ISO50 - why?

Started Aug 23, 2009 | Discussions
jerrytlr New Member • Posts: 23
ISO50 - why?

Can somebody give me some pointers as to when I should be using ISO50 on my DP1 or SD14? I assumed it would give me higher dynamic ranges for shots with difficult lighting conditions but so far it hasn't really seemed to make much difference... I suspect my use of ISO50 is wrong and I need some education!!

I generally use 100 and occasionally 200 or even 400 - where the results I get are as I would expect (higher noise) - so I think I sort of understand what is going on...

Cheers

Jerry

mike earussi Veteran Member • Posts: 8,577
Re: ISO50 - why?

ISO 50, by giving you one more stop of exposure, makes the sky and shadow areas less noisy, but it can also have the effect of lowering your dynamic range in the highlights making them easier to blow out. I use it a lot if I know that I will be using + fill in SPP as that function lightens the shadows and lowers the highlights. But lightening the shadows also has the side effect of making whatever noise is in them more visible, hence the need for the one stop extra exposure for the shadow areas to cut down on potential noise. The extra one stop exposure also makes the midtones smoother and slightly increases resoultion. So ISO 50, if used correctly, is a very good ISO, in fact my favorite, but it does take some practice to use correctly.

MOD Kendall Helmstetter Gelner Forum Pro • Posts: 20,586
Re: ISO50 - why?

jerrytlr wrote:

Can somebody give me some pointers as to when I should be using ISO50 on my DP1 or SD14? I assumed it would give me higher dynamic ranges for shots with difficult lighting conditions but so far it hasn't really seemed to make much difference... I suspect my use of ISO50 is wrong and I need some education!!

Actually it's slightly lower, as highlights blow out a little more easily in ISO 50.

I use it for cleaner skies, or sometimes for colorful subjects if the scene contrast is low.

Or of course to drag the shutter for something like water or motion blur...

 Kendall Helmstetter Gelner's gear list:Kendall Helmstetter Gelner's gear list
Sigma 70-200mm F2.8 EX DG OS HSM Sigma 85mm F1.4 EX DG HSM Sigma 1.4x EX DG Tele Converter Sigma 24-70mm F2.8 EX DG HSM Sigma 8-16mm F4.5-5.6 DC HSM +4 more
Roland Karlsson Forum Pro • Posts: 28,650
Re: Over exposure

ISO 50 is actually a one stop over exposure that (probably digitally) is lowered one stop in the file. This means that you have more detected photons -- i.e. less noise - but at the same time lower dynamic range as the highlights clips easier.

-- hide signature --
 Roland Karlsson's gear list:Roland Karlsson's gear list
Sony RX100 III Sigma DP3 Merrill Sigma dp2 Quattro Pentax K-3 Pentax K-1 +14 more
OP jerrytlr New Member • Posts: 23
Re: Over exposure

Thanks for that very helpful explanation - my assumption was always that lower ISO meant less gain on the detector chip so less noise but lower sensitivity - but it would appear that it is a software simulation of ISO50 rather than altering the detector's electrical characteristics at all - which I guess makes sense given that ISO50 became available as a result of a firmware upgrade!

So to check my understanding - ISO50 would be useful where you have large areas of low light in the image where noise might become a problem, the sacrifice being a risk of overexposure in the bright areas of the image - so it is of most use with a relatively 'flat' image without extremes of dark and bright areas.

Final question then - is ISO50 effectively the same as shooting at ISO100 in RAW format, but overexposing by +1 stop, then using SPP to adjust the exposure back down?

Cheers

Jerry

mike earussi Veteran Member • Posts: 8,577
Re: Over exposure

The Foveon chip has only one base ISO, after that gain is applied in camera. Though Sigma doesn't officially list what that base iso is, based on the experiments of several photographers here at the Sigma forum we have pretty much agreeded that it is most likely around iso 64. So when shooting at iso 50 the camera is probably attenuating the signal a little. Therefore, shooting at anything higher than 64 gain would be applied.

Getting the best combination of low noise with the best dynamic has been an ongoing experiment for many of us, for instance, my latest experiments seem to indicate that the best way of achieving both is to set the camera iso at 50 and then underexpose by one stop. That way the gain is being applied by the more sophisticated SPP software in the computer than the, by necessity, simpler software in the camera. Later, after I've done more tests, I'll post my results so everyone can judge for themselves.

mike earussi Veteran Member • Posts: 8,577
Re: ISO50 - why?

To illustrate my last point, here are a couple of shots. The first is at iso 100 and the second is at iso 50 with one stop of exposure added in SPP. Both exposures are the same and taken seconds apart. You can see that the second shot has lower contrast, especially in the highlights. For the sake of clarity these are just the straight X3F files with no further processing of any kind.

