| Author | Thread |
|
|
12/15/2005 02:34:48 PM · #26 |
| iirc the canon 200L/1.8 is the sharpest of any of the lenses rated on the site! |
|
|
|
12/15/2005 02:48:20 PM · #27 |
One thing conspicuously absent from the Canon lens line-up are new versions updated for use with digital cameras. Both Sigma and Tamron have replaced some of their better selling high end lenses with versions that have anti-reflection coatings to deal with the increased reflection from a sensor as compared to film.
|
|
|
|
12/15/2005 02:58:30 PM · #28 |
Originally posted by coolhar:
One thing conspicuously absent from the Canon lens line-up are new versions updated for use with digital cameras. Both Sigma and Tamron have replaced some of their better selling high end lenses with versions that have anti-reflection coatings to deal with the increased reflection from a sensor as compared to film. |
While I agree that digital sensors do in fact imply that some different optimization of coatings, etc. may well be beneficial, still, as they say, the proof of the pudding is in the eating. The Canon 70-200/2.8 is the premier performer in that focal length range, bar none. Not saying it couldn't be improved even further, but put yourself in Canon's place... you have a market-leading product. Are you chomping at the bit to spend more in R&D for further improvements? |
|
|
|
12/15/2005 03:11:12 PM · #29 |
Originally posted by kirbic:
While I agree that digital sensors do in fact imply that some different optimization of coatings, etc. may well be beneficial, still, as they say, the proof of the pudding is in the eating. The Canon 70-200/2.8 is the premier performer in that focal length range, bar none. Not saying it couldn't be improved even further, but put yourself in Canon's place... you have a market-leading product. Are you chomping at the bit to spend more in R&D for further improvements? |
Couldn't agree more. Just in this thread people have called it "perfect" and "the standard". There isn't much incentive for Canon to improve upon it.
|
|
|
|
12/15/2005 03:18:04 PM · #30 |
Originally posted by coolhar:
Originally posted by kirbic:
While I agree that digital sensors do in fact imply that some different optimization of coatings, etc. may well be beneficial, still, as they say, the proof of the pudding is in the eating. The Canon 70-200/2.8 is the premier performer in that focal length range, bar none. Not saying it couldn't be improved even further, but put yourself in Canon's place... you have a market-leading product. Are you chomping at the bit to spend more in R&D for further improvements? |
Couldn't agree more. Just in this thread people have called it "perfect" and "the standard". There isn't much incentive for Canon to improve upon it. |
Actually, there is.
This applies to both Canon and Nikon. The ability to resolve images with the current lenses is reaching it's limits. We already see that the Canon 1Ds Mark II reolves more than 35 mm film.
Resolution for 35mm is going into unchartedered territories.
What happens when Nikon or Canon come out with a 22 megapixel or higher chip and the results show that...it has no effect on quality at 16 x 20 prints and..when blown up larger...doesn't get any better because the lens simply cannot resolve the pixels?
These camera makers might not be making changes today but I would be willing to bet any amount of money that they are thinking about this in their R&D divisions. |
|
|
|
12/15/2005 03:20:06 PM · #31 |
New from Galoo!
1mm-1000mm f/1.2 IS lens
And you too could have one for only $10,000,000,000,000 |
|
|
|
12/15/2005 03:33:40 PM · #32 |
Originally posted by hokie:
...Actually, there is.
This applies to both Canon and Nikon. The ability to resolve images with the current lenses is reaching it's limits. We already see that the Canon 1Ds Mark II reolves more than 35 mm film.
Resolution for 35mm is going into unchartedered territories.
What happens when Nikon or Canon come out with a 22 megapixel or higher chip and the results show that...it has no effect on quality at 16 x 20 prints and..when blown up larger...doesn't get any better because the lens simply cannot resolve the pixels?
