MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8

Started Feb 16, 2014 | Discussions
Alex Notpro
Contributing MemberPosts: 826Gear list
Like?
MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
Feb 16, 2014

I've been wanting to get rid of redundant photo gear, so today I decided to do a quick test to see if there was really that much difference between FF and mFT in the 50mm FOV. I was expecting to find equivalent results at f/2.8 on FF versus f/1.4 on mFT. However, that's not what I found.

I'm slightly baffled by this, as "equivalence" theory says these two photos should result in the same DOF and background blur. However, the full-frame 50mm shot at f/2.8 clearly out-blurs the background versus the Micro Four Thirds 25mm lens shot at f/1.4.

Just for kicks I also attached the shot with the full-frame lens at f/1.4, which I expected to look different from mFT @ f/1.4, but wanted to see just how different it would be.

Micro Four Thirds 25mm @ f/1.4

Full Frame 50mm @ f/2.8

Full Frame 50mm @ f/1.4

Does anybody have a rational explanation for the difference in background blur between the MFT/25/1.4 and the FF/50/2.8 shots?

 Alex Notpro's gear list:Alex Notpro's gear list
Olympus E-M1 Panasonic Leica Summilux DG 25mm F1.4 Olympus 12-40mm F2.8 Leica Nocticron 42.5mm Apple iPhone 6 +10 more
Alex Notpro
Contributing MemberPosts: 826Gear list
Like?
some additional info
In reply to Alex Notpro, Feb 16, 2014

Cameras were setup 6'3" from the subject, 10' from background (3'9" subject to background). The images were cropped for easier comparison.

 Alex Notpro's gear list:Alex Notpro's gear list
Olympus E-M1 Panasonic Leica Summilux DG 25mm F1.4 Olympus 12-40mm F2.8 Leica Nocticron 42.5mm Apple iPhone 6 +10 more
Reply   Reply with quote   Complain
Willing
Contributing MemberPosts: 655Gear list
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

Looks like the subject 'flash unit' is not really in focus in image1 and image3 and image1 could have a little back focus problem and could have made less background blur than full frame at f2.8!!

Alex Notpro wrote:

I've been wanting to get rid of redundant photo gear, so today I decided to do a quick test to see if there was really that much difference between FF and mFT in the 50mm FOV. I was expecting to find equivalent results at f/2.8 on FF versus f/1.4 on mFT. However, that's not what I found.

I'm slightly baffled by this, as "equivalence" theory says these two photos should result in the same DOF and background blur. However, the full-frame 50mm shot at f/2.8 clearly out-blurs the background versus the Micro Four Thirds 25mm lens shot at f/1.4.

Just for kicks I also attached the shot with the full-frame lens at f/1.4, which I expected to look different from mFT @ f/1.4, but wanted to see just how different it would be.

Micro Four Thirds 25mm @ f/1.4

Full Frame 50mm @ f/2.8

Full Frame 50mm @ f/1.4

Does anybody have a rational explanation for the difference in background blur between the MFT/25/1.4 and the FF/50/2.8 shots?

 Willing's gear list:Willing's gear list
Panasonic Lumix DMC-FZ1000 Olympus E-M1 Olympus OM-D E-M10 Panasonic Lumix DMC-GF7 Olympus M.Zuiko Digital 45mm 1:1.8 +6 more
Reply   Reply with quote   Complain
Fygaren
Regular MemberPosts: 169
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

I see you have cropped the E-M1 pictures into 3:2, this makes the crop factor more like 2.1 and should give less background blur. If you would have cropped the D700 to 4:3 the crop factor would have been more like 1.85 and the E-M1 picture should have had more background blur. Focus has to be exactly on the same spot of course.

http://howmuchblur.com/#compare-1x-50mm-f2.8-and-1.85x-25mm-f1.4-and-2.1x-25mm-f1.4-on-a-0.9m-wide-subject

Reply   Reply with quote   Complain
Olfdee
Forum MemberPosts: 95Gear list
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

Apart from some focusing difference, processing ISO3200 (D700) vs ISO200 is also likely to soften the former image.

Also mind that even with the same physical depth of field, quality of defocused area (bokehliciousness) won't be equal and lens specific additions may affect reception - here you have mostly chromatic abberation on wide open Pana, but it still pronounces wire baskets, giving impression of better defined (less blurred) edge.

 Olfdee's gear list:Olfdee's gear list
Panasonic Lumix DMC-ZS3 Panasonic Lumix DMC-FZ1000 Nikon 1 V1 Olympus PEN E-PM2 Olympus E-M1 +4 more
Reply   Reply with quote   Complain
Daiken
Regular MemberPosts: 138Gear list
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

While I'm no expert, perhaps the equivalence theory only really applies to the depth of field being the same, and not the amount of blur you see.

