leica m lenses on m43 system?

[gteague]

Advertisement (gone after registration)

i know this isn't directly cl related, but you guys are my 'homies' on this forum!

i never thought i'd be using any manual lenses on my cl, but i got a 50 and a 75 and they both work brilliantly on the cl and on my s1. but since i've gotten those two cameras, my gh5 is growing cobwebs in a corner and even with the premium video features (i rarely shoot video) i haven't used it in over a year.

but when we got the firmware update with the tele-converter function i started wondering how my 50/2 m lens would perform on the gh5 which is a superb stills camera. using the panasonic tele-converter function should yield a 200/2 equivalent which would be fantastic. so i ordered a metabones adapter from b&h. the size should be ideal for the gh5 and that camera has at least as good peaking as the cl and s1.

i can't wait to try it and am charging up some gh5 batteries now in anticipation. have any of you guys been using m lenses on an m43 system?

thanks, /guy

[jaapv]

All the time without problems. My main problem with MFT is that the images tend to get gritty due to sensor noise quite easily.

You will be down to 10 MP, though, which is a bit limiting for prints. Sufficient for A4, acceptable for A3, possible but demanding for larger sizes.

[gteague]

52 minutes ago, jaapv said:

All the time without problems. My main problem with MFT is that the images tend to get gritty due to sensor noise quite easily.

You will be down to 10 MP, though, which is a bit limiting for prints. Sufficient for A4, acceptable for A3, possible but demanding for larger sizes.

be ok for facebook i'll bet and that's my main outlet for my photos nowadays. i don't think i've printed anything since the 80s.

my adapter should be here monday and i'll come back with a few samples i hope.

thanks! /guy

[ramarren]

With my Olympus E-M1 and various other recent mFT cameras, I find the default raw sharpening in Lightroom Classic is too heavy handed. These cameras have a very light to no anti-alias filter and the default sharpening in LR/Camera Raw is about double the optimum. My defaults for processing E-M1, L1, and G1 exposures cuts the default sharpening provided by LR to about a third. Doing this can solve the grainy sky issues in many instances.

The mFT sensors also have a rather deep sensor stack which can affect imaging at the corners and edges with short focal length lenses. 50mm and longer are generally no problem, 35mm can be on the edge depending on the specific lens design.

Using the E-M1 with a Novoflex MFT/LEM with a Summicron-M 50mm f/2 lens nets an equivalent 100mm f/2. Using digital zoom (aka Digital Teleconverter in Olympus nomenclature) and comparing it to normal, I get a 2x magnification factor so equivalent 200mm f/2. With the Olympus, the digital zoom is exclusively applied to the JPEG output and doesn't affect the raw (.ORF) file at all. I made two exposures, one with and one without digital zoom: it was obvious that the zoomed JPEG is simply upscaled to the full pixel resolution, but in the raw file you can see that it is purely a crop of the field of view. 

G

[gteague]

good info. thanks! maybe i'll get some more life out of my gh5 yet and it deserves it. before the s1 it was the best camera i'd owned in 55 years of photography and i hardly used any of the video features which were the main selling point. maybe when i can upgrade to equipment which will support a 4k/60p workflow i can get back into it.

well, the best digital camera--film cameras have different standards. /guy

[jaapv]

36 minutes ago, ramarren said:

With my Olympus E-M1 and various other recent mFT cameras, I find the default raw sharpening in Lightroom Classic is too heavy handed. These cameras have a very light to no anti-alias filter and the default sharpening in LR/Camera Raw is about double the optimum. My defaults for processing E-M1, L1, and G1 exposures cuts the default sharpening provided by LR to about a third. Doing this can solve the grainy sky issues in many instances.

The mFT sensors also have a rather deep sensor stack which can affect imaging at the corners and edges with short focal length lenses. 50mm and longer are generally no problem, 35mm can be on the edge depending on the specific lens design.

Using the E-M1 with a Novoflex MFT/LEM with a Summicron-M 50mm f/2 lens nets an equivalent 100mm f/2. Using digital zoom (aka Digital Teleconverter in Olympus nomenclature) and comparing it to normal, I get a 2x magnification factor so equivalent 200mm f/2. With the Olympus, the digital zoom is exclusively applied to the JPEG output and doesn't affect the raw (.ORF) file at all. I made two exposures, one with and one without digital zoom: it was obvious that the zoomed JPEG is simply upscaled to the full pixel resolution, but in the raw file you can see that it is purely a crop of the field of view. 

G

I found the 16 MP sensor smoother than the 20MP one.  

