Panasonic GH7

[PPNS]

3 hours ago, mercer said:

Haha, okay. Take a close up with a wide angle lens and then one with a standard lens and report back what happens. There is more field curvature in a wide angle lens due to the ... lens design of a wide angle lens. This field curvature will round out your subject. Just because the FOV changes, the focal length doesn't.

the fov is dependent on the focal length as well as the imager size. 21mm is not inherently a wide angle focal length. It’s

- very wide on MF

- wide on FF and s35

- a widish medium on mft

- mild telephoto on s16

- regular telephoto on B4 broadcast.

lf you had an f/2 20mm lens on ff, and an f/1 10mm on mft, with similarly designed optics, your shot would look the same if taken from the same spot. 

https://www.yedlin.net/NerdyFilmTechStuff/MatchLensBlur.html
 

 

 

[zlfan]

according to my experience, 

putting on a mf lense like 50 mm on a ff cam like 5d3ml, is a little different than putting on 50mm sim is on 5d3ml. even the same focal length, medium format lenses are easier to get bokur even at f4. mf lenses have smoother rendition but less sharper. 

[mercer]

1 hour ago, PPNS said:

lf you had an f/2 20mm lens on ff, and an f/1 10mm on mft, with similarly designed optics, your shot would look the same if taken from the same spot.

A lot of IFs in that statement. The problem with the all else equal argument is that it's rarely applicable because all else is rarely, if ever, equal.

Even the same lens on the same camera will have a different look with a couple steps forwards or backwards.

With a 24mm at f/2, even 2.8, on FF you can still get some subject separation, I have never noticed that with an equivalent aperture on a 12mm lens on mft.

What lenses do you currently use on your projects? I assume, since you're referencing Yedlin, that you shoot in 1080p?

[PannySVHS]

I love a 0.71 speedboosted Canon FD 28mm F2.8 lens on a full mft sensor. Maybe I should sell all my stuff and keep the 28, the 50mm F1.4 and the 100mm F2.8. Maybe the Vivitar 90 F2.5 as well.:) It really is about lens and sensor combinations. In theory equivalence is helpful and by that simple translation it yields same results. It yedlins same results, so to say.:)

In practice the specific combinations still paint light differently. I remember listening to the DP of "Moonlight" at a beautiful festival in Poland when he expressed his exitement of using the then new Alexa LF with a 50mm instead of using a 35mm on the S35 sensor with an equivalent, larger Fstop. Both perspectives combined allow an understanding of my lens choices. And restrictions by money.😊

[Emanuel]

28 minutes ago, PannySVHS said:

(...) It really is about lens and sensor combinations. In theory equivalence is helpful and by that simple translation it yields same results. It yedlins same results, so to say.:)

In practice the specific combinations still paint light differently. (...)

This is the secret sauce, not the brand or whatever marketing sells for ; )

In a word or in a single line, art (because cinema is an art form) is a combo, no less.

- EAG

[PPNS]

59 minutes ago, mercer said:

A lot of IFs in that statement. The problem with the all else equal argument is that it's rarely applicable because all else is rarely, if ever, equal.

Even the same lens on the same camera will have a different look with a couple steps forwards or backwards.

With a 24mm at f/2, even 2.8, on FF you can still get some subject separation, I have never noticed that with an equivalent aperture on a 12mm lens on mft.

What lenses do you currently use on your projects? I assume, since you're referencing Yedlin, that you shoot in 1080p?

The ifs have a grammatical function. I’m explaining to you how optics work. A 24 at 2.8 on a 36x24mm sensor and a 12mm at 1.4 on a 18x12mm sensor will look exactly the same when placed at the same spot, with exactly the same separation.

that is also the reason i linked yedlins article: you get to see example images where he matched the look of 1 certain FOV and DOF on different imager sizes and lenses. this is just pure math that you’re trying to disprove.

“Even the same lens on the same camera will have a different look with a couple steps forwards or backwards.”

not sure what you’re trying to say here. Flaring and a different focus point tend to have an influence on the image, sure.

I agree that 1.4 lenses on mft are more rare, but that’s a different aspect in this argument (manufacturers not wanting to make high quality mft gear anymore). 

if the client is broke, i use my pocket 4k and meike lenses or olympus zoom. If theyre not, i rent an alexa and some lenses (i like zeiss super speeds a lot). I don’t see why resolution is relevant to this discussion though. are we trying to move goalposts again?

[Emanuel]

21 minutes ago, PPNS said:

(...)

