Irrational views on SLT part duex

Started Jan 22, 2014 | Discussions
JimmyMelbourne Senior Member • Posts: 2,458
Re: Had To Join This Great Discussion (my samples)

Is it just me or does the iso 6400 photo look less noisy?

Im not sure how this can be a solution for anything, when shooting at iso 100 the image turns black so reviewing the image after the shot will be difficult.

 JimmyMelbourne's gear list:JimmyMelbourne's gear list
Sony Alpha DSLR-A300 Sony SLT-A77 Sony SLT-A37 NEX-5T Sony Alpha a7 +20 more
sybersitizen Forum Pro • Posts: 10,645
Is you mind open or not?
3

Shield3 wrote:

If it were truly "ISO-less" then the results shouldn't change based on exposure time.

You do not yet understand the term 'ISO-less'. That's okay, you will.

It's not. That's why on all cameras you see the dynamic range and shadow noise go to pot at high ISO's.

You'll never get me to think otherwise.

Then why are you discussing anything?

I do appreciate you continuing to say "not sure if you're trolling" to somehow add credibility to your claim though. Bonus points.

A properly exposed ISO6400 shot then shot at ISO 100 would certainly mean the 100 shot is going to get blown out.

Again, you do not yet understand the term 'ISO-less'. All it means is that every shot can be made at base ISO and then 'amplified' after the fact instead of in the camera.

But, again, if they were truly ISO-less one should be able to do what I've done in this thread and the shots be very similar. Why aren't they?

Because you exposed one shot for 1/100 second and the other for 1/8000 second.

If cameras were truly "ISO-less" then the inverse would be true - a "properly exposed" ISO 100 shot could be re-shot after setting the camera to ISO 6400 (with nothing else changing) and the images could be recovered to look identical in post.

You're just making stuff up. Nobody has to take that made-up statement into account. But the reason it won't work is that the ISO 6400 image will already be amplified internally, thus permanently losing most highlight data. Again, you do not yet understand the term 'ISO-less'. All it means is that every shot can be made at base ISO and then 'amplified' after the fact instead of in the camera.

MediaArchivist
MediaArchivist Veteran Member • Posts: 4,512
Good sample scene

That was actually a pretty good sample scene; shadows and highlights, different colors, angled lines, etc.

-- hide signature --

Want a roXplosion!?

 MediaArchivist's gear list:MediaArchivist's gear list
Sony a99 II Sony 135mm F1.8 ZA Carl Zeiss Sonnar T* Sony 135mm F2.8 (T4.5) STF Sony 24mm F2 SSM Carl Zeiss Distagon T* Sigma 35mm F1.4 DG HSM Art +22 more
William Curtindale
William Curtindale Veteran Member • Posts: 8,026
JimmyMelbourne: Probably Best To Discuss W/ Nordstjernen
1

JimmyMelbourne wrote:

Is it just me or does the iso 6400 photo look less noisy?

Im not sure how this can be a solution for anything, when shooting at iso 100 the image turns black so reviewing the image after the shot will be difficult.

Jimmy, this is beyond me to explain (you need to discuss with Nordstjernen, I know he can explain).

Yes, my ISO 100 sample appears to have more noise (could something I did in PP - don't think so be I don't really know).

I find this experiment very interesting and know there is more to learn. For now I will continue to adjust Shutter, aperture and ISO (old fashioned thinking). At least I can see a benefit of understanding how much forgiveness there is in using too low of an ISO (I think that's good to know).

I also find it very interesting in this discussion to think about exposure in terms shutter and aperture and of how much light hits the sensor (disregarding the sensor sensitivity - not totally sure but I think I said the right). Yeah, I need to get my head around this stuff.

-- hide signature --

A99, A900, A700 w/Zeiss + Sony G Glass
http://curtindale.net/ http://www.youtube.com/TheWillybug

MediaArchivist
MediaArchivist Veteran Member • Posts: 4,512
Low light shooting

I do a lot of low light shooting. Learning to use the widest apertures at my disposal (f/1.4 to f/2) was key, and often involves delicate gymnastics to get the focal plane lined up exactly where I want it. Based on the scenes I usually shoot, with barely lit subjects and unlit (almost black) backgrounds, I need a good combination/compromise of shutter speed and low grain. I could set the ISO to 100 and pump it all in PP, but I do need to chimp a little to make sure I am getting what I need in focus with a very thin DoF.