[ramarren]

Advertisement (gone after registration)

7 hours ago, jaapv said:

I found the 16 MP sensor smoother than the 20MP one.  

That could be... I'm happy enough with 16 Mpixel on this format that I never saw much benefit to upgrading.

G

[gteague]

got the metabones adapter in today and mounted my 50 summicron on the gh5. i took my stock test shots across a dim room at normal fl, 1.4x, and 2.0x and the results of all of them were disappointingly soft compared to that lens on other bodies. also i noted that unlike with the cl and the s1, i get no exif for aperture or focal length or effective focal length and that might be because of the leica m-adapter used on those bodies.

i'll go outside for a short walk later and see if i can get better results. i did focus using the 2x setting and then kept the same focus point as i moved to 1.4x and 1x. it might be possible the peaking doesn't work as well on the gh5, but since it's a very video-centric camera, i doubt that's the case. any tips appreciated--is there perhaps a setting in the menu i'm missing? when i mounted the lens it prompted for a focal length and i put in 50mm. 

/guy

Edited August 24, 2020 by gteague

[gteague]

ok, had my short walk. what i found over the course of 12-15 images is that _if_ you get focus right, at 50mm (100 equiv) you can get a sharp image. whether it's as sharp as the same lens on the cl or the s1, i doubt although i haven't put it under a magnifier or zoomed all the way in. but if you use the tele-ex 1.4x/2.0x function, the images get first soft, then very soft. at 200mm equivalent 2.0x the images are sometimes smeary. i shot at f2, f4, and f5.6 to give myself focus leeway, but to tell you the truth, only about one image out of the 15 was acceptable to me.

seems i've wasted $100 on an adapter for this experiment. i have still the 75mm lens to try though. 

/guy

[nicci78]

Why bother ? Years ago,  Lens rental proved that m4/3 were the worst candidate for M lenses. Due to über thick 4mm stack sensor filters. 
Digital M sensor filter is a little bit less than 1mm. Whereas SL is a little bit over 1mm. 
Leica uses the thinnest stack filters for their sensors. Combine with special micro lenses array and automatic software correction to ensure the best results with M lenses. 

Of course other manufacturers have no incentive to make their cameras working well with competitor lenses. So they do not bother to do so. 
 

NB : at close distance, M long lenses can performed quite well with third party cameras. But at infinity, everything starts to fall appart. With lots of smearing. 
Even SL required 2016 tweaked Elmarit-M 28 and Summicron-M 28 to work best with wide angle lenses. 

Edited August 25, 2020 by nicci78

[nicci78]

By the way CL in 1.3x crop mode provide around the same field of view as M lenses upon m4/3. But in 3:2 aspect ratio instead of 4/3. 
So really why bother ? IBIS ? Smearing will certainly void any steady image benefits. 

[gteague]

it's just that i have been so delighted with using the 50 summicron on my other two bodies i thought it might make a nice size match to the gh5. i have about a dozen m43 lenses and 3-4 around the same focal length. 

so what i'm going to do, hopefully today, is mount my m43 leitz nocticron (pana branded, but i doubt it has any panasonic dna in it at all--it seems pure leitz and may still be the highest rated lens in the dxomark database although the oly 75mm is right up there as well) and walk the exact same route and take the exact same images and then compare. after seeing the disappointing results from the summicron, i suspect the 42.5mm nocticron will easily beat it. for some reason i've only used this lens one or two times over the years as i have too many choices!

/guy

[nicci78]

Don't worry Nocticron 42.5mm is 100% Panasonic, with Leica stamps upon it. Leica only send engineers to certified the process. But they have nothing to do with lens design. 

Olympus 75mm is a Sigma design and manufacturing. Just like the new Olympus 100-400mm f/5-6.3 lens. Which is just the m4/3 variant of DG DN 100-400mm available for L-mount. So it is designed and made by Sigma in Olympus mount and fashion. 

Panasonic engineers are very competent. The über huge and expensive Leica APO-Vario-Elmarit-SL 90-280mm f/2.8-4 patent is granted to Panasonic. But manufactured in Wetzlar Germany. 

Actually Leica only designed the TL and SL prime lenses. TL and SL Vario zooms are either designed by Konica-Minolta or Panasonic. Relative patents can be found online. 

 

[gteague]

28 minutes ago, nicci78 said:

Don't worry Nocticron 42.5mm is 100% Panasonic, with Leica stamps upon it. Leica only send engineers to certified the process. But they have nothing to do with lens design. 

