Sailor Ted
Well-known
fgianni said:
Hmmm yes well certainly not with the same zeal or mean spiritedness (at least with many) as the bashing is progressing for the M8. Let's see. The M8 is essentially very sound mechanically (we assume at this point) DRF whereas the R-D1 and R-D1s are not (their primary flaw- mechanically unreliable to say the least), the M8 takes much sharper pictures with a wider dynamic contrast of color, and the only real strike against the M8 is IR sensitivity to which the R-D1 suffers to a similar degree. Both the R-D1 and the M8 solve this problem in exactly the same way- with IR cut filters so no problem.
Honestly I don't see what is wrong with the M8. Sure I'd like a film advance lever but I'd really like a RF mechanism on my R-D1s that did not go out of alignment within the first three days of use (no kidding) backed by a company that will back the product (at least Leica solved their issues and stood behind the product- Epson could give a ****).
No I'm a shooter and at the moment my tool is the R-D1s. I will continue to make trips and take pictures with my camera and post them here and on my gallery- for better or for worse :angel: . When the M8 becomes available I will pick up a copy and likewise start posting pictures and we shall see which camera is more capable- IR filters and all
.
Last edited:
Bas
Dough!
With regard to the problem on the pictures... to me, it doesn't seem so strange. The paint looks like that metalized paint (I don't know the exact name in english... but a lot of expensive cars are painted in that way). The paint looks like composed of small crystals inside which reflect light in different directions, with different strenths... you know which kind of paint I mean. I guess we'd need to see the 5D picture, so we can really understand which is painting on the car. I see a rare metal gray. At least to my humble eyes, that doesn't look so strange. Bur of course, I wasn't there looking at the car
Now, with regard to the M8 problems... yes, it has come out with more image problems than other cameras. However, for what it does, it seems to be a good camera (I like more the RD1, but that's me). I don't know why someone asked how to do macro with the M8, when no one has ever thought to do macro work with a rangefinder. The closest to that I think it was the Summicron DR. For versatility, get a 5D, there are zillions of accesories for the EOS system for anything you'd like to do. Rangefinder shooters, we know and we live with the fact that we can't focus closer than 1 meter (3'3") and all the nuances of RFs ;-)
Back to the picture... it seems normal to me. I repeat, I think we'd need to see the 5D shot.
Cheers!
Sebastian.
Now, with regard to the M8 problems... yes, it has come out with more image problems than other cameras. However, for what it does, it seems to be a good camera (I like more the RD1, but that's me). I don't know why someone asked how to do macro with the M8, when no one has ever thought to do macro work with a rangefinder. The closest to that I think it was the Summicron DR. For versatility, get a 5D, there are zillions of accesories for the EOS system for anything you'd like to do. Rangefinder shooters, we know and we live with the fact that we can't focus closer than 1 meter (3'3") and all the nuances of RFs ;-)
Back to the picture... it seems normal to me. I repeat, I think we'd need to see the 5D shot.
Cheers!
Sebastian.
jlw
Rangefinder camera pedant
I think that in the second picture you're seeing an interesting example of aliasing.
Metallic car finishes contain reflective flakes that look just like those in the image, only smaller. Since the M8 famously lacks an anti-aliasing filter, I think that the irregular pattern of the flakes is interacting with the regular pattern of the sensor to produce an enlarged irregular pattern -- exactly the same way that film's random grain pattern interacts with the regular pattern of a CCD sensor to produce the enlarged pattern we call "grain aliasing."
Metallic car finishes contain reflective flakes that look just like those in the image, only smaller. Since the M8 famously lacks an anti-aliasing filter, I think that the irregular pattern of the flakes is interacting with the regular pattern of the sensor to produce an enlarged irregular pattern -- exactly the same way that film's random grain pattern interacts with the regular pattern of a CCD sensor to produce the enlarged pattern we call "grain aliasing."
jaapv
RFF Sponsoring Member.
