Main weakness of m43 (according do me)

PakkyT

Mu-43 Hall of Famer
Joined
Jun 20, 2015
Messages
4,102
Location
Massachusetts, USA
"I love the size and quality of M43 lenses, but I hate the way they get those results for me."? 🤷‍♂️

To be fair to the original poster, I think it was just something he was thinking about that if they did a better job optically there would be less need for SW correction and therefore less theoretical IQ reduction. I am not sure he actually SEES that reduction in his photos so more of a theoretical discussion. Then again, he hasn't returned since the original post to follow up, so maybe he does think he sees the IQ reduction. We'll have to wait for his return. :coffee-30:
 

BDR-529

Mu-43 Top Veteran
Joined
Jun 27, 2020
Messages
630
To be fair to the original poster, I think it was just something he was thinking about that if they did a better job optically there would be less need for SW correction and therefore less theoretical IQ reduction.

Getting totally rid of pincussion/barrel distortion and vigneting by optical design alone seems to be both extremely difficult and expensive.

These distortions used the be the ones which were always tested in lens reviews during analog era because they indicated the overall quality of lens (and were very difficult to correct in post processing). Only the most expensive pro lenses were able to deliver optical quality with almost non-existent vigneting and barrel/pincussion. Having them removed by optical design would mean substantial increase in lens price to start with.

Since lens design is always a compromize between several optical properties, something else had to give when these were optimized. So, having less SW correction here would mean more need to correct something else if price point is fixed.

Anyone who is not able to spend $thousands€ on every lens should be absolutely happy with digital corrections because they allow budget lens designers to allocate limited resources to some other optical properties that can't be fixed for free with SW.
 

bargainguy

Mu-43 Regular
Joined
Apr 21, 2015
Messages
137
Location
Milwaukee, WI
Real Name
Don
Sigma addresses in-camera correction as part of their lens design philosophy.

In the Art series, the optical design is maximized without the use of any in-camera correction. Especially difficult the wider you go.

In the Contemporary series, in-camera correction is relied upon to produce the best result.

My Sigma 16/1.4 falls in the Contemporary line. Having said that, it's an outstanding lens. Sharp wide open, best at f/2.8, with very low levels of linear distortion.

It doesn't matter to me that the 16 was designed with in-camera correction. Whatever Sigma is doing, I hope they keep doing it.
 

ABFoz

Mu-43 Regular
Joined
Jan 27, 2021
Messages
70
Location
Tamaki Makaurau, Aotearoa
Real Name
No estoy listo para esto todavía.
Welcome to the forum, @Aloyz!

Almost (?) all of them (even the high-end ones) have a significant distortion (5-6%) which needs to be corrected by software.
Are you talking about barrel distortion? How do you get 5-6%? That's rather high. Are you over-correcting? Which software are you using? Can you specify the lenses you are having significant distortion problems with?

Usually the smaller the lens, the bigger the distortion but many MFT lenses are well-designed to the point that even the consumer-grade Olympus 25mm F/1.8 just has a 0.5% barrel distortion. Bigger pro lenses are bigger because lens corrections are done optically just like the Olympus 25mm F/1.2, which has virtually no barrel distortion. The 5-6% barrel distortion that you usually get can be seen on small but wider-angled lenses like the Panasonic 12-32mm, Olympus 12mm F/2.0, etc.

Distortion is present even in APS-C and small format ILC lenses, especially with consumer-grade ones. With Sony, if you want the same almost distortion-free performance as Olympus Pro lenses, you will have to go with the G-master series which costs much more than the best camera + lens kits MFT can offer.

Do you also see distortion as a serious weakness?
No, because I am more into lighting, exposure, composition and photography being the art of subtraction so some technicalities do not matter to me. Cheers.
 

pdk42

One of the "Eh?" team
Joined
Jan 11, 2013
Messages
7,447
Location
Leamington Spa, UK
On m43, the correction info is built into the lens’s firmware and will be automatically applied, even on raw files using most post processing software (e.g. Adobe Lightroom). You’d have to try hard to over correct.
 