Olympus 75mm is a Sigma design and manufacturing. Just like the new Olympus 100-400mm f/5-6.3 lens. Which is just the m4/3 variant of DG DN 100-400mm available for L-mount. So it is designed and made by Sigma in Olympus mount and fashion. 

Panasonic engineers are very competent. The über huge and expensive Leica APO-Vario-Elmarit-SL 90-280mm f/2.8-4 patent is granted to Panasonic. But manufactured in Wetzlar Germany. 

Actually Leica only designed the TL and SL prime lenses. TL and SL Vario zooms are either designed by Konica-Minolta or Panasonic. Relative patents can be found online. 

 

i owned the oly 75 (f1.8 i think) at one time years ago. and it was indeed a sharp lens. but i lived in a dense urban area at the time and that focal length (150mm equivalent) wasn't useful at all, so i traded it off. today in my semi-rural area it would be perfect and indeed i have the sigma 70mm macro which nearly lives on my s1 and more often than not used with the tele-ext function at 2.0x. 

btw, i have a dozen or more m43 lenses including a half dozen leica stamped and pro 'g' lenses and  the nocticron is _nothing_ like  those lenses--there's no comparison with any of them. it is heavy duty and all metal and heavy and solid in a way none of them are which is what led to my (ass)umption that panasonic merely branded a 'real' leitz lens or produced it in a leitz factory. i can't remember if it's made in japan or wetzlar though--likely japan. 

/guy

Edited August 25, 2020 by gteague

[jaapv]

I completely fail to see the issues you are having with M lenses on MFT. I cannot even imagine a scenario in which M lenses could underperform when just mounted on a smaller-sensored camera, there is simply nothing between lens and sensor to impede the light rays.

 

 

Summicron 40 C on Panasonic GX8

 

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!

[carbon_dragon]

I'm not saying M lenses would have an issue, BUT I have seen people rating lenses on YouTube where a lens works pretty well on a full frame camera and when you put it on a smaller sensor camera, it actually only uses the center part of the lens and is thus more demanding of the lens than the full frame camera is. So I don't *think* it's crazy to think that a lens might not do as well on a small sensor. I've used my 35/2 ASPH and my 50/2 summicron (not the APO) on my Nex-7 and they seemed to work well for me.

[jaapv]

It is just the other way around - you use the best part of the lens - the center, which can only enhance the quality of the lens. "people" on YouTube are not immune to being idiots.

[carbon_dragon]

Yes I know in the vast majority of cases the lens is best in the center, but you are still trying to "get more pixels" out of that center than a full frame would. With "cheaper" lenses (I've seen Chinese lenses reviewed) the full frame test chart is fine and the APS-C one is pretty bad. Those lenses are not M lenses though. They're either cheap lenses or old lenses or both.

[gteague]

it's not a huge difference and at first i thought i was missing focus or moving, but i have no such problems on the cl which lacks ibis and the gh5 has industry leading ibis. even the best shots i got were noticeably not as good as the same subjects in the same light when taken on the cl. i have the advantage that i walk around about a 3 sq mile area time after time after time and i can shoot the exact same things with different equipment.

in comparing, i had to throw out the shots with the tele-ext engaged as they were very soft. but we've seen that this function on the cl and the s1 result in barely any degradation. all the shots without tele-ext were fine, just not quite as good as the cl and i suspect that's exactly the expected results. the only reason i was trying the experiment was to bring the gh5 out of retirement and perhaps to have the equivalent of a 200/2 in a very compact form. but i got a result (as the mythbusters are fond of saying!) and this result tells me that the m43 lenses perform better than the m lenses despite the latter being full frame. and, even though the tele-ext function works fine with m43 lenses, it seems to fall apart with the m lenses. 

still, as i said, i have the nocticron mounted now and come magick hour i'll go back out and confirm results. this lens is about 5x the size and weight of the 50/2, but if the tele-ext function works with it as a i remember, it'll be about a 170/1.2. 

/guy

Edited August 25, 2020 by gteague

[Advertisement]

Hallo,

Look here: Leica L-Mount

[jaapv]

20 minutes ago, carbon_dragon said:

Yes I know in the vast majority of cases the lens is best in the center, but you are still trying to "get more pixels" out of that center than a full frame would. With "cheaper" lenses (I've seen Chinese lenses reviewed) the full frame test chart is fine and the APS-C one is pretty bad. Those lenses are not M lenses though. They're either cheap lenses or old lenses or both.

Well, just have a look at the image I posted from a fifty-year old Summicron on MFT... And once again: sensor resolution vs lens resolution is NOT a weakest-link situation.