I think we need a direct head-to-head comparison:
D800 with a good wide-angle lens (say the Zeiss 21?), vs a 5D3 with TS-E 24, stitched to the same field of view and same 36MP.
I'll volunteer, anyone want to post me your cameras and lenses? (I promise i'll post them back ... eventually)
An awful lot of electrons were terribly inconvenienced in the making of this post.
Gear Photos
I read this post yesterday, and keep wondering the same thing. If you've been waiting for years for Canon to fix the low-ISO problem, why give up now, possibly within days of knowing if the next rev (5DIII and or 1DX) shows improvement?Originally Posted by DavidEccleston
Not very rational, was it? I guess I just got caught up in the emotional aspect of it.
As it turns out, I think it was the right choice. I found some raw files today that confirm the 5D3 still has a significant temporal FPN problem:
http://www.kleptography.com/dl/5diii/raw/02.cr2
http://www.kleptography.com/dl/5diii/raw/
They will open in rawdigger.
Hmmm.... just as I suspected
Which pic(s) show the problem? I looked at pics1, 2, 12, and 13 (low iso pics with lots of shadows) and didn't see any pattern noise. But the colors were all screwed up (I guess I don't know how to use rawdigger). Maybe I'm just not seeing it or looking in the wrong place.
Try 02 with the new 6.7 DNG converter. Here is what comes out of LR4 with default settings, enable lens corrections, and a 2/3 stop push:
And no, that's not supposed to be a pinstripe suit.
Okay, I see what you mean- without the push I didn't notice it. Still, 2/3 stop is pretty modest. I'm guessing with a real camera (like the D800) one could push it off a cliff and not see stripes
That's a real bummer. I guess it is hard to see if there has been any improvement at all without some quantitative measurements.
I just waltzed into Best Buy this morning and bought a D800. I feel sorry for all of the folks that are at the bottom of a pre-order list somewhere -- if the world was fair they would have got theirs before me. From a quick peek at their inventory screen, it appears that each location in the Portland area only got two D800's today.
Let the Canon bashing begin.....
I was more hoping for a low-light suit picture, brightened in post for comparison, but Canon bashing is good too.
I did see a ISO 25600 comparison image taken with 5DmkIII and D800, in the same controlled test setup, full-sized jpegs. The entire ISO series was there, but I only checked the one high ISO. I found a test tool where you can see this yourself here, looking at the still-life set of shots at the bottom. Also, looking again, I'm not sure how controlled the test setup is. The lighting appears to be different. A different shade, and shadows moved. Still, from what I can see, the 5D looks significantly less noisy at 100%, which is expected due to the lower pixel count. However the Nikon noise is fairly severe. Straight edges didn't look quite straight anymore, with lots of splotchyness. See the crayons, and the edges of the color palette squares. It looks like the Nikon would require more post work. I'd rather have the Canon.
Still, both blow away my T1i, and I could gratefully accept either one as a gift.
Are you going to leave us know and start a Nikon site like Bryan's site?LOL
Congrats on the new camera.
Mark
Posting Permissions
Reply With Quote