Joined
May 12, 2019
Messages
1,328
Hi all,

After using m43 for more than 2 years I feel like the main disadvantage of the system is the way that lenses (especially the wide ones) are designed. Almost (?) all of them (even the high-end ones) have a significant distortion (5-6%) which needs to be corrected by software. Do you know why such a compromise was taken? It influences image quality. Even though the lenses are a bit wider to compensate it, still the matrix size is the same and details are deteriorated on edges. In landscape photography, where wide angle lenses are mostly used, one wants to have sharpness across the frame.
Apart from distortion, I'm pretty satisfied with the system and hope it will live long...

Do you also see distortion as a serious weakness?

Cheers.
You are welcome to use legacy and adapted lenses from other camera systems on m43 bodies. Many of us did that in the early days when manufacturers had few lenses. We took advantage of in body IS to use legacy glass with IS. Since those lenses were designed for larger sensor formats, they have different kinds of distortion.
 

Aloyz

New to Mu-43
Joined
Feb 24, 2021
Messages
5
I am sorry for posting such a rather controversial subject and disappearing for some time. First of all, I would like to thank you all for the feedback which let me better understand the problem. I lived with the impression that software distortion correction is unique to m43, which seems not to be the case at all. I do not have anything against software correction as long as it does not introduce any side effects.

However, in my case, I can see impact on quality. The fact is that I am comparing it for my kit lens 12-32, but this is the widest lens I have. Maybe its quality at the edges is not sufficient to correct distortion without IQ impact? At 12mm end this lens has around 6% distortion, but even the so called PRO lenses, like Oly 12-40 or Pana 12-35 have around 5-6% distortion at the wide end. Still their edge resolution might be better to allow such operation.

Let me share with you some comparisons that I quickly made. These are just screen shots, but I hope it will be sufficient for you to see. The first file (comparison.png) shows 100% crops of the very edge of the same photo:
- the left one is JPG developed in darktable software with lens correction applied (in darktable distortion correction is based on lensfun db)
- the middle one is JPG developed in darktable without correction
- the right one is OOC JPG

The second file (comparison2.png) contains other 100% crops of another photo. Both were developed in darktable, one with corrections applied, while the other without. I think that it is easy to guess which is which.

Maybe the 12-32 lens quality at the edges is the cause, maybe darktable correction is not the best or maybe I am a pixel addict ;-).

Time to start weekend, take a camera and ..... it has just started raining like cats and dogs.
 

Attachments

  • comparison.png
    comparison.png
    1.1 MB · Views: 53
  • comparison2.png
    comparison2.png
    966 KB · Views: 48

RS86

Mu-43 Top Veteran
Joined
Mar 26, 2019
Messages
959
Location
Finland
Real Name
Riku
I am sorry for posting such a rather controversial subject and disappearing for some time. First of all, I would like to thank you all for the feedback which let me better understand the problem. I lived with the impression that software distortion correction is unique to m43, which seems not to be the case at all. I do not have anything against software correction as long as it does not introduce any side effects.

However, in my case, I can see impact on quality. The fact is that I am comparing it for my kit lens 12-32, but this is the widest lens I have. Maybe its quality at the edges is not sufficient to correct distortion without IQ impact? At 12mm end this lens has around 6% distortion, but even the so called PRO lenses, like Oly 12-40 or Pana 12-35 have around 5-6% distortion at the wide end. Still their edge resolution might be better to allow such operation.

Let me share with you some comparisons that I quickly made. These are just screen shots, but I hope it will be sufficient for you to see. The first file (comparison.png) shows 100% crops of the very edge of the same photo:
- the left one is JPG developed in darktable software with lens correction applied (in darktable distortion correction is based on lensfun db)
- the middle one is JPG developed in darktable without correction
- the right one is OOC JPG

The second file (comparison2.png) contains other 100% crops of another photo. Both were developed in darktable, one with corrections applied, while the other without. I think that it is easy to guess which is which.

