ISO discussion, digital SLR compared to film SLR

[TimmyD11]

Well I have a lot to absorb and process here.

I have a few point and shoot cameras but I am researching cameras now to purchase a good one, a DSLR probably but perhaps a mirrorless camera. I am trying to figure out if I get an APSC DSLR or full frame DSLR, and if I should stick with Canon based on having some older early EF lenses from when I had an EOS A2 film camera. If I stick with Canon and go APSC I have to figure out how much camera is enough...t7? 77D? 80D? Or make the few hundred dollar leap to the 6D full frame (I think?)

OR cut my old lens losses and be free to go mirrorless APSC...

But I like what I have read about ISO invariance sensors and according to someone on another thread Canon sensors are STILL lagging behind other manufacturers, even with the latest improvements in their sensors...Is this true?

Wanna make a case for what direction I should go in?

I'm a nature / great outdoors fanatic that likes to photograph what I see on my journeys.

I am just getting back into taking photography a bit more seriously and I want to get a nice camera but I have to know how much camera may be the right amount for me and how much camera may be too much camera for me. This is the main reason I have not purchased anything yet. I have to figure out what the right camera for me is so I don't have buyer's remorse and wish I spent a few more bucks for a better camera. At the same time I don't want to go so crazy that I spend too much and get something I may not take full advantage of.

Also, if I decide to get into astralphotography in the mountains can APSC handle that?

The thread subject will drift but that's fine by me if you guys don't mind.

 

[Ysarex]

Well I have a lot to absorb and process here.

I have a few point and shoot cameras but I am researching cameras now to purchase a good one, a DSLR probably but perhaps a mirrorless camera. I am trying to figure out if I get an APSC DSLR or full frame DSLR, and if I should stick with Canon based on having some older early EF lenses from when I had an EOS A2 film camera. If I stick with Canon and go APSC I have to figure out how much camera is enough...t7? 77D? 80D? Or make the few hundred dollar leap to the 6D full frame (I think?)

Yes the 6D is FF.

OR cut my old lens losses and be free to go mirrorless APSC...

I would take the loss on the old lenses and start fresh even if you do go with Canon.

But I like what I have read about ISO invariance sensors and according to someone on another thread Canon sensors are STILL lagging behind other manufacturers, even with the latest improvements in their sensors...Is this true?

In terms of ISO invariance yes. You want to look at Nikon, Sony and Fuji if you're interested in ISO invariance. I wouldn't rank that variable too high though -- Canon makes some nice and very usable hardware. But, yeah I'd go Nikon or Fuji. Nikon if you want FF and Nikon or Fuji for APS.

If you want mirrorless then Fuji with the caveat that Fuji's X-Trans CFA is a unique tech and special handling is indicated. Research that before you decide Fuji -- I can help. Sony could be an APS mirrorless fall back.

Wanna make a case for what direction I should go in?

I'm a nature / great outdoors fanatic that likes to photograph what I see on my journeys.

In the great outdoors you face the problem of not being able to control the lighting. With difficult lighting like very high contrast, the ISO invariance ability of some newer cameras can prove a real positive -- you're better able to take full advantage of the sensor DR. But I'm starting to get technical there and we don't want to start another round of "don't worry stay stupid and be happy." ;-) Otherwise I fall back on my earlier comment of not ranking it too high.

I am just getting back into taking photography a bit more seriously and I want to get a nice camera but I have to know how much camera may be the right amount for me and how much camera may be too much camera for me. This is the main reason I have not purchased anything yet. I have to figure out what the right camera for me is so I don't have buyer's remorse and wish I spent a few more bucks for a better camera. At the same time I don't want to go so crazy that I spend too much and get something I may not take full advantage of.

Also, if I decide to get into astralphotography in the mountains can APSC handle that?

The thread subject will drift but that's fine by me if you guys don't mind.

To get a better response from the group at large a new thread would be a good idea with a more appropriate title.

Joe

 

[Braineack]

But I like what I have read about ISO invariance sensors and according to someone on another thread Canon sensors are STILL lagging behind other manufacturers, even with the latest improvements in their sensors...Is this true?

yes.