Full size here:

http://www.pbase.com/mikeearussi/image/116431024

http://www.pbase.com/mikeearussi/image/116431023

SandyF Forum Pro • Posts: 14,941
link: Sigma ISO50 text
 SandyF's gear list:SandyF's gear list
Sigma DP1 Sigma DP2 Sigma DP2 Merrill Sigma DP3 Merrill Sigma DP1 Merrill +7 more
arny978 Forum Member • Posts: 75
Re: ISO50 - why?
 arny978's gear list:arny978's gear list
Fujifilm X-T20 Fujifilm X-Pro1 Fujifilm X-E2 Samyang 50mm F1.2 Fujifilm XF 23mm F1.4 R +4 more
Ed_S
Ed_S Veteran Member • Posts: 8,078
Re: ISO50 - why?

Jerry,

I don't use ISO 50 extensively but have used it occasionally - often on a tripod - where I'm looking for low noise, improved color, etc. The advice to this point IMHO is very good. The native ISO of the Foveon chip seems to be no higher than 100 and perhaps a bit lower as one or more have suggested. Sigma for whatever reason chose to implement ISO in the post processing - kind of like pushing film - vs. modifying the characteristics of the sensor electronically in camera. For that reason, ISO 50 seems to uniquely (among available ISOs) have limited headroom. Conversely, all ISOs of 100 or above give a bit more headroom but with the potential for increased noise as ISO increases. That said, we've seen excellent examples of low noise/higher ISO images from some contributors.

Using ISO 50 has been likened to using an excellent lower ISO color transparency film. With appropriate attention to exposure and post-processing, the results can be quite good. Not an exceptional example, but one in indirect window light I had handy.

http://www.pbase.com/ecsquires/image/92540113

and one outdoor oldie

http://www.pbase.com/ecsquires/image/87757303

Enjoy experimenting with it,
--
Ed_S
http://www.pbase.com/ecsquires

Serhan2 Senior Member • Posts: 1,357
Re: ISO50 - why?

Love ISO 50 for clean look and extra stop gives for those slow shutter speeds but I think dynamic range suffers when you switch to ISO 50 or 100 for cameras with ISO 100/200 as a base ISO. Here is an example shot with DP1:

3s f/10.0 at 16.6mm iso50

Serhan
http://www.pbase.com/sc_20170/

Lucas Jarvis Regular Member • Posts: 475
Re: Over exposure

mike earussi wrote:

several photographers here at the Sigma forum we have pretty much agreeded that it is most likely around iso 64. So when shooting at iso 50 the camera is probably attenuating the signal a little. Therefore, shooting at anything higher than 64 gain would be applied.

That's an interesting bit of information. I always thought that 100 ISO would have been the base ISO with no gain added. Do they make it around 64 ISO so that there is a bit of highlight headroom at ISO 100?

You think they would include the base iso as an option, say ISO 64 instead of ISO 50. That way you can expose to the right and get the absolute max out of the shot without clipping.

-- hide signature --

Regards,
Lucas

Lucas Jarvis Regular Member • Posts: 475
Re: ISO50 - why?

mike earussi wrote:

To illustrate my last point, here are a couple of shots. The first is at iso 100 and the second is at iso 50 with one stop of exposure added in SPP. Both exposures are the same and taken seconds apart. You can see that the second shot has lower contrast, especially in the highlights. For the sake of clarity these are just the straight X3F files with no further processing of any kind.

I have had a look at the high res. They look very close. The white balance looks a little different but that doesn't really matter to illustrate your point. The highlights look clipped at equal locations. With the RAWS, if you put on your highlight clipping warnings and slide the exposure bars up, do they clip at exactly one stop from each other or does one give you a bit more room?

I really think that ISO 50 at -1 is the same as ISO 100, but your posts do show there is a slight difference. The difference may only be how it maps contrast. Do you have any information as to whether there is a noise advantage (Probably the main reason to use ISO 50) or a highlight clip advantage?

Regards,
Lucas

Lucas Jarvis Regular Member • Posts: 475
Re: Over exposure

jerrytlr wrote:

Thanks for that very helpful explanation - my assumption was always that lower ISO meant less gain on the detector chip so less noise but lower sensitivity - but it would appear that it is a software simulation of ISO50 rather than altering the detector's electrical characteristics at all - which I guess makes sense given that ISO50 became available as a result of a firmware upgrade!

So to check my understanding - ISO50 would be useful where you have large areas of low light in the image where noise might become a problem, the sacrifice being a risk of overexposure in the bright areas of the image - so it is of most use with a relatively 'flat' image without extremes of dark and bright areas.

Final question then - is ISO50 effectively the same as shooting at ISO100 in RAW format, but overexposing by +1 stop, then using SPP to adjust the exposure back down?

Cheers

Jerry

It sounds like you have a pretty good grasp on it. What you have said is my understanding as well. I believe that overexposing ISO 100 by +1 is the same as ISO 50, but it's been stated later in the post that ISO 100 may actually have some gain added to it, while ISO 50 hasn't. Using SPP on the ISO 50 shot may have an advantage over in camera gain to equal the shots out. I struggle to see much grain at these low ISO's so I think the advantage of one over the other may be minimal/trivial.

