Jump to content

Reidreviews.com on G1


nugat

Recommended Posts

Advertisement (gone after registration)

I wonder if this is fall-off in resolution is linked to the maze-like artefacts that some have noticed and commented on? Here's a link to my posting on the subject :

 

G1 : Problem with "maze pattern" artefacts with 3rd Party lenses - The GetDPI Workshop Forums.

 

A contributor to the thread noted that they don't occur when Silkypix is used for the RAW conversion. I checked this out. They don't.

 

I find the optical explanation for the issue to be quite compelling - but perhaps the maze-like structures are aalso a consequence of this?

Link to post
Share on other sites

  • Replies 49
  • Created
  • Last Reply

But the adapter isn't just a spacer, especially for a Leica M lens; there is the potential (actually, high probability) for light hitting the sides of the adapter and scattering - could give all sorts of strange effects.

 

Sandy

 

Hi Sandy,

 

Actually, a spacer is exactly what it is. But that doesn't mean that your latter theories are wrong. There are a lot of possible reasons for what I observed but the key thing is that, whatever the causes, the result exists.

 

Replicate....do you have a G1 and an adapter?

 

Cheers,

 

Sean

Link to post
Share on other sites

I wonder if this is fall-off in resolution is linked to the maze-like artefacts that some have noticed and commented on? Here's a link to my posting on the subject :

 

G1 : Problem with "maze pattern" artefacts with 3rd Party lenses - The GetDPI Workshop Forums.

 

A contributor to the thread noted that they don't occur when Silkypix is used for the RAW conversion. I checked this out. They don't.

 

I find the optical explanation for the issue to be quite compelling - but perhaps the maze-like structures are aalso a consequence of this?

 

I don't read that forum but the moire one sees in some G1 files isn't tied to RF lenses but rather to the cameras (wonderfully) low level of AA filtering. The G1 is in the M8's spirit in that respect.

 

Cheers,

 

Sean

Link to post
Share on other sites

... the moire one sees in some G1 files isn't tied to RF lenses but rather to the cameras (wonderfully) low level of AA filtering. The G1 is in the M8's spirit in that respect.

 

Cheers,

 

Sean

 

Sean,

 

I don't think that the patterning that some have been getting is moire. It has a distinct rectangular structure and - in many of the cases I have seen - it appears in featureless sky - usually in the upper right hand corner of the frame. I'm well-used to seeing moire on the output from my M8. And it isn't at all like this effect.

 

Regards,

 

Geoff

Link to post
Share on other sites

Sean,

 

I don't think that the patterning that some have been getting is moire. It has a distinct rectangular structure and - in many of the cases I have seen - it appears in featureless sky - usually in the upper right hand corner of the frame. I'm well-used to seeing moire on the output from my M8. And it isn't at all like this effect.

 

Regards,

 

Geoff

 

Interesting...thanks for the additional information.

 

Cheers,

 

Sean

Link to post
Share on other sites

Advertisement (gone after registration)

I wonder if this is fall-off in resolution is linked to the maze-like artefacts that some have noticed and commented on? Here's a link to my posting on the subject :

 

G1 : Problem with "maze pattern" artefacts with 3rd Party lenses - The GetDPI Workshop Forums.

 

A contributor to the thread noted that they don't occur when Silkypix is used for the RAW conversion. I checked this out. They don't.

 

I find the optical explanation for the issue to be quite compelling - but perhaps the maze-like structures are aalso a consequence of this?

 

That maze pattern is absolutely a raw converter issue; it's well known as a problem if the parameters of the demosaicing algorithm aren't well matched to the sensor. In engineering speak its a kind of an "undamped oscillation" that happens because when you demosaic, every pixel depends on the one next door. But it's a problem that can be "provoked" by an alternating pattern in the underlying sensor data. I've seen Capture One do this on noise, or close to sharp transitions. So the two problems could be related; if somehow you're getting almost a double image, that could cause the resolution loss, and provoke the maze pattern. Kind of points to the AA filter. But I'm speculating.....