I agree that 1.4 lenses on mft are more rare, but that’s a different aspect in this argument (manufacturers not wanting to make high quality mft gear anymore). 

(...)

 

I think the point is exactly this one. Without mention, the larger the more light you can play with.

Look is made by light, after all.

Low light performance as reference, not the same as no lighting, OK? But room for.

- EAG

[TomTheDP]

Maybe it could improve with firmware updates. As of now though its a cool implementation but it's not going to make me buy this camera. Just got a BM Micro though 😅 

I might end up getting a GH6, but I was going to get one anyways. IBIS and internal prores, long battery life and tilting screen. Great for when I just want to get out of the car and grab a shot with no fuss. I'll probably get the LOGC upgrade for it. 

[PannySVHS]

Looking forward to seeing both, Gh6 and Bmmcc put to action by you. @TomTheDP I've been using my og Bmpcc for six days in a row last week just for relaxation and fun. So much fun, had it coupled with a 25mm Tevidon and some other times with a 0.7 speedboosted Pentax M 28mm F2 non Hollywood version. Colours are magical. During daytime beyond diffraction limit.:)

[mercer]

17 minutes ago, PPNS said:

The ifs have a grammatical function. I’m explaining to you how optics work. A 24 at 2.8 on a 36x24mm sensor and a 12mm at 1.4 on a 18x12mm sensor will look exactly the same when placed at the same spot, with exactly the same separation.

The ifs only have a grammatical function if all else is equal, which they are not, so some aspect of the image, no matter how small will be different.

I understand the optics element of your point. This argument has been around forever. What I am saying is the problem with your argument is that it requires all elements to be equal to be valid but they seldom are... if ever.

Sure you can create a similar image with a crop sensor camera, but something in the image will be different. Or you'll have to make a compromise with the larger sensor camera to get them close to matching. Which you wouldn't do if you were using the camera/lens in a practical way.

But even if you look at Yedlin's example, the subject's face is slightly thinner in the Imax and slightly fuller in the Alexa. Obviously, Yedlin went to great lengths to prove his point... that's the point of his article. And I am not saying you can't get it close, I'm saying that something will be different.

As I said in my previous reply... use a fast 12mm lens on mft and then a fast 24mm lens on FF and tell me that they have the same separation.

With that said...

37 minutes ago, PPNS said:

“Even the same lens on the same camera will have a different look with a couple steps forwards or backwards.”

not sure what you’re trying to say here. Flaring and a different focus point tend to have an influence on the image, sure.

I thought it was clear, but my apologies. My point is that even if you use the same lens, on the same camera, and take a single step forward or backwards, you can hit a sweet spot of the lens and have more pop or separation.

In the Yedlin article, it is clear that he stopped down the lenses to create less shallow depth of field. In fact, in one of the images, the bookshelf is almost in focus with the subject... so again not the same.

In my original "nonsense" post, I was suggesting it was difficult to make these comparison videos because the LF would always have an advantage. Not only does it have a bigger and better sensor, better color science, it has a lens advantage as well... which is obvious since he had to stop down to f/4 or 5.6 so he could get a match with the lens on the GH7. That guy did a good job making them pretty close, but the GH7 looks more compressed and her face is slightly less angular. Is it enough to affect his test... not at all.

You are correct though, I mistakenly referred to the 21mm lens as a wide angle on M43.

1 hour ago, PPNS said:

(i like zeiss super speeds a lot)

I'm a fan of them as well. I haven't used the Super Speeds, but I had a small set of the Zeiss Rollei lenses which were supposedly modeled after the Standard Speeds from the late 60s. The 3D pop from Zeiss lenses is something else. Some say it's micro contrast and others say it is more obvious with larger formats... all I know is that it looks amazing when you pull that focus and hit it.

1 hour ago, PPNS said:

I don’t see why resolution is relevant to this discussion though. are we trying to move goalposts again?

Not moving any goal posts and probably not relevant, I was just curious if you were a believer/practitioner in his resolution theories.

[Emanuel]

19 minutes ago, mercer said:

(...)

Not moving any goal posts and probably not relevant, I was just curious if you were a believer/practitioner in his resolution theories.

LOL : ) Who reads you here for a while... more than a decade now, can only send a good laugh but you know something? Don't kill the messenger later on when I've proposed this and most part of you love the game:

 

[PannySVHS]

After Moonlight James Laxton was again Dop for Barry Jenkins. He photographed "Beale Street", which was filmed with an Alexa 65. He must love the look of a 50mm lens on a larger format.:) There are some desired qualities of larger formats for the look beyond the logic of focal length and f-stop equivalence.