-- hide signature --

Regards,
Lucas

mike earussi Veteran Member • Posts: 8,577
Re: ISO50 - why?

Lucas Jarvis wrote:

mike earussi wrote:

To illustrate my last point, here are a couple of shots. The first is at iso 100 and the second is at iso 50 with one stop of exposure added in SPP. Both exposures are the same and taken seconds apart. You can see that the second shot has lower contrast, especially in the highlights. For the sake of clarity these are just the straight X3F files with no further processing of any kind.

I have had a look at the high res. They look very close. The white balance looks a little different but that doesn't really matter to illustrate your point. The highlights look clipped at equal locations. With the RAWS, if you put on your highlight clipping warnings and slide the exposure bars up, do they clip at exactly one stop from each other or does one give you a bit more room?

I really think that ISO 50 at -1 is the same as ISO 100, but your posts do show there is a slight difference. The difference may only be how it maps contrast. Do you have any information as to whether there is a noise advantage (Probably the main reason to use ISO 50) or a highlight clip advantage?

Regards,
Lucas

There seems to be slightly less noise in the iso 50 shot, especially in the shadows, but you really have to look for it. My purpose was just to see if I could extend the dynamic range a little in the highlights. If you compare the roofs you'll see that the iso 50 shot has a little lower contrast and slightly more detail that the straight iso 100 shot. But there isn't a huge difference, so for most purposes there probably isn't any reason to shoot at iso 50 -1 to equal 100.

Lucas Jarvis Regular Member • Posts: 475
Re: ISO50 - why?

The noise issue really comes into play when shots are severely underexposed. It might be a good experiment to then compare an ISO 50 shot at -4 to an ISO 100 shot at -3. Then pull them up and see if the shadow noise is equal. This should be a good indication if there is a big difference between the two.
--
Regards,
Lucas

mike earussi Veteran Member • Posts: 8,577
Re: ISO50 - why?

I also tried a two stop push and it didn't work very well, it was actually more contrasty. My main use in 50 is when I want a grainless sky or plan on using the + fill option which can really make the shadows look really grainy if shot at 100 and up. The best thing for you to do is your own experiments and then you'll start to get a feel for when it's best to use 50.

dlj Regular Member • Posts: 303
Re: ISO50 - why?

ISO50 is my 'default' or starter ISO, if I can get away with it. I find it is a little compressed for dynamic range, and IMHO (no truly systematic tests here) it has slightly lower noise than ISO100. It is very much like transparency film. In super low-light shots (available light night shots, where long exposures permit), the smoothness of pixel-to-pixel colour consistency is like butter, and really suits up-sampling if required. I just love it. Yet it almost seems better at capturing the most subtle and accurate colour variations in an otherwise 'solid' background, but only when there should be some degree of variation. Not stemming from noise or some demosaicing artifact.

Now & then in post processing, I'll question myself as to whether I would have been better off with the speck wider dynamic range, but the exposure range you do get with ISO50 never disappoints. I hope that offers a somewhat specific response, without getting too airy-fairy about the whole opinion.

spalbird Senior Member • Posts: 2,493
Re: Headroom, but not dyn range.

Roland Karlsson wrote:

ISO 50 is actually a one stop over exposure that (probably digitally) is lowered one stop in the file. This means that you have more detected photons -- i.e. less noise - but at the same time lower dynamic range as the highlights clips easier.

Roland, the dynamic range will not chance with your explaination. Dependent on the levels the dynamic range is defined, e.g. from clipping level of the first color in the color triple down to a value at a certain noise level, this will not change with overexposure.

Overexposure of one stop will decrease the headroom by one stop. But what do you understand by overexposure? For maximum dynamic range the highlight should be just below clipping level for correct exposure. Overexposure by one stop in this case will clip all highlights down to -1 stop by optimal exposure.

The correct explaination should be:

ISO50 just effects the metering. The metering is done in a way that the exposure is done 1 stop closer to the clipping level compared with ISO100.

Best regards
wolfgang

-- hide signature --

...conservative is not the opposite of progressive...

 spalbird's gear list:spalbird's gear list
Sigma DP1s Sigma SD10 Sigma SD14 Sigma 10-20mm F4-5.6 EX DC HSM Sigma 18-125mm F3.8-5.6 DC HSM +6 more
Roland Karlsson Forum Pro • Posts: 28,650
Re: dyn range!

spalbird wrote:

The correct explaination should be:

ISO50 just effects the metering. The metering is done in a way that the exposure is done 1 stop closer to the clipping level compared with ISO100.

No - it does not only affect the metering. It also shifts the resulting values one stop to get an image with correct brightness, i.e. it makes the image one stop darker. This will remove one bit from the available dynamic range.

If you expose more than base ISO you must always get lower dynamic range.

-- hide signature --
 Roland Karlsson's gear list:Roland Karlsson's gear list
Sony RX100 III Sigma DP3 Merrill Sigma dp2 Quattro Pentax K-3 Pentax K-1 +14 more
Keyboard shortcuts:
FForum MMy threads