same settings (shutter speed, aperture, ISO) using as similar as possible lens and then treated with the same processing (i.e. not in camera) is the only real test.
Even then real world useage will vary as a lot depends on your technique and how you expose the subject.
specialman
Suspended / Banned
- Messages
- 8,193
- Name
- Pat MacInnes
- Edit My Images
- Yes
Looks like Splog is having a digital meltdown......
specialman
Suspended / Banned
- Messages
- 8,193
- Name
- Pat MacInnes
- Edit My Images
- Yes
does anyone actually shoot at 1600 ?
All the photos I have seen at 1600 were shot at that iso to give a grainy noise to the composition ....... surely then the best camera is the one with the worst noise ratio
I sometimes shoot at 1600 and often shoot at 800/1000 but I can't remember the last time i used those kinds of ISOs for any aesthetic reason. They come into play when I need to get the shot and regular film speeds (sub-ISO 400) just can't cut it.
Splog
Suspended / Banned
- Messages
- 6,257
- Name
- Steve
- Edit My Images
- No
Looks like Splog is having a digital meltdown......
Oh b****r! ........ My phone likes to contribute too, even when I'm not using it
petersmart
Suspended / Banned
- Messages
- 5,000
- Edit My Images
- Yes
does anyone actually shoot at 1600 ?
All the photos I have seen at 1600 were shot at that iso to give a grainy noise to the composition ....... surely then the best camera is the one with the worst noise ratio
These shots were all taken using my 350D at 1600 ISO and then processed in Neat Image using my own profiles.
This one was taken on the 350D and underexposed by 2 stops to give the equivalent of 6400 ISO:
Obviously like with like would have to be compared. I never said the opposite?? Thats why I asked if someone had both cameras so conditions could be the same.
does anyone actually shoot at 1600 ?
All the photos I have seen at 1600 were shot at that iso to give a grainy noise to the composition ....... surely then the best camera is the one with the worst noise ratio
I am sorry but you will have to explain
Thanks
Those photos are great considering they were taken at ISO 1600. Just one thing though underexposing photos make them darker and increasing the ISO makes photos brighter so how do you make this to be ISO 6400. I don't know much about image sensors and quality etc. Thanks
Last edited:
petersmart
Suspended / Banned
- Messages
- 5,000
- Edit My Images
- Yes
If you set your camera to manual and set the exposure to give a correctly exposed photograph then close down the aperture by 2 stops and take another pic it will be dark.
Now set the ISO to four times the setting it was and the picture will again look normal.
This is because the ISO setting is like a volume contol and turning it up will, in effect, increase the sensitivity of the sensor to give a normal brightness picture.
The pics here were taken with the camera set at 1600 ISO (the 350D doesn't go any higher) and the exposure was decreased by 2 stops.
The originals were very dark but were then edited to give a normal looking pic.
So they were, in effect, taken at 6400 ISO with the ISO settings turned up.
.
Yes that is all in a nutshell thank you. I wonder is the 60D similar. Surly.