All this is interesting and informative, and maybe this is for another forum, and I would love to have a D800 or D800E, but after owning a D700 and selling it (dissatisfied with the 12 MP) , I keep hearing that Sony is leading in the DSLR arena now. Maybe it's time to think in another direction, even though I have several FX Nikkor lenses? Is Nikon losing it, and all this is moot?
If you understand why you were dissatisfied with 12 mpx, that knowledge will guide you to your next decision.
I have 36, 24, 18 and 12 mpx available to me and use the 12 at least as often as the others.
Nobody is 'leading' in the DSLR space (certainly not SONY, pellicle mirrors 'lose' 1/3 of the light, and if all are using the same laws of physics, is a disadvantage that simply cannot be overcome).
Nikon and Canon are so competitive, and overlap / leapfrog one another, while individual bodies and lenses can be compared, trying to decide if one brand is 'ahead' of the other is a fool's errand.
.... H
D810, D3x, 14-24/2.8, 50/1.4D, 24-70/2.8, 24-120/4 VR, 70-200/2.8 VR1, 80-400 G, 200-400/4 VR1, 400/2.8 ED VR G, 105/2 DC, 17-55/2.8. Nikon N90s, F100, F, lots of Leica M digital and film stuff.
@Harold; Then I take it you think "mirror-less is not the future? Also, I do have my better images on PAD, and I tend to do some cropping that higher pixels would help. I've seen how much you can digital zoom or crop from a D800 image before "pixelating". Sometimes my 70-300 isn't enough and I would like to crop during PP with PhotoShop. I did like most of the images I got from the D700, but I guess I expected too much just because it was FF. I probably out-grew my Minolta SRT101, but am still learning the D90. Msmoto, seeing we're from the same creation date (42), I must be old also, and therefore agree!"
Actually, I think mirrorless is the future, it just is not the present where high performance is needed. There is no fundamental reason for a swinging mirror in a digital camera.
What mirrorless needs to replace DSLR's is:
Full (24 X 36 mm) frame sensors Better and faster EVF's Better and faster autofocus Phase detect autofocus to allow adaption of SLR lenses. Electronic shutter with no rolling shutter effect.
Mirrorless will be smaller, lighter, have fewert moving parts, and not have the minimum back/retro focus limitation for design of lenses with a focal length less than 58 mm, as current FF SLR's do.
... H
D810, D3x, 14-24/2.8, 50/1.4D, 24-70/2.8, 24-120/4 VR, 70-200/2.8 VR1, 80-400 G, 200-400/4 VR1, 400/2.8 ED VR G, 105/2 DC, 17-55/2.8. Nikon N90s, F100, F, lots of Leica M digital and film stuff.
- There's no technical reason why a global shutter cannot be implemented in CMOS sensors. In fact some CMOS sensors already have them. - Even without a global shutter, there's no technical reason a CMOS rolling shutter speed cannot be fast enough to minimize artifacts for all but extremely high speed movements.
It's all about $$$... as technology continues to improve, eventually a CMOS global shutter or a very fast rolling shutter will be affordable and "trickle down" to consumer devices.
So I was covering a charity fashion launch party this week with my D800E, and to my surprise, moire!!
Actually moire on two spots. Capture NX2 cleaned up one area with ease but not the second.
Here's the image post NX2. Can you see where the two moire spots are?
(MUA, model and the fashion designer, left to right)
This scene has been the only one from about 1,000 pictures I've taken so far with any hint of moire. And maybe 950 of those 1,000 frames were fashion event shots with lots of different kinds of fine fabrics.
I'm actually impressed how well NX2 cleaned up the moire. Even if not 100% clean, the image is still very useable.
Absent a 100% image, it is difficult (possibly impossible) to differentiate moire in the image from aliasing of your display. I am not seeing any in this image. I would suspect the fine checker scarf first but I do not see it at this magnification.