 

Sandy

Link to post
Share on other sites

OK, so its a matter of definition.

 

But, more importantly, you have a scientific mind....want to do some testing yourself and see what you find?

 

Cheers,

 

Sean

 

No G1 to test with. Happy to take a close look at the your original images and see if I spot anything interesting though, if you can upload them.

 

Regards,

 

Sandy

Link to post
Share on other sites

That maze pattern is absolutely a raw converter issue; it's well known as a problem if the parameters of the demosaicing algorithm aren't well matched to the sensor. In engineering speak its a kind of an "undamped oscillation" that happens because when you demosaic, every pixel depends on the one next door. But it's a problem that can be "provoked" by an alternating pattern in the underlying sensor data. I've seen Capture One do this on noise, or close to sharp transitions. So the two problems could be related; if somehow you're getting almost a double image, that could cause the resolution loss, and provoke the maze pattern. Kind of points to the AA filter. But I'm speculating.....

 

Sandy

 

OK: totally unscientific samples. Conslusion: LR produces more of a mazepattern than Silkypix. Upper right corner magnified. Would indeed not be surprised if this contributes to softness.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

Sean,

 

The images that appear to be reversed are the pair right after the ISO heading, B&W G1 on left and M8 on right but image on right is far more magnified than one on left.

 

Gotta duck out but will check and get back to you later today.

 

Cheers,

 

Sean

Link to post
Share on other sites

OK: totally unscientific samples. Conslusion: LR produces more of a mazepattern than Silkypix. Upper right corner magnified. Would indeed not be surprised if this contributes to softness.

 

More "unscientific samples" that illustrate the same phenomenon using the Zeiss 25mm f/2.8 Biogon on the G1 and also M8 for comparison. Same maze pattern and distortion in corners, but also apparently more pronounced CA in the G1 upper left corner sample in 2nd image. I'm also seeing this with some other wide M lenses but not with longer focal lengths nor with Canon FD lenses, all on the G1.

 

Lower right corner:

original.jpg

 

Upper left corner:

original.jpg

Link to post
Share on other sites

Seems to me that there is a ovious difference in technology here... after all, the 4:3 system have always been centered around lenses projecting the light more "straight" at the chip, and also the chip is made for this light arriving at a right angle... this have always been part of the 4:3 claim to fame.

 

The Leica lenses works at a much steeper angle, the result of great center image rendering and poor edges seems very consistent with the theory expectations to the match of G1 and M lenses.

 

I'm in the possibly unique position of being able to test a "rangefinder" design lens on a 4:3 camera other than the G1. As some here know, I have been using the (now NLA) Nikon F-mount versions of the CV 12mm and 15mm lenses on my DMC-L1 for nearly two years now and I have not noticed any of the problems Sean observed in his G1 review with either of them (provided they're stopped down to F8, that is, as there is a bit of softness in the corners when they're used wide-open). As I understand it, the optical elements of these F-mount CV lenses are identical to the LTM versions, the only difference being that they're recessed into the mount in order to position them at the proper flange-to-sensor distance when mounted on a Nikon F body.

 

Since the L1 doesn't (to my knowledge) rely upon RAW processing software to correct lens distortions and uses a similar sensor design as the G1, perhaps it would be possible to isolate the effects of the software by comparing RAW images shot with the same lenses on both the L1 and G1?

 

I would volunteer to do this myself, but I've let Mike Johnston at TOP borrow my G1, so it's out of my hands at the moment, but I may be persuaded to make my lenses available to Sean (or someone else similarly qualified) if they wish to pursue this.

Link to post
Share on other sites

Sean,

Before you go (and have a great time!):

1. Which corner you picked?

2. What software you used?

We'll take it from there.

Piotr

 

Hi Piotr,

 

1) It doesn't matter - all four showed the same change.

 

2) As described in the review, CS 4.

 

Thanks for the good wishes on the trip.

 

Cheers,

 

Sean

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...