Author Topic: Noisetest at ISO100 and 200 (Read 2518 times)

exzrael · « **on:** September 04, 2010, 11:25:08 PM »

Just tested my Olympus PEN E-PL1 with noisefilter on and off at ISO100 and 200, just for fun. And I was amazed how much noise there is in the darker parts - have I missed something here or do you guys experience the same thing? That just can't be right. I know this isn't the best way of testing ISO-noise and stuff but still, I did it my way.

And one more thing - is there a gradation-setting in E-PL1? ~~If so, I just can't find it~~.(found it!)

Here is a link for the images and here is a crop from all four 4:3 testimages.

Brus av = noisefilter off
Brus pÃ¥ = noisefilter on (standard)

Panther · « **Reply #1 on:** September 05, 2010, 05:13:16 AM »

Hmmm...
Looks like there's more noise with the filter on...

voyager · « **Reply #2 on:** September 05, 2010, 11:55:57 AM »

That is because the base ISO of the E-P1 starts at ISO 200. ISO is simply an ISO 200 photo that is digital downgraded, which is why there's so much highlight loss.

mikmas101 · « **Reply #3 on:** September 05, 2010, 12:30:28 PM »

Quote from: exzrael on September 04, 2010, 11:25:08 PM

have I missed something here or do you guys experience the same thing?

Don't know if you are using the 14-42 kit lens for these but if so - yes, it can be damn noisy and at any ISO setting. One of the main reasons I dropped it in favour of MF lenses.

I took the kit lens with me for a family outing this week for the first time in months and I regretted it as soon as I had chance to view the images on my home PC. The majority were soft to very soft and with very noisy shadow areas. Even with strong sunlit subjects the results were disappointing to say the least.

My suspicion is that the in-lens corrective filters have much way too much input

Panther · « **Reply #4 on:** September 06, 2010, 09:43:07 PM »

Hmmm....
I've heard of sharp & soft lenses, fast & slow lenses, but noisy lenses?

lisandra · « **Reply #5 on:** September 06, 2010, 10:30:26 PM »

www.google.com, it's this radical new site!! try it out, it's free!

adash · « **Reply #6 on:** September 06, 2010, 10:32:14 PM »

Try to keep a polite voice, please!

Ian Tindale · « **Reply #7 on:** September 06, 2010, 10:41:54 PM »

Yes, whatâ€™s wrong with bing?

lisandra · « **Reply #8 on:** September 06, 2010, 10:54:02 PM »

Quote from: adash on September 06, 2010, 10:32:14 PM

Try to keep a polite voice, please!

sorry...um um.... must be medication time...

mink70 · « **Reply #9 on:** September 06, 2010, 11:18:22 PM »

The main reason I've refrained from buying an E-PL1 is the way the camera handles noise. Compared to the E-P1/2, there appears to be more aggressive noise reduction applied to the jpegs in-camera, and as a result the noise looks more digital and processed, and to me, more objectionable. There's a lot of talk about the E-PL1 having the best jpeg engine among the PENs, but not to me. Anyone else have the same experience?

lisandra · « **Reply #10 on:** September 07, 2010, 09:50:45 AM »

Quote from: mink70 on September 06, 2010, 11:18:22 PM

Anyone else have the same experience?

nope. I'm gonna go ahead and predict you're alone in that camp

mikmas101 · « **Reply #11 on:** September 07, 2010, 11:29:58 AM »

- take away all the nice algorithms and such that your onboard gizmos are applying to the image and you will find that the plastic barrel you have stuck on the front of your panny, oly or whatever is a pile of crap

barrel distortion, chromatic aberration et al...

Unfortunate side effect from all this correction is that each stage can introduce is own little evil demons - noise, artefacts, micro distortions, loss of detail....the list goes on.

So yes - the more compromised a lens is, the noisier it can be - and the Oly kit lens is NOISY....

mink70 · « **Reply #12 on:** September 07, 2010, 11:34:47 AM »

Quote from: lisandra on September 07, 2010, 09:50:45 AM

Quote from: mink70 on September 06, 2010, 11:18:22 PM
Anyone else have the same experience?
nope. I'm gonna go ahead and predict you're alone in that camp

Well, thanks for the sentiment, but perhaps the first message in this post suggests otherwise?

mikmas101 · « **Reply #13 on:** September 07, 2010, 12:10:56 PM »

http://www.dpreview.com/articles/distortion/

<<Kazuto Yamaki, Chief Operating Officer for lens maker Sigma explains how this can work in more detail: 'compromise in distortion and chromatic aberration is a very effective in making lenses more compact and reducing cost. In general, compromising the distortion at the wide-angle end affects the compactness of the lens and compromising the CA influences cost reduction.' It's not without its drawbacks, he says: 'it's also true that doing this deteriorates the image quality if we correct them with image processing. It may not be very obvious if you look at the images in small format, but it becomes recognizable when enlarged. In the case of users who are very, very insistent on top image quality, they may not be 100% satisfied with such images.' As a result, no Sigma lenses currently require software correction.>>

voyager · « **Reply #14 on:** September 07, 2010, 12:50:20 PM »

This thread is making me uncomfortable. Please don't give me headaches. Thanks.

