Does Moire=Sharp=Good ?

Started May 28, 2011 | Discussions
rovingtim
Veteran MemberPosts: 7,925
Like?
I think history is playing its part here
In reply to MaxB1, May 31, 2011

Big Ga was using Oly for pro work. However, when the E3 came out he noticed some rather serious issues, most importantly, focusing unreliability. As a friend of this forum, he dutifully reported those issues (along some others) as he found them and he got slammed for his efforts. He (and some others, including myself) call accused of making it up, unable to use the camera properly, as a troll and a liar.

IN the end, everything he originally said and got slammed for proved to be true.

This made him ill disposed to people he 'love' Oly to such an extent they refuse to listen to anything bad, or cannot understand a rational argument.

In this thread, the intention seems to be a discussion on the moire/sharpness trade-off. He has used Nikon since 2008 and Nikon has been using weak AA's since the D70 so he is well aware of the effects of a weak AA and wanted to discuss it here because its new for Oly users and they have a direct comparison in the E3 (which this forum declared as sharp before the E5 came out).

I'm also interested. I'm also curious why no one is talking about the huge reversal of Oly's camera design philosophy that the E5 represents (optically correct everything for superior IQ).

Reply   Reply with quote   Complain
pris
Senior MemberPosts: 2,144
Like?
Re: I think history is playing its part here
In reply to rovingtim, May 31, 2011

rovingtim wrote:

In this thread, the intention seems to be a discussion on the moire/sharpness trade-off.

But there is no trade-off. Sharpness is there, moire is not - at least not to any degree that makes it a problem. Do we have any evidence that E-5 is so moire-prone that it's really a trade-off? 2 or 3 threads with samples of more since E-5 release? Shrug.

. I'm also curious why no one is talking about the huge reversal of Oly's camera design philosophy that the E5 represents (optically correct everything for superior IQ).

I for one don't see it that way. AA filter as a method to prevent moire was always a compromise solution; if Olympus found the way to give more shine to its lenses (optical excellence, as you say) by applying different solution to that marginal problem, kudos to them. If that solution happened to be software-based, it's far from being a "huge reversal of philosophy" - rather opposite, they put emphasis on hardware solutions for better IQ while dealing with rare side-effect with whatever works. How is it all a bad thing?

Reply   Reply with quote   Complain
Rriley
Forum ProPosts: 21,843Gear list
Like?
Re: I think history is playing its part here
In reply to rovingtim, May 31, 2011

rovingtim wrote:

Big Ga was using Oly for pro work. However, when the E3 came out he noticed some rather serious issues, most importantly, focusing unreliability. As a friend of this forum, he dutifully reported those issues (along some others) as he found them and he got slammed for his efforts. He (and some others, including myself) call accused of making it up, unable to use the camera properly, as a troll and a liar.

this situation is nothing like that. Ive been sitting here watching this BS thread circulate for days, where a bunch of people attempt to reason why E5 is as it is, not one of them owned or used an E5, not one of them got to the truth of 'Does Moire=Sharp=Good ?'

People left this thread well alone, and they were wise to do so as its just more nonsense. Yet the thread lived b/se the trolls kept the thing alive, nothing else really happened until around the time stirling posted crops others dont seem to be able to duplicate, not that this in itself is significant

IN the end, everything he originally said and got slammed for proved to be true.

This made him ill disposed to people he 'love' Oly to such an extent they refuse to listen to anything bad, or cannot understand a rational argument. Even then though he isnt 'slammed' as you put it for that alone.

The repeated circumstance that his ideas alone are right, and people that actually own and use the thing are wrong isnt on message, its trolling, and thats pretty well all he's become known for.

the notion that Does Moire=Sharp=Good ? isnt right, as many have attempted to point out. So either he and his message are wrong or he refuses to listen. None of that makes him and his opinion either useful true or rational

In this thread, the intention seems to be a discussion on the moire/sharpness trade-off.

there isnt one, you get moire on occasion like any other camera, and when you do you deal with it like anything else. Moire is something like 5% of digital imaging, actually I think its rather less, that doesnt suggest a trade-off at all, it suggests an incident which BTW occurs on E3 too

He has used Nikon since 2008 and Nikon has been using weak AA's since the D70 so he is well aware of the effects of a weak AA and wanted to discuss it here because its new for Oly users and they have a direct comparison in the E3 (which this forum declared as sharp before the E5 came out).