D810, D3x, 14-24/2.8, 50/1.4D, 24-70/2.8, 24-120/4 VR, 70-200/2.8 VR1, 80-400 G, 200-400/4 VR1, 400/2.8 ED VR G, 105/2 DC, 17-55/2.8. Nikon N90s, F100, F, lots of Leica M digital and film stuff.
Before Capture NX2's moire removal, false color artifacts were clearly visible on the fashion designer's left shoulder, and to a lesser extent near his suit pocket + left arm.
Shoulder at 100%
Pocket + Left Arm at 100%
I use Aperture (which doesn't have good moire control) so for this one frame the workflow was:
NEF -> Capture NX2 -> TIFF -> Aperture -> JPG
When zooming to 100%, Capture NX2 removed virtually all of the false colors from the shoulder area but the result wasn't perfect. However at normal zoom levels it looks just fine.
There was actually no sensor moire on the scarfs at all. Any artifacts seen were introduced when Aperture downsized the image from 36mp to a web-sized JPG.
@Ade: It would be good to see what Lightroom can do with that moire, if you are interested I would be happy to check it out if you forward me a full size copy. If you are interested, PM me.
OK, well that was interesting. The moire is present at high magnifications and the effect seems to reduce/go away at smaller sizes on my (decent) screen. Look at the guy on the rights jacket shoulder and cuff for the most moire.
I used the brush in LR4 in an attempt to remove the moire, but find that it does not remove moire, it apparently just seems to alter the colour patterning towards B&W (on this image anyway) other instances of moire on different materials may differ. The slider starts at a centre zero position (I can't think why one would want to add colour to moire by sliding it to the left), and sliding to the right reduces/alters the colour leaving the pattern. I deliberately brushed over the guys lapel and scarf slightly and it is plain to see the colour has been altered where there is no moire.
Here it is, you will have to go to flickr and see the 2048 size for the effect to show:
I think we're seeing the two aspects of aliasing: 1) false color artifacts; 2) the moire pattern. Looks like LR did a great job with the false color correction. I think the pattern itself is best removed in Photoshop if needed be, but since we don't usually see images at 100% zoom the slider controls are probably sufficient most of the time.
I have seen more moire effects in video than in stills. I also think the point Ade makes is right on target. I do think the D7100 in actual normal video use has less moire or about the same as the D7000 video. One thing the D7100 video reduces compared to the D7000 video is pulsing of light caused by the rolling shutter. I am very sure the D7100 camera was designed and outfitted to address many video weaknesses of the D7000. From what i have seen in actual practice, that has been accomplished quite nicely. I have yet to see moire effect in stills or video. It does seem that the D7100 video is advanced enough to now require better techniques to achieve very good quality video. A staggering amount of after market gear is becoming available to maximize D7100 video quality. Some of it seems pretty pricey and even questionable to me. But as to moire? Not a problem I can detect. Only use in many situations will let us know how competent the D7100 is. Right now my guess is Nikon sure knew what they were doing by eliminating it. I do think the D800E was a wonderful test camera to lead in that direction. Unfortunately most D800E users are mostly shooting stills. My D7100 shoots about 50% stills and 50% video. At this point I'd like to have used the D800E a lot more and especially in video mode.
I have been noticing a LOT of moire on our Sony HD TV lately. I wonder if there has been a shift towards further reducing AA filters on sensors of video cameras? You would think the wardrobe department would have been told to avoid herringbone and small checks but there they are fizzing away. I even see it among women's hair.
Video cameras usually have stronger AA filters than DSLRs, because moire is more evident in video than still images. It's more likely that you are seeing chromatic aberrations, which I see in TV shows and documentaries all the time.
If I take a good photo it's not my camera's fault.
Video cameras usually have stronger AA filters than DSLRs, because moire is more evident in video than still images. It's more likely that you are seeing chromatic aberrations, which I see in TV shows and documentaries all the time.
Nooo, it is moire! There is no mistaking moire - and this is the most violent, flashing fizzing moire!
The point is that I have had HD TV for a few years now, but only in the last three months or so am I seeing such bad moire effects. It's like they have just stopped trying to avoid it with filters or choice of clothing. One jacket was flashing like a neon sign it was so bad.
