GH5 Big Glass vs Small Glass

[David W. Jones]

That's the bottom line.... It's all about the look.

Once upon a time I was a lens whore, with well over 70 lenses. and having owned many more over the years.
Sure I have looked at charts, MTF numbers and the like. But always ended up choosing a lens because of A: it was the right tool for the job at hand, and B: it produced the look I was after.

Some of my favorite lenses I have used with the Panasonic GH series cameras... Cooke Speed Panchro's, Cooke 20-100 & 18-100 zooms, Zeiss Standard & Super Speeds, Zeiss Contax.

All delivering an extremely pleasing look.

Your mileage way vary.

 

[combatentropy]

I must admit, there is something to be said for the character (flaws) of certain old glass. This afternoon I was shooting with a pristine Lumix lens and then towards the end I pulled out an old Contax lens and ended-up with my favorite stuff from the day.

In these photos, I don't see any flaws. In both of their names is the word "contax," so I assume they are both from the Contax.

Can you elaborate on what flaws you see in them and why you liked them better than the ones from the Lumix? If you can post stills from the Lumix and try to help me see the differences, that would be nice.

 

[joe12south]

In these photos, I don't see any flaws. In both of their names is the word "contax," so I assume they are both from the Contax.

Can you elaborate on what flaws you see in them and why you liked them better than the ones from the Lumix? If you can post stills from the Lumix and try to help me see the differences, that would be nice.

No, actually one is the Lumix 25mm f/1.7 manipulated in Davinci Resolve to match the look of the Contax.

The "flaws" in the Contax:
1. Low contrast (the most obvious difference from modern lenses.)
2. Light "bloom"
3. Very susceptible to flaring
4. A "warm" color cast.
5. A complicated/fussy bokeh (luckily, so does the Lumix 25mm, because this would be very hard to match in post.)

For comparison, here's an unaltered shot from the Lumix lens at the same time of day...

The Lumix lens renders a more accurate representation of the scene. Because the lens is more "neutral", it gives me more room to push it around in post to create whatever "look" I want. Conversely, the Contax has a "look" baked-in. It's personality would be more difficult to remove in post.

I like both lenses for different reasons. The Lumix is technically superior (and it's not even a very good lens by today's standards) but the Contax has a look that elicits a certain emotional response. Different tools for different use cases.

 

[icarusfilm]

What I meant is that contax Zeiss and Leica R's were the top notch lenses of their time and not SOO long ago. They are by no means "exotic choices" They have great color palates and very good micro contrast that renders images in a very specific way (resolution isn't everything) . At the same time everything is a little too flary and low contrast too often not quite as annoying as the teal-orange nightmare but a little much.

A sweeping statement. Granted, there is probably a lot of bad glass out there, but there are also some real gems.

 

[combatentropy]

So:

Lumix	Contax

?

 

[Greg_E]

So:

View attachment 133558	View attachment 133557
Lumix	Contax

?

Lumix untouched and Contax after adjustments is the way I read it.

 

[Shirozina]

No, actually one is the Lumix 25mm f/1.7 manipulated in Davinci Resolve to match the look of the Contax.

The "flaws" in the Contax:
1. Low contrast (the most obvious difference from modern lenses.)
2. Light "bloom"
3. Very susceptible to flaring
4. A "warm" color cast.
5. A complicated/fussy bokeh (luckily, so does the Lumix 25mm, because this would be very hard to match in post.)

As an owner of a lot of Contax lenses I have to fundamental disagree with the first 3 observations - they are not low contrast and often quite the opposite to the extent that it can become a problem of too much contrast. Naturaly these lenses are now very old and low contrast examples exist due to a build up of dirt on the elements either side of the aperture mechanism which also causes lens 'bloom' The Ziess T * coating was and still is superb at controling flare. They possibly have a warm cast (but compared to what) but that's easy to calibrate out and yes they do have fussy Bokeh.

 

[Shirozina]

What I meant is that contax Zeiss and Leica R's were the top notch lenses of their time and not SOO long ago. They are by no means "exotic choices" They have great color palates and very good micro contrast that renders images in a very specific way (resolution isn't everything) . At the same time everything is a little too flary and low contrast too often not quite as annoying as the teal-orange nightmare but a little much.

Contax and Leica R have quite different renderings. Leica R tends to be of lower contrast with less micro contrast than Zeiss but has better Bokeh and a softer more natural colour rendering. Leica coatings are also much less effective than the Zeiss T* coatings at controlling flare. Very distinct renderings that are infact greater than differences between modern brands.

 

[joe12south]

As an owner of a lot of Contax lenses I have to fundamental disagree with the first 3 observations - they are not low contrast and often quite the opposite to the extent that it can become a problem of too much contrast. Naturaly these lenses are now very old and low contrast examples exist due to a build up of dirt on the elements either side of the aperture mechanism which also causes lens 'bloom' The Ziess T * coating was and still is superb at controling flare. They possibly have a warm cast (but compared to what) but that's easy to calibrate out and yes they do have fussy Bokeh.