I'm also interested. I'm also curious why no one is talking about the huge reversal of Oly's camera design philosophy that the E5 represents (optically correct everything for superior IQ).

since for whatever reason you need it spelled out AGAIN

having less intrusive AA filter means more detail gets beyond the lenses to the imaging engine be-it RAW or jpeg. MTF scores are significantly higher than E3 which did not have the same technology, without that technology to deal with instances of moire in the jpeg shooting regime would be unacceptable to Olympus IMO.

is the overall package of E5 better than E3, yes

case closed
--
Riley

any similarity to persons living or dead is coincidental and unintended

 Rriley's gear list:Rriley's gear list
Canon EOS 5D Olympus E-3 Olympus E-5 Panasonic Lumix DMC-GH2 Olympus E-M1
Reply   Reply with quote   Complain
JohnWheeler
Forum MemberPosts: 91
Like?
A newbie to DPReview and my perspective on Moire
In reply to rovingtim, May 31, 2011

Hi all

I almost hate this to be my first post on DP Review. Amateur Photographer for 40 years, Nikon based yet better at Photoshop and retouching than the consummate Photographer (more left brained than right brained). Joined DP Review because I have not upgraded by camera body for 5 years and thought these would be good forums to start gathering information. This seemed like an active thread. Read through the whole thing (wow) and still having a hard time acutally distilling down the essence.

Given that the ReTouching forum on DP Review is pretty sparse I thought I probably would not be making any posts yet I do know some things about Camera Sensor based Moire based on the physics of the interactions with the Bayer Sensor. I even even posted on my blog recently about a new way to remove the luminosity component for some types of Moire patterns. My real comments are about the original question about Moire and sharpness so will give you my thoughts.

If you are going to buy a camera and its the "dog" and "Moire" is the tail, you want the dog wagging the tail, not the other way around. Moire is quite difficult to objectively measure as a purchase criteria, unless there is overwhelming evidence the camera creates real problems for the types of images you take, I say it should be low on the list of considerations especially given the post processing techniques to fix any issue that shows up. Here are just some of he considerations that impact Moire

  • AA filter on sensor

  • The offending image pattern frequency must be close to the frequency of the 2x2 Bayer sensor pattern (or sub harmonic or super harmonic). This by itself makes Apples to Apples comparison hard

  • Related to the prior point, the exact distance to the offending image also makes a huge difference. Are the testers accurately controlling the distance to the image.

  • Moire is highly dependent on the de-mosaicing algorithm/software processing. Even the best will have its own Achilles heal. Trying to do apples to apples comparison would assume you have figured out each algorithms Achilles heal. And that software can change over the life of the camera. You can also use other Raw Converters and get a different result.

  • We have seen how the algorithm using its best efforts will actually introduce its own very strange patterns. This is not a camera issue, rather more a software/algorithm issue. Those can be updated.

  • The angle of the offending image pattern relative to the sensor plays a big role if the image pattern will fool the de-mosaicing software and allow the false striations (color or luminosity) to get by the software.

  • Is the focus tac on the offending image. Lack of focus is like a stronger AA filter. Big imact on any testing

  • Secondarily and usually not an issue is the lens quality for a similar reason as the previous post. Stopping down too far introduces diffraction which can impact the results

Trying to compare just the camera portion of the above list is problematic due to all the variables, lack of controlling all those variables in testing and not considering that each sensor typically has a different pixel density so different trigger point for the Moire issues.

That on top of do you even shoot items very often that can create the problem, will it be a distraction in the image, and how hard is it to repair.

If really concerned, add in the practice to bracket either shooting distance a tad or just the focal zoom point a tad if using a zoom. If one shot has it typically the other may not (I think this practice is overkill in general yet depends on how critical the shot)

Again, Moire is the tail on the dog IMHO of course.

So unless I read somewhere that a camera body/software is extremely susceptible to Morie, it would be way way down my list of decision factors for a camera body.

