Camera industry in the dumpster - article

[jjj]

Signal levels and noise levels are topics of interest for audio and image people alike. "why are my ISO50000 shots so noisy??

As I have already mentioned I and indeed many others like grainy images. However try and find anyone who like hissy recordings.

"Musicians have probably used "lowfi" recording as a stylistic thing for nearly as many decades as they have been able to record at all. The most prominent example is perhaps overdriven tube amplifiers and crappy 12" speakers used as sound shapers for electric guitars. A more recent thing is the popularity of early 8/12-bit samplers and drum machines, vinyl record simulation, etc.

The question is not if "poor quality" can be used for artistic purpose (of course it can).

You are confusing things here. You were talking about recording quality, not the sound being produced that is then recorded. Making lo-fi music is quite different from a bad recording of a lo-fi styled output. I have plenty of music that is distorted, has feedback etc, but the reproduction of that 'mess' is still high quality.

The question is if stribing for "high quality" can be a worthwhile thing to do.

Don't think anyone is arguing against that. But thinking that is the most important part [of photography] is however misguided. My observation is that those who tend to be obsessed with this aspect of photography take pictures that are technically very good and that's about it. There's rarely any art or beauty in the photos.

[Isaac]

That observation has been made some time ago by Emil.

Interesting, although I don't see any suggestion in Mr Reichmann's 2003 article about raising ISO to ETTR; just "slower shutter speeds and/or larger apertures".

If one has the option to lower the ISO and the shutter speed (or widen the aperture), the highest S/N for the image is obtained by increasing the exposure, pushing the right end of the histogram right up to the upper edge of the range of exposure on the horizontal axis. This is the usual ETTR philosophy.

[bjanes]

Interesting, although I don't see any suggestion in Mr Reichmann's 2003 article about raising ISO to ETTR; just "slower shutter speeds and/or larger apertures".

There has always been some confusion as to whether or not ETTR applies to above base ISOs. Exposure is in terms of lux seconds and is determined by the f-stop and shutter speed (for a given scene). Raising the ISO while keeping the shutter speed and aperture constant is useful for most Canon dSLRs when one is constrained by shutter speed and aperture considerations. This does not change the exposure, but does move the histogram to the right. With most CCD sensors and recent Sony CMOS sensors, raising the ISO yields minimal benefit and compromises highlight headroom.

Bill

[jjj]

I think what needs to be noted about ETTR is that like many tools is that sometimes it's useful, but most of the time it's irrelevant to either the photographer or the situation.

[BJL]

So the entire logic that refers to the limited number space at the left side of the histogram, compared to the right, is flawed.

Worrying about spacing between raw levels (quantization noise) might make sense with a digital camera that has ADCs that at best match the DR of the sensor, so that quantization noise is a significant worry, and that might still be an issue with some very good sensors when used at base ISO speed (say 14 bit ADC with a "14-stop sensor".)  But it is becoming irrelevant, and the best way to eliminate that worry entirely is doing ADC with a few bits of headroom relative to sensor DR -- as DMF has been doing for years, using 16 bit ADCs with 12 to 13 stop sensors.  And at elevated ISO speed, it has probably always been irrelevant: the extra analog gain pushes sensor noise well above quantization noise.

The real issue these days is getting as much exposure as possible to the sensor without highlight damage, and even this is only significant when dealing with scenes of unusually high subject brightness range, higher than the ten stops or so present in the vast majority of scenes.   And this "sensor exposure optimization" sounds like a job for good old-fashioned exposure bracketing.

[Hans Kruse]

Worrying about spacing between raw levels (quantization noise) might make sense with a digital camera that has ADCs that at best match the DR of the sensor, so that quantization noise is a significant worry, and that might still be an issue with some very good sensors when used at base ISO speed (say 14 bit ADC with a "14-stop sensor".)  But it is becoming irrelevant, and the best way to eliminate that worry entirely is doing ADC with a few bits of headroom relative to sensor DR -- as DMF has been doing for years, using 16 bit ADCs with 12 to 13 stop sensors.  And at elevated ISO speed, it has probably always been irrelevant: the extra analog gain pushes sensor noise well above quantization noise.

The real issue these days is getting as much exposure as possible to the sensor without highlight damage, and even this is only significant when dealing with scenes of unusually high subject brightness range, higher than the ten stops or so present in the vast majority of scenes.   And this "sensor exposure optimization" sounds like a job for good old-fashioned exposure bracketing.

