m9_moire

[jaapv]

Advertisement (gone after registration)

Quick answer - normally not.

Longer answer. Moire is a pattern created by a matrix. As a sensor is regular it will produce such a matrix. Film is random, so it will not, unless you shoot through a mesh or something like that.

[mjh]

Based on my eperience with 5D II and with M9 - the ratio of pictures with moire is a lot higher in case of M9.

Of course it is. The EOS 5D Mark II has an antialiasing filter while the M9 has not. Traditionally, vendors of APS-C and FF DSLRs have always opted for an antialiasing filter to get rid of moiré whereas MF vendors have, as a rule, chosen to do without and to deal with (color) moiré during raw conversion. Consequently those raw converters targeted mostly at APS-C and FF DSLRs don’t do much (if anything) to avoid moiré whereas moiré removal or avoidance is an important issue for raw converters catering for the needs of MF photographers. Both Capture One and Phocus excel here.

 

Now in the light of these facts, bundling the M9 with Lightroom may not have been the best decision, but that’s how it is. You have a camera designed to leave dealing with moiré to the raw converter; there is no way around this fact.

[Jerry_R]

Low pass filters affects resolution. Effective resolution from 12 Mpx u43 GH1 is better than 14 Mpx 500D. Why? GH1 has weak AA filter, 500D regular one, used in dSLRs.

 

Fortunately soft becomes better with each version.

You can have a look here on some comparison: M9, part 8B1

 

Hmmm, have a look too: Logical Designs SharpRaw Image Processing Software

and here (1st and 2nd picture): Logical Designs SharpRaw Image Processing Software

[AlanG]

Just to give a rough comparison of moire. Here is an overall shot that has similar chars to an example on the first page of this thread. (This isn't much of a shot and I did some quick conversions that aren't matched.)

 

Shot with a 5DII

 

Overall and a crop using DXO (no moire reduction.)

 

And a crop using C1 which automatically removed the moire without me having to adjust the slider.

 

So obviously the raw converter makes a difference.

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!

[jaapv]

That, Alan, is a very convincing example.

[Jerry_R]

Isnt't 2nd shot desaturated? Maybe better would be to allign saturation first.

 

Anyway - good example, topic worth now posting on LR forum on its wish list thread

[AlanG]

Advertisement (gone after registration)

Isnt't 2nd shot desaturated? Maybe better would be to allign saturation first.

 

Anyway - good example, topic worth now posting on LR forum on its wish list thread

 

That is a good point and I was really in a hurry when I converted them. But the examples are valid. C1 suppressed that moire pretty well and DXO did not. It may not do as well with more severe examples of moire.

 

I'm away from my office but I did a screen grab on each, removed all color and turned up the contrast and you can clearly see that the moire pattern was removed and not just the color. (They're probably a bit degraded from being screen grabs and then jpegged again.)

 

I'm really busy shooting the next two days then I'm traveling for about 2 weeks. This is just to say that it might take me a while to do some moire tests comparing ACR, C1, and some other converters I own. Getting a handle on using more than one converter really well can be quite challenging.

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!

[brill64]

With all due respect, I have extremely extensive experience in many aspects of photographic printing. Having owned a custom color printing business (Ciba, Type R, CP) and I currently own and use two large format digital printers.

 

However I only had reason to print a very tiny percentage of my photographs. Most of my commercial photography was produced on large format transperencies as that was required by architects, magazines, and ad agnecies for my architectural and some commercial work. Medium format transparencies and slides were used for other applications. All of my stock submissions were transparencies or slides. Occassionally I shot color neg along with the transparencies when prints were needed. And later, I used NPS/NPL due to the unique abilities of those films under mixed light... and had transparencies made from them.

 

The point is that there usually would be no opportunity for me to adjust the image after the shoot so I had to get it very accurate at the time of exposure. (Later we could scan and correct digitally.) This is true for most commercial and advertising photographers even today. Considering that interiors and architectural subjects frequently contain patterns that can result in moire, I really need to use a camera/software system that avoids or suppresses the moire as much as possible since having to correct many images is time consuming and costly. And a lot of very good photographers don't even have the skills to retouch and may not want to learn them.

 

Jobs often need to be shot, converted and posted in a very short time window. Let's say that all of the images from a series has moire, you won't want to retouch all of them. So if it ends up that one of the images the client selects from that series has moire, it could be a problem for you as the client will then want you to fix it in a hurry and you may be away doing another project at that point.

 

A lot of working photographers simply can't spend the time to do any more adjusting to their images than is absolutely necessary, let alone have to look out for moire or other problems and regularly retouch or correct them. It is a completely different situation if your end result is just a few images rather than a few hundred or a few thousand images that get submitted. This is why I think the exclusion of an AA filter in the M9 was the wrong decison. (Cameras with AA filters can still exhibit moire but it seems to be less often and less obvious on them. Perhaps there is also a difference in camera firmware in this regard.) A lot of work with MF digital is shot tethered so you can spot the moire in advance and perhaps shoot from closer or further if the software won't eliminate it.

 

good point but to be fair, not everyone's in advertising and if they were, they probably wouldn't be using just an m9. probably not yet. i like your post, though, it's informed. changing the camera angle or lens or adjusting additional lighting certainly helps if you have the eye to spot it. kodak must know about moire issues intimately by now, they've been in digital development for quite some time..maybe a little younger but so must have leica, however they decided in their combined wisdom not to go with the aa filter. that decision would have have been an informed and educated one. every digital back or camera ccd exhibits moire to some extent or another, again to be fair. although i've used it for such, the m9 is not really designed with shooting interiors in mind but if you do use it, check your images preferably on a laptop. after all, shooting interiors is a very careful craft..my feeling is that this is going to be a software fix as with other cameras/digital backs.