Sorry about such a long post and if I have stepped over any of the social norms of the forums, apologies. I was more than anything after spending a good deal of time reading this thread, trying to summarize the essence of what I got out of it especially given what I know about Camera Sensor based Moire patterns.

So by the way, I am John from Colorado. Hi. If you want me to go away just let me know and no hard feelings. Thanks for listening. Hope it was useful.

-- hide signature --

John Wheeler

Reply   Reply with quote   Complain
Vaughan Williams
Contributing MemberPosts: 734
Like?
Stick around
In reply to JohnWheeler, May 31, 2011

JohnWheeler wrote:

So by the way, I am John from Colorado. Hi. If you want me to go away just let me know and no hard feelings. Thanks for listening. Hope it was useful.

Knowledgeable, articulate and polite: this forum needs more people like you!

-- hide signature --
Reply   Reply with quote   Complain
boggis the cat
Veteran MemberPosts: 6,267Gear list
Like?
LR demosaicing algorithm at fault
In reply to jim stirling, May 31, 2011

I looked at developing the E-5 raw in Lightroom, and it does have issues with colour artefacts. This appears to be a weakness in the demosaicing algorithm used.

This shows the output from the E-5 raw using RawTherapee with different demosaicing algorithms -- the part within [] of the filename is the method selected:

The "amaze" algorithm shows no colours or artefacts at all when viewed locally. The recompression or other fiddling that DPR seems to use has made it appear that there is some.

There is still some evidence of the angular aberration between the 1" and 80" markings on the scale, but it is simply grey when produced from the "amaze" algorithm.

So, the question is: does anybody know how to change LR demosaicing?

 boggis the cat's gear list:boggis the cat's gear list
Olympus E-5 Olympus E-M1 Olympus Zuiko Digital ED 12-60mm 1:2.8-4.0 SWD Olympus Zuiko Digital ED 50-200mm 1:2.8-3.5 SWD Olympus 12-40mm F2.8 +10 more
Reply   Reply with quote   Complain
Sergey_Green
Veteran MemberPosts: 6,130Gear list
Like?
Have I misunderstood you?
In reply to Rriley, May 31, 2011

Rriley wrote:

is the overall package of E5 better than E3, yes

So that means you do not see moire, or does it mean you do see it but not enough to be bothered by it. If you have been watching this thread long enough ...

It is always good to have an (honest) opinion from those who own the camera. And the more such threads exist the better it is for those who are about to own (or not to own) it. Personally, although not interested in E-5 specifically, I do find the thread rather informative.

-- hide signature --

- sergey

 Sergey_Green's gear list:Sergey_Green's gear list
Sigma 18-35mm F1.8 DC HSM
Reply   Reply with quote   Complain
Rriley
Forum ProPosts: 21,843Gear list
Like?
Re: Have I misunderstood you?
In reply to Sergey_Green, May 31, 2011

Sergey_Green wrote:

Rriley wrote:

is the overall package of E5 better than E3, yes

So that means you do not see moire, or does it mean you do see it but not enough to be bothered by it. If you have been watching this thread long enough ...

my statement was there is no more moire in E5 than E3

It is always good to have an (honest) opinion from those who own the camera. And the more such threads exist the better it is for those who are about to own (or not to own) it. Personally, although not interested in E-5 specifically, I do find the thread rather informative.

be informed by this

all RAW crops for respective dpr reviews

Olympus E5

Fuji X100

Nikon D5100

Canon 60D

Panasonic GH2

Nikon D7000

Nikon D3100

Pentax K5

Sony SLT A55

Canon 50D

Nikon D300s

now theres nothing to 'misunderstand'

-- hide signature --

Riley

any similarity to persons living or dead is coincidental and unintended

 Rriley's gear list:Rriley's gear list
Canon EOS 5D Olympus E-3 Olympus E-5 Panasonic Lumix DMC-GH2 Olympus E-M1
Reply   Reply with quote   Complain
coudet
Veteran MemberPosts: 3,602
Like?
Moire=Yuck!=Bad /NT
In reply to Big Ga, May 31, 2011

no text here, senor

Reply   Reply with quote   Complain
jim stirling
Veteran MemberPosts: 5,553Gear list
Like?
Re: LR demosaicing algorithm at fault
In reply to boggis the cat, May 31, 2011

boggis the cat wrote:

I looked at developing the E-5 raw in Lightroom, and it does have issues with colour artefacts. This appears to be a weakness in the demosaicing algorithm used.