[zlfan]

the master was filmed with 65mm film. special look. I especially like the sea scenes in this movie. but there are some other scenes very compelling. the actors performed very well too. a true master piece. 

 

[Aaron Smith]

I'm just wondering how this will compare to my GH5S's in low light. The high megapixel count worried me but I heard that this one has a brand new sensor design so I guess we'll see. 

[PannySVHS]

Should equal or better the GH5S all in all but especially how it holds onto resolution and colour in lowlight. There is a comparison between GH6 and GH5s demonstrating an advantage for the GH6, though possibly demonstrating more luma noise but less mush and preserving more visual information. @Aaron Smith

[mercer]

1 hour ago, PannySVHS said:

After Moonlight James Laxton was again Dop for Barry Jenkins. He photographed "Beale Street", which was filmed with an Alexa 65. He must love the look of a 50mm lens on a larger format.:) There are some desired qualities of larger formats for the look beyond the logic of focal length and f-stop equivalence.

Now that you mention it, someone should tell Yedlin about Yedlin's test... if he had known about it, he could have shot Knives Out on a m4/3 camera instead of the Alexa 65...

[PPNS]

3 hours ago, mercer said:

Sure you can create a similar image with a crop sensor camera, but something in the image will be different. Or you'll have to make a compromise with the larger sensor camera to get them close to matching. Which you wouldn't do if you were using the camera/lens in a practical way.

But even if you look at Yedlin's example, the subject's face is slightly thinner in the Imax and slightly fuller in the Alexa. Obviously, Yedlin went to great lengths to prove his point... that's the point of his article. And I am not saying you can't get it close, I'm saying that something will be different.

As I said in my previous reply... use a fast 12mm lens on mft and then a fast 24mm lens on FF and tell me that they have the same separation.

i dont get this argument? you can get essentially a 95% match across most image formats, but since you can get just that tiny bit shallower with larger sensors that makes it all moot?

the reason to get a gh or fx or pocket camera is because its a shitty camera to own. It’s for personal projects where theres no budget, or as a bcam in certain situations, either as an extra angle during scenes, or to rig it on a car so the good expensive camera doesnt break. If you have to use it, a few compromises should be made. In the case of mft that’s mostly not being able to be super shallow on wide lenses, but for normal use cases, you’ll be fine. this stuff should be liberating to know?

If there’s budget, why wouldnt you rent a mini/mini lf/35/venice? Its obvious better cameras have advantages!

3 hours ago, mercer said:

In the Yedlin article, it is clear that he stopped down the lenses to create less shallow depth of field. In fact, in one of the images, the bookshelf is almost in focus with the subject... so again not the same.

 

there’s a certain other test by manuel luebbers where he conducts a similar test with a mini and 65, where he shoots more shallow. the same conclusion arises: the large format look is just super shallow depth of field when the lenses are wide open. If theyre not, the look can be matched with different formats.

 

3 hours ago, mercer said:

I thought it was clear, but my apologies. My point is that even if you use the same lens, on the same camera, and take a single step forward or backwards, you can hit a sweet spot of the lens and have more pop or separation.

 

i think theres about a million things more important than hitting the sweet spot of separation in real life scenarios. that being said, with the right choice of focal lengths and distance from camera, you can get pretty shallow dof from any format!

 

3 hours ago, mercer said:

I'm a fan of them as well. I haven't used the Super Speeds, but I had a small set of the Zeiss Rollei lenses which were supposedly modeled after the Standard Speeds from the late 60s. The 3D pop from Zeiss lenses is something else. Some say it's micro contrast and others say it is more obvious with larger formats... all I know is that it looks amazing when you pull that focus and hit it

i just think they hit a great balance of being sharp and slightly soft at the same time. The vignette you get at t/1.4 is nice too. Zeiss CP2 set is garbage tho, so i’m not going to generalise statements about their brand.

 

3 hours ago, mercer said:

Not moving any goal posts and probably not relevant, I was just curious if you were a believer/practitioner in his resolution theories.

i think 1080p is a great delivery resolution, but i’m not always inclined to shoot it. Someone in the editorial suite is probably going to zoom on something without my intention, and then they’ll get mad at me for not shooting in a higher resolution. I’d prefer to avoid that.

 

2 hours ago, PannySVHS said:

After Moonlight James Laxton was again Dop for Barry Jenkins. He photographed "Beale Street", which was filmed with an Alexa 65. He must love the look of a 50mm lens on a larger format.:) There are some desired qualities of larger formats for the look beyond the logic of focal length and f-stop equivalence.