Maybe the 12-32 lens quality at the edges is the cause, maybe darktable correction is not the best or maybe I am a pixel addict ;-).

Time to start weekend, take a camera and ..... it has just started raining like cats and dogs.

With wider end, and if you want best results, it seems you have to get FF or one of those Laowa M43 wide lenses with Zero-D (not totally zero distortion, but very good in that area) in the name. I think it has been a long-known fact that FF is better for wide to normal angle shooting. Wide lenses are easier to design for a bigger sensor?

Someone more knowledgeable might tell us how much for example M43 12mm distortion corrections will crop from the resolution. Even the PL 12mm f/1.4 has -5.39 % distortion, and that is a big prime lens.

"As our test concerns the performance of the lens itself we also checked how the optics dealt with distortion without any help; we did it by developing RAW files with the dcraw program which doesn’t correct automatically the geometric flaws of the lens. In that case, at the wide angle, you have to take into account a huge level of barrel distortion, reaching a level of –6.32%. For comparison, at the same focal length the Panasonic 12-35 mm f/2.8 got –6.54%, a tad higher result than that of the Olympus. With the increase of the focal length the distortion decreases: at 14 mm it still remains high, amounting to –5.11%, and at 25 mm it reaches just –0.60%. At higher focal lengths there is a sign change and at 40 mm you already see pincushion distortion of 0.80%.

It is an interesting fact that the Olympus, featuring a wider focal range that the Panasonic 12-35 mm, has better distortion results almost across the range. Unfortunately, though, it is clear the constructors didn’t even try to control distortion effectively, ceding that task completely to the software. We definitely don’t like such an approach because correcting such a high distortion level entails significant cropping of images."

https://www.lenstip.com/392.6-Lens_...Digital_12-40_mm_f_2.8_ED_PRO_Distortion.html
 

ABFoz

Mu-43 Regular
Joined
Jan 27, 2021
Messages
70
Location
Tamaki Makaurau, Aotearoa
Real Name
No estoy listo para esto todavía.
I am sorry for posting such a rather controversial subject and disappearing for some time.
Don't be. I am also new here but have been browsing for some time before signing up and noticed that it's full of good photographers who are more than happy to discuss instead of just compare systems.

I understand, though, what you mean. I just haven't seen the distortion to be an issue because, as I've said, I am interested in photography than technicalities but I am not new to technical discussion because my brother who lives in EU now is a proper gearhead who has around 20 lenses per system.

At 12mm end this lens has around 6% distortion, but even the so called PRO lenses, like Oly 12-40 or Pana 12-35 have around 5-6% distortion at the wide end. Still their edge resolution might be better to allow such operation.
Yes, these zooms have significant distortion at the wide end but even when corrected, they still produce sharp images. I guess they have better optics, generally? That's for an optical engineer to answer, really.

It's hard to think how the Panasonic 12-32mm can correct distortion optically, being the tiny lens that it is.

You can try Olympus primes, Pro and consumer, especially the ones in the standard focal length. They are already corrected optically. Manual ones from Kowa Prominar and Voigtlander have very low distortion values.

- the left one is JPG developed in darktable software with lens correction applied (in darktable distortion correction is based on lensfun db)
- the middle one is JPG developed in darktable without correction
- the right one is OOC JPG
I also use Darktable and also found out that the corrections are less aggressive vs SooC JPEGs just like in your example.

The middle photo demonstrates MFT lenses' telecentricity. They are really sharp.

Someone more knowledgeable might tell us how much for example M43 12mm distortion corrections will crop from the resolution. Even the PL 12mm f/1.4 has -5.39 % distortion, and that is a big prime lens.
Agree. This is optical engineering zone.

