![]() |
|
|
|||||||
| Register | iSpy | Wiki | gXboxLive | FAQ | Members List | Social Groups | Calendar | Search | Today's Posts | Mark Forums Read |
| Welcome to the FinalGear.com Forums! | |
| Photography Think you have a knack for photography? Post your stuff here. |
![]() |
|
|
LinkBack | Thread Tools | Search this Thread |
|
|
#1 |
|
I decided to see just how much difference ISO makes in terms of quality and to find the highest "useable" ISO (I'm not that big of a geek, this is just better than writing papers). Thought I'd share. All shots are cropped to 100%. Here are the results:
ISO 100 ![]() ISO 200 ![]() ISO 400 ![]() ISO 800 ![]() ISO 1600 ![]() 200 was decent; 400 useable but certainly noisy; 800 and 1600 were terrible and even more terrible, respectively. Someone on this forum (don't remember who, sorry) suggested a program called NoiseNinja once. I've been using it and I've been satisfied with the results so I thought I'd try it for this little comparo and see just how good it is. NoiseNinja'd edits: ISO 100 - pretty much no difference ![]() ISO 200 - definitely better quality ![]() ISO 400 - less noise than original shot at 200 ![]() ISO 800 - big improvement over the original; a bit more noise than original at 200 (NoiseNinja sharpens shots a little too, so the Queso can looks much better in this Ninja'd shot at 800 than it does in the original at 200; however the dark background is clearly noiser in this 800 than the orig 200) ![]() ISO 1600 - again, big improvement over the original; not as good as orig at 400; a little better than orig at 800 ![]() Then I ran the Ninja'd 1600 shot through once more. The noise particles are a lot smaller than in the original shots and detail is lost. ![]() Conclusions? ISO 200 is the highest I'd want to go; anything higher produces a lot of noise. When using NoiseNinja, ISO 800 becomes useable; Ninja'd 1600 shots lose a lot detail, while at 800 details are retained. Thus, up to ISO 200 is good straight out of the camera; up to ISO 800 is good if you use NoiseNinja; ISO 1600 is good only if you absolutely have to use it. Notes: Canon Digital Rebel XT; 18-55mm IS; f/5.6, aperture priority; all shots at 55mm from a tripod positioned about 4 feet away from the Queso can, using a remote; once it focused the first time, I switched it to MF. This was done at about 8:30pm; the room was lit only with an overhead flourescent light; white balance was left on auto.
__________________
^Don't worry he's a closet flickr . . . photography + UMass Motorsport |
|
|
|
|
|
|
#2 | |
|
Have you been high today?
|
Neat. IMO that 400 is fine, and 800 is acceptable.
Quote:
![]() Your exposures aren't all the same, mostly the ISO 100 shot (which is 2/3 stops darker than the ISO 200, par example). You should've used manual (obviously you'd just be halving the shutter speed on every ISO speed increase). If I could make a suggestion, putting the original and NoiseNinja'd images side by side would be more effective than making us scroll up and down. Anywho, you didn't say, so just out of curiousity, what method are you using with Noise Ninja? Are you just doing the auto profile, or are you using the noise profiles for your camera?
__________________
The post is made out of aluminum-magnesium alloys. Last edited by Ramseus; November 5th, 2009 at 2:38 AM. |
|
|
|
|
|
|
#3 |
|
Yeah, I realized that I should have set the exposure manually while uploading the shots. Oh well.
I used NoiseNinja's profiler; then these settings:
__________________
^Don't worry he's a closet flickr . . . photography + UMass Motorsport |
|
|
|
|
|
|
#4 |
|
Have you been high today?
|
Eh, nevermind, I just had a look at some noisy as balls ISO 1600 shots of mine and profiling the image or using a downloaded profile, exact same result more or less. Whatever, it's all good.
__________________
The post is made out of aluminum-magnesium alloys. |
|
|
|
![]() |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
|
||||||
| Thread Tools | Search this Thread |
|
|