Wanna make a case for what direction I should go in?

the video posted in post #31: ISO discussion, digital SLR compared to film SLR

watch between 7:15 and 9:20. That's the case alone for me.

 

[Gary A.]

Thanks for hanging there TimmyD11. (I'm not a techie ... so I hope I get this right ...)

For me, the bottom line in all this, was Joe saying that with the latest and greatest Sony sensors (used in Sony, Nikon and Fuji cameras), that changing the ISO in-camera doesn't matter because the same basic/technical process is used to lighten the image in the camera as on a computer. There are two advantages to manipulating the image on a computer:

1) a computer is more powerful and is capable of rendering a better manipulated image than in-camera manipulation (i.e. raising the ISO); and
2) Raising the ISO will often results in the camera blowing out highlights, while manipulation on a computer tends to reduce the chances of 'clipping the highlights'.

Thank You Joe.

 

[TCampbell]

Just remember that noise is "additive" in that it always results in a photosite reporting a HIGHER numeric value than it should have reported. There is no "anti-noise" where the photosite reported a LOWER numeric value (at least if there is, I have yet to encounter it.)

huh? no

take a picture of a white piece of paper at high ISO, zoom in on it, and observe in astonishment.

discussing noise is extremely difficult and not one i would dare approach in a page or two. there are many sources of noise each with their own properties and each with their own appropriate models for analysis and so on. sadly it's also critical to understanding in any real
technical detail any of the issues here, which is why all the efforts to nupack what ISO settigs actually do are so hilarious and yet dull

I'm not sure where you're trying to go with this. I mentioned that there are lots of reasons for noise (including quantum uncertainty) but I wasn't going to attempt to pick each one apart.

It looks like you are responding without having read what I wrote.

Noise is much easier to notice in "dark" areas than in "light" areas. Shoot a wedding and look at the bride's "white" wedding dress next to the groom's "black" tuxedo. Black tux has lots of noise. White dress... not so much.

When I do astrophotography (but really any photography) I more aggressively treat noise in dark tonal regions than in light tonal regions.

It's easy to find examples of this.

Now... can you severely under-expose an image and then boost ISO high enough that something that should be dark only appears bright under extreme ISO boost? Sure. Would you see noise if you did that? Sure. Why would you do that if you knew it was going to be a problem?

 

[Gary A.]

As I shoot mirrorless, I imagine the ISO dial is useful to gauge if a potential image, in a low light situation, can be acceptably manipulated to a successful image. Sorta like a preview.

 

[SCraig]

...It's easy to find examples of this.

Just Google "Expose To The Right".

 

[TCampbell]

But back to that "upstream" amplification... this results in a gain being applied BEFORE digital conversion (before ADC) and as a result it doesn't lose much in the way of dynamic range.

If you were to test the dynamic range of a camera (using a test target) what you'd probably notice is that as you boost ISO, you don't seem to be losing much dynamic range... but there's a limit to this... and then suddenly you hit an ISO where you get a linear drop off in dynamic range for each boost in ISO beyond that point.

What you're describing here is peculiar to Canon engineered cameras. This is a Canon only behavior that does not apply to the rest of the photo world. You need to preface this with that disclaimer. The basic rule is DR drops linearly with increasing ISO. Here's a visual on that looking at the Canon 5D III and Nikon D810: Photographic Dynamic Range versus ISO Setting

Joe

Hi Joe,

The graph varies by model and whether or not they employ the use of upstream gain.

Here's the profiles for several Nikon cameras:

Best ISO values for Nikon cameras

Notice the D3100 (first camera & graph on the list). There's almost no significant loss of dynamic range as they boost ISO (comparing ISO's 100, 200, & 400). At 800 there's a more noticeable drop... but it's not huge. Then you get to 1600 and suddenly the graph takes on a linear slope. (and the camera follows the graph just like the example you cited above.)

That indicates that the D3100 is doing "upstream" amplification and somewhere around ISO 800 it begins using a moderate blend of upstream + downstream amplification... but by the time you get to ISO 1600 it's all downstream amplification (digital side) and no additional upstream gain is applied.

