Jump to content
lct

Clean CL at 3200 iso?

Recommended Posts

I find the focus on ISO3200 grain and lifting shadows 100% fascinating. Most probably the grain in above ISO3200 crop won't even show up in a print. As for lifting shadows, I find myself resisting the temptation many times too. Then I remember film days. Blacks in the picture used to be black. Do we need to see more details on the dark suit in the following photograph!

http://public.media.smithsonianmag.com/legacy_blog/around-the-mall-yousuf-karsh-churchill-cropped-big.jpg

Share this post


Link to post
Share on other sites

α7R III with the 50 Apo Summicron-M and CL with the 35 Summilux-TL

 

Only WB adjusted a bit to try to match, Vignetting corrected on the α7R III, Exposure +0.10 on the CL, +30 NR, and LR default sharpening for both.

 

Uncompressed JPEGs here: https://www.smugmug.com/gallery/n-qDx8fd/

 

α7R III

ΙΣΟ 5000, f/2.0 @1/3200 sec.

 

 

CL

ΙΣΟ 3200, f/1.4 @1/3200 sec.

Share this post


Link to post
Share on other sites

Starting to look hard for that ISO 3200 CL noise now.

 

Uncompressed JPEGs here: https://www.smugmug.com/gallery/n-qDx8fd/

 

α7R III

ΙΣΟ 5000, f/2.0 @1/3200 sec.

 

 

CL

ΙΣΟ 3200, f/1.4 @1/3200 sec.

Edited by Chaemono

Share this post


Link to post
Share on other sites

With JPEG being a lossy compressed format, what is an "uncompressed JPEG"?

 

 Yes, it could be construed as misleading. The statement has to be seen in the context of the post where it's found. Below the statement follow the LUF compressed JPEGs. The link simply leads to less compressed JPEGs. Maybe I'll say it more accurately this way in the future. Thanks.

Share this post


Link to post
Share on other sites

These comparisons are interesting ....... but there is a world of difference between high ISO pictures in good daylight and those at night. 

 

Poor, diffuse illumination and lack of contrast show up the issues more readily and reflect the sort of situations where iso 3200 would really be used.

 

Whilst usable, I generally cannot tolerate the muddiness and lack of detail that intrudes even at quite modestly raised ISO levels. Ok for posting at forum resolution or small prints but not much else. 

 

I'd never go above 800 ...... unless all the other alternatives allowing lower than this have been exhausted..... as my results are invariably flat and lifeless. 

Edited by thighslapper

Share this post


Link to post
Share on other sites

Never going above 800 iso is simply not possible indoor in available light for me. It's 1600 or 3200 isos at least and the shutter speed can rarely be higher than 1/50s then. Even at f/2 or f/1.4, shooting indoor at moving subjects can hardly be achieved without disturbing noise with my digital Leicas so far. 

Edited by lct

Share this post


Link to post
Share on other sites

Not to mention that in many night scenes the DR is quite high (due to bright lights and deep shadows). Trying to recover shadows or protect highlight in PP is very much limited with high ISO (since DR cuts down drastically with rising ISO). This is just fact of life and is not going to change much with sensor technology (there are not many photons). Only solution is to add more photons (long exposure at low ISO) or add additional light (flash etc.).

Edited by jmahto

Share this post


Link to post
Share on other sites

Thanks for the comparisons. But many images look underexposed from the beginning which is bad for noise.

Overll on my display I am surprised how good the CL holds against the Sony.

Even though I believe there isway to much discussion about x000 ISO comparisons and too less about color, tones at the main important ISO range (IMO ) of 100-800.

However, my impression is the Cl does quite well at higher ISO

Share this post


Link to post
Share on other sites

It has been raining, cold and grey here all day, so outdoor photographic opportunities were limited.  Trying to explore two of the CL zooms indoors this evening, I am working mostly in the 1600-3200 ISO range.

 

C1000345 by scott kirkpatrick, on Flickr 11-23@18mm,

 

and here's a grab shot at 11 mm:

 

C1000338 by scott kirkpatrick, on Flickr

Share this post


Link to post
Share on other sites

Thanks for the comparisons. But many images look underexposed from the beginning which is bad for noise.

Yes, yes, often intentionally done so to lift the Shadows by +100 and then look for noise. AND the processed CL JPEGs and their crops here are from upsized 42.5 MPx files to even the playing field to the Sony also in terms of noise. Obviously, this works against the CL. I have many more but haven't had time to post them, yet.

Share this post


Link to post
Share on other sites

I disagree with the "getting to the point where sensors won't be able to improve anymore" assertion.

There are many sensor improvements I'm waiting for to become commercially available:

 

curved sensors (already patented by Sony, Microsoft, Canon, etc)

native ISO <100

16-bit ADCs

non-Bayer filter image sensors (Foveon, etc)

 

There's probably more if I spent more time thinking about it.

 

Toshi

 

 

I think you are taking my statement out of context a bit.  I was specifically referring to quantum efficiency levels, read noise, and thermal noise.  Those are near the theoretical limits.  That doesn't mean there will be no improvements of any kind.  Even so...