This shows the output from the E-5 raw using RawTherapee with different demosaicing algorithms -- the part within [] of the filename is the method selected:

The "amaze" algorithm shows no colours or artefacts at all when viewed locally. The recompression or other fiddling that DPR seems to use has made it appear that there is some.

There is still some evidence of the angular aberration between the 1" and 80" markings on the scale, but it is simply grey when produced from the "amaze" algorithm.

So, the question is: does anybody know how to change LR demosaicing?

Thanks for taking the trouble to look at the issue, I don’t think it is such a big deal and I am seriously considering getting an E-5 as I am pretty impressed with the low iso images coming from it, I have FF for low light high ISO stuff and I think using it at low ISO the results from the E-5 are excellent. I am only interested in the moiré issue as years ago I had a D70 and had a real problem with moiré it appeared a lot in any shots with fine mesh and it was a royal pain to deal with if you actually wanted the detail it was obscuring, it looks like the E-5 is a lot less likely to display it. And the jpeg engine seems to do a decent job of getting rid of most of it.
Cheers
Jim

 jim stirling's gear list:jim stirling's gear list
Olympus C-8080 Wide Zoom Panasonic Lumix DMC-GF1 Nikon D800 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GX7 +13 more
Reply   Reply with quote   Complain
mujana
Senior MemberPosts: 2,868Gear list
Like?
John
In reply to JohnWheeler, May 31, 2011

Please stay; very informative post from you.
Thnxs for contributing!

 mujana's gear list:mujana's gear list
Olympus OM-D E-M5 Olympus M.Zuiko Digital ED 9-18mm 1:4.0-5.6 Olympus M.Zuiko Digital 45mm 1:1.8 Olympus M.Zuiko Digital ED 75mm 1:1.8
Reply   Reply with quote   Complain
boggis the cat
Veteran MemberPosts: 6,267Gear list
Like?
Re: LR demosaicing algorithm at fault
In reply to jim stirling, May 31, 2011

jim stirling wrote:

Thanks for taking the trouble to look at the issue, I don’t think it is such a big deal and I am seriously considering getting an E-5 as I am pretty impressed with the low iso images coming from it, I have FF for low light high ISO stuff and I think using it at low ISO the results from the E-5 are excellent.

If you want an E-5 due to better portability over your larger system then you may be better to wait a bit and see if the "E-50" materialises. You won't get 100% VF or bomb-proof construction, but you may (should!) get a new sensor and should get at least as much detail as from the E-5. Even if you still decided on an E-5, a new E-50 should force the price to drop a bit.

You need to have worthwhile lenses to use with the body as well. I only have HGs, but they seem to perform well on the E-5 -- easily as sharp as the E-510, and better than the E-620, and a lot more useful detail. Contrast seems better, too, due to the detail (or vice versa, maybe).

I am only interested in the moiré issue as years ago I had a D70 and had a real problem with moiré it appeared a lot in any shots with fine mesh and it was a royal pain to deal with if you actually wanted the detail it was obscuring, it looks like the E-5 is a lot less likely to display it. And the jpeg engine seems to do a decent job of getting rid of most of it.

Olympus must have a good demosaicing algorithm. Note that I have found that you can eke out even more detail using RT and the "amaze" option. Unfortunately, I can't get the colours right in RT.

 boggis the cat's gear list:boggis the cat's gear list
Olympus E-5 Olympus E-M1 Olympus Zuiko Digital ED 12-60mm 1:2.8-4.0 SWD Olympus Zuiko Digital ED 50-200mm 1:2.8-3.5 SWD Olympus 12-40mm F2.8 +10 more
Reply   Reply with quote   Complain
boggis the cat
Veteran MemberPosts: 6,267Gear list
Like?
Beware of the demosaic dragon...
In reply to Rriley, May 31, 2011

As I posted up-thread:
http://forums.dpreview.com/forums/read.asp?forum=1022&message=38554801

All of those DPR resolution charts use ACR to process the raw files. ACR is what Lightroom uses, and Lightroom certainly has some issues getting good results out of E-5 files (as you can see from Jim Stirling's posted crops).