Reply   Reply with quote   Complain
Fygaren
Regular MemberPosts: 169
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Fygaren, Feb 16, 2014

Example:

When you frame the picture between the long edges (like OP) there is a noticable difference in background blur depending on what aspect ratio you use.

Both pictures below are framed between the bottom of the farting pillow and white mark on monitor with focus being on the power button.

25mm f2 - framed with 3:2

25mm f2 - framed with 4:3

Reply   Reply with quote   Complain
unknown member
(unknown member)
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

Some sensors show a strong fall-off in response to light rays incident at large angles off-axis, resulting in an effective aperture for the lens + sensor combination that is less than the geometrical optics aperture of the lens alone. I've no idea if this is significant in the effect you are seeing...

Joe

-- hide signature --

"Well, it's too late to die young now..."

 Joe Pineapples's gear list:Joe Pineapples's gear list
Ricoh GR
Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,792
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Fygaren, Feb 16, 2014

Fygaren wrote:

Example:

When you frame the picture between the long edges (like OP) there is a noticable difference in background blur depending on what aspect ratio you use.

The images are cropped, so we can't se how they were framed. What matters here is that the shooting distance is held constant. If it is, then the vertical and horizontal FoV will vary with the cameras native aspect ratio, but the diagonal FoV should be the same.

Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,792
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Joe Pineapples, Feb 16, 2014

Joe Pineapples wrote:

Some sensors show a strong fall-off in response to light rays incident at large angles off-axis, resulting in an effective aperture for the lens + sensor combination that is less than the geometrical optics aperture of the lens alone. I've no idea if this is significant in the effect you are seeing...

Joe

-- hide signature --

"Well, it's too late to die young now..."

Yes, it could theoretically be the 'f-stop blues' :

http://www.dxomark.com/Reviews/F-stop-blues

According to DxO the t-stop of the 25/1.4 on E-M1 is 1.7, but not quite clear whether that affects the DoF too. Also possible that the E-M1 image is a bit back-focused.

Reply   Reply with quote   Complain
bluelemmy
Contributing MemberPosts: 867Gear list
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

Shot 1, the flash is fuzzy, shot 2 it sharp. Shot 1 is at low ISO, 2 at 3200 ISO so the comparisons are invalid - plus any difference in cropping makes the comparison invalid.

For a given distance, DoF is governed by the size of the hole the light comes through. 50/2.8 = 17.86mm hole, ditto 25mm/1.4. That's it.

But, the focal length of a lens is only nominally correct at infinity (and even then a 50mm is often 51.5mm or 49mm or something similar). When focal point is closer than infinity, there is an effective change in the focal length of the lens and that can differ according to lens construction and design.

In other words, the two lenses you test are identical in theory but not necessarily so in practise.

-- hide signature --
Reply   Reply with quote   Complain
zcello
Junior MemberPosts: 42Gear list
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

I am thinking if this could be a possible explanation: The m4/3 use quite a few internal correction to correct the distortion etc, so when we see a 25mm lens, it could be in fact a slightly wider lens such as 23mm, the sensor picks up a wider distorted picture and the body makes some crop and transform and the modified picture is what we see equivalent to 25mm fov. If that's the case, the Dof of 25mm/1.4 would be slightly deeper than the true 50mm/2.8.

 zcello's gear list:zcello's gear list
Nikon D300 Olympus PEN E-PL1 Olympus OM-D E-M5 Nikon AF-S Nikkor 300mm f/4D ED-IF Olympus M.Zuiko Digital ED 9-18mm 1:4.0-5.6 +4 more
Reply   Reply with quote   Complain
Fygaren
Regular MemberPosts: 169
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Steen Bay, Feb 16, 2014

I assume he cropped only one picture to get them "equivalent", the E-M1. Shooting with 50mm FF and 25mm mFT the shooting distance will only be the same if the diagonal FoV is the same when FF uses 3:2 and the mFT uses 4:3, crop 1 and 2.

When cropping a crop 2 4:3 sensor to 3:2 you change the diagonal FoV so that the crop factor becomes 2.1

When cropping a crop 1 3:2 sensor to 4:3 you change the diagonal FoV so that the crop factor becomes 1.1

Comparing crop 1.1 to crop 2 is the same as comparing crop 1 to crop 1.85

Reply   Reply with quote   Complain
markintosh13
Senior MemberPosts: 1,648
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Alex Notpro, Feb 16, 2014

Alex Notpro wrote:

Does anybody have a rational explanation for the difference in background blur between the MFT/25/1.4 and the FF/50/2.8 shots?

Focus points and ISO settings.