Usually, I will shoot at a fixed aperture (almost always wide open) and set my ISO on auto and adjust the auto range to 400-1600. I then set my EV to -3. I vary this a little sometimes, and sometimes even mid shoot. In aperture priority mode, I have the rear thumb wheel set to adjust EV, so that makes it pretty easy. The eventual effect of my EV setting and ISO range is to limit the range of the shutter speed. If I could choose an speed range (high and low limits, like with auto ISO) I would prefer to do it that way, but in fairness the camera gives me enough tools at my disposal to get what I want.

I don't use spot metering. I probably could, but when I try I never get the results I want. In theory, spot or area metering ought to give me better results, so at this point I consider this to be an area I need more experience with.

Later, I can choose to pump the whole image, just a certain area, just the highlights, or just a particular hue. Lightroom makes this pretty easy but I would guess other, similar programs do, too. So far, this technique works— at least in the sense that I get photos I would not any other way. Whether or not those photos are any good is a subject for another post

-- hide signature --

Want a roXplosion!?

 MediaArchivist's gear list:MediaArchivist's gear list
Sony a99 II Sony 135mm F1.8 ZA Carl Zeiss Sonnar T* Sony 135mm F2.8 (T4.5) STF Sony 24mm F2 SSM Carl Zeiss Distagon T* Sigma 35mm F1.4 DG HSM Art +22 more
cosmonaut
cosmonaut Senior Member • Posts: 2,223
Rodeo, 8000 ISO

Here are some action shots at an indoor rodeo 8000 ISO. I am pleased. Would my D800 do better? Probably but I like using the a99 more.

-- hide signature --

www.gregmccary.com

MediaArchivist
MediaArchivist Veteran Member • Posts: 4,512
Another issue

The Canon 5D3 does not use an ISOless sensor, so the test I suggest might not have the results we see with ISOless sensors (see examples further down this thread).

-- hide signature --

Want a roXplosion!?

 MediaArchivist's gear list:MediaArchivist's gear list
Sony a99 II Sony 135mm F1.8 ZA Carl Zeiss Sonnar T* Sony 135mm F2.8 (T4.5) STF Sony 24mm F2 SSM Carl Zeiss Distagon T* Sigma 35mm F1.4 DG HSM Art +22 more
MediaArchivist
MediaArchivist Veteran Member • Posts: 4,512
Re: Rodeo, 8000 ISO

Sweet! What lens were you using?

-- hide signature --

Want a roXplosion!?

 MediaArchivist's gear list:MediaArchivist's gear list
Sony a99 II Sony 135mm F1.8 ZA Carl Zeiss Sonnar T* Sony 135mm F2.8 (T4.5) STF Sony 24mm F2 SSM Carl Zeiss Distagon T* Sigma 35mm F1.4 DG HSM Art +22 more
cosmonaut
cosmonaut Senior Member • Posts: 2,223
Re: Rodeo, 8000 ISO

MediaArchivist wrote:

Sweet! What lens were you using?

Most of it was shot with the Sony 70-400mm and a few at the end with the Sony 70-300mm.

-- hide signature --

www.gregmccary.com

OP Shield3 Senior Member • Posts: 2,440
Re: Is you mind open or not?

sybersitizen wrote:

Shield3 wrote:

If it were truly "ISO-less" then the results shouldn't change based on exposure time.

You do not yet understand the term 'ISO-less'. That's okay, you will.

No such thing.  ISO matters.  Again, for it to be true, the inverse would have to be true.  IE a shot properly exposed for ISO 100 then taken @ ISO 6400.  Good luck getting all those highlights back that you just blew out.  Not to mention that cameras are ELECTRONIC devices, and gain/ISO does add noise, reduce midtones, lower dynamic range.

For a "true" ISO-less camera the SNR and dynamic range chart would have to be linear; they're not.  Even in the shots above I see vast differences; the shadow detail is gone when you underexpose by 6! stops just like the highlights are gone when you UNDERexpose by 6.

