I saw a recent review here on FM for the 70-200 f/2.8L IS, and on the 1DsMkIII, the person says its not sharp compared to the f/4L version due to the sensor out resolving this lens.
rsg_1 wrote:
I saw a recent review here on FM for the 70-200 f/2.8L IS, and on the 1DsMkIII, the person says its not sharp compared to the f/4L version due to the sensor out resolving this lens.
Would the same hold true for the 5DMkII?
Well, it is pretty much the same sensor, so logic would dictate...
I think the 70-200/2.8LIS is just fine on the 1Ds3, especially when you need the 2.8 and IS. Nobody views images at 100% in real life, and when you scale it down even for an 13x19"" I am quite sure you won't be able to tell the difference. The reason why I'd get the 70-200/4LIS (in addition to my 2.8LIS) is for the size & weight, not the sharpness. Sharpness by itself is overrated.
WoW thanks a lot, I think that these videos are so much more interesting than the one from vincent laforet !!!
more interesting, LONGER ! and better understanding of the effect of lenses on video
rsg_1 wrote:
Then I wonder what would happen with faster lenses at f/2, 1.8, 1.4 and 1.2? A 100% crop wouldn't be useful when wide open?
I am not sure I understand the question. With any lens, with any camera, do you frequently use 100% crops for final work? Or only to see "what if"?
I have all four sub-2 L primes, shoot them pretty much exclusively below f2, and crop happily. Never really had to go to 100%, nor do I see any purpose in it - again, outside of academic pixel peeping.
I saw a recent review here on FM for the 70-200 f/2.8L IS, and on the 1DsMkIII, the person says its not sharp compared to the f/4L version due to the sensor out resolving this lens.
No, he doesn't understand what's happening.
First, realize that the 21mp sensor isn't even yet the sharpest "sensor" that's been put behind the 70-200 lens. It's still not as sharp as high-resolution, thin-emulsion black and white films like Adox and Kodak Technical Pan. This is nothing particularly new. This phenomenon has already been known and worked with...years ago.
The lens and sensor are mutual-limiting components in the complete system. The total system resolution is always lower than the total capability of either component. Theoretically, you can never max out either component.
What happens is that as you raise the capability of one component, it raises the total system capability closer to the maximum of the weaker component...but you never get to 100 percent. I read years ago that a film would have to have three times the resolution of the lens to raise the total system to 99 percent of lens' theoretical maximum capability.
A lens is never going to produce worse results because of a higher-resolving sensor. The results will always be better (although with continually diminishing returns). Yes, with greater resolution you may then be able to specify aberrations that were just part of the general blur before, but that is still a better image.
Just as a point of interest. My company produced a professional Kodachrome line. In working with kodak on this, we discussed the various parameters of the film.
I was told by Kodak, later on, when digital had become tolerable,that Kodachrome 25 film had the equivalent of about 24 MP. As a practical matter though, they said that the actual resolution would be closer to 18 MP, because of the lens/film limits.
An interesting article from Harold Merklinger from 2000, in response to Michael Reichmann Canon D30 tests.
A more technical article on the ultimate limits of resolution that everyone should take a gander at. This is very interesting, and comes as close as anything I've read outside of an optics text, to a definitive explanation.
Scroll down to the bottom of the 100% crop pic and click the link. You'll have top copy and paste the whole link as the FM parser doesn't like the ( ) in the above.
Canon has just posted the official 5D MARK II Instruction Manual!
I know we are getting close to production release now. I haven't read the entire manual but it has some great new features. If anyone finds something interesting, please share.
when I was playing with the 5D2 demo here, I noticed three holes in front and thought that it's for the internal mic. Is it possible to weather-seal a camera like this?
If so, good. im not really familiar with sound recording and pro video cams and weather sealed external connections and ports.
if not, will it find it's(video) way to 1 series?
some crops from tonight(forgive me for poor crops-not too much time for borrowing):
p.s. I also tried the α900. that crazy "salesperson" says its better than 5D2 because of the 3 extra mp's. Over here the α900 costs about US$600 more than 5D2. at iso400, lets just say I returned the camera to him and walked away.
The ISO 800 shot looks plasticky and the ISO 1600 shot has horrible noise. Sure hope this is not representative of the 5D II as I pre-ordered one on the very first day.
markarce wrote:
when I was playing with the 5D2 demo here, I noticed three holes in front and thought that it's for the internal mic. Is it possible to weather-seal a camera like this?
If so, good. im not really familiar with sound recording and pro video cams and weather sealed external connections and ports.
if not, will it find it's(video) way to 1 series?
The 1-series cameras have had an internal mic for quite a while and they are weather sealed, so I'd say there would be no problem weather sealing a camera like this.
I expect all Canon cameras will have video added over the next revision, including the 1-series cameras.
don't worry about getting plasticky and noisy photos. I have seen the same demo 5D2 produce excellent photos and videos that day.
here's a scaled down photo and a crop ( iso800. please don't link this photo to any other sites as it's not my photos. I just had testers use my memory card. I asked them nicely)
Mel Gross wrote:
Just as a point of interest. My company produced a professional Kodachrome line. In working with kodak on this, we discussed the various parameters of the film.
I was told by Kodak, later on, when digital had become tolerable,that Kodachrome 25 film had the equivalent of about 24 MP. As a practical matter though, they said that the actual resolution would be closer to 18 MP, because of the lens/film limits.
That is not my experience. Even Kodachrome 25 drum scanned at 8000 ppi is well below the 12 MP of a 5D in usable resolution shot with the same lens. A 16.7 MP 1Ds2 really is astonishingly better than K25 and a 21 MP 1Ds3 just widens the gap.
Heheh, true for you Stanj. I worked in Bausch and Lomb for seven years both making contact lenses and in the IT dept later on. The pixelated pattern near the edge of the lens is the biggest give away, well that and you can see the edge in the white of the eye too!
jcbenner wrote:
That is not my experience. Even Kodachrome 25 drum scanned at 8000 ppi is well below the 12 MP of a 5D in usable resolution shot with the same lens. A 16.7 MP 1Ds2 really is astonishingly better than K25 and a 21 MP 1Ds3 just widens the gap.
i don't know if it is true, but a few years ago i read that the sharpest 35mm color film maxed (some color print film) out at about 15MP equivalent, but that some B&W films were some degree more than that (not specified).
stanj wrote:
Given that the eyes are blue and she has more than just a strain of Asian genes in her, I am pretty sure the color is not natural
also given that i've never seen an iris with printer banding before
or such uniform and unusually shaped radial lines either.
maybe she's a terminator....
