Fast apertures on the GH5 = Full frame

[Don Kotlos]

Yep, perspective is only affected by the distance of the objects to the lens. The focal length & sensor size do not affect perspective in any way. A 25mm on a m4/3 sensor with an aperture of f/1.4 will have the same perspective and DoF as the 50mm on a FF sensor with an aperture of f/2.8. The other aspects of the image like distortion or busyness of the bokeh depend on the design of the lens and not the focal length/aperture/sensor size. 

[Grégory LEROY]

Good article. What about the out of focus transition you were mentioning on your article about medium format ? Why doesn't it apply to FF vs m43.

I watch a lot of M43 pics on flickr, most of the time I find them flater than FF.

 

 

[Nathan Gabriel]

Has anyone compared the voigtlander to the much cheaper SLR Magic 17mm t1.6? I've found one lab test, but I'd love to hear from someone who has real world experience. I use the SLR Magic all the time and it's really difficult for me to imagine needing shallower depth of field. However I've never really spent time shooting full frame wide open. I will say that I've used 4x5 box cameras where it is easy to get razor thin depth of field and it looks terrible because almost nothing is in focus. I always thought high iso performance rather than shallow depth of field was the bigger issue for MFT. I do have the voigtlander 25mm f0.95 and I rarely feel the need to shoot wide open with it. Here's the comparison for the 17mm's that I mentioned: 

 

[wolf33d]

I have known that from the day I started photography.

BUT as you say, 1.2 and 1.4 on FF cannot be replicated on M43 and i do not agree there is a big difference between a FF 1.4 and 2.0. 

Also I don’t agree with DR in stills. Giant leap between M43 and FF, and yes I have owned a GH4, GH5 and Olympus as well as FF DSLRs and A7RII. Massive real world difference. In fact as you say super 35 is not bad. A Nikon D7200 for exemple boasts a very good usable DR better even that some FF like Canon shit. But M43, not at all.

For video DR is less impacted. In fact FF is not crazy important for me in video especially because having a 1.4 look in video is pretty hard to manage. 

I returned the GH5 not because of M43, but because of that god damn shitty AF. GH5 with great AF is all I need for video and I guess it’s gonna be a A7S3

i would not mind paying less for body and lens and loose FF. Again, for video only.

 

[psoetio]

Agreed that super fast lenses can mostly make up for the full frame advantage in terms of low light performance and shallow dof, although still leaving m43 lagging behind in dynamic range. However the two reasons  I'm looking very seriously at switching to the Sony A7iii from the Panasonics are video autofocus and cost. Autofocus especially. It's simply not reliable for video on the Panasonics, and you get that jittery background even when it does work, and af is not available on the super fast lenses and is totally unacceptable when using speed boosters. It's a serious issue when shooting events like wedding receptions. Cost, because while ff lenses are expensive you don't have to get the most expensive, super-fast ones, and, with clear image zoom and aps-c mode you can get away with having fewer lenses..

[noone]

Great article.

To me it has always been about if there are lenses to match though I have never actively tried to match systems and have just tried to use what was appropriate for the system I am using/used (FF, APSC, M43 and Pentax Q).

In the last couple of years there has been an increasing number of faster options for smaller sensors (and FF too).

  I am sure there will be situations that I would have to go higher ISO with FF than M43 but not as many as is often implied and is focal length and subject dependent and sometimes it might just mean a different shutter speed.

The thing that I sometimes have an issue with is "matching DOF".   

FF lenses get infinite DOF pretty quickly with wide angle lenses so in many cases there will be nothing to match once you are at infinity anyway.

Longer lenses like 300mm also I find the typical subject will have enough DOF even at 2.8 FF

Maybe around 50mm FF versus 25mm M43 for my uses there could be an issue FF for me with very fast lenses but while I like fast lenses, I use them wide open fairly rarely (at times a few 50 1.2 and an 85 1.2 and a 24 1.4) but then it is pretty hard to match those fast speeds FF with M43 anyway.

 

 

[scotchtape]

For non-narrative where you do want shallower DOF, it's still way easier on FF since there are tons of F2.8 Zooms.

For M43 you're getting F5.6 equivalent bokeh and there's really nothing you can do if you want the flexibility of a zoom.

Yes you can use the sigma but it's impossible to balance on smaller gimbals, so now you need a bigger heavier gimbal, and the lens itself is pretty heavy, negating the whole small and light m43 thing, plus the AF is 

I've used GH5 for the last year but I really miss how easy it is to get shallow DOF on FF.

