[loco]

I just got a DA 35mm/f2.4 lens and in using it, I've noticed that the noise level in my shots at higher ISO is either much less or more palatable than in other lenses I've used. By more palatable, I guess I mean it's a bit easier to address in Lightroom.

I have the kit 18-55, the DA L 55-300, and the A 50 1.7. The K-r handles noise admirably anyway, but I just feel like it's even better with the DA 35 attached.

I'm a newb, and I assumed that noise was the camera body's responsibility. Does focal length play into it? I'm much more likely to let this lens shoot at higher ISO than my others, which makes it more enjoyable and more versatile. Even though the DA35/2.4 is slower than the A50/1.7 (which I also love!), my comfort with higher ISO makes it seem faster, if that makes sense.

I admit I have done no testing to confirm this. This is just my quick observation after owning the lens for several days and taking probably 250 shots.

I'm very pleased by this, but just curious why I'd see a noticeable difference with this particular lens. Thanks for any thoughts you might have on this.

[newarts]

Physics says the lens should have no effect on noise levels at a particular ISO.

Noise depends on how many light photons are received by the sensor and what the camera does to them - it doesn't matter how they got there.

Say you take a photo with your lens set at 2.8 & there is a certain amount of noise; now keep the iso the same, increase the f-stop to 5.6 and re-take the photo - the noise will be very close to the same. The exposure will take 4 times as long because you are letting only 1/4 as much light into the camera - but the total amount of light collected will be the same for the two images so the noise will be very similar.

I had to stick in some hedge words because there may be small secondary causes of noise depending on how long the shutter is open ....for example the longer the shutter is open the warmer the sensor gets which in principle can add a tiny bit of noise.

[DaveHolmes]

Even though the DA35/2.4 is slower than the A50/1.7 (which I also love!), my comfort with higher ISO makes it seem faster, if that makes sense.

That makes sense to me... I just sold the M50/1.7 as I get much better results from my Tamron 17-50/2.8 at higher ISO's... That said, I recently purchased the FA50/1.4 and its hardly left my camera and performs in a totally different way to the M50/1.7...

DA 35 Limited macro and DA 40 limited are next on my 'want' list...

[loco]

Thanks for the replies. I appreciate the explanation, newarts. I may have to look at this more scientifically, but I really feel there is a difference with this lens. Maybe it's placebo effect.

However, it's nice to know that Dave saw a similar effect with different lenses. At least I'm not alone!

[newarts]

Thanks for the replies. I appreciate the explanation, newarts. I may have to look at this more scientifically, but I really feel there is a difference with this lens. Maybe it's placebo effect.

However, it's nice to know that Dave saw a similar effect with different lenses. At least I'm not alone!

In situations like this controlled experiments are the thing to do.

I do not doubt your observations and impressions; but they may be only correlated with the lens not caused by it... the scenes you've chosen (light backgrounds show far less noise than dark) or your self reinforcement of an observation can have big effects on such things.

It would be great to hear the results of any testing you do! Especially if this lens consistently shows less noise.

[loco]

Thanks newarts, I'll certainly share if I find anything of interest.

[JohnBee]

I'm very pleased by this, but just curious why I'd see a noticeable difference with this particular lens. Thanks for any thoughts you might have on this.

I think its very simple really... Since noise and detail go hand in hand and both are present in an image, we are ultimately drawn to detail.

Therefore, a sharper lens(resolving capacity) the more detail it will render, and the detail to noise threshold increases respectively.

Which in turn yields healthier images.

Hope this helps.
JohnB

[Ron Kruger]

It took me a couple of decades to learn the importance and effects of different lenses. Everything is better with better lenses--period--and I don't doubt you have less noise (and more detail) with your 35mm.

[Digitalis]

one other thing to keep in mind is that some lenses can be slightly faster or slower than their rated apertures,E.G the SMCP-K 55mm f/1.8 is actually F/1.78 - slightly faster than it's rating, and this will have an influence on overall exposure levels. The bottom line is that digital doesn't perform well with underexposure but slight overexposure helps keep the noise from swamping detail in images.

[newarts]

I think its very simple really... Since noise and detail go hand in hand and both are present in an image, we are ultimately drawn to detail.