This thread is hilarious! Nobody looked at the photograph.Those are different renderings of different parts in there!The car has paint on metal and on plastic surfaces. Both surfaces react differently. Ask any car spraying firm. Our eye compensates when we see it nor can the 5D record that.It gets lost in the noise filtering. But the M8 can and will differentiate those different reflections.....
Last edited:
fgianni
Trainee Amateur
Sailor Ted said:
Indeed, mainly because R-D1 owners did not start from the point of view that anything Epson makes is perfect by definition and anyone that does not agree is just an Epson basher.
I can't disagree with thisSailor Ted said:
I bloody hope so, 2.5 years newer technology and 1.5 times the price when new, even the 5D I am sure takes much sharper pictures than the Canon D30.Sailor Ted said:
At least to my eyes (at least from the pictures I see posted since I don't own an M8 yet) the M8 suffers from the problem to a much higher degree than the R-D1, definitely the R-D1 IR sensitivity is much more tolerable, almost no one that owns it uses IR cut out filters with it.Sailor Ted said:
Leica tried to hide the issue, to the point of sending very rude letters to the reviewers that noticed it (see Damien Demolder's experience) , and only came up with a solution when even my cat noticed itSailor Ted said:
And a camera that needs IR filtration on the lens, does not need it on the sensor side as well, so if it was all part of the design, why didn't they simply leave out the hot mirror completely, increasing the definition?
Sailor Ted
Well-known
fgianni said:R-D1 owners did not start from the point of view that anything Epson makes is perfect by definition and anyone that does not agree is just an Epson basher.
At least to my eyes (at least from the pictures I see posted since I don't own an M8 yet) the M8 suffers from the problem to a much higher degree than the R-D1, definitely the R-D1 IR sensitivity is much more tolerable, almost no one that owns it uses IR cut out filters with it.
And a camera that needs IR filtration on the lens, does not need it on the sensor side as well, so if it was all part of the design, why didn't they simply leave out the hot mirror completely, increasing the definition?
Why would you assume the M8 could be perfect? What is perfect, in your estimation in the photographic arena let alone anywhere else in this world? Put another way, why should Epson owners expect perfection from Leica and not from other cameras? Odd, very.
And to retort in a nut shell, both the Leica and the Epson shift to red to an unacceptable degree on black for any serious work- can you imagine delivering images with a shift to red on black to paying wedding clients? The point is that since both cameras share this trait both must use IR cut filters if true black is to be achieved. That one may exhibit this to a greater or lesser degree when compared to the other is not the point- both cameras exhibit this and both must be corrected in the same manner. When the proper IR cut filters are used the problem is solved 100%. Period.
Now for the real issue: The Epson has a SERIOUS problem that CANNOT be fixed-it's overwhelming tendency to flare when light of a high dynamic range exists in a shot. This issue cannot be addressed by any means I know of and is not an issue with the Leica.
Lastly the M8 is in a whole different league in terms of image sharpness and dynamic range of color when compared to not only the Epson but to most digital cameras on the market today.
See the images below for examples of both the Epson’s tendency to shift to red on black as well as flare caused by the thick filter placed in front of the R-D1's digital sensor. The first image exhibits a shift to red on a black jacket- in natural lighting no less. The second images exhibits strong filter/sensor flare (due to that thick IR filter that some feel Leica should have used) on bare light bulbs. The third image exhibits flare on a back lit subject near the top of the bridge for the same reason as the second image- try and fix these last two images with a lens filter!
Attachments
Last edited:
jlw
Rangefinder camera pedant
jaapv said:
I do talk occasionally to car-spraying firms and have sprayed a couple myself.
If I got such a different "flop" (car-shooter term for alignment of the metal particles in metallic paint) on metal and plastic surfaces, it would mean either that I had used the wrong primer/surfacer or that I had done a very incompetent job!