So just because a result looks poor from one raw development system it does not necessarily follow that the camera has worse moire/artefact issues than another.

Olympus' JPEG engine seems to handle the issue extremely well, but on the flip side it doesn't get the most detail from the files.

 boggis the cat's gear list:boggis the cat's gear list
Olympus E-5 Olympus E-M1 Olympus Zuiko Digital ED 12-60mm 1:2.8-4.0 SWD Olympus Zuiko Digital ED 50-200mm 1:2.8-3.5 SWD Olympus 12-40mm F2.8 +10 more
Reply   Reply with quote   Complain
jim stirling
Veteran MemberPosts: 5,553Gear list
Like?
Re: LR demosaicing algorithm at fault
In reply to boggis the cat, May 31, 2011

boggis the cat wrote:

jim stirling wrote:

Thanks for taking the trouble to look at the issue, I don’t think it is such a big deal and I am seriously considering getting an E-5 as I am pretty impressed with the low iso images coming from it, I have FF for low light high ISO stuff and I think using it at low ISO the results from the E-5 are excellent.

If you want an E-5 due to better portability over your larger system then you may be better to wait a bit and see if the "E-50" materialises. You won't get 100% VF or bomb-proof construction, but you may (should!) get a new sensor and should get at least as much detail as from the E-5. Even if you still decided on an E-5, a new E-50 should force the price to drop a bit.

The E-5 is light enough for me relatively speaking I think you have to wait anyway as they are usually on backorder , I would prefer to wait and find one in a shop for a hands on. FF frame gives me the advantages i need for high ISO etc. I principally use Nikon gear and just recently sold off some Canon equipment I just was not using , and I have a good few bob in the kitty , it is really just for pleasure .

You need to have worthwhile lenses to use with the body as well. I only have HGs, but they seem to perform well on the E-5 -- easily as sharp as the E-510, and better than the E-620, and a lot more useful detail. Contrast seems better, too, due to the detail (or vice versa, maybe).

I agree about the lenses , I only have the 12-60 and 50 macro at the moment, I would look at the 35-100 f2 and the 14-35 { though it is very pricey here in the UK}. I would certainly be keeping the 50mm macro as it delivers great results. To be honest if I thought that Olympus or Panasonic for that matter was going to step up with a more rugged mFT body and a couple of high grade lenses in the near future I would probably opt for that . But it all seems to be rumors and maybes at the moment. I do like the mFT system but it is not at the level of performance of the DSLR's yet and I find AF of the Olympus FT lenses a bit slow for my liking
i

I am only interested in the moiré issue as years ago I had a D70 and had a real problem with moiré it appeared a lot in any shots with fine mesh and it was a royal pain to deal with if you actually wanted the detail it was obscuring, it looks like the E-5 is a lot less likely to display it. And the jpeg engine seems to do a decent job of getting rid of most of it.

Olympus must have a good demosaicing algorithm. Note that I have found that you can eke out even more detail using RT and the "amaze" option. Unfortunately, I can't get the colours right in RT.

I haven't tried RT and tend to stick with the software I am comfortable with , I will have a wee look at it . The Olympus jpeg engine does a great job all round.

Jim

 jim stirling's gear list:jim stirling's gear list
Olympus C-8080 Wide Zoom Panasonic Lumix DMC-GF1 Nikon D800 Panasonic Lumix DMC-GH3 Panasonic Lumix DMC-GX7 +13 more
Reply   Reply with quote   Complain
papillon_65
Forum ProPosts: 15,665Gear list
Like?
Just what we need....
In reply to coudet, May 31, 2011

on this forum, another drive by from a Nikon numpty, it must be a brand specific thing, it even happens on the m4/3's forum.
--
It's a known fact that where there's tea there's hope.
Tony
http://the-random-photographer.blogspot.com/

 papillon_65's gear list:papillon_65's gear list
Canon PowerShot G1 X Sigma DP1 Merrill Sigma DP2 Merrill Panasonic Lumix DMC-FZ200 Canon EOS 5D Mark II +4 more
Reply   Reply with quote   Complain
pris
Senior MemberPosts: 2,144
Like?
Re: A newbie to DPReview and my perspective on Moire
In reply to JohnWheeler, May 31, 2011

JohnWheeler wrote:

Hope it was useful.