F2.8 isn't crazy shallow but it's hella lot better than F5.6 with the M43 zooms. I had to work so hard for it on m43 it just wasn't worth it especially with the useless AF-C

[TwoScoops]

10 hours ago, Nathan Gabriel said:

Has anyone compared the voigtlander to the much cheaper SLR Magic 17mm t1.6? I've found one lab test, but I'd love to hear from someone who has real world experience. I use the SLR Magic all the time and it's really difficult for me to imagine needing shallower depth of field. However I've never really spent time shooting full frame wide open. I will say that I've used 4x5 box cameras where it is easy to get razor thin depth of field and it looks terrible because almost nothing is in focus. I always thought high iso performance rather than shallow depth of field was the bigger issue for MFT. I do have the voigtlander 25mm f0.95 and I rarely feel the need to shoot wide open with it. Here's the comparison for the 17mm's that I mentioned: 

 

 

The Voigt is much better than it looks in that comparison, it's an incredible lens. One of the few I really regret selling.

I posted a still from my portfolio taken with the GH4/Voigt 17.5 here: https://www.eoshd.com/comments/topic/6396-lenses/?page=162&tab=comments#comment-233871 I felt like the Voigts coupled with the GH4 had the most mojo of any non medium format camera I've used for raw stills, and they made the video look a lot nicer too.

 

[Sanka]

Sure, with a series of fast prime lenses, it is possible to achieve that "Full Frame" look on M43 sensor, but the big advantage of FX is that you can get almost everything in 1 single zoom lens such as a 24-70 mm f/2.8, which offer a nice combination of traditional field of views: 24mm, 28mm, 35mm, 50mm.

On M43 you will need to own and change several prime lenses to get that package, such as : 12 mm f/1.4 + 14 mm f/1.4 + 17.5 mm f/1.4 + 25 mm f/1.4.

If for video, I prefer the overall package offered by Panasonic (GH series), for photography I prefer to use my Nikon D750 with a 24-70 mm f/2.8.

[Timotheus]

15 hours ago, Inazuma said:

I wish you'd start a Youtube show Andrew to combat all the misinformation (esp from Northrup) out there

Kidding? This is exactly the point Northrup has been making all along (i.e. apply crop factor to both focal length and aperture to understand how lenses behave on a specific sensor size). Happy everyone is one the same page now

Now let's see whether we ever get an EOS-M speedbooster... The M5 / M50 are not that hot for video, but boosting 'em up to fullframe in such a small package...should be very nice indeed.

[Thpriest]

Voigtlander vs SLR Magic:

I have the 25mm 0.95 Voigtlander and it's amazing. Only thing is it has a bit of bloom wide open but it's useable for video. The image is lovely.

The SLR Magic 12mm T1.6 is nice enough and a lot cheaper. It has a natural look (not like the Lumix lenses) and is a bit warmer than the Voigt. It's a bit harder to focus as well. What I really feel is that T1.6 is not fast enough for me at times on the GH5.

M43 vs APSC/FF

I have the GH5, a Speedbooster Ultra and the Sigma 18-35mmm 1.8 (1.2 with the Speedbooster) and there are good things to say about the combo. What I really like is the IBIS as it means I can go handheld in run'n'gun situations. The 4K 10bit and 4K 50p are great and give a lot of flexibility in post (edited in 1080p).

But....it has 4 weaknesses for me for video. Coming from a Canon C100 mk1, the GH5 really isn't great in lowlight. With the same Sigma lens I could go to 6400-8000 iso with the Canon and with the GH5 it's pushing it at 3200. Whilst the Voigtlander improves things I don't feel it's able to really compete. Another problem is focussing. The LCD and EVF are miles better in the GH5 but being able to punch in to check focus is a godsend in the C100. In the GH5 if you haven't nailed it you are punished (we won't even go into AF!). Ergonomics is another issue. The GH5 is a vast improvement over the Canon DSLRs I have used before but on a C100 everything is in the right place and works. I now find myself thinking about buying a loupe (Vary-i) or mini rig for the GH5 which is not why I bought it! I wanted to go lighter! The other thing I notice is that somehow (voodoo?) the small Canon (WDR) files seem to have far more dynamic range than the GH5 (even 10bit). The Canon is somehow able to make the image work and handles complicated lighting (colour temp or lack of light) much better. At least that's my opinion!

