andrewd01 wrote:
Do you prefer the results from CS3 over SilkyPix? I have a G1 but haven't tried processing in CS3. Wasn't overly impressed from the results from SilkyPix.
Having used SilkyPix in the past, it didn't even get installed when I got the G1. I'm not willing to deal with its poor performance and interface when I already had CS3 and knew it did a good job with conversions.
The 20mm really couldn't be much better, IMO. Sharp, contrasty, buttery-smooth bokeh, close focusing and it's tiny? What more could you ask for? Well, one thing - a bayonet-mount, reversible hood would have made it perfect.
Sam Bennett wrote:
The 20mm really couldn't be much better, IMO. Sharp, contrasty, buttery-smooth bokeh, close focusing and it's tiny? What more could you ask for? Well, one thing - a bayonet-mount, reversible hood would have made it perfect.
I'd want it to be at least one stop faster and even smaller, which should be possible. The barrel is huge compared to the entrance pupil. Ugly, to say the least.
Thanks! Some pics of the lens on the body. Many have complained about the looks of the 20 on the E-P1, but I find it quite nice. It looks better in person than it does in pictures, though.
I tried the same lens on the Panasonic GF1. There is no doubt in my mind that this is a very nice lens for this type of camera. The cam is interesting as well, though I am debating whether to spend the price for it plus a short tele, because that is getting to be expensive
I know you mean "technically." But $400 for a normal FOV, f/1.8 lens -- yikes. If it were f/1.4 maybe, but f/1.8? Me thinks the vendors is perhaps recouping its R&D costs.
kevinsullivan wrote:
I know you mean "technically." But $400 for a normal FOV, f/1.8 lens -- yikes. If it were f/1.4 maybe, but f/1.8? Me thinks the vendors is perhaps recouping its R&D costs.
Note the build quality on the 20 is excellent, it's much better built than your average kit lens.
Comparing to closest equivalent for 35mm FF, which would be the CV Ultron 40/2 rather than one of the cheap AF 50/1.8's (which are cheap due to poor build and massive sales volumes), it's a little lighter built, a half-stop faster, adds AF and costs about the same in the US.
I'm curious how telecentric this lens is. Could somebody measure the diameter of aperture as it appears from the back and from the front of the lens and provide the ratio of the first to the second?
The focal length multiplied by this ratio is the distance from the exit pupil to the focal plane. The larger this distance is, the more normal is the angle of the light hitting the sensor. For my Olympus 17mm this distance is ~35mm. This is a lot for such a thin lens. It seems that Olympus take the telecentricity seriously. What about Panasonic?
kevinsullivan wrote:
I know you mean "technically." But $400 for a normal FOV, f/1.8 lens -- yikes. If it were f/1.4 maybe, but f/1.8? Me thinks the vendors is perhaps recouping its R&D costs.
Or miniaturization and early adoption have a cost per se.
If we were talking about a standard Leica lens, $400 would be a 'yikes bargain' lens
fourfa wrote:
Just checked with my calipers and my friend's lens: from the front, 0.433", from the back, 0.572".
Ratio: 1.32
Times focal length (assuming a true 20mm): 26.4mm
That is quite interesting. There are quite a lot of reports that wide angle rangefinder lenses don't perform well on mFT. The common explanation is that the digital sensors are not designed for oblique light angles. CV 28/2 is very often mentioned as a poor performer because of this. For this symmetrical lens the distance between the exit pupil and the focal lane should be about 28mm which is practically the same as for the Panasonic. The angles should be pretty much the same. I'm puzzled
Alex wrote:
That is quite interesting. There are quite a lot of reports that wide angle rangefinder lenses don't perform well on mFT. The common explanation is that the digital sensors are not designed for oblique light angles. CV 28/2 is very often mentioned as a poor performer because of this. For this symmetrical lens the distance between the exit pupil and the focal lane should be about 28mm which is practically the same as for the Panasonic. The angles should be pretty much the same. I'm puzzled
Alex
People are generalizing too much. There's issues with a number of RF wide-angle designs on m43, but a number of other designs do not have the same issues. The CV 28/2 is ust about the only CV wide with this issue. Even the CV 28/1.9 generally doesn't suffer for it and none of the slower CV wides and ultra-wides have the issue. Many Leica wides have problems however. Distance from the rear element to the sensor seems to be the primary determinant.
mawz wrote:
People are generalizing too much. .... Distance from the rear element to the sensor seems to be the primary determinant.
I think that's why Alex is puzzled. The distance from the rear element to the sensor with the CV28/2 (a poor performer on mFT) and the Pany 20/1.7 (an excellent performer on mFT) are almost the same. If the distance from the sensor is the primary determinant this result doesn't make sense. If anything the 20/1.7 is at a disadvantage...it's wider, faster and slightly closer to the sensor.
My suspicion is that Alex's approach is valid, but perhaps only for lenses not designed for mFT. It has been suggested that the mFT lenses have been designed to be corrected by the software. The performance of this lens may be an indication of how well this approach to lens design can work.
CVickery wrote:
I think that's why Alex is puzzled. The distance from the rear element to the sensor with the CV28/2 (a poor performer on mFT) and the Pany 20/1.7 (an excellent performer on mFT) are almost the same. If the distance from the sensor is the primary determinant this result doesn't make sense. If anything the 20/1.7 is at a disadvantage...it's wider, faster and slightly closer to the sensor.
My suspicion is that Alex's approach is valid, but perhaps only for lenses not designed for mFT. It has been suggested that the mFT lenses have been designed to be corrected by the software. The performance of this lens may be an indication of how well this approach to lens design can work....Show more →
Actually Alex has been calculating based on exit pupil rather than rear element. The rear element of say the Oly 17 is around 20mm from the sensor, not 35mm. The same goes for the 28/1.9. I'm not sure about the 20/1.7 or CV 28/2 as I don't have those in front of me.
OK, I admit I'm way over my head here and it probably shows. It would seem to me that the distance from the sensor to the exit pupil, rather than the rear element, would be what detemines the angle at which light would hit the sensor (no doubt there can be other factors, perhaps the diameter of the exit pupil??).