This, for me, is an interesting thread because I haven't really thought about distortion, at all, before this. Haha. It still doesn't bother me as I am happy with the stories of my photos.
 

saladin

Mu-43 Hall of Famer
Joined
May 29, 2015
Messages
2,198
Location
Melbourne
Real Name
jason
The only criticism I can make of the MFT lens ecosystems are practical ones:

1) the 1.8 / 1.7 primes are not weather sealed. They should be by now.
2) ideally, Oly bodies would recognise aperture rings and we'd have more lenses with them.

The first makes 100 percent sense given the big promotion MFT likes to make about "small and rugged" . Especially for the Em5 series , and doubly especially for the 5iii. Why have a premium camera that emphasizes how light and weatherproof it is, an online mantra of "breakfree with olympus" and then force prime users to use the much larger Pro lenses to keep the integrity of the weather sealing? You're undermining a key aspect of the camera. Yes, the 12-45 Pro makes total sense for the 5iii. But f/4 is a long way from f/1.8 . Eight years since the release of the Em-5 and the most obvious lenses for that line still don't exist. Perplexing.

The 2nd one is purely for my tactile wants, lol.


I'm not bothered in the slightest by software correction of lens aberation.
 

John King

Mu-43 Hall of Famer
Joined
Apr 20, 2020
Messages
2,571
Location
Beaumaris, Melbourne, Australia
Real Name
John ...
@Aloyz , I suspect that more mainstream software such as Photoshop and Capture One Pro adheres more closely to the manufacturer supplied correction tables.

I do not see this as any kind of problem.

It's rather like removing a lens element from an optically corrected lens (such as the FTs 14-54 MkII), then complaining that it shows distortion ...

The software lens correction profiles are an essential 'lens element'.
 

stevedo

Mu-43 Top Veteran
Joined
May 12, 2012
Messages
606
Location
UK
Real Name
Steve
First of all, I would like to thank you all for the feedback which let me better understand the problem.

There you go again. What problem? As I stated earlier, there are 1,000's of great photos in this forum. Nobody ever said that a single one of them would be much better if there was no software correction.

I challenge you to point all of us to any photos on this forum which are worse for having been partially software corrected (there is always some optical correction as well).
 

BDR-529

Mu-43 Top Veteran
Joined
Jun 27, 2020
Messages
630
However, in my case, I can see impact on quality. The fact is that I am comparing it for my kit lens 12-32, but this is the widest lens I have. Maybe its quality at the edges is not sufficient to correct distortion without IQ impact? At 12mm end this lens has around 6% distortion, but even the so called PRO lenses, like Oly 12-40 or Pana 12-35 have around 5-6% distortion at the wide end. Still their edge resolution might be better to allow such operation.

The fact that pro lenses have nowadays practically the same amount of distortion as kit zooms goes to prove that some of it just isn't a problem in digital imaging as long as it's the only IQ defect.

The main difference between this 12-32 kit lens and pro counterparts is that kit IQ would not be that good around the corners even if there was no distortion at all. You'll see pretty low sharpnes, low contrast and visible colour fringing the further you go away from the center.

All these other IQ problems become pronounced when distortion correction is applied on the very same areas which had the lowest image quality to start with.

That being said, if you really have a need for high quality wide angle, the only way to go is a wide prime which might be surprisingly inexpensive if you are able to use manual focus. Zoom lenses, even pro-ones, are never that good on either end.
 

John King

Mu-43 Hall of Famer
Joined
Apr 20, 2020
Messages
2,571
Location
Beaumaris, Melbourne, Australia
Real Name
John ...
@BDR-529 You have patently obviously never used a FTs 7-14 or 11-22, I surmise ...

Your comments do not accord with my personal experience of either the kit level lenses, Pro level lenses, FTs HG and SHG lenses. I own a whole lot of these lenses, and have used some of them for over 13 years.

Please list the specific lenses that you own, and your specific assessment/s of them.
Not reviews, your personal experience.
 