Stop drinking the cool-aid and use your eyes and your logic.  You'll get it. 

It's not. That's why on all cameras you see the dynamic range and shadow noise go to pot at high ISO's.

You'll never get me to think otherwise.

Then why are you discussing anything?

Why does anyone discuss anything?

I do appreciate you continuing to say "not sure if you're trolling" to somehow add credibility to your claim though. Bonus points.

A properly exposed ISO6400 shot then shot at ISO 100 would certainly mean the 100 shot is going to get blown out.

Again, you do not yet understand the term 'ISO-less'. All it means is that every shot can be made at base ISO and then 'amplified' after the fact instead of in the camera.

I do get raw and the ability to do that to a certain degree.  Not 6 stops, and not ISO 100 to 6400.  I'm not saying from an arm's length you can't get a similar result from a web-sized shot, but up close and you'll begin to see all the problems (shadow noise, blown highlights).

 Shield3's gear list:Shield3's gear list
Canon EOS 5D Mark III Canon EF 35mm F1.4L USM Canon EF 85mm f/1.2L II USM Canon EF 135mm F2L USM Canon EF 70-200mm F2.8L IS II USM +2 more
OP Shield3 Senior Member • Posts: 2,440
Re: Rodeo, 8000 ISO

cosmonaut wrote:

MediaArchivist wrote:

Sweet! What lens were you using?

Most of it was shot with the Sony 70-400mm and a few at the end with the Sony 70-300mm.

I think those shots are fine for downscaled web viewing.  I always see examples on here but rarely does anyone show the full size picture.

My goal is to try to figure out how to get large, printable shots (at least 13x19) of my son's basketball.  Even with a 5d3 and a 135/2 I'm not super thrilled with the results.  I also hate being that wide open due to barely any DOF.

Perhaps I'll bump it down to 1/640 and just have a blurred ball.  Shrug.

 Shield3's gear list:Shield3's gear list
Canon EOS 5D Mark III Canon EF 35mm F1.4L USM Canon EF 85mm f/1.2L II USM Canon EF 135mm F2L USM Canon EF 70-200mm F2.8L IS II USM +2 more
EinsteinsGhost
EinsteinsGhost Forum Pro • Posts: 11,977
Re: Rodeo, 8000 ISO

Shield3 wrote:

cosmonaut wrote:

MediaArchivist wrote:

Sweet! What lens were you using?

Most of it was shot with the Sony 70-400mm and a few at the end with the Sony 70-300mm.

I think those shots are fine for downscaled web viewing. I always see examples on here but rarely does anyone show the full size picture.

My goal is to try to figure out how to get large, printable shots (at least 13x19) of my son's basketball. Even with a 5d3 and a 135/2 I'm not super thrilled with the results. I also hate being that wide open due to barely any DOF.

Perhaps I'll bump it down to 1/640 and just have a blurred ball. Shrug.

No blurred ball here, at 1/640s:

f/2.8, ISO 800, 1/640s

 EinsteinsGhost's gear list:EinsteinsGhost's gear list
Sony Cyber-shot DSC-F828 Sony SLT-A55 Sony Alpha NEX-6 Sigma 18-250mm F3.5-6.3 DC OS HSM Sony 135mm F2.8 (T4.5) STF +12 more
sybersitizen Forum Pro • Posts: 10,645
Re: Is your mind open or not?

Shield3 wrote:

sybersitizen wrote:

Shield3 wrote:

If it were truly "ISO-less" then the results shouldn't change based on exposure time.

You do not yet understand the term 'ISO-less'. That's okay, you will.

No such thing. ISO matters. Again, for it to be true, the inverse would have to be true. IE a shot properly exposed for ISO 100 then taken @ ISO 6400. Good luck getting all those highlights back that you just blew out. Not to mention that cameras are ELECTRONIC devices, and gain/ISO does add noise, reduce midtones, lower dynamic range.

For a "true" ISO-less camera the SNR and dynamic range chart would have to be linear; they're not. Even in the shots above I see vast differences; the shadow detail is gone when you underexpose by 6! stops just like the highlights are gone when you UNDERexpose by 6.