These camera makers might not be making changes today but I would be willing to bet any amount of money that they are thinking about this in their R&D divisions. |
Remember that the 20D today has a pixel pitch that is as aggressive as a 22Mpx FF camera. Todays lenses, particularly sharp primes but excellent zooms as well, are capable of supporting this pixel pitch, at least in the central area. The corners are, of course, another matter. |
|
|
|
12/15/2005 03:41:37 PM · #33 |
Originally posted by kirbic:
Remember that the 20D today has a pixel pitch that is as aggressive as a 22Mpx FF camera. Todays lenses, particularly sharp primes but excellent zooms as well, are capable of supporting this pixel pitch, at least in the central area. The corners are, of course, another matter. |
I would be interested to see a real study of the 20d pixel pitch. I wonder if, like many other digital cameras that use really small sensors with small pixel pitches if that pixel size is extrapolated in camera so you ultimately get a resized pixel output.
My point is....I would find it unusual that cameras like the 20d or even point and shoot digitals with very small pixels are resolving finer detail than a Canon 1ds Mark II.
Message edited by author 2005-12-15 15:42:06. |
|
|
|
12/15/2005 03:48:16 PM · #34 |
Originally posted by kirbic:
...still, as they say, the proof of the pudding is in the eating. |
Oh, thank you thank you thank you! This is the FIRST time I have ever seen this old saw used correctly online :-) Most people say "the proof is in the pudding" and that's absolutely meaningless...
You just gained major points in my estimation, not that you needed any more, but...
Robt. |
|
|
|
12/15/2005 03:51:57 PM · #35 |
Originally posted by hokie:
Originally posted by kirbic:
Remember that the 20D today has a pixel pitch that is as aggressive as a 22Mpx FF camera. Todays lenses, particularly sharp primes but excellent zooms as well, are capable of supporting this pixel pitch, at least in the central area. The corners are, of course, another matter. |
I would be interested to see a real study of the 20d pixel pitch. I wonder if, like many other digital cameras that use really small sensors with small pixel pitches if that pixel size is extrapolated in camera so you ultimately get a resized pixel output.
My point is....I would find it unusual that cameras like the 20d or even point and shoot digitals with very small pixels are resolving finer detail than a Canon 1ds Mark II. |
Would the 20D resolve finer detail if fitted with the same lens as the 1Ds II, with the shot being taken from the same place? It seems to me that the 20D, with its crop factor, would, indeed resolve more, but at the expense of the picture being 'less' of the scene being viewed (i.e. a section would be magnified over the 1Ds's shot). Or is my logic more screwed up than I think?
|
|
|
|
12/15/2005 04:01:49 PM · #36 |
Originally posted by hokie:
...my point is....I would find it unusual that cameras like the 20d or even point and shoot digitals with very small pixels are resolving finer detail than a Canon 1ds Mark II. |
There is really no debate that the pixel pitch of the 20D is about 6.4µm, where the 1DsMkII is more like 8µm. The 20D therefore is in theory capable of extraacting more detail out of a *portion* of the image circle, assuming the lens is up to the task. The 1DsMkII, however, makes use of MUCH more of the image circle (2.5 times the area, in fact) and so easily surpasses the 20D in TOTAL detail rendered. |
|
|
|
12/15/2005 04:04:05 PM · #37 |
| resolution limitation on the lens side doesn't mean you can get more resolution out of a camera. The total resolution of a camera is based on the whole system. Increasing the MP will still give you more resolution even if it's outresolving the lens. |
|
|
|
12/15/2005 04:11:20 PM · #38 |
Originally posted by kyebosh:
resolution limitation on the lens side doesn't mean you can get more resolution out of a camera. The total resolution of a camera is based on the whole system. Increasing the MP will still give you more resolution even if it's outresolving the lens. |
Well, you get more pixels, but not more detail, so what would be the point?
MTFs are multiplicative, so, if at 50 lines per mm a lens has an MTF of 0.5, and the sensor has an MTF of 0.9, the system MTF is about 0.5 * 0.9 = 0.45, so the lens dominates the system performance if it has lower resolving power than the sensor. When the resolving power of the lens and sensor are nearly equal, they contribute nearly equally. Whether the lens or the sensor has the greatest impact on system MTF is not like flipping a switch, more like turning a dimmer... |
|
|
|
12/15/2005 04:12:11 PM · #39 |
Originally posted by AJAger:
Would the 20D resolve finer detail if fitted with the same lens as the 1Ds II, with the shot being taken from the same place? It seems to me that the 20D, with its crop factor, would, indeed resolve more, but at the expense of the picture being 'less' of the scene being viewed (i.e. a section would be magnified over the 1Ds's shot). Or is my logic more screwed up than I think? |
The section is not "magnified": if you took the actual sensor images and overlaid them in registration, they'd match up perfectly; the larger sensor would just have "more" picture in areas the smaller sensor did not cover.