There are a couple of Mazda 3s in my neighborhood, and none has such a drastic difference in finish on various parts.
I still think we're seeing "aliasing" here.
Sailor Ted
Well-known
jlw said:
With all due respect and a fair amount of conjecture, cars do have fender benders. If this car has had such a mishap perhaps it was repainted on the bumper, or around the bumper? If so the repair could have hit the color spot on when viewing straight on but could show a different cast when viewed from a different angle. I had just such a problem on a car of mine involved in a bender some time back and I did not notice it until a couple of years later when this was pointed out by an auto body repair guy.
Again just conjecture but possible?
jlw
Rangefinder camera pedant
Sailor Ted said:
With mutual respect, certainly it's possible. Usually, though, these mismatches are pretty easy to see, especially in good light. You probably would have noticed yours if you had parked the car out in the sun and taken a close look at it, moving your eye around to check how the metallic particles reflect from different angles.
No, normally I don't look at my car that closely either (I'm not a very good painter, so I don't want to look all that closely!) but professional body men and car-show judges notice this kind of stuff immediately. You'll be looking at a car that you think looks absolutely sensational, and then a judge or an old-timer will walk up to it and immediately wave a hand at a certain area and grunt, "Ungh! What happened there?" Once you get used to looking for it, it's not hard to see.
(Those people look at cars kind of the way all of us look at pictures. You know how it goes -- a non-photographer will show you a picture he's taken of his kid or whatever and say, "Wow, isn't that a great shot?" and you'll think, gee, the horizon is tilted and there's a tree growing out of her head and the color balance is 'way too cyan and it looks like he focused about two feet behind her. But of course if you're smart, you smile enthusiastically and say "Great! What a cute kid!")
Anyway, even if the photographers didn't notice it at the time -- which I agree is plausible -- I think the EOS 5D would have had as good a chance to pick up such a paint variation as the M8 would.
So to figure out why it would show up in the M8 picture but not the 5D picture, we have to look at what the M8 has that the 5D doesn't or vice-versa, that might cause this type of effect.
I think the two most likely candidates for producing this kind of artifact are:
1) aliasing -- the 5D has an antialiasing filter and the M8 doesn't, so the M8 would be at risk to produce this effect via the subject detail interacting with the sensor.
2) noise -- this one just occurred to me as I was writing this. The pattern of flecks in metallic paint does look a bit like the pattern of pixel noise, and it's possible to imagine a camera with aggressive noise reduction "thinking" that the paint speckles are noise pixels that need to be suppressed. There's a pretty broad consensus that the 5D has stronger noise reduction than the M8, so it might have smoothed out the paint pattern to a greater extent. (In this case, it would be the M8 that was producing the more "realistic" rendition and the 5D that was producing the more "processed" rendition.)
As to why the speckled effect is visible only in the highlight area and not in the other areas, I suspect that's because of a characteristic of metallic paints. In a good metallic paint job, the metal flecks lie more or less parallel to the car body (you have to wait a certain amount of time between coats to let them settle) and that causes the metallic appearance to vary depending on the angle at which the light strikes it -- that's what gives a metalllic such a "lively" appearance, the way it changes as you move around the car. (There are even some finishes now with dimensional particles that provide different colors of lustre when the light strikes the car from different angles!)
Anyway, I wouldn't categorize this particular effect as an M8 "defect" -- I'd call it more of a "quirk." I suspect every digital camera has some quirks in how it renders particular types of "extreme" subject matter, types that a different camera might render completely normally. We normally don't notice them because these types of anomalous subjects are rare, but somebody who does a lot of one specific type of photography might find that a particular otherwise-great camera has trouble with that subject matter, while another does not.
It's kind of like the situation that arises with slide films, in which (for example) a particular kind of flower looks great on Film A but looks "off" on Film B... which in turn does a better job on a different kind of flower than Film A does. We've gotten used to it over the years with films... with digital, everyone's still crawling up the learning curve!