I have two copies in pristine condition. Compared to any Lumix lens I own, compared to other modern lenses like the Sigma 18-35, it's low contrast, bloomy and prone to flare.

PS. This is a flocked, mechanical adapter. No other glass in the path. No adjustments made other than my standard rec709 LUT.

Not hard to add back that contrast and balance the color, though, revealing a pretty damn detailed image that still retains some of that old glass character...

 

[icarusfilm]

Youvare right i was referring to the Zeiss

Contax and Leica R have quite different renderings. Leica R tends to be of lower contrast with less micro contrast than Zeiss but has better Bokeh and a softer more natural colour rendering. Leica coatings are also much less effective than the Zeiss T* coatings at controlling flare. Very distinct renderings that are infact greater than differences between modern brands.

 

[combatentropy]

So far the differences I've seen between lenses are subtle. Some are within the bounds of the mildest of color grading, things like contrast and color.

Now there are differences that can't be graded away, things like bokeh. Side by side, I can tell that the Contax's bokeh is busier than the Lumix's, as has been said. And some of you have mentioned other differences, which I'm not seeing in these samples, but I have seen them in other comparisons, things like flare, ghosting, and other things that might all be categorized as things that happen when you let the source of light shine down the barrel of the lens. This doesn't happen in every shot, and in the early days DPs would jump through hoops to avoid it. Nowadays flare is not taboo, it's embraced.

Although I am seeing some of these little differences, I'm not getting the magnitude implied by some of these comments. For example, whatever difference in warmth there is between the Lumix and the Contax, it is swamped by the difference in the quality of light between the two shots. If you had taken two identical shots, maybe I could make it out, but as it stands I can't be sure of any difference in that regard. I certainly don't see any difference that warrants the adjective "exponential," as filmguy123 said in his first post:

just by virtue of attaching BIG GLASS the quality jumps exponentially

The only difference I've ever seen between two lenses at the same focal length and same f-stop has not been exponential, always microscopic.

 

[icarusfilm]

Also the optical quality and 'look" of a lens is only half of the story the mechanics and ability to control focus is just as important.
These are a lot of lenses that "look" great but have fly by wire focusing and are to me inoperable. In the entire native M43 lineup only the Olympus PRO line and the Voigtlanders and Veydras.
I see no point in having any lenses that are not both optically and mechanically useable.

 

[jonpais]

Also the optical quality and 'look" of a lens is only half of the story the mechanics and ability to control focus is just as important.
These are a lot of lenses that "look" great but have fly by wire focusing and are to me inoperable. In the entire native M43 lineup only the Olympus PRO line and the Voigtlanders and Veydras.
I see no point in having any lenses that are not both optically and mechanically useable.

Hard to argue with this, particularly with a system with notoriously poor AF. The focus transition feature on the GH5 is pretty neat though, and works great for planned repeatable actions.

 

[joe12south]

Also the optical quality and 'look" of a lens is only half of the story the mechanics and ability to control focus is just as important.
These are a lot of lenses that "look" great but have fly by wire focusing and are to me inoperable. In the entire native M43 lineup only the Olympus PRO line and the Voigtlanders and Veydras.
I see no point in having any lenses that are not both optically and mechanically useable.

And then you pair that "inoperable" wire focus with some of the new wireless electronic follow focuses (like on the Ronin-S) and the game completely changes. Tuneable electronic focus will soon equal or better to mechanical in terms of smoothness, accuracy and flexibility. Fun times.

 

[Shirozina]

I have two copies in pristine condition. Compared to any Lumix lens I own, compared to other modern lenses like the Sigma 18-35, it's low contrast, bloomy and prone to flare.

PS. This is a flocked, mechanical adapter. No other glass in the path. No adjustments made other than my standard rec709 LUT.

Not hard to add back that contrast and balance the color, though, revealing a pretty damn detailed image that still retains some of that old glass character...

What lens is this exactly and have you got any side by side shots in thr same lighting?

 

[joe12south]

What lens is this exactly and have you got any side by side shots in thr same lighting?

The Carl Zeiss Planar 1,7/50 T*

All of the shots are from within 30 minutes of each other, but since I wasn't doing a lens test, I made no effort to exactly match set-ups. Here's where I switched lenses, though, so the two shots are pretty close:

There's been no effort to match the above shots, and I think many people would consider these two very different lenses (separated by decades), yet they're really not that different. If I tweaked the contrast, color balance and added a bit of diffusion to the Lumix, they'd be very close, indeed.

Now, I know some are going to complain that neither of these are particularly good lenses - and that's true - so if someone wants to put up some comparison images that show some sort of magic from a Milvus, or better yet if someone wants to give me a Milvus, I'm open to learning what this full frame magic really is.