Compare that to a D3200 ... it's all one linear slope. This suggests that the D3200 is doing only downstream (digital) amplification and no upstream gain is applied at all.

If you go through the full list, you'll see some Nikon models do apply some upstream amplification... many apply none and it's all downstream.

Canon seems to always use upstream amplification for low amounts of ISO boost, then switches to downstream for high amounts of ISO boost.

 

[Ysarex]

But back to that "upstream" amplification... this results in a gain being applied BEFORE digital conversion (before ADC) and as a result it doesn't lose much in the way of dynamic range.

If you were to test the dynamic range of a camera (using a test target) what you'd probably notice is that as you boost ISO, you don't seem to be losing much dynamic range... but there's a limit to this... and then suddenly you hit an ISO where you get a linear drop off in dynamic range for each boost in ISO beyond that point.

What you're describing here is peculiar to Canon engineered cameras. This is a Canon only behavior that does not apply to the rest of the photo world. You need to preface this with that disclaimer. The basic rule is DR drops linearly with increasing ISO. Here's a visual on that looking at the Canon 5D III and Nikon D810: Photographic Dynamic Range versus ISO Setting

Joe

Hi Joe,

The graph varies by model and whether or not they employ the use of upstream gain.

Here's the profiles for several Nikon cameras:

Best ISO values for Nikon cameras

Notice the D3100 (first camera & graph on the list). There's almost no significant loss of dynamic range as they boost ISO (comparing ISO's 100, 200, & 400). At 800 there's a more noticeable drop... but it's not huge. Then you get to 1600 and suddenly the graph takes on a linear slope. (and the camera follows the graph just like the example you cited above.)

That indicates that the D3100 is doing "upstream" amplification and somewhere around ISO 800 it begins using a moderate blend of upstream + downstream amplification... but by the time you get to ISO 1600 it's all downstream amplification (digital side) and no additional upstream gain is applied.

Compare that to a D3200 ... it's all one linear slope. This suggests that the D3200 is doing only downstream (digital) amplification and no upstream gain is applied at all.

If you go through the full list, you'll see some Nikon models do apply some upstream amplification... many apply none and it's all downstream.

Canon seems to always use upstream amplification for low amounts of ISO boost, then switches to downstream for high amounts of ISO boost.

Having a deja vu experience -- we've done this before. So I see the 3100 graph and when I check it against Bill's data he shows the same thing but not to the same degree. All these test sites show data pulled from raw files (post ADC) and camera manufacturers aren't talking so I'm not prepared to speculate exactly what they're doing. When I get a chance I'll pass this one by Illiah Borg and see what he has to say. He should have an answer. In the meantime the majority of cameras tend to show a pretty linear raise the ISO a stop/drop a stop of sensor DR behavior and I'd call that the basic rule and other behavior an exception. I'll ask Illiah.

Joe

 

[Derrel]

Why are we addressing the Nikon D3100 here?

From WIkipedia: "The Nikon D3100 is a 14.2-megapixel DX format DSLR Nikon F-mount camera announced by Nikon on August 19, 2010. It replaced the D3000 as Nikon's entry level DSLR. It introduced Nikon's new EXPEED 2 image processor and was the first Nikon DSLR featuring full high-definition video recording with full-time autofocus and H.264 compression, instead of Motion JPEG compression. It was also the first Nikon DSLR to provide high-definition video recording at more than one frame rate.[3]"

The D3100 is a VERY outdated camera...

 

[fmw]

Why are we addressing the Nikon D3100 here?

From WIkipedia: "The Nikon D3100 is a 14.2-megapixel DX format DSLR Nikon F-mount camera announced by Nikon on August 19, 2010. It replaced the D3000 as Nikon's entry level DSLR. It introduced Nikon's new EXPEED 2 image processor and was the first Nikon DSLR featuring full high-definition video recording with full-time autofocus and H.264 compression, instead of Motion JPEG compression. It was also the first Nikon DSLR to provide high-definition video recording at more than one frame rate.[3]"

The D3100 is a VERY outdated camera...