I totally agree. Good old fashioned bracketing is what I do to postpone the decision of what the best exposure is for high DR scenes to post processing time and to automate the shooting process as much as possible to avoid human errors. Sometimes in landscape shooting a lot happens within a few minutes and mistakes are typically made in sub situations. I have clouds open up and close again within minutes and the shot of the year was within these minutes and the last thing one wants to do is staring at the LCD.

I shoot both Canon 5D III and Nikon D800E. For both the bracketing is to protect essential highlights and especially for the Canon it is to protect the shadows. Sometimes also on the Nikon and both require occasionally an HDR blend which I do using either Photoshop CC or the Photomatix 32bit plugin and then tonemap in Lightroom. This works well most of the time but at times it falls apart. I'm still dreaming of my dream camera mentioned earlier 

[Isaac]

There has always been some confusion as to whether or not ETTR applies to above base ISOs.

There's always some confusion :-)

[Isaac]

The real issue these days…

Please, please, please, may I have this in my next entry-level camera --

What one would like is to somehow be able to use ISO 100 to keep all the highlights, while at the same time using ISO 1600 to recover all the shadows. But how can one have two ISO settings at once? By having two separate amplifiers fed from the same sensor data, running in parallel.

[LKaven]

Please, please, please, may I have this in my next entry-level camera --

It made sense for the Canon, and an earlier generation of Nikon architecture.  But I think the technology has moved beyond that. 

[jjj]

There's always some confusion :-)

[stevierose]

My two passions are photography and fly fishing. Regarding gear I see several strong similarities. People buy camera gear for many reasons--producing great photographs is only one of them. There is the whole gear head and gear acquisition thing (because cameras are cool gadgets, or collectible, or whatever). There is also a large group of folks who confuse the technical capabilities of gear with the ability to produce work product with the gear--great photographs or, in the case of fly fishing, catching fish. Fishing gear went through a decade of technical improvements and sales coinciding with the change from bamboo or fiberglass rods to graphite based rods as well as the popularization of the past time in the media by movies such as "A River Runs Through It". During this time period I acquired new equipment nearly every year and lusted after each new innovation. However, around ten years ago or so most fly fishing equipment reached a point of "very likely good enough" and I stopped buying new gear. It was at that point that I fully realized that any problems I was having with making effective casts was not due to the rod, it was due to me.  I mean a great caster can pick up broom stick and cast the complete line without a problem...it's a matter of timing and skill. At the same time, domestic rod manufacturers started seeing decline in sales of expensive American made rods because of much cheaper imports using last generation (read: one or two year old) technology made in Asia. So the industry saw a decline. Actually, you can fill in the blank...a great golfer can hit a straight drive with a ten year old club. Same thing.

I think that cameras hit the point of "very likely good enough" a few years ago from a technical standpoint and recent improvements have been mainly in the area of usability (Fuji 100s, etc) and portability (mirrorless). I switched to Micro 4/3's because I like the lenses and the ability to carry what I need with me without rupturing a spinal disc. I now choose bodies based on how intuitive they are and reject bodies that are difficult to use to make photographs. Today any of the available cameras are more than capable of producing technically successful images. If my photographs aren't to my satisfaction, the problem is me, not the camera. But, then again, this has always been true, yes?

[hjulenissen]

As I have already mentioned I and indeed many others like grainy images. However try and find anyone who like hissy recordings.

I have never heard about anyone that prefer the look of a DSLR at ISO50000 to that of ISO100. Do you?

There seems to be quite a few people who like to listen to flawed vinyl when there are available more accurate music distribution means. Much like there are people who prefer grainy film to more accurate techniques. I believe that many have a liking for "good old" errors (typically analog), but I'd speculate that this is a culturally learned thing.

You are confusing things here.

I know exactly what I intend to say, thus I am not confused about my contribution. If you think that I am interpreting your posts erroneously, please tell me in what way.

You were talking about recording quality, not the sound being produced that is then recorded.

Samplers are recording devices. Cameras are recording devices. Both can be used to make art.

I have plenty of music that is distorted, has feedback etc, but the reproduction of that 'mess' is still high quality.

The art of making contemporary music cannot easily be distinguished from the art of recording music. The analogy for photography would be music making and recording. If you are interested in the technology of music distribution, that would be something like shipping JPEGs of a wedding to your client.

