Question about GH1 ISO Settings

D

dvbrother

Well-known member
I don't believe this has been discussed before.
Setup: I've read about how certain Canon DSLR ISO settings are preferred (160, 320, 640, etc.) vs. the "whole stop" ISO settings like 200, 400, 800, etc. because apparently the latter apply digital gain to create the image and therefore reduce dynamic range.

Question: Does anyone know if the same applies to the GH1? It seems smarter to me that the "native" ISOs would be 100, 200, 400, 800, 1600 as these are the default whole stop settings in the menu.

In terms of the Canon, there seems to be conflicting information, as Philip Bloom and Co. subscribe to the 160, 320 etc. settings as the good ones, but I also read a very detailed convincing blog article stating that 200, 400, 800 etc. were the correct ones. Wow, lots of potential misinformation.
 
Watch the latest zacuto episode, apparently they tested all the cameras for native isos and yes the gh1 does have the same native marks
 
I had been shooting GH1 at the 100, 200, 400, etc...but now I am using the 160 multiples. I have seen enough tests on the Canons to convince me that these are the natives and if they say they ran the tests on the GH1 and found that out, then I believe them..but some clarification would be nice.
 
I'm skeptical that they actually tested the GH1, judging by the video. I guess I may try doing my own tests since I have access to a DSC labs chart.
Another thing that doesn't make sense to me is this: in other tests, it has been shown that the GH1 and the Canon ISO's are not equivalent, exposure-wise. In other words, a GH1 and a Canon 7D set to 400 ISO (shutter and f-stop set the same) will yield differently exposed images. Perhaps even a one-stop difference. But the latest Zacuto test shows the GH1 to look about the same as the others. It's starting to hurt my brain.
 
Upon surfing the net on this subject, I'm convinced that "nobody knows anything". Knowledgeable, reputable people seem to be saying two different things about the "native" ISO's. Some say 160, 320, etc. Others say 200, 400 etc. Both arguments seem to come from the same style of test. Who's right?
 
dvbrother said:
Upon surfing the net on this subject, I'm convinced that "nobody knows anything". Knowledgeable, reputable people seem to be saying two different things about the "native" ISO's. Some say 160, 320, etc. Others say 200, 400 etc. Both arguments seem to come from the same style of test. Who's right?
Click to expand...

For the Canons 160.
 
MicroMacro said:
Watch the latest zacuto episode, apparently they tested all the cameras for native isos and yes the gh1 does have the same native marks
Click to expand...

I was wondering the original posters question as well but in the Zacuto shootout they never reveal these "native" ISO's though do they?
 
Multiples of 160. Or so they claim.

Actually from what I understand about the underlying tech, these aren't actually the native ISOs but rather are being digitally pulled back from native to be cleaner at a cost of dynamic range...as opposed to digitally pushed and made noisier. I don't know the exact numbers but I guess the idea would be that 160 is digitally pulled back from the 200 native, and that something like 250 is digitally pushed.
 
Ben_B said:
Multiples of 160.
Click to expand...

Ah so thats what they refer to, I actually thought they meant what the sensor is rated at. Does anyone know whats the optimal ISO to shoot at with the GH1?

Im doing a night shoot in 2 weeks and would like to know whats the ISO I should use for it. Im thinking 320 at the moment but would like to get more info on this.
 
Ben,

You just nailed it for me. I had a Eureka moment that may be obvious to you already. The "native ISOs" ARE 200, 400, etc. in terms of best Dynamic Range and the fact that the increments are analog/sensor -based. The "good" ISOs for lowest noise are the ones right below those because of a digital pad going on. So I think the two camps are approaching the data from two different angles. Philip Bloom and Co. want the lowest noise in the shadows, thus 160, 320, 640 etc. If you want maximum dynamic range regardless of noise, the 200, 400, 800 settings are best. The ISOs to avoid at all costs are the multiples of 125. So 250, 500, 1250 should be avoided because they add digital gain and therefore noise.
Which sucks because I shot most of my short film at 250 ISO. And I DID notice issues in the shadows that I didn't when I shot at ISO 320. It's all becoming clear to me now!
 
dvbrother said:
Ben,

You just nailed it for me. I had a Eureka moment that may be obvious to you already. The "native ISOs" ARE 200, 400, etc. in terms of best Dynamic Range and the fact that the increments are analog/sensor -based. The "good" ISOs for lowest noise are the ones right below those because of a digital pad going on. So I think the two camps are approaching the data from two different angles. Philip Bloom and Co. want the lowest noise in the shadows, thus 160, 320, 640 etc. If you want maximum dynamic range regardless of noise, the 200, 400, 800 settings are best. The ISOs to avoid at all costs are the multiples of 125. So 250, 500, 1250 should be avoided because they add digital gain and therefore noise.
Which sucks because I shot most of my short film at 250 ISO. And I DID notice issues in the shadows that I didn't when I shot at ISO 320. It's all becoming clear to me now!
Click to expand...

Yup! Actually 1250 and its multiples are fine I think because, like with shutter speed, the numbering multiples change partway through. 1250 is the next in line after 640 because 1280 isn't there. It's digital pull from 1600 or something.
 
"Actually 1250 and its multiples are fine I think because, like with shutter speed, the numbering multiples change partway through. 1250 is the next in line after 640 because 1280 isn't there. It's digital pull from 1600 or something."

Good to know. I'll rarely be in those ISOs anyway, as I usually have some form of lighting. Or a fast lens. :-)
 
Most of this is with research I did cause of a shoot with the 5d...never have had to push the gh1 that high; its 400 seems like canon 1250.
 
Ben_B said:
Most of this is with research I did cause of a shoot with the 5d...never have had to push the gh1 that high; its 400 seems like canon 1250.
Click to expand...

Is that true? I remember back in the day (i.e. last summer) Illya Friedman said to rate a light meter at ISO 2500 when shooting at ISO 400 on the GH1. I thought he was insane. Do you think this is related? FWIW, my Sekonic light meter is fairly close to my GH1's meter. I wonder if there's a lot of variance in the sensitivity between individual GH1s. What I mean is: maybe MY GH1 is less sensitive than your GH1, and this variance exists among GH1 users.
 
actually I think Illya was right. there's a thread on here that showed the the Gh1's ISO ratings. it's buried under all the "should I get a Gh1 or canon camera threads"
 
Car3o said:
actually I think Illya was right. there's a thread on here that showed the the Gh1's ISO ratings. it's buried under all the "should I get a Gh1 or canon camera threads"
Click to expand...

This is the one thing that has really stumped me. I'm not a guy who uses a light meter all the time (I come from the video world, not the film world), but I think I'm fairly smart. With my GH1, when I shoot a gray card and have the histogram and exposure meter indicate proper exposure, my Sekonic light meter is within 1/3 stop of reading the same thing. Certainly not 2 and 2/3 stops difference like Illya would recommend. On this one, I have to defer to my own empirical results. That's why I wonder if different GH1's could have different sensitivities. I suppose my meter could be the culprit, but I've checked it with other meters, and they show the same thing. Weird.
 
You must log in or register to reply here.
Back
Top