In film terms, using a 4x5/8x10 interchangeable-back view camera, if I shot with a 210mm lens on 4x5 film, then swapped in the 8x10 back and shot again (the image circle of that lens covers the 8x10 format easily) then i coulkd overlay the 4x5 film over the center of the 8x10 film and they would register precisely.
The pixel pitch on the 20D is smaller, however, so as Kirbic points out it is in theory capable of resolving finer detail within that limited frame.
R. |
|
|
|
12/15/2005 04:13:07 PM · #40 |
Originally posted by kirbic:
Originally posted by hokie:
...my point is....I would find it unusual that cameras like the 20d or even point and shoot digitals with very small pixels are resolving finer detail than a Canon 1ds Mark II. |
There is really no debate that the pixel pitch of the 20D is about 6.4µm, where the 1DsMkII is more like 8µm. The 20D therefore is in theory capable of extraacting more detail out of a *portion* of the image circle, assuming the lens is up to the task. The 1DsMkII, however, makes use of MUCH more of the image circle (2.5 times the area, in fact) and so easily surpasses the 20D in TOTAL detail rendered. |
I read an article that compared the Nikon D2x and the Canon 1ds Mark II and this topic was explored. The Nikon has a 5.?um pixel pitch and..whne examining the area of the center of the lens..the Nikon did resolve at a finer "grain".
I find this all very interesting. I think we will find that pixel pitches will converge as 35mm format for all manufacturers move to a similar size sensor and resolution.
I know this...my lenses feel like they are at their limit. Two of us in the studio shoot with the Nikon D2x and 2 shoot with the Canon 1Ds MarkII...We all feel that the lenses are becoming the limiting factor.
Maybe improvements with dynamic range will open up resolution a bit..but that may not happen for years. |
|
|
|
12/15/2005 04:30:54 PM · #41 |
| Anybody comment on the fact that the 1ds out resloves all other Canon digitals? |
|
|
|
12/15/2005 04:36:46 PM · #42 |
Originally posted by Brent_Ward:
Anybody comment on the fact that the 1ds out resloves all other Canon digitals? |
how does the 1Ds still outresolve the 1DsII? can you show this, or point out some research? |
|
|
|
12/15/2005 05:13:58 PM · #43 |
Originally posted by bear_music:
Originally posted by AJAger:
Would the 20D resolve finer detail if fitted with the same lens as the 1Ds II, with the shot being taken from the same place? It seems to me that the 20D, with its crop factor, would, indeed resolve more, but at the expense of the picture being 'less' of the scene being viewed (i.e. a section would be magnified over the 1Ds's shot). Or is my logic more screwed up than I think? |
The section is not "magnified": if you took the actual sensor images and overlaid them in registration, they'd match up perfectly; the larger sensor would just have "more" picture in areas the smaller sensor did not cover.
In film terms, using a 4x5/8x10 interchangeable-back view camera, if I shot with a 210mm lens on 4x5 film, then swapped in the 8x10 back and shot again (the image circle of that lens covers the 8x10 format easily) then i coulkd overlay the 4x5 film over the center of the 8x10 film and they would register precisely.
The pixel pitch on the 20D is smaller, however, so as Kirbic points out it is in theory capable of resolving finer detail within that limited frame.
R. |
I apologise for the loose use of the word 'magnified'. I meant it in the sense of the final picture size. Although the sensor size of the 20D is smaller than that of the 1DsII, one would still view the final images at the same size, be it on screen or in print. It is in this sense that I meant that a portion of the 1Ds image would appear to be magnified in the 20D image.
I don't really explain myself very well, do I?