RS86

Mu-43 Top Veteran
Joined
Mar 26, 2019
Messages
959
Location
Finland
Real Name
Riku
That being said, if you really have a need for high quality wide angle, the only way to go is a wide prime which might be surprisingly inexpensive if you are able to use manual focus. Zoom lenses, even pro-ones, are never that good on either end.

To me it seems the Pro zooms have great corner sharpness at the wide-end, from wide-open. You're right about the tele-end.

Olympus 12-40mm f/2.8 Pro corner resolution chart:
3989_roz_brzeg.jpg
Subscribe to see EXIF info for this image (if available)


https://www.lenstip.com/392.4-Lens_...l_12-40_mm_f_2.8_ED_PRO_Image_resolution.html

PL 12mm f/1.4 corner resolution (this seems not like a good copy for sure, but in any case, it's not much better than Olympus Pro zoom.)
189188_roz2.png
Subscribe to see EXIF info for this image (if available)

https://www.lenstip.com/505.4-Lens_...mmilux_12_mm_f_1.4_ASPH_Image_resolution.html
 
Last edited:

archaeopteryx

Gambian sidling bush
Joined
Feb 25, 2017
Messages
1,646
Maybe the 12-32 lens quality at the edges is the cause, maybe darktable correction is not the best or maybe I am a pixel addict ;-).
I think all three apply to some extent. @junkyardsparkle has posted extensively about lensfun corrections here so I would suggest searching up those posts. With respect to aberrations, it seems likely you're either mistaking coma for distortion or at least partially conflating the two.

Possibly the most substantial thing to recognize is coma is often a primary cause of lower MTF in the corners of (ultra)wide lenses. Correcting barrel distortion increases effective pixel density along an image's sides, nominally acting to increase MTF, so it may also be useful exercise to work through the details of resampling a step function versus the pulse trains used in MTF measurements. Another source of useful data would be to convert to DNG, extract the distortion correction opcodes, and compare to lensfun.
 

Aloyz

New to Mu-43
Joined
Feb 24, 2021
Messages
5
I challenge you to point all of us to any photos on this forum which are worse for having been partially software corrected.

Please, do not take it too personally. I clearly stated that I do not have anything against software corrections as long as these are not introducing side effects. In my case these side effects are visible as I hope I showed you in the attachments. As you will read other comments, you will notice that I am not alone. There was a comment from a PRO lens owner who sometimes disables sw correction to maintain quality.

This is my personal view which I presented here and I do hope that personal views are allowed at this forum...

@archaeopteryx: Thank you for the tips. I will check them.
 

John King

Mu-43 Hall of Famer
Joined
Apr 20, 2020
Messages
2,571
Location
Beaumaris, Melbourne, Australia
Real Name
John ...
The 12-100 tests.

According to Lenstip, only beaten by a tiny margin by a couple of the best primes they have ever tested ...

https://www.lenstip.com/497.4-Lens_...ED_12-100_mm_f_4_IS_PRO_Image_resolution.html

Centre:

1614492178241.png
Subscribe to see EXIF info for this image (if available)


Edges:

1614492381770.png
Subscribe to see EXIF info for this image (if available)


Is it a 'perfect' lens? Of course not. There is no such thing.

Would I choose the primes they mention over it?
No.
 

Attachments

  • 1614492076623.png
    1614492076623.png
    2.7 KB · Views: 10
  • 1614492382503.png
    1614492382503.png
    2.7 KB · Views: 9
Last edited:

BPCS

Mu-43 Regular
Joined
Oct 15, 2017
Messages
146
A few percent distortion software corrected is neither here nor there. Have a look how surprisingly well the Oly 8mm Pro fisheye goes to rectilinear if defished. That must be from 100% distortion to 0%. The end result is almost acceptable on the edges professionally. If a RAW software correction for distortion from even a "massive" 8% results in a fully professionally sharp edge, what is there to complain about?
 
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Forum GIFs powered by GIPHY: https://giphy.com/
Copyright © Amin Forums, LLC
Top Bottom