But thinking that is the most important part [of photography] is however misguided. My observation is that those who tend to be obsessed with this aspect of photography take pictures that are technically very good and that's about it. There's rarely any art or beauty in the photos.

I believe that you (like many others) are constructing a false contradiction between being into technique and producing great art (whatever that is). Some great artists were (for all intents and purposes) nerds. Most nerds are not great artists. But then, most people are not great artists, either.

I believe that everyone in this discussion have by now gotten the point that ETTR might not matter and that other things might matter more for any particular shot. Everyone agrees on this, so please lets move on.

-h

[Isaac]

Some great artists were (for all intents and purposes) nerds.

Ansel Adams: something of an artist; and something of a nerd ("The Camera", "The Negative",  "The Print").

"If there was a new route to excellence in a photographic technology, Ansel either led the climb or pushed others up it."

David Brower, 1938

[Isaac]

But I think the technology has moved beyond that.

I'm quite willing to believe you; but it's difficult to keep track of where the technology has reached for a particular camera model, and what "wisdom" no longer applies.

[Hans Kruse]

I believe that everyone in this discussion have by now gotten the point that ETTR might not matter and that other things might matter more for any particular shot. Everyone agrees on this, so please lets move on.

-h

I certainly disagree. Optimum exposure does matter.

[jjj]

I have never heard about anyone that prefer the look of a DSLR at ISO50000 to that of ISO100. Do you?

Never said that, try reading post properly. Many people like grain, however try and find anyone who like hissy recordings.
Besides I already said I used to develop Kodak recording film in a speed enhancing developer. At it's base ISO of 1000 it was rather low quality, at 3200 ISO as I used it, it would give pixel peepers dyspepsia.
And the new Nikon at 50,000 ISO is in fact better quality than how I used Recording film. So 

There seems to be quite a few people who like to listen to flawed vinyl when there are available more accurate music distribution means.

I think you'll find they listen to vinyl because it sounds better to them. This obsession with accuracy tends to make people not see the wood for the trees.

Much like there are people who prefer grainy film to more accurate techniques. I believe that many have a liking for "good old" errors (typically analog), but I'd speculate that this is a culturally learned thing.

Or far more, it's likely their innate taste.

I know exactly what I intend to say, thus I am not confused about my contribution. If you think that I am interpreting your posts erroneously, please tell me in what way.

As I said above, you conflated making music with the recording of music. They are not the same thing.

Samplers are recording devices. Cameras are recording devices. Both can be used to make art. The art of making contemporary music cannot easily be distinguished from the art of recording music. The analogy for photography would be music making and recording. If you are interested in the technology of music distribution, that would be something like shipping JPEGs of a wedding to your client.

Are you indulging in some Burroughs-esque random cutting and pasting style of posting now?
Actually shipping JPEGs is exactly what wedding photographers do.

I believe that you (like many others) are constructing a false contradiction between being into technique and producing great art (whatever that is). Some great artists were (for all intents and purposes) nerds. Most nerds are not great artists. But then, most people are not great artists, either.

It's not about being a great artist, they are outliers anyway. I was talking about those who are obsessed with technical perfection and long experience of such people has shown that with very few exceptions they do not produce interesting work. Or frequently show no work at all.
Those who do actually produce creative work however do not tend to be so hung up on that sort of thing. They learn enough to master their craft and get on with it and not only do they not mind if if something isn't perfect, but may actively seek the opposite.

The other similar truism is those that slag off photoshop/post processing etc tend to have no ability in that area of photography.

[jjj]

I certainly disagree. Optimum exposure does matter.

It's defining what optimum exposure is, that is the sticking point though.

[Hans Kruse]

It's defining what optimum exposure is, that is the sticking point though.

In general yes and implementing automatic ETTR will reveal that, I'm sure. For the individual shot in a bracketed sequence that stretches from no clipping to clipping, it is easy to determine in post processing.

[jeremyrh]

I'm about to buy some Sugru

Strange synchronicity - I had never heard of this before, and then almost immediately saw an article about it in the Grauniad:
http://www.theguardian.com/business/2014/mar/16/sugru-inventor-jane-ni-dhulchaointigh
Looks handy.

Monkeys, shakespeare, an' all that.

[Isaac]

Perhaps jjj also read that Guardian story.