So last night I got very little sleep as I contemplated selling the GH5 and buying a C100mk2!

Why o why hasn't Canon produced a 4k 50p C100 mk3!!!

[Inazuma]

2 hours ago, Timotheus said:

Kidding? This is exactly the point Northrup has been making all along (i.e. apply crop factor to both focal length and aperture to understand how lenses behave on a specific sensor size). Happy everyone is one the same page now

Now let's see whether we ever get an EOS-M speedbooster... The M5 / M50 are not that hot for video, but boosting 'em up to fullframe in such a small package...should be very nice indeed.

The last video I saw of his (a couple years ago), he was claiming that m43 lens manufacturers were mis-marketing their products and that they should be calling their f2.8 lenses f5.6. And seemed to forget that other companies including Canon have been making lenses for crop sensors for years even before m43. Should Canon be calling their 17-55mm f2.8 an f4 instead?

[Andrew Reid]

At the end of the day you can project the same image into a smaller space.

That's physics.

What Northrup was suggesting was wrong.

An F2.8 aperture is always an F2.8.

The optics don't physically change shape when you have a smaller sensor behind it.

A lens that covers a smaller sensor at F2.8 is still a wider opening than F5.6 - with the same brightness as a full frame F2.8.

So no, manufacturers should not be calling a Micro Four Thirds F2.8 lens a F5.6 lens at all.

Using equivalent aperture to calculate depth of field vs full frame, is the only useful use of it. It's for us in our heads to use. It doesn't apply to other aspects.

2 hours ago, Sanka said:

Sure, with a series of fast prime lenses, it is possible to achieve that "Full Frame" look on M43 sensor, but the big advantage of FX is that you can get almost everything in 1 single zoom lens such as a 24-70 mm f/2.8, which offer a nice combination of traditional field of views: 24mm, 28mm, 35mm, 50mm.

On M43 you will need to own and change several prime lenses to get that package, such as : 12 mm f/1.4 + 14 mm f/1.4 + 17.5 mm f/1.4 + 25 mm f/1.4.

If for video, I prefer the overall package offered by Panasonic (GH series), for photography I prefer to use my Nikon D750 with a 24-70 mm f/2.8.

If you put the Canon 24-70mm F2.8 on the Speed Booster XL 0.64x on 1.86x GH5S, you get...

A 1.19x crop sensor, practically full frame.

An F1.8 effective aperture.

15-45mm focal length on a 1.86x crop sensor, so an equiv. range of 28-84mm.

Not bad I'd say.

You forgot the Sigma 18-35mm F1.8 on a Speed Booster 0.71x as well. That is a 28-52mm F2.7 equiv. on the M43 SB and Super 35.

[KnightsFan]

3 hours ago, Inazuma said:

The last video I saw of his (a couple years ago), he was claiming that m43 lens manufacturers were mis-marketing their products and that they should be calling their f2.8 lenses f5.6. And seemed to forget that other companies including Canon have been making lenses for crop sensors for years even before m43. Should Canon be calling their 17-55mm f2.8 an f4 instead?

Assuming we're talking about the same video, he was saying that manufacturers were erroneously marketing their lenses giving equivalent focal lengths, but not equivalent apertures. Like marketing an 14-42 f2.8 as a 28-84 f2.8. His point was that IF you change one number, you have to change the other. And he did specify that he was talking about equivalent depth of field - his title card for talking about it is "Aperture & Depth of Field." He was using ISO to talk about exposure, which he explained early in the video. In his logic, once you compensate for exposure with ISO, you have to compensate with aperture as well.

It is a roundabout logic, but it does account for the lower light gathering power of a smaller surface area, and thus the need for more gain (and thus more noise or lower resolution) to reach the same ISO.

 

3 hours ago, Andrew Reid said:

At the end of the day you can project the same image into a smaller space.

True, but the light gathering power of the space is directly related to its size. The SNR ratio of a smaller sensor will be lower than a larger sensor (assuming similar tech), given the same image scaled down. Hence the f0.95 doesn't actually have any low light advantage over the f2.0, if you look at the system as a whole.

 

[Inazuma]

@KnightsFan Ahh right, I misunderstood his point then. Fair enough.

[Andrew Reid]

47 minutes ago, KnightsFan said:

True, but the light gathering power of the space is directly related to its size.