Panther · « **Reply #15 on:** September 07, 2010, 03:59:23 PM »

I don't think CA is the same as noise....

lisandra · « **Reply #16 on:** September 07, 2010, 06:04:16 PM »

CA is not noise, and correcting for it doesn't bring it on either. Barrel distortion is not noise either, and correcting for it also doesn't create noise, a smudge detail or loss of it maybe, but no, noise won't come from it. So in the end regardless of how crappy a lens is noise can't be born from it. Besides, what doesn't make sense in all of this is that two lenses on the same camera give different noise levels, JPEG output is the same for all lenses given the same camera.

lisandra · « **Reply #17 on:** September 07, 2010, 06:07:26 PM »

And...I apologize if I insulted anyone on this thread, it's not my intention, really, so I'm sorry. I kind of have a short fuse when it comes to wrong information being shown as facts, that's all. Again, I'm sorry

lisandra · « **Reply #18 on:** September 07, 2010, 06:11:55 PM »

Quote from: mink70 on September 07, 2010, 11:34:47 AM

Quote from: lisandra on September 07, 2010, 09:50:45 AM
Quote from: mink70 on September 06, 2010, 11:18:22 PM
Anyone else have the same experience?
nope. I'm gonna go ahead and predict you're alone in that camp

Well, thanks for the sentiment, but perhaps the first message in this post suggests otherwise?

I say that because I think the EPL1 handles the noise way better than the other PENs, and all reviews seem to agree with that. In terms of detail and quality, I imagine noise reduction has to take its toll on something, so on that part well you're right

mink70 · « **Reply #19 on:** September 07, 2010, 09:33:26 PM »

Quote from: lisandra on September 07, 2010, 06:11:55 PM

Quote from: mink70 on September 07, 2010, 11:34:47 AM
Quote from: lisandra on September 07, 2010, 09:50:45 AM
Quote from: mink70 on September 06, 2010, 11:18:22 PM

I say that because I think the EPL1 handles the noise way better than the other PENs, and all reviews seem to agree with that. In terms of detail and quality, I imagine noise reduction has to take its toll on something, so on that part well you're right

My point isn't that the EP-L1 has more or less noiseâ€”the sensor is the sameâ€”just that the engineers at Olympus decided to suppress it more aggressively. This makes the noise appear lower in some instances, especially at lower ISOs, but also makes the shadows look blotchier and murkier. To my eye, they E-PL1 photos also have a slightly more "plastic" look, a la Canon.

The reason I like using my E-P1 so much is not its resolution or freedom from noiseâ€”in these respects it can't touch the Nikon 300D I sold to get itâ€”but how esthetically pleasing its photos can look. The way it handles noise reminds me of shooting Tri-X. (And of course I love the colors.) I was ready to get the lighter, cheaper E-PL1, but the images look more conventionally digital, and therefore disappointing, to me. They just don't have the analog look of the E-P1/2, which reminds of the E-1 I had for years. Cameras are rather personal, aren't they?

mikmas101 · « **Reply #20 on:** September 08, 2010, 03:07:52 PM »

Quote from: lisandra on September 07, 2010, 06:04:16 PM

CA is not noise, and correcting for it doesn't bring it on either. Barrel distortion is not noise either, and correcting for it also doesn't create noise, a smudge detail or loss of it maybe, but no, noise won't come from it.

And that is exactly where I am disagreeing with you.

Correction of the magnitude required to cope with the sometimes extreme barrel distortion noted by the dp review (and others) of the Oly kit lens is unlikely to come free. To correct this requires a very significant intrusion into the image data required to build the whole map, not just the edges. In certain cases this intrusion (even when coupled with noise reducing algorithms) can amplify existing image noise - shadows being particularly vulnerable to this.

But if I'm completely wrong, how else would you account for the acute lack of noise I get when using a wide range of manual lenses - which will not be corrected - on the same camera body with the same sensor and the rubbish that litters the shadows when using the 14-42? (and I have plenty of samples) Rubbish that is less or absent when using the Panny kit lens for instance.

You were completely correct at one point - lens elements do not produce noise - so what is?

Panther · « **Reply #21 on:** September 08, 2010, 04:06:19 PM »

I love it when these degrade into a "who's right" contest.....

mikmas101 · « **Reply #22 on:** September 08, 2010, 04:41:10 PM »

Panther - without debate there is no knowledge.

I would rather constructively disagree than sullenly acquiesce......

Panther · « **Reply #23 on:** September 08, 2010, 04:50:26 PM »

I debate at work all the time without it degrading to a "who's right/wrong" situation.....

adash · « **Reply #24 on:** September 08, 2010, 10:13:14 PM »

Every lens has vigneting besides distortion. Coping with vigneting requires brightening peripherial areas, which may bring noise even at low ISO. This should not affect the center of the image however. That's what I can speculate from what I already know.

Author Topic: Noisetest at ISO100 and 200 (Read 2518 times)

Panther

Panther

Ian Tindale

Panther

Panther

Panther

Related Topics

Recent Topics

Recent Gallery