Message edited by author 2005-12-15 17:18:32.
|
|
|
|
12/15/2005 05:32:51 PM · #44 |
Originally posted by kyebosh:
Originally posted by Brent_Ward:
Anybody comment on the fact that the 1ds out resloves all other Canon digitals? |
how does the 1Ds still outresolve the 1DsII? can you show this, or point out some research? |
I'm looking for the specific info I read, but I looked at some other sites. Anyway, I might be eating some words if I can't find it. ;o)
Message edited by author 2005-12-15 17:33:02. |
|
|
|
12/15/2005 06:01:52 PM · #45 |
| Thanks bear and kyebosh for straightening me out on the non-existent weatherproofing on the 70-200/f4L! It's a sturdy bugger, though. |
|
|
|
12/15/2005 07:23:23 PM · #46 |
While we're talking zooms in this range, does anybody know why the 70-300mm f/4.5-5.6 DO IS is not carry a "L" designation? I wish Canon had something with 300 at the top with L.
If I get a zoom, I doubt I'll shoot sports, but I would like to do some birding. It seems a fair number of people don't particularly like the 100-400 for various reasons. However, to get up to the 400 range, there aren't many other options. The 70-200 with a 2x is one, but it is just as expensive and inferior at the 400 range. I thought about a 70-300 with a 1.4 converter. However, nothing in this range is fast and nothing carries the "L".
I have the cheapo 80-200 f4.5-5.6 II and find that 200 just really isn't enough to capture most birds. But I don't know where to go above.
Ideas? |
|
|
|
12/15/2005 07:28:38 PM · #47 |
Here is my opinion/hypothesis.
There is something inherent in the optics design that does not allow for excellent 70-300 zoom range. If you look at consumer 70-300 zooms from Canon and Sigma from Photozone.de, they all have excellent sharpness from 70-200 (atleast in the center) but the performance falls off at 300. That's why I don't think any maker makes a pro level 70-300 zoom b/c it can't deliver uniformly sharp resolution throughout the whole range.
Just my hypothesis.
P.S. If you look at Photozone.de, between the 80-200 and the 70-200 f2.8L lenses, the biggest area of improvement came in corner sharpness. That maybe a significant improvement for some shooters out there.
Message edited by author 2005-12-15 19:34:01.
|
|
|
|
12/15/2005 07:57:58 PM · #48 |
ok another question.
last year I took my D60 and put it on a tripod. Used a 28-200mm lens. I took 2 photos, 1 at 28mm the other at 200mm.
then I put my sister in laws Canon Rebel on my tripod, used the same 28-200 lens and shot the exact same shots as above.
had the film and digital prints developed at the exact same place.
Why is it that the film camera at 28mm was much wider than my D60., then at 200mm my D60 was "zoomed" in much futher than the film camera.
there was noticible size differance in the objects in the photos. I could not over lay them and "register" them to match perfectly, or even close.
So why is that
James |
|
|
|
12/15/2005 07:59:31 PM · #49 |
Originally posted by jab119:
ok another question.
last year I took my D60 and put it on a tripod. Used a 28-200mm lens. I took 2 photos, 1 at 28mm the other at 200mm.
then I put my sister in laws Canon Rebel on my tripod, used the same 28-200 lens and shot the exact same shots as above.
had the film and digital prints developed at the exact same place.
Why is it that the film camera at 28mm was much wider than my D60., then at 200mm my D60 was "zoomed" in much futher than the film camera.
there was noticible size differance in the objects in the photos. I could not over lay them and "register" them to match perfectly, or even close.
So why is that
James |
crop factor on the digital camrea. |
|
|
|
12/15/2005 07:59:38 PM · #50 |
Originally posted by DrAchoo:
While we're talking zooms in this range, does anybody know why the 70-300mm f/4.5-5.6 DO IS is not carry a "L" designation? I wish Canon had something with 300 at the top with L.
If I get a zoom, I doubt I'll shoot sports, but I would like to do some birding. It seems a fair number of people don't particularly like the 100-400 for various reasons. However, to get up to the 400 range, there aren't many other options. The 70-200 with a 2x is one, but it is just as expensive and inferior at the 400 range. I thought about a 70-300 with a 1.4 converter. However, nothing in this range is fast and nothing carries the "L".
I have the cheapo 80-200 f4.5-5.6 II and find that 200 just really isn't enough to capture most birds. But I don't know where to go above.
Ideas? |
sigma ex 100-300/4
sigma ex 120-300/2.8 |
|