Therefore, a sharper lens(resolving capacity) the more detail it will render, and the detail to noise threshold increases respectively.

Which in turn yields healthier images.

Hope this helps.
JohnB

It took me a couple of decades to learn the importance and effects of different lenses. Everything is better with better lenses--period--and I don't doubt you have less noise (and more detail) with your 35mm.

How to reconcile these observations with physics? I believe that both the physics prediction that the lens doesn't matter regarding noise and that observations that the lens does matter are both true.

Because the overall effect of noise on our perception of an image has to do with both the quantity of noise and the characteristics of the image.

I believe JohnBee comes closest when he says "noise and detail go hand in hand"; but I don't know how to quantify that.

For example, the same quantitative amount of noise on a hazy, fuzzy image will likely have a different visual impact than on a crisp, high contrast image.

[Marc Sabatella]

I believe JohnBee comes closest when he says "noise and detail go hand in hand"; but I don't know how to quantify that.

The technical term in other fields would be "signal-to-noise ratio", but that's still not exactly quantifiable when it comes to image detail versus noise. Or rather, while there might be a way to do so, there is no commonly used unit of measurement.

I agree this is likely what's going on. Depending on the nature of the scene and the detail with which it is represented, our subjective impression of noise does indeed vary, a lot. But do a controlled test - same scene, focal length, same distance, same exposure, same PP - and you have a much better sense of what is going on.

[loco]

Interesting observations, everyone. Thanks for the feedback. I don't think I can do it today, probably not until the weekend, but I do want to try testing this out. It may just be all perception, but now I'm sufficiently curious.

The only way I can get the same focal length is to test the 18-55 kit lens and the DA 35/2.4. I can also test the 55-300 and A 50/1.7 with all other parameters equal. If I get a chance to do this, I'll report back.

[reeftool]

I have noticed also that my high ISO shots look better with some lenses. Noise can be an issue with the K10D. Getting exposure right on is important with higher ISO shots and some lenses meter better than others. The tech guys can probably explain this but I can get noticeable differences in exposure with the same scene and a different lens even though the settings are the same. My K10D tends to underexpose slightly and any high ISO scene that you to have to add + to the exposure in PP will bring out the noise like crazy.

[Designosophy]

Loco, I have noticed the same thing (though I don't have the particular lens you are talking about), and I think that JonhBee & newarts have hit on something. I'm going to take a guess at causes as well.

It seems to me that shadows tend to have more noise than highlights. Of course, noise also is amplified whenever exposure is pushed in post-processing. Furthermore, it seems that better lenses have better distribution of light and shadow — when exposed properly, resulting photos need less adjustment in post-processing. So, if a lens produces more accurate/pleasing distribution of lights and shadows, and noise occurs in shadows and areas of the image that are exposure-pushed, then you may see more noise when using a lens of lower quality, even at the same aperture and ISO.

[newarts]

Loco, I have noticed the same thing (though I don't have the particular lens you are talking about), and I think that JonhBee & newarts have hit on something. I'm going to take a guess at causes as well.

It seems to me that shadows tend to have more noise than highlights. Of course, noise also is amplified whenever exposure is pushed in post-processing. Furthermore, it seems that better lenses have better distribution of light and shadow — when exposed properly, resulting photos need less adjustment in post-processing. So, if a lens produces more accurate/pleasing distribution of lights and shadows, and noise occurs in shadows and areas of the image that are exposure-pushed, then you may see more noise when using a lens of lower quality, even at the same aperture and ISO.

All true I think.

It would be helpful if people would post examples of the lens effect on perceived noise/image quality.

The underlying problem is that a lens does not create noise nor does it affect noise in the light that passes through it.

So the effect of a lens on image quality from a noise standpoint must be related to some other aspect of the image created by the lens. A low contrast image with noise likey looks different than a high contrast version of the same image with the same amount of noise.

Other defects like chromatic abberation, while not noise, might be being lumped in with noise.

It is an interesting topic & what I've read so far about the effects of noise on human visual perception implies lens contrast in certain resolution ranges is important.

While one might expect that noise on an image with no contrast at all (a flat gray image) wouldn't be objectionable, it sounds like the reverse might be true; maybe we forgive noise if there is detail to see.