Stop drinking the cool-aid and use your eyes and your logic. You'll get it.

Okay, you're just being dumb, and I don't care what you choose to believe. Bye now.

EinsteinsGhost
EinsteinsGhost Forum Pro • Posts: 11,977
Re: Is you mind open or not?

Shield3 wrote:

IE a shot properly exposed for ISO 100 then taken @ ISO 6400. Good luck getting all those highlights back that you just blew out. Not to mention that cameras are ELECTRONIC devices, and gain/ISO does add noise, reduce midtones, lower dynamic range.

You're thinking in reverse. The point being made is that you can shoot at ISO 100, where you will get better DR than you would at ISO 6400. In fact, and while I didn't really need it (was on trial basis), the basketball photographs you've seen from me were all taken at a lower ISO than "Auto ISO" would choose. I just posted one, and you might notice ISO 800, which is the ISO I set instead of using ISO 1600 or letting the camera choose 1600. During RAW conversion, +1EV was applied. The DR is for ISO 800, rather than ISO 1600.

 EinsteinsGhost's gear list:EinsteinsGhost's gear list
Sony Cyber-shot DSC-F828 Sony SLT-A55 Sony Alpha NEX-6 Sigma 18-250mm F3.5-6.3 DC OS HSM Sony 135mm F2.8 (T4.5) STF +12 more
MediaArchivist
MediaArchivist Veteran Member • Posts: 4,512
Re: Is you mind open or not?
1

Shield3 wrote:

sybersitizen wrote:

Shield3 wrote:

If it were truly "ISO-less" then the results shouldn't change based on exposure time.

You do not yet understand the term 'ISO-less'. That's okay, you will.

No such thing. ISO matters.

On a 5D3, you are correct. The ISO setting affects the analog gain applied to the sensor signal, with higher numbers applying more gain. The resulting RAW file reflects this amplification, in that the raw data is, in fact, different for different ISO settings. On other cameras, with ISO-less sensors, ISO does not alter the raw data, other than to add a meta-tag as a hint for a RAW converter. If that converter is the in-camera JPEG engine, then this technical fact is hidden from you (this means the JPEG header on RAW files is affected, as is the preview— perhaps this is what is throwing you off?).

Again, for it to be true, the inverse would have to be true.

It is.

IE a shot properly exposed for ISO 100 then taken @ ISO 6400. Good luck getting all those highlights back that you just blew out.

You did not blow them out, all the raw data is still there. Lower the exposure in whatever RAW converter you are using and it all comes back. Unless, of course...

Not to mention that cameras are ELECTRONIC devices, and gain/ISO does add noise, reduce midtones, lower dynamic range.

... you are using a 5D3, where the analog gain adds noise and reduces midtones and DR.

You are not doing your homework. Perhaps you have never extensively used a camera with an ISO-less sensor, or if you did it was all in JPEG mode?

-- hide signature --

Want a roXplosion!?

 MediaArchivist's gear list:MediaArchivist's gear list
Sony a99 II Sony 135mm F1.8 ZA Carl Zeiss Sonnar T* Sony 135mm F2.8 (T4.5) STF Sony 24mm F2 SSM Carl Zeiss Distagon T* Sigma 35mm F1.4 DG HSM Art +22 more
JimmyMelbourne Senior Member • Posts: 2,458
Re: JimmyMelbourne: Probably Best To Discuss W/ Nordstjernen

William Curtindale wrote:

JimmyMelbourne wrote:

Is it just me or does the iso 6400 photo look less noisy?

Im not sure how this can be a solution for anything, when shooting at iso 100 the image turns black so reviewing the image after the shot will be difficult.

Jimmy, this is beyond me to explain (you need to discuss with Nordstjernen, I know he can explain).

Yes, my ISO 100 sample appears to have more noise (could something I did in PP - don't think so be I don't really know).

I find this experiment very interesting and know there is more to learn. For now I will continue to adjust Shutter, aperture and ISO (old fashioned thinking). At least I can see a benefit of understanding how much forgiveness there is in using too low of an ISO (I think that's good to know).