It's anything but a direct relationship.

I have an old Hasseblad H3DII for instance which is 39MP, Medium Format, much larger sensor than my A7R III which is a similar megapixel count (42MP).

The A7R III is FOUR stops more sensitive, with a smaller sensor.

That's because the quantum efficiency of the chip is higher, the micro lenses are larger, the active surface area of the chip is larger, and the readout circuit is less noisy.

The CCD in the Hasselblad has a lot of front-side wiring taking up space on the front of the sensor where the photo sites gather light.

So yeah, no direct relationship between sensor size and light gathering power. It's more of a positive correlation with pixel size if anything, rather than total surface area.

51 minutes ago, KnightsFan said:

Assuming we're talking about the same video, he was saying that manufacturers were erroneously marketing their lenses giving equivalent focal lengths, but not equivalent apertures. Like marketing an 14-42 f2.8 as a 28-84 f2.8. His point was that IF you change one number, you have to change the other.

But nope... You don't. They are two completely different things.

When that video came out it was just plain confusing and a swipe at the Micro Four Thirds market.

It is actually written 14-42mm on the lens any way so who says they are marketing it misleadingly?

[no_connection]

A speedbooster does not change the focal length or the aperture of a lens. All it does is changing the apparent sensor size. And so does a teleconverter.

If you look down the barrel of a teleconverter you see the sensor appearing at a smaller size. If you look at it this way it's no longer a problem to grasp the concept.

 

The only two things that change DoF at the lens level is apparent front element size and focal/object distance.
The only factor that remain is viewing size. Or viewing angle as seen by the eye. (iPhone screen at arms length will have longer dof than the same image viewed with the phone pressed to your nose). Or circle of confusion or what ppl normally call it.

Sensor size and the needed focal length and corresponding f-stop is only a resulting number that depends on another number and as shown can be interchanged with the same result.

And on a side note, the light gathering is directly related to apparent front element size. The focal length determines how large area that light appears on. And if you know math, it should not come as a surprise that regardless what you do, it's the same amount of light for any chosen lens diameter and field of view, just spread out over a different sensor area.

[KnightsFan]

@Andrew Reid Yeah, "directly related" isn't the correct wording, but we mean the same thing. As sensor size increases, you can either increase resolution with the same pixel size, or increase pixel size at the same resolution. You said so yourself in your disclaimer at the end: "For example you can have higher megapixel counts because there’s simply more real-estate on the chip surface to add more pixels."

40 minutes ago, Andrew Reid said:

The A7R III is FOUR stops more sensitive, with a smaller sensor...It's more of a positive correlation with pixel size if anything, rather than total surface area.

That is why I said "assuming similar tech." The a7r3 has better tech. I should have specified photo-sensitive area, rather than simply "surface area" and would perhaps have been clearer.

 

44 minutes ago, Andrew Reid said:

But nope... You don't. They are two completely different things.

Wait, you just posted an  article about how the 17.5mm f0.95 is comparable to the 35mm f2.0. If you don't have to change both numbers, why aren't you saying the 17.5mm is comparable to a 35mm f0.95?

But anyway, I'm just explaining Northup's argument. He claims there is marketing material on their websites that is misleading, idk if that's true but his crop factor equivalence works for the comparison he's making, despite being poorly worded and convoluted.

[Robert Collins]

It seems to me the price of getting bigger sensors into cameras is falling faster than the cost of getting bigger faster lenses to make up for the smaller sensors.

[noone]

1 hour ago, no_connection said:

A speedbooster does not change the focal length or the aperture of a lens. All it does is changing the apparent sensor size. And so does a teleconverter.

If you look down the barrel of a teleconverter you see the sensor appearing at a smaller size. If you look at it this way it's no longer a problem to grasp the concept.

 

 

I would say it does change the lens in that it becomes part of the lens as a teleconverter does.     It isn't part of the camera anyway.

I believe some faster zoom lenses were actually made by adding a focal reducer to a normal lens in the lens as part of the construction.

In any event, this is what Metabones who makes the Speedbooster says (Speedbooster is a focal reducer).

"Mounted between a mirrorless camera and a SLR lens, Speed Booster® increases maximum aperture (hence its name), increases MTF and makes lens wider. Optics designed by Caldwell Photographic in the USA (patent pending)."

http://www.metabones.com/products/?c=speed-booster