Now, now Derrel, if the idea is to make good images, a 14mp DX camera is hardly outdated. It just doesn't have all the features of current models. It is capable of stunning photographs in the hands of someone who makes stunning photographs. I know you could use one to create any image you like and be happy with the result. I know I could. Sometimes we confuse need with want.

 

[Derrel]

Holding up the D3100 as an example of what "Nikon does" as far as image signal processing is not appropriate in the context of discussion how a camera maker's image processing strategies are currently implemented...that's the issue. Just calling out a bit of shade-throwing that somebody else was trying to do.

I've occasionally shot the old D2x over the past couple of years; inage quality is very "2005". Like the now-old D2x, the five years newer D3100 has easily been bettered; the current D3400 is a vastly better iange maker from a dynamic range point of view. As far as what "Nikon deoes" today, let's look to their D7200, which represents an actual and CURRENT Nikon camera. In the D7200 the dynamic range is 14.6 EV. The now-old D3100 with the same-sized APS-C sensor has 11.3 EV of dynamic range; that is a HUGE improvement, of three and one-third MORE stops's worth of DR.

In terms of overall sensor score, the D3100 is at 67, the D7200 is at 87; right up there in what used to be "full-frame" territory as far as technical image quality.

A sensor score of 67 is now considered poor in Nikon circles; that score and that bad dynamic range is more like m4/3 mini-sensor territory. An improvement of 3.3 EV in DR is major; holding up the D3100 as an example of modern sensors or signal processing as done by Nikon "today" is disingenuous, and that's what I wanted to call attention to.

Ohhhhh--and the D7200 is for all practical purposes, ISO Invariant.

 

[KmH]

How does a "fixed" sensor in a digital camera become more or less sensitive to light by just selecting a different ISO?

It doesn't. As said before in the thread, the voltage developed by the pixel gets amplified (gain added) after the exposure has been made but before the analog voltage is converted to a digital number in the Analog to Digital (A/D) converter circuitry.
But if you want to know more about how a CMOS pixel works: Active pixel sensor - Wikipedia

 

[TCampbell]

Why are we addressing the Nikon D3100 here?

[snip]

The D3100 is a VERY outdated camera...

Derrel, follow the link and look at all the graphs. The D3100 is simply the first camera they list and it does upstream amplification (and is is probably the most extreme example of it), then the D3200 doesn't do it all... then the D3300 does it again (but not to the extreme of the D3100) and you can keep going through the list of all the models.

I think the author says they pulled the data from DxO (and I'm always extremely skeptical of anything that comes from DxO) but I do personally know people who have tested their own individual cameras. A friend of mine did a test to compare several of his bodies because he's an astrophotographer and wanted to find the optimal ISO prior to graphs like this being available.

 

[fmw]

Holding up the D3100 as an example of what "Nikon does" as far as image signal processing is not appropriate in the context of discussion how a camera maker's image processing strategies are currently implemented...that's the issue. Just calling out a bit of shade-throwing that somebody else was trying to do.

I've occasionally shot the old D2x over the past couple of years; inage quality is very "2005". Like the now-old D2x, the five years newer D3100 has easily been bettered; the current D3400 is a vastly better iange maker from a dynamic range point of view. As far as what "Nikon deoes" today, let's look to their D7200, which represents an actual and CURRENT Nikon camera. In the D7200 the dynamic range is 14.6 EV. The now-old D3100 with the same-sized APS-C sensor has 11.3 EV of dynamic range; that is a HUGE improvement, of three and one-third MORE stops's worth of DR.

In terms of overall sensor score, the D3100 is at 67, the D7200 is at 87; right up there in what used to be "full-frame" territory as far as technical image quality.View attachment 144776

A sensor score of 67 is now considered poor in Nikon circles; that score and that bad dynamic range is more like m4/3 mini-sensor territory. An improvement of 3.3 EV in DR is major; holding up the D3100 as an example of modern sensors or signal processing as done by Nikon "today" is disingenuous, and that's what I wanted to call attention to.

Ohhhhh--and the D7200 is for all practical purposes, ISO Invariant.

So what? Can you make great images with it or not? Isn't that what matters?