I also find it very interesting in this discussion to think about exposure in terms shutter and aperture and of how much light hits the sensor (disregarding the sensor sensitivity - not totally sure but I think I said the right). Yeah, I need to get my head around this stuff.

I get the general principle and do feel it supports the notion of correct exposure as the driving principle. I am just not sure how practical it is to try and shoot lower ISO (to the point of blacking out the scene) all the time to reduce noise when other advantages are lost, such as seeing the output after the shot is taken rather than trying to correct in post processing. As a test its quite interesting.

 JimmyMelbourne's gear list:JimmyMelbourne's gear list
Sony Alpha DSLR-A300 Sony SLT-A77 Sony SLT-A37 NEX-5T Sony Alpha a7 +20 more
Nordstjernen
Nordstjernen Veteran Member • Posts: 6,876
Re: Irrational views on SLT part duex

Shield3 wrote:

I thought you said "if both shots were exposed properly". Not "same amount of light hitting the sensor". Big difference.

Went back and read it. You said "If you first shoot a well exposed frame at ISO 6400 and then shoot at ISO 100 with exactly the same exposure settings, the processed pictures will look almost identical - same amount of shadow detail and same amount of noise!"

The PROCESSED pictures!!!

In my above shots the exact or very near the same amount of light was hitting the sensor. The exposure time was shorter in one than the other, but all other things were equal.

No, you did NOT. When using 1/8000 and 1/100 sec and the same aperture value, there is a HUGE difference on how much light that hits the sensor. Many, many stop difference!

As said, ISO setting has NOTHING to do with exposure (the amount of light falling on the sensor). The only two valid factors are shutter speed and aperture value. When changing any of these, or a combination of both, the exposure will be different.

 Nordstjernen's gear list:Nordstjernen's gear list
Sony Alpha a99 Sony Alpha a7
Jeadm Senior Member • Posts: 1,744
Re: Had To Join This Great Discussion (my samples)

JimmyMelbourne wrote:

Is it just me or does the iso 6400 photo look less noisy?

Im not sure how this can be a solution for anything, when shooting at iso 100 the image turns black so reviewing the image after the shot will be difficult.

It's not just you.

I've quietly hung in with this discussion because it's intriguing and enjoy learning something new.  However I'm not quite there yet when trying to discern if there are real world practical advantages to doing this.  I'm no expert by any means but the ISO6400 image still looks better to me, despite the promise of low ISO benefits.  If that's the case, then the point of all this is...?  Am I reading too much into the examples?

It's almost as if what's being said here is that shutter and aperture are irrelevant because you can just shoot ISO100 and push PP everything (keeping the same exposure settings as the higher ISO that if used gives you the preferred shutter speed & f-stop), and I know that can't be the case.  6-stops seems extreme; is there a more reasonable expectation on how far one can reasonably push this technique?  What are the differences between APS-C and FF sensors in that regard, if any?

 Jeadm's gear list:Jeadm's gear list
Sony a77 II Tamron SP AF 70-200mm F/2.8 Di LD (IF) MACRO Tokina AT-X Pro 11-16mm f/2.8 DX Sony 85mm F2.8 SAM Sony DT 35mm F1.8 SAM +15 more
phaedin
phaedin Senior Member • Posts: 1,634
Re: Had To Join This Great Discussion (my samples)

I think that this is useful when you look at taking the photo from the other side.
So you "over expose" the image so you allow more light to hit the sensor and then pull the image back in PP so that you can get less noise in the shadows (as long as you don't blow the highlights out)
I think this is the basic principle behind the Expose To The Right theory

 phaedin's gear list:phaedin's gear list
Nikon D7100 Nikon D7500 Sigma 17-50mm F2.8 EX DC OS HSM Tamron SP 70-200mm F/2.8 Di VC USD Sigma 50mm F1.4 DG HSM | A +1 more
Wolfgang Fieger Regular Member • Posts: 151
Re: Irrational views on SLT part duex
1

Except for the added grain in the 6400ISO variant there is not much difference between the 2 shots: Exactly the same dull, contrast lacking & out-of-focus scenery....

If you really want to prove something, start with better shots

-- hide signature --

see you,
Wolfgang

Keyboard shortcuts:
FForum MMy threads