snapsy wrote:
Yep. This same JPEG exercise was done when the 5DM3 came out, with the same conclusions by some that DR was improved, which turned out to be wrong once the raws could be processed. I'm not saying DR hasn't been improved on the 6D, just that we can't tell if that's the case from JPEGs.
Good to know, thanks
Ah... No wonder there's a heavy posterization. This was pushed from a JPG? The image may look ok visually but the histogram probably looks like a broken comb.
As for the sharpness of those ISO8000 images, I think they captured enough details considering the high ISO. Sample it down to 18MP and it'll be more than acceptable.
n0b0 wrote:
Ah... No wonder there's a heavy posterization. This was pushed from a JPG? The image may look ok visually but the histogram probably looks like a broken comb.
As for the sharpness of those ISO8000 images, I think they captured enough details considering the high ISO. Sample it down to 18MP and it'll be more than acceptable.
Nope. Looks completely normal (histogram that is).
Sampling down from 20mp to 18mp isn't going to do too much ^_^
You can't tell anything from this sample unfortunately. 5d3 also looks fine pushing at least this much, in a web sized sample. Really in web sized samples you can push 5d3 2-3 stops or so and be okay. So it's no biggie. In order to really know DR we need some black frames at various ISOs and we can figure out if it is actually improved ( i bet no ).
someone needs to remove the lens, put on the body cap, go into a pitch black room, set 1/8000th shutter, set ISO100, set RAW format (turn off all the vignetting and other settings), snap a black frame and then snap another frame at ISO100 in RAW with all the whatevers turned off and exposed so that a good chunk of the scene is absolutely utterly blown out in all three channels beyond any doubt. Both shots are pretty easy to manage .
and then post the RAW files and the DR can quickly, easily and reliably compared to the other cameras boom just like that. it is that simple. you'll have real solid data in 30 seconds as opposed to spending hours trying to read tea leaves as is being done.
Gochugogi wrote:
I can tell that Canon improved shadow and low mids a lot. No way I can lift the lows that much with my 5D2 and not see gritty noise and banding (not that I normally want or need to do so).
skibum5 wrote:
and then post the RAW files and the DR can quickly, easily and reliably compared to the other cameras boom just like that. it is that simple. you'll have real solid data in 30 seconds as opposed to spending hours trying to read tea leaves as is being done.
Does not happen very often. But hear I am with you, "bummi"
My local dealer here in Portland, OR is scheduled to get them in Monday 12/3.
Unless you have LightRoom 4 or DPP and a 6D shooting RAW files we will never know what it is capable of.
So if anyone wants to send me $2100.00 to get one I'll be glad to pick up one and do some serious testing with it. Oh and yes - I'll eventually send you the camera.
Sigh... year after year, camera after camera... we have all been hoping for improved sensor performance from Canon... it's almost hard to believe how hard the king has fallen, the king who once had no equal in the realm of sensor performance...
Let's hope the 6D does not disappoint. The 6D brochure mentions new photodiode design and microlens array to improve signal to noise as well as dynamic range... mere marketing talk? I hope not.
Ralph Conway wrote:
But I still wait for some (high ISO) raw samples not overdeveloped, noise reduced and oversharpened jpegs, smeared in detail even at ISO 800.
Could somebody purchase this camera and do some shots to post?
I'll save you the suspense - it will likely be identical to the 5DM3's performance
snapsy wrote:
I'll save you the suspense - it will likely be identical to the 5DM3's performance
I got the hunch that you are right, Adam. That will be the end of the world for a lot of people (read: the end of the Canon world for a lot of shooters) . Frankly, I am really enjoying my 5D Mark III .
skibum5 wrote:
well then there is this method, which takes 0.1 seconds to accomplish
hehe. I figure Canon couldn't stomach a sensor that would outperform the 5DM3 in High ISO, let alone the 1DX. Low ISO is a tossup. I'm more interested in why they didn't use the 5DM3 sensor and whether this new sensor comes off the newer rumored fab tech that Chipworks discussed.
snapsy wrote:
hehe. I figure Canon couldn't stomach a sensor that would outperform the 5DM3 in High ISO, let alone the 1DX. Low ISO is a tossup. I'm more interested in why they didn't use the 5DM3 sensor and whether this new sensor comes off the newer rumored fab tech that Chipworks discussed.
It is quite bizarre it doesn't use the 5D3 sensor. I don't think they'd use the new fab on it. 5D3+1DX would go ape if the little came out a few months later and blew those cams away for lower ISO performance I'd think. MAYBE for the 7D2?? Maybe??
I hate it when an upgraded body or lens comes out because it degrades the quality of the previous version(s) it supersedes... rendering the superseded version inadequate for anything other than snapshots unless in the hands of the most capable 1% of photographers...IMHO of course...
.
. Frankly, I am really enjoying my 5D Mark III 