It really was. Informative and competent, what more do we need.

John, I'd be interested to hear your take on this method of removing rainbow component of moire: http://forums.dpreview.com/forums/read.asp?forum=1022&message=38549300

Seems to be working very well on every sample I tried, but I haven't found that many to try it on.

Reply   Reply with quote   Complain
John Mason
Veteran MemberPosts: 5,616Gear list
Like?
John - don't worry about it - it is a clear troll post
In reply to JohnWheeler, May 31, 2011

troll post being one that poses a fake question leaving out known information on purpose to get a rise out of people

That said - your first post was very cogent. But trolls love cogent posts as they like to try to get you to argue with them.

I usually just pop on one or two statements then go away on a troll post. It's my contribution to get these waste of time posts up to 150 sooner so they die and go away.
--
John Mason - Lafayette, IN

http://www.fototime.com/inv/407B931C53A9D9D

 John Mason's gear list:John Mason's gear list
Olympus E-M1 Panasonic Lumix G 20mm F1.7 ASPH Olympus Zuiko Digital ED 7-14mm 1:4.0 Olympus Zuiko Digital ED 14-35mm 1:2.0 SWD Panasonic Leica Summilux DG 25mm F1.4 +5 more
Reply   Reply with quote   Complain
Rriley
Forum ProPosts: 21,843Gear list
Like?
Re: A newbie to DPReview and my perspective on Moire
In reply to pris, May 31, 2011

pris wrote:

Seems to be working very well on every sample I tried, but I haven't found that many to try it on.

I'd check out the Nikon forums, there should be any number of good examples of moire over there

D3100

-- hide signature --

Riley

any similarity to persons living or dead is coincidental and unintended

 Rriley's gear list:Rriley's gear list
Canon EOS 5D Olympus E-3 Olympus E-5 Panasonic Lumix DMC-GH2 Olympus E-M1
Reply   Reply with quote   Complain
illy
Forum ProPosts: 12,160Gear list
Like?
Re: A newbie to DPReview and my perspective on Moire
In reply to Rriley, May 31, 2011

Rriley wrote:

pris wrote:

Seems to be working very well on every sample I tried, but I haven't found that many to try it on.

I'd check out the Nikon forums, there should be any number of good examples of moire over there

D3100

-- hide signature --

Riley

any similarity to persons living or dead is coincidental and unintended

how would anyone tell, all aps-c have fuzzy edges...everybody knows that
--
Smoke me a kipper....i'll be back for breakfast

 illy's gear list:illy's gear list
Nikon D200 Panasonic Lumix DMC-GF1 Nikon D5100 Nikon AF-S DX Nikkor 55-200mm f/4-5.6G VR Nikon AF-S DX Nikkor 18-55mm f/3.5-5.6G VR +4 more
Reply   Reply with quote   Complain
papillon_65
Forum ProPosts: 15,665Gear list
Like?
Re: A newbie to DPReview and my perspective on Moire
In reply to illy, May 31, 2011

illy wrote:

Rriley wrote:

pris wrote:

Seems to be working very well on every sample I tried, but I haven't found that many to try it on.

I'd check out the Nikon forums, there should be any number of good examples of moire over there

D3100

-- hide signature --

Riley

any similarity to persons living or dead is coincidental and unintended

how would anyone tell, all aps-c have fuzzy edges...everybody knows that

They certainly have some stellar performers in the Moire stakes, very impressive lol. It's nice to see the GH2 with its tiny sensor kicking some serious ass in there :).

-- hide signature --

Smoke me a kipper....i'll be back for breakfast

-- hide signature --

It's a known fact that where there's tea there's hope.
Tony
http://the-random-photographer.blogspot.com/

 papillon_65's gear list:papillon_65's gear list
Canon PowerShot G1 X Sigma DP1 Merrill Sigma DP2 Merrill Panasonic Lumix DMC-FZ200 Canon EOS 5D Mark II +4 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads