Still subject : High iso little speed or Low iso long exposure ?

[steflaurent974]

This may seem a stupid question. But I'd like to know...technically if I choose low ISO in natural light I will go for 1s to severall seconds of exposure regardless the choosen aperture.
:lecture:
Low Iso gives me less noise but long exposure times brings the noise too, am I wrong ?

So wich is the best choice , I can imagine a curve "amount of noise/exposure time" against the curve LOW ISO/HIGH ISO .
Is there an exposure time limit , over this limit it's better to go for higher Iso.

Assuming that I'm working with a P25+, I don't have much choice for exploitable high ISO 400 would be the maximum...

These are my reflexion and questions...do I miss something ?

Stephane.

 

[Stuart Richardson]

For Sinar at least, the official answer is to use the lowest ISO up to the maximum time of the back (30 seconds for sinar), and then increase ISO. I don't think there is much replacement for sheer amounts of light hitting the sensor with these backs -- the more light, the better, so all things being equal, the longer the exposure at the base ISO, the better.

 

[steflaurent974]

My error was to believe that the more the exposure the more noise you get... thank you.

 

[carstenw]

I believe that you are right in that, otherwise all backs would allow much longer exposures. However, the noise seems to increase much slower with time than with ISO.

 

[Stuart Richardson]

Well, I think any ISO other than base ISO is more or less a software amplification -- just taking the existing data and multiplying the luminance while trying to stifle the noise -- I don't think most of the different ISO's in digital backs are different hardware sensitivities, but of course I could be completely wrong. But in my photographic experience, I find that longer exposures at lower ISO's look better than short exposures at high ISO, at least assuming a tripod and a still subject.

Another advantage of a longer exposure versus a high ISO is that the subsequent dark frame is probably more likely to eliminate noise than simply a shot at a higher ISO.

 

[carstenw]

IIRC from Panopeeper's posts on LL, most backs don't have different ISOs, but some do, like the P45+.

 

[Guy Mancuso]

Just a very extreme example here at 32 minute exposure , now all the Phase backs take a dark frame which is equal to the exposure to help eliminate noise( 64 minutes total time). This is at ISO 100 which is the lowest for the P30+. Now this is extreme and there is a little noise but if I went a higher ISO and maybe a little shorter time I would actually get more noise even with a shorter time. In general pretty safe to say the higher the ISO the higher the noise factor regardless of time. Most backs will do at least 30 seconds minimum. Now the second shot of the Lake is ISO 100 at 8 seconds and not even a sign of noise. So in general there is a time point where noise will start to play on those very long exposure times but my experience up to 30 seconds or longer with some backs the time will not be a factor until you start doing stuff like the first image. So you have a pretty big window here

 

[carstenw]

So in summary it means that for many backs, high ISO with short time is actually a low ISO with short time, pulled up in (in-back) post. No wonder there is more noise.

 

[Paratom]

So in summary it means that for many backs, high ISO with short time is actually a low ISO with short time, pulled up in (in-back) post. No wonder there is more noise.

I think at least our backs Carsten. However IMO the final result (noise) is more important than the question how the result is generated.

 

[steflaurent974]

Well, on my P25+ (the one you own before Guy ?) in many case I wasn't happy with the 400iso ...Even on the little LCD of the back it looks crappy (maybe capture One has to reduce this...)

If no clouds tonight I'll try an astrophoto shot (startrail) at 50 iso and I'll see.

Carsten is right it really seems that the higher Isos are only algorythms to produce .....noise

 

[carstenw]

Stephane, the results look at *lot* worse on the back of my back (heh), than on my laptop, so don't judge by your back. I have a Sinar eMotion 54 LV, but I expect the current Phase backs are similar in that regard.

 

[Guy Mancuso]

Yes don't go by the LCD on really any of the Phase backs when doing these very long exposures it's a noisy mess on the LCD which is not the case for real in C1. The P25 + should be able to handle 30 minutes. I know the P45+ can do the whole hour since I seen it in real life but don't quote me and maybe Doug can fill in the safe zone for the P25+ and P30+ is 30 minutes. But do not go by the LCD. Nice little trick if you can do it. Is take a quick shot at the highest ISO for a sort time see where that exposure is than start doing the math for the ISO 100 shot . Bring that 1600 or 800 in C1 on location than go from there. But you are correct ISO 400 on long exposures is going to be somewhat noisy use ISO 50 or 100 on the P25+ better off going more time than higher ISO. Try to stay within the 30 minute window for the P25+ it may do longer but I think the results are better shorter than 30 minutes otherwise the noise may start in. Like I mentioned you may exceed that noise window going longer

 

[Guy Mancuso]

Stephane, the results look at *lot* worse on the back of my back (heh), than on my laptop, so don't judge by your back. I have a Sinar eMotion 54 LV, but I expect the current Phase backs are similar in that regard.

Exactly , flat out horrible on the LCD. Do not go by the look but just use as a exposure guide

 

[Jack]

Back to Stepahane's original post, I'll offer a simple answer: Test it.

The test should be simple to do indoors, making sure some shadow areas are in teh image. Shoot some higher ISO under 1 second exposures and compare them to the same image taken at lower ISO multiple second exposures. Compare them side-by-side and you will have your answer for YOUR back. Probably the best way to know for certain...

I can tell you that with my P45+ back -- and it is different than other backs in this regard -- is that if I use ISO 100, there is very little visible noise even up to 30 second exposures. Even ISO 200 noise is usably low at 30 seconds. ISO 400 at 30 seconds is visible, but not objectionable. Any of these ISO's look like maybe one or 1-1/2 ISO higher at 30 MINUTE exposures when compared to the 30 second exposure....

 

[Dale Allyn]

Jack's suggestion is important because backs (models and makes) will vary, but also because we all live in different locations where differing ambient temperatures have an effect on our individual results vis-a-vis noise.

I, too, have a P25+, and I understand that the ISO is actually more or less fixed such that the change in settings are a software push. Like Stephane, I tend to use up to ISO 400, but prefer to use 50 or 100. Jack's P45+ is said to have "real ISO" adjustments at least for some levels like ISO 400. This is also why Jack's suggestion is important – we each must test.

Finally, this gets mentioned occasionally, but I feel that it gets glossed over in some discussions: some noise in an image when viewed on screen can be acceptable in the final print (if that is your end goal). Noise on screen does not necessarily mean noise in the print. Printers dither this data differently and in some images it actually adds to the image IMO. This all depends on your final output requirements.

 

[gogopix]

Testing makes sense, esp when there are so many factors.

However, from a practical POV, lower ISO longer time should always be better. That is because the 'signal' the light from outside is not where the noise is coming from (in radio eg noise is a combo of thunderstorms, random radiation and the broadcast).
In sensors, the noise is more like the random electronic noise of a radio; if you increase the volume where there is a weak braodcast, the noise gets louder.
In a long exposure you have something called 'coherent processing, that is the image gets better and better IN THE SAME PLACE, while the noise jumps around. Also the noise is NOT amplified (that happens at higher ISO) but is spread. Ultimately noise may get less, as it spreads randomly all over the image while the image builds and builds over time.

Victor

 

[Guy Mancuso]

Dale you brought up a extremely good point HEAT. The colder the ambient the better the noise is the general theme or maybe better stated the longer your exposure can go. Heat can be a big issue. My shot above was in Monument Valley and from memory about 60 degree's. Here in the Phoenix area at night I'm lucky to dip below 80 and that would cause me some issues with the really long exposures as the sensor heats up the noise is induced. Great point Dale