He likes it because it was twice as shallow as S35. Just like the 2x anamorphic used on moonlight are twice as shallow as spherical s35

“We shot Beale Street on the Alexa 65, a large-format camera. You get a much more shallow depth of field than you would with a Super 35 sensor, or 35mm film if you're shooting film. It's what helps that rack focus to the foreground on Brian to be as dramatic as it is.”

https://filmmakermagazine.com/106532-laxton/
 

 

 

[mercer]

55 minutes ago, PPNS said:

dont get this argument? you can get essentially a 95% match across most image formats, but since you can get just that tiny bit shallower with larger sensors that makes it all moot?

I didn't say it was moot, I merely said that the LF had an advantage. If anything, I was giving the GH7 some leeway in those tests. I mean, he had to stop down the 40mm lens on the LF to 5.6 and set the shutter angle to 45 degrees on the GH7 to get the shots to match. Obviously, he didn't want to pollute the test by putting an ND on the LF when the GH7 didn't need one.

56 minutes ago, PPNS said:

the reason to get a gh or fx or pocket camera is because its a shitty camera to own. It’s for personal projects where theres no budget, or as a bcam in certain situations, either as an extra angle during scenes, or to rig it on a car so the good expensive camera doesnt break. If you have to use it, a few compromises should be made. In the case of mft that’s mostly not being able to be super shallow on wide lenses, but for normal use cases, you’ll be fine. this stuff should be liberating to know?

I agree, I am not knocking m4/3 or any shitty camera... I still shoot ML Raw on a 5D3. I'm planning on shooting a short film this summer on a 1" Canon camcorder. I had a GH6 last year and may buy another one, or a GH7 eventually.

1 hour ago, PPNS said:

i think theres about a million things more important than hitting the sweet spot of separation in real life scenarios. that being said, with the right choice of focal lengths and distance from camera, you can get pretty shallow dof from any format!

It depends on what you're going for and what impact you want your shot to have based on the story you're trying to tell. Everyone has different styles.

Years ago when camcorders were used for indie films, we used to intentionally zoom in and frame something in the foreground to look blurry.

All that said, I think there is more to larger sensors than shallow depth of field, there's separation between the foreground and background which invokes more of three dimensional element to the image. With smaller sensors everything is more compressed.

 

[eatstoomuchjam]

24 minutes ago, mercer said:

I didn't say it was moot, I merely said that the LF had an advantage.

This is just factually wrong.  The Super 35 sensors in my C70 and K-X have a FOV/DOF indistinguishable from FF cameras if I use a focal reducer with them.  They can also work great with glass made for S35 film.  This could be seen as an advantage.  I also have the option to remove the focal reducer and get a second set of focal lengths from my lenses.  That's also neat.

Sensors larger than VV have disadvantages as well - a lot of wider lenses made for 24x36mm format don't cover well all the way to the edge.  Extreme shallow DOF?  Sounds great until you are missing eye focus constantly because the talent moved 0.01mm from when you acquired focus.  Without using eye tracking AF, the Canon 85/1.2L is completely unusable on my GFX 100 (when wide open).  When the DOF is that shallow, it should be considered a special effects lens.  The background gets so blurred as to be unrecognizable as anything other than a series of color splotches.

Almost every format has some advantages over the others and some drawbacks.  That includes M43.  

[mercer]

3 hours ago, eatstoomuchjam said:

This is just factually wrong.  The Super 35 sensors in my C70 and K-X have a FOV/DOF indistinguishable from FF cameras if I use a focal reducer with them.  They can also work great with glass made for S35 film.  This could be seen as an advantage.  I also have the option to remove the focal reducer and get a second set of focal lengths from my lenses.  That's also neat.

Sensors larger than VV have disadvantages as well - a lot of wider lenses made for 24x36mm format don't cover well all the way to the edge.  Extreme shallow DOF?  Sounds great until you are missing eye focus constantly because the talent moved 0.01mm from when you acquired focus.  Without using eye tracking AF, the Canon 85/1.2L is completely unusable on my GFX 100 (when wide open).  When the DOF is that shallow, it should be considered a special effects lens.  The background gets so blurred as to be unrecognizable as anything other than a series of color splotches.

Almost every format has some advantages over the others and some drawbacks.  That includes M43.  

Holy Shit... it has an advantage for that test! What the fuck does the C70 have to do with it?

And I never said anything about shallow depth of field as being an advantage for full frame.