It will be interesting when I get my first DSLR with reduced/no AA filter to see if I see it in my images more.
Comments
I have 36, 24, 18 and 12 mpx available to me and use the 12 at least as often as the others.
Nobody is 'leading' in the DSLR space (certainly not SONY, pellicle mirrors 'lose' 1/3 of the light, and if all are using the same laws of physics, is a disadvantage that simply cannot be overcome).
Nikon and Canon are so competitive, and overlap / leapfrog one another, while individual bodies and lenses can be compared, trying to decide if one brand is 'ahead' of the other is a fool's errand.
.... H
Nikon N90s, F100, F, lots of Leica M digital and film stuff.
Old people agree.....LOL )
There is no fundamental reason for a swinging mirror in a digital camera.
What mirrorless needs to replace DSLR's is:
Full (24 X 36 mm) frame sensors
Better and faster EVF's
Better and faster autofocus
Phase detect autofocus to allow adaption of SLR lenses.
Electronic shutter with no rolling shutter effect.
Mirrorless will be smaller, lighter, have fewert moving parts, and not have the minimum back/retro focus limitation for design of lenses with a focal length less than 58 mm, as current FF SLR's do.
... H
Nikon N90s, F100, F, lots of Leica M digital and film stuff.
- Even without a global shutter, there's no technical reason a CMOS rolling shutter speed cannot be fast enough to minimize artifacts for all but extremely high speed movements.
It's all about $$$... as technology continues to improve, eventually a CMOS global shutter or a very fast rolling shutter will be affordable and "trickle down" to consumer devices.
Actually moire on two spots. Capture NX2 cleaned up one area with ease but not the second.
Here's the image post NX2. Can you see where the two moire spots are?
(MUA, model and the fashion designer, left to right)
This scene has been the only one from about 1,000 pictures I've taken so far with any hint of moire. And maybe 950 of those 1,000 frames were fashion event shots with lots of different kinds of fine fabrics.
I'm actually impressed how well NX2 cleaned up the moire. Even if not 100% clean, the image is still very useable.
Nikon N90s, F100, F, lots of Leica M digital and film stuff.
When enlarged on my monitor I do not see any moire....
Shoulder at 100%
Pocket + Left Arm at 100%
I use Aperture (which doesn't have good moire control) so for this one frame the workflow was:
NEF -> Capture NX2 -> TIFF -> Aperture -> JPG
When zooming to 100%, Capture NX2 removed virtually all of the false colors from the shoulder area but the result wasn't perfect. However at normal zoom levels it looks just fine.
There was actually no sensor moire on the scarfs at all. Any artifacts seen were introduced when Aperture downsized the image from 36mp to a web-sized JPG.
I used the brush in LR4 in an attempt to remove the moire, but find that it does not remove moire, it apparently just seems to alter the colour patterning towards B&W (on this image anyway) other instances of moire on different materials may differ. The slider starts at a centre zero position (I can't think why one would want to add colour to moire by sliding it to the left), and sliding to the right reduces/alters the colour leaving the pattern. I deliberately brushed over the guys lapel and scarf slightly and it is plain to see the colour has been altered where there is no moire.
Here it is, you will have to go to flickr and see the 2048 size for the effect to show:
I think we're seeing the two aspects of aliasing: 1) false color artifacts; 2) the moire pattern. Looks like LR did a great job with the false color correction. I think the pattern itself is best removed in Photoshop if needed be, but since we don't usually see images at 100% zoom the slider controls are probably sufficient most of the time.
The point is that I have had HD TV for a few years now, but only in the last three months or so am I seeing such bad moire effects. It's like they have just stopped trying to avoid it with filters or choice of clothing. One jacket was flashing like a neon sign it was so bad.
It will be interesting when I get my first DSLR with reduced/no AA filter to see if I see it in my images more.
One viewing unit vs. video recording industry... I think you just found yourself an excuse to buy a new TV!
... And no time to use them.