  • I'm not sure how much of a benefit we will get from curved sensors, as even a curved sensor would need a mating lens.  Are all lenses going to be designed for a particular radius of curvature?  While there is potential here, I'm worried about the long term practicality.  If one manufacturer adopts a standard radius of curvature, that will make certain lenses easier to make and more compact, but certainly not all of them.  I suppose we will see.
  • Native ISO < 100--How?  I certainly wouldn't want this at the cost of lower QE.  So how would one get a lower native ISO?  Larger full well capacity?  I suppose that would be fine.  I don't know how close we are to what physics allows for full well capacity for a given pixel size--whether that can improve significantly or not.  It's certainly not an area where we have seen significant improvements over the last few generations.  Maybe this is a good candidate?  Not sure.
  • 16-bit ADC's--I understand this being attractive for larger chips with bigger pixels and high full well capacities, but I'm not sure how it would help for your typical 4/3" or APS-C chip whose full well capacity is in the 10,000 e- range.  As megapixel counts continue to climb (and I certainly don't see them reversing), the benefits to a higher bit depth on a smaller form factor chip are vanishingly small.  What good is a 16-bit converter on a chip with a 10,000 e- full well?  I think the 16-bit ADC's will be reserved for medium format or at least full frame chips.  It just doesn't have any benefit with smaller pixels.
  • Foveon style sensors--five years ago I'd have been in total agreement, but as megapixels have continued to climb, most of my reasons for wanting a Foveon type sensor have gone away.  Anti-aliasing filters have already been removed from many high-megapixel cameras, and aside from the occasional picture with moire I feel like the resolution is all I could ask for.  Literally.  I wouldn't want more resolution without a larger chip.  Even with a 24 megapixel chip on an APS-C camera, we are at the point that the resolution is limited by diffraction at f/8.  On a 4/3" camera it's limited at f/5.6.  The interpolation I always hated on Bayer chips has basically gone away by just pumping up the megapixel count, so I no longer need, long for, or even want a Foveon type sensor.

Yes, I'm sure there will be improvements that you didn't happen to mention.  I don't mean to imply that human creativity stops when certain milestones are reached.  And while I don't see the particular improvements you pointed out as being particularly important or likely, I'm sure there will be lots of other improvements we haven't even thought of.  My main point was just that we are reaching a "wall" in terms of QE and noise.  We are used to each generation of chip showing obvious and dramatic improvements in terms of high ISO performance.  I think those days are past.

Share this post


Link to post
Share on other sites

> I think you are taking my statement out of context a bit.  I was specifically referring to quantum efficiency

> levels, read noise, and thermal noise.  Those are near the theoretical limits.  That doesn't mean there will

> be no improvements of any kind.  Even so..

 

OK, in that context maybe.

 

> Native ISO < 100--How?  I certainly wouldn't want this at the cost of lower QE.  So how would one get a

> lower native ISO?  Larger full well capacity?  I suppose that would be fine.  I don't know how close we are

> to what physics allows for full well capacity for a given pixel size--whether that can improve significantly

> or not.  It's certainly not an area where we have seen significant improvements over the last few

> generations.  Maybe this is a good candidate?  Not sure.

 

The newer Nikon full-frame sensors have a native ISO of 64.

Check out the Nikon D810 and D850.

 

Toshi

  •  
Edited by TMorita

Share this post


Link to post
Share on other sites

With quantum efficiency already high and the MP count saturating, the questions about lower ISO and more bits our of the ADC are linked, not separate.  It would be nice if the competition between smaller pixels making up a 24 MPx APS-C chip and the larger pixels in full frame chips was addressed by making the APS-C wells deeper, so that they could hold enough electrons to justify outputting a 14 or 16 bit signal.  But that would require operating at lower ISO.  You have to wait longer (or open the lens to a wider f-stop) to get those extra photons in to the smaller cells.  It's a tradeoff that we manage across the huge range of ISOs now offered.

Share this post


Link to post
Share on other sites

10,000 ISO  Daylight

 

I'm pleased with the quality of this image at this ISO.

 

/applications/core/interface/imageproxy/imageproxy.php?img=http://derekburgess.com/leica_cl/L1000184.jpg&key=512ad7966dd0d41e47a325b7cd5ed642678e1d14eb230769676ff1443c5e4106">

 

/applications/core/interface/imageproxy/imageproxy.php?img=http://derekburgess.com/leica_cl/100_crop.JPG&key=21e602ab3cab649a446e7d902b4bd36e17752ef6129a9fb7c2ba55e8580e4441">

Share this post


Link to post
Share on other sites

So far, and it’s early in my use of this camera, if you have lots of light, high ISO is good. If it’s dark it is not so good. Although there are strategies for dealing with this problem.

Share this post


Link to post
Share on other sites

Even after sunset, with huge DOF (f16) and shutterspeed to freeze walking people (1/125) you need only ISO 3200. What are we shooting with ISO 10,000 in day light or good light?

Share this post


Link to post
Share on other sites

The camera was set to Auto ISO.  I didn't take notice until after the fact, but I was happy with the results.

 

Normally as other poster's have said, if I am walking around in daylight I try not to exceed 800.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...