[35mmSummicron]

To be fair, CCD sensor technology has come a long with in terms of suppressing instances of moire. Just take a look at a Phaseone P25 (moire would show up often) vs their newer backs like the P30, P40, P65. MUCH less moire prone.

 

Yes raw software definitely comes into play, which is partly why I have ALWAYS used C1 Pro.

 

The M9 is for sure prone not immune to moire, however I would like to restate my previous questions--why am I seeing more instances of moire in C1 Pro from M9 compared to M8 raw files, and why are the moire tools in C1 less effective on M9 files?

 

 

 

good point but to be fair, not everyone's in advertising and if they were, they probably wouldn't be using just an m9. probably not yet. i like your post, though, it's informed. changing the camera angle or lens or adjusting additional lighting certainly helps if you have the eye to spot it. kodak must know about moire issues intimately by now, they've been in digital development for quite some time..maybe a little younger but so must have leica, however they decided in their combined wisdom not to go with the aa filter. that decision would have have been an informed and educated one. every digital back or camera ccd exhibits moire to some extent or another, again to be fair. although i've used it for such, the m9 is not really designed with shooting interiors in mind but if you do use it, check your images preferably on a laptop. after all, shooting interiors is a very careful craft..my feeling is that this is going to be a software fix as with other cameras/digital backs.

[wlaidlaw]

As someone who chose to stay with C1 and paid the full price to get C1 V5 Pro, I think this whole thread points out the problems having to go with LR, at I suspect quite a late stage in the M9's development. If Leica knew early on that they were going to bundle LR, which has no Moiré reduction tool, would they have put on an AA filter - maybe.

 

I have been sent a few DNG's by forum members in the last day or so, to run through C1 and I think I have at least one member, who is going to buy C1 as a result. I know Jaap's methodology works well for limited areas of Moiré and I might do that for exhibition prints but for normal stuff, the C1 sliders do a good job, without too much degradation of the image.

 

Wilson

[diogenis]

I always loved C1 myself as well. And still using it with my M8.

Aren't there some special tools to reduce the moire? One has to buy and learn the whole C1 package for it? Nik s/w or other shouldn't they make addon tools for this?

[wlaidlaw]

I always loved C1 myself as well. And still using it with my M8.

Aren't there some special tools to reduce the moire? One has to buy and learn the whole C1 package for it? Nik s/w or other shouldn't they make addon tools for this?

 

I don't think the C1 plug in is available any more but Neat Image does one. It also does noise reduction. I don't know how good it is at either function Neat Image plug-in for Photoshop 64-bit /Win :: overview as I use C1 and Noise Ninja. Descreen also does a plug in but it is designed for a different purpose, getting rid of moiré from scanned half tones. It may work for normal moiré. Finally there is a moiré reducer on the Fred and Miranda package. Not cheap but all F&M products have a very good reputation.

 

Wilson

[george_b]

. . .Finally there is a moiré reducer on the Fred and Miranda package. Not cheap but all F&M products have a very good reputation.

 

Wilson

All the Fred Miranda packages have not yet been upgraded to CS4, I have the moire one which i will load on my laptop later which has CS3. I'll report back.

[wlaidlaw]

All the Fred Miranda packages have not yet been upgraded to CS4, I have the moire one which i will load on my laptop later which has CS3. I'll report back.

 

F&M will miss the boat - CS5 due out later this year.

 

Wilson

[brill64]

I don't think the C1 plug in is available any more ...

 

Wilson

 

after capture one 3.7.2, they stopped including a moire tool. you can still find it as a free raw plug-in for cs2 only in the archive of capture one previous editions.

[AlanG]

...The M9 is for sure prone not immune to moire, however I would like to restate my previous questions--why am I seeing more instances of moire in C1 Pro from M9 compared to M8 raw files, and why are the moire tools in C1 less effective on M9 files?

 

The posted examples of moire from the M9 looks much more severe than what I have seen from Canon cameras. C1 can only do so much. It can't generate detail that has been wiped out by moire. I'd bet the lack of the AA filter is the principal reason for the M9's moire being so strong. But for all I know there could also be better moire suppression in the Canon firmware. Cameras do have to generate an RGB file from the Bayer pattern afterall. And that require a lot of interpolation.

[george_b]

F&M will miss the boat - CS5 due out later this year.

 

Wilson

Interestingly after I loaded the FM action onto my laptop so I knew how it worked I then loaded it onto my 64bit Vista with CS4 P/C it seems to work OK. I get similar results to Wilson from the same test image I had sent him. I would find it a pain to have to do it routinely but it will rescue images for me on the hopefully few essential occasions.

[wlaidlaw]

The posted examples of moire from the M9 looks much more severe than what I have seen from Canon cameras. C1 can only do so much. It can't generate detail that has been wiped out by moire. I'd bet the lack of the AA filter is the principal reason for the M9's moire being so strong. But for all I know there could also be better moire suppression in the Canon firmware. Cameras do have to generate an RGB file from the Bayer pattern afterall. And that require a lot of interpolation.

 

Alan,

 

I suspect that the moiré tool on C1 does a similar job to an AA filter. I feel the point is that 95%+ of your photos will not need the moiré sliders moved off their bottom stops, so that you can benefit from the additional sharpness conferred by having no AA filter on the M9. When you need moiré reduction, you are no worse off than having an AA filter. If you have C1 V5 Pro with an M9, you get to have your cake and eat it.

 

Wilson