 

[wwjd]

sooooo.. in conclusion to this thread... any differences are microscopic, definitely subjective to personal tastes of desired looks, and m43 is not really inferior at all? just different

 

[Thomas Smet]

I really feel like we are grasping at elusive straws here desperately trying to achieve some sort of magical slam dunk instant perfect look. At best everything we are discussing here is subtle and I guarantee outside of the video nerd club for geeks nobody and I mean nobody can tell a difference or give a darn. Of course flare, contrast and corner sharpness are still very valid factors but when we have a m43 and FF lens that handle all of those well we start to get into very subtle differences. Lenses do have character and that is important but character is just that, character. Not quality and it’s very difficult to say character makes one lens superior to another lens. That’s like saying the blue crayon is superior to the red crayon. While it’s a fascinating argument to have with a three year old for about five minutes it really is just as pointless to debate such a thing.

 

[Shirozina]

Joe, The CZ 50 1.7 shouldn't flare like that so what adapter did you use as even with flocking light can still bounce off the material. I've got a couple of them and have had about 3 others pass through my hands over the years. 2.8 is the sweet spot for smooth Bokeh as under that it looks a bit busy. Stopped down to F8 on a full frame sensor like the Canon 5Ds or Sony A7r2/3 and they have astonishing resolution right into the corners ( better than the 1.4) but for video this is not a concern with 4k / 8mp. I must admit I haven't used it on M43 as the Sigma 18-35 covers it and I like the flexibility of zooms. They made an AE and MM version and in terms of outright resolution on high MP full frame sensors I found the MM versions better but again for video this shouldn't be an issue. I have used the 1.4 for video and have to say that wide open it was a bit on the soft side and I reckon a native M43 fast prime at the same speed would easily better it. Now if I wanted a flattering but detailed look from a 50mm lens I'd probably reach for my Leica 50mm F2 Summicron-R 2nd version.

 

[filmguy123]

I know I owe the thread a full up with some side by sides when I get a chance (probably December!).

That said, I do see how far this thread diverged and I should have predicted such. When starting, I came in with the ASSUMPTION that "big glass" rendered a superior image to "small glass" and was inquiring about as to WHY that is. As this has progressed, I see it's sort of followed the same trend as usual and things aren't so clear. Perhaps larger glass is not inherently better.

I want to remind that the original intent was primarily a question as to why, not a discussion as to if.

Now, that said, I've done a few more shoots and I can't shake it. My Sigma and Milvus lenses consistenly produce to me more pleasing images than my Lumix 12-35 or Voigtlander primes. I dunno. I am not running science charts here. They just look good. Maybe it has ZERO relation to the fact that they are FF lenses. Maybe it's just that they are better lenses and would be the same for a smaller m43 image circle, too.

I hate to be unscientific as I know it exasperates our engineers, and I know there are a lot of variables. But some things seem to a recurring pattern here. To my eye, the BMD camera looks better than the GH5/GH5s hands down. S35 cameras often seem to look better to my eye than m43 cameras, I started noticing this years ago - I can usually spot it if I watch a video, but not always, but this relates to my work as well. But the FF glass on m43 closed the gap considerably which is what started this question for me - I have found using a Milvus on GH5 elevates it to near EVA1 levels (but not quite - the EVA1 still "wows" more consistently). But I honestly cannot say I can always tell in a controlled shoot-out, still grab, science chart. I don't think this invalidates that, and I don't think its placebo effect or a better DP or higher production value. I think overly controlled tests don't show the whole story, and while I think there's a scientific explanation I don't know it.

I do think people more than video nerds notice because from my experience screening work to lay people, when I switched from native m43 glass to voigtlanders people noticed. They didn't know what they noticed and could NEVER pinpoint it like some could here, but they just knew it looked better. Same thing when I swapped to Milvus. I got comments noticing the difference but not knowing what. If I did an A/B test with those people they'd all fail, but if I show a video side by side they'll prefer the look of the one with the better glass as sort of a wholistic impression.

Ok - I am not trying to make an argument here, I am only looking for understanding. Maybe this recurring theme is all placebo and esoteric. My experience and eyes make me doubt that. But, it's besides the point. At least here, there seems to be no clear consensus on what might make a larger piece of glass better than a smaller piece of glass. Or a larger sensor better than a smaller sensor.

I'm not trying to waste people's time, just hoping to gain understanding and I also must say I don't have a point to prove! As it is, I own a ton of m43 gear and I really like the format so I'm not incentivized to see it "less than", I am just trying to figure out what I am seeing because I am the same shooter, shooting similar-ish content, removing those essential variables.

As others have posited, perhaps its just as simple as Sigma 18-35 and Milvus Set > Voigtlander/Olympus/Lumix.