04-22-2012, 06:38 PM
I didn't really acknowledge that "only base iso boost" trend before your reply. So I give you a point for the incorrect assumption but don't go assuming solely based on the D800.
I thought and I still think that progress in this domain is going to pull the curve up, thus actually providing an operational benefit beyond the "spec jockey".
Even though, the D800 seems to be converging to 5D mk III, I invite you to do more comparisons. 5D vs 5D II vs 5D III.
You'll see other cases :
http://www.dxomark.com/index.php/Cameras...2%29/Canon
Between 5DmkI & 5DmkII : parallel(okay, curves can't be parallel) progress -> this is what I'm talking about. .
Betweek mkII & mkIII : progress in higher ISO. Surely you won't say 1 stop is insignificant.
So I'm seeing something different than you just by looking at different cameras comparisons. I'll concede the exercise is probably more significant with same generation cameras but that's only valid assuming sensor designers would all follow the same trends?
Another interesting thing, even on my "assumption", you still have until ISO 800 to converge with the 5D mkII. You really read it *your* way : "it gets on par" when I read "I'll have an advantage until ISO 800". A case of half-full vs. half-empty glass I guess <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='
' />.
My mistake for talking in undefined terms but the general idea is that progress on that matter isn't only a "spec jockey" advantage and that it's going to provide operational advantages, even now.
Your interpretation of my auto iso is wrong, I think. I'll favour an automatic ISO change regardless of the DR reduction, not the other way around. So if DR has improved on the whole iso range, it's bonus. That's all.
[quote name='Brightcolours' timestamp='1335077752' post='17684']
The thing is that the very high base ISO does NOT bring much better DR at higher ISO. There, at high ISO, the advantage disappears.
When you for instance actually look at high ISO results from the 18mp Canon APS-C sensor and the high ISO results from the 16mp Sony APS-C sensor used in the D7000/K5, you will see they are basically op par.
Similar to the 5D mk III and D800... the 5D mk III seems to have nicer high ISO results at very high ISO's and results that are basically on par with medium settings.
So we are only Talking about lower read noise here, not the noise introduced by lower light capture with higher gain. As I said before, if the high DR range from base ISO would be that important, no one would consider using ISO 400 and ISO 800, let alone ISO 3200. But, in reality, we DO consider with with sensors like the D800 and D4 have. Why? Because the results just are so good that we just can. Just to illustrate how the 14EV DR is more of a spec list thing than anything else.
Just to be clear... at ISO 800 the 5D mk III is on par with the D800.
http://www.dxomark.com/index.php/Cameras...kon#tabs-2
http://home.comcast.net/~NikonD70/Charts/PDR.htm
So, what you are assuming is wrong on two points then... You assume the higher DR difference remains the same over the entire ISO range.. and this is a false assumption. Only the read noise is making the difference, so only at base ISO and a bit above. And when you want to have ISO to be totally "auto" and transparent, this then means that you have no control over the higher "DR advantage" as you only have that in the lower ISO settings.
In reality, these Canon sensors (22mp FF, 18mp APS-C), Sony sensors (36mp FF, 16mp APS-C) and Nikon sensors (16mp FF) are very close in performance, noise wise, when you step up the ISO setting. The bit lower DR at base ISO really is not that big a deal, with no real impact on real photography.
To place things into perspective, positive slide film has 6EV DR at best. Colour negative film has 10EV DR with the better films. BW negative has about 14 EV for the top films.
And that is just looking at the film,not about the different photo papers and how they reacted/compressed DR.
[/quote]
I thought and I still think that progress in this domain is going to pull the curve up, thus actually providing an operational benefit beyond the "spec jockey".
Even though, the D800 seems to be converging to 5D mk III, I invite you to do more comparisons. 5D vs 5D II vs 5D III.
You'll see other cases :
http://www.dxomark.com/index.php/Cameras...2%29/Canon
Between 5DmkI & 5DmkII : parallel(okay, curves can't be parallel) progress -> this is what I'm talking about. .
Betweek mkII & mkIII : progress in higher ISO. Surely you won't say 1 stop is insignificant.
So I'm seeing something different than you just by looking at different cameras comparisons. I'll concede the exercise is probably more significant with same generation cameras but that's only valid assuming sensor designers would all follow the same trends?
Another interesting thing, even on my "assumption", you still have until ISO 800 to converge with the 5D mkII. You really read it *your* way : "it gets on par" when I read "I'll have an advantage until ISO 800". A case of half-full vs. half-empty glass I guess <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='
' />.My mistake for talking in undefined terms but the general idea is that progress on that matter isn't only a "spec jockey" advantage and that it's going to provide operational advantages, even now.
Your interpretation of my auto iso is wrong, I think. I'll favour an automatic ISO change regardless of the DR reduction, not the other way around. So if DR has improved on the whole iso range, it's bonus. That's all.
[quote name='Brightcolours' timestamp='1335077752' post='17684']
The thing is that the very high base ISO does NOT bring much better DR at higher ISO. There, at high ISO, the advantage disappears.
When you for instance actually look at high ISO results from the 18mp Canon APS-C sensor and the high ISO results from the 16mp Sony APS-C sensor used in the D7000/K5, you will see they are basically op par.
Similar to the 5D mk III and D800... the 5D mk III seems to have nicer high ISO results at very high ISO's and results that are basically on par with medium settings.
So we are only Talking about lower read noise here, not the noise introduced by lower light capture with higher gain. As I said before, if the high DR range from base ISO would be that important, no one would consider using ISO 400 and ISO 800, let alone ISO 3200. But, in reality, we DO consider with with sensors like the D800 and D4 have. Why? Because the results just are so good that we just can. Just to illustrate how the 14EV DR is more of a spec list thing than anything else.
Just to be clear... at ISO 800 the 5D mk III is on par with the D800.
http://www.dxomark.com/index.php/Cameras...kon#tabs-2
http://home.comcast.net/~NikonD70/Charts/PDR.htm
So, what you are assuming is wrong on two points then... You assume the higher DR difference remains the same over the entire ISO range.. and this is a false assumption. Only the read noise is making the difference, so only at base ISO and a bit above. And when you want to have ISO to be totally "auto" and transparent, this then means that you have no control over the higher "DR advantage" as you only have that in the lower ISO settings.
In reality, these Canon sensors (22mp FF, 18mp APS-C), Sony sensors (36mp FF, 16mp APS-C) and Nikon sensors (16mp FF) are very close in performance, noise wise, when you step up the ISO setting. The bit lower DR at base ISO really is not that big a deal, with no real impact on real photography.
To place things into perspective, positive slide film has 6EV DR at best. Colour negative film has 10EV DR with the better films. BW negative has about 14 EV for the top films.
And that is just looking at the film,not about the different photo papers and how they reacted/compressed DR.
[/quote]