It was just done in lightroom with no post processing. For sharpening I had 25,0,25. I had noticed some ca on the white wheel disk on the wagon and used -15 on the red cyan shift. This is probabably what you are seeing. The shift of the is probably not even across the frame so shifting to fix the white wheel disk may have negative effects on the other parts of the scene.
The white is a flare or bloom or something of that effect. It is also in the white tire hubs of the wagon. It was there regardless of using the cyan shift function.
I was thinking some more about the blur in the clothes line image, and I think it may be motion blur. It might have been windy that day, with the clothes flicking around on the line.
Andi - I linked to one way back, that I took on holiday. It was the post where I introduced my second crap flawed test. I will try to find the link again. The corners definitely go a bit soft at f2.8 on the 5D although I am not sure I am using the optimal adapter. I haven't seen the R19 f2.8 results on a 5D. Does it stay sharp all the way out?
> ...Here it is again, no Reduce fringe correction in Lightroom.
It is odd. I am still seeing the bleed exactly as in the crops above. It doesn't seem to be a general blurring of colour. Crop 3 above shows that the sky blue has spread across the branch in a very specific direction: tangential.
> ...I was thinking some more about the blur in the clothes line image, and I think it may be motion blur. It might have been windy that day, with the clothes flicking around on the line.
I estimated the throw of that white clothes peg to be about an inch based on traditional clothes peg sizes. With a shutter speed of 1/2000 second that means it must have been moving at over 100 miles an hour. Since only a few per cent of its illumination occurs above the main impression, that means it was travelling at thousands of miles an hour and has stopped faster than a ufo. As you pointed out, there is also tangential ghosting of the white hub caps. How fast were they moving in an almost vertical direction?!?! ;-) I think we can rule out motion.
Edited by brainiac on Jan 09, 2007 at 08:00 PM GMT
The footprints shot was definitely at f2.8. You can see a bit of mushiness in the corners.
Thanks I remember now, did not know it was f 2.8, great!
I posted recently some L19 shots in another thread. I was not sure if my adapter was 100% so I posted only f5,6 and f11 shots. Now I see Roberts image of his lovely backyard I can say it was right. This lens must be stopped down to f5,6 to cover the corners and needs a bit more sharpening and contrast than something like my 28mm Zeiss. Overall also a good lens
I looked at you 19mm f2.8 sample. I think some of what you are seeing is the sensor not the lens. I have had my 19mm since film days and it wasn't that soft in the corners on film. The DMR has angled microlenses and ROM correction that might make it perform better than on the Canons. I don't think the Canon cameras have this.
I also had similar experiences with the 35-70mm f4, where people thought it was soft on a 1Ds MkII, while I had plenty of sharp slides shot with it and the results with the DMR were also excellent.
In the end, if a lens doesn't perform to the corners on the Canon, that is what counts if you are using it on a Canon.
I think we see DOF here at f2,8 and I removed the first file now, I have not made that many shots I can post with that lens, I was finaly too scared to shave it for my 5D
Rob, as you can see the 5D with Richards 21 has nice corners even at f2,8
I think my 21 is a little mushy right in the very corner at f2.8. But by f5.6 it is razor. Looking at your f11 shot I would say your 19 performs well in the bottom right corner where subject is distant, but the other corners look less good. The problem with the Zeiss is the waveform distortion which you can easily see in my holiday beach shots. It is bad enough that I found myself putting the horizon through the middle of the picture to avoid it.
Here is a European sky, Kodachrome 64 It was this trip to France in 1998 that got me back into Leica. I had bought a used R4 with 35mm Summicron and brought it along with my Canon gear. I shot the whole trip in K64. The Leica slides just jumped off the slide table compared to the Canon shots.
As a primarily nature/wildlife photographer these days, I have to say I disagree with your analyses of Rob's sky colors. These shots are very similar to what I see in clear areas here in Tennessee, especially when out of the city. Less cyan when in town or in summer. This cyan is also less pronounced in the American West and Southwest, and less pronounced in Florida.
So Jeff, does the sky in this chateau shot look too purple to you? It looks more natural to me than the lawn shot, despite the posterised graduation in the water. I have driven across the US once by a North route, and twice South. I have visited Florida, NY, LA, Boston, the Bahamas many more times. I once worked for 6 weeks in the Mojave desert. Sure the sky can be hazy on the West coast due to vapour and smog as Carsten said, and clear in the desert. But I have never seen it go green like the lawn shot looks ON MY MONITORS. I accept that this could be a hardware/software thing rather than the camera itself, but that is still not my prime suspect. I rarely see blue->cyan compression from other cameras ON MY MONITORS. I have travelled enough to have grounds to doubt that the sky is routinely that green in some locations. By my recollection, greenish skies started appearing with the advent of the DMR.
The last two were film. Chantilly was Kodachrome 64. I don't think it is the DMR has the wrong skies, but that it has colour response similar to film. The Canon digitals tend to have a different blue to the sky. I have owned every Canon professional digital starting at the D2000 and ending at the 1D MkII and their colour never had the Kodachrome look of the DMR.