Set the M43 to ISO 200, the D700 to ISO 400 (not 900 or 3200), use a tripod and manual focus. Then use the same shutter speed on both.

I think you'll then find the shots more similar.

Reply   Reply with quote   Complain
Alex Notpro
Contributing MemberPosts: 826Gear list
Like?
Fygaren is on to something here ... FL and Crop!
In reply to Fygaren, Feb 16, 2014

Thanks all for your responses and trying to help me figure this out. I think the effective(actual) focal length combined with cropping must be the source of the difference. (Other explanations offered don't make sense from an engineering standpoint)

The Nikkor lens has a small but noticeable amount of Focus Breathing, so it's hard to know where exactly it behaves as a 50mm as opposed to a 48mm or 52mm lens. Also, I can't find any manufacturer's specs for either lens, to see if their stated focal lengths have been rounded for marketing purposes. According to this article (http://photo.stackexchange.com/questions/13603/how-to-test-actual-focal-length) I should be able to measure the FL... perhaps I'll try that later.

Here are the crops used for posting:

  • MFT 25/1.4 original 4608x3456 cropped to 1055x1583
  • FF 50/2.8 original 4256x2832 cropped to 988x1484
  • FF 50/1.4 original 4256x2832 cropped to 958x1440

I quoted the crops above in case I'm missing something, but I don't think the crops matter by themselves. If the lens had the exact same FL, than these shots taken from the same distance and cropped down to the same framing would be exactly equivalent! It's only in combination with a difference in FL, that when cropped to match the framing, causes the equivalent aperture on the MFT lens to creep up, i.e. if the 25mm is really a 23mm, then the extra cropping would make it a 50mm f/3.0 equivalent.

 Alex Notpro's gear list:Alex Notpro's gear list
Olympus E-M1 Panasonic Leica Summilux DG 25mm F1.4 Olympus 12-40mm F2.8 Leica Nocticron 42.5mm Apple iPhone 6 +10 more
Reply   Reply with quote   Complain
Alex Notpro
Contributing MemberPosts: 826Gear list
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Daiken, Feb 16, 2014

Daiken wrote:

While I'm no expert, perhaps the equivalence theory only really applies to the depth of field being the same, and not the amount of blur you see.

I'm pretty sure equivalence theory says that two equivalent lenses at the same aperture and shutter speed (with ISO adjusted for sensor size) would take "the same photo".

 Alex Notpro's gear list:Alex Notpro's gear list
Olympus E-M1 Panasonic Leica Summilux DG 25mm F1.4 Olympus 12-40mm F2.8 Leica Nocticron 42.5mm Apple iPhone 6 +10 more
Reply   Reply with quote   Complain
Ido S
Senior MemberPosts: 1,457Gear list
Like?
Look at the shutter speeds
In reply to Alex Notpro, Feb 16, 2014

This may not be relevant to the subject matter, but I'll type and publish it anyway.

To my eyes, the blur is similar enough. It's not so different. But I noticed something really cool. With the Nikon D700, you used ISO 3200 and a shutter speed of 1/80 sec., at f2.8. With the Olympus OM-D E-M1, you used ISO 200 (base ISO, if I'm not mistaken), and a shutter speed of 1/20 sec. at f1.4.

Sometimes, 50mm at f1.4 gives a depth of field that's too shallow for a lot of uses. Using that aperture value on a 25mm lens, gives more depth of field, yet on a µ4/3 camera, it gives a very similar -- if not identical -- field of view. Yet it still is f1.4 in light gathering (though T-stops vary between one lens to another), letting you use a lower ISO for better image quality, and the OM-D's 5-axis in-body image stabilization lets you use very slow shutter speeds, so even if the light drops, you can shoot at base ISO with the aperture set to f1.4, and get a sharp photo with the exposure time close to a full second!

That's one of the advantages of µ4/3 to me: more depth of field for the same field of view and aperture. I don't shoot a lot of portraits (if at all), focusing (pun-intended) more on landscape imagery. And being able to shoot at 12mm, while not including too much in the frame, is very nice. It lets me fight diffraction, and sometimes even allows me to leave my trusty Manfrotto at home.

 Ido S's gear list:Ido S's gear list
Olympus OM-D E-M5 Olympus M.Zuiko Digital ED 9-18mm 1:4.0-5.6 Olympus M.Zuiko Digital ED 12-50mm 1:3.5-6.3 EZ Olympus M.Zuiko ED 75-300mm 1:4.8-6.7 II Olympus M.Zuiko Digital 25mm F1.8 +4 more
Reply   Reply with quote   Complain
labe
Veteran MemberPosts: 3,238Gear list
Like?
Prospective difference!
In reply to Alex Notpro, Feb 16, 2014

its the one and only time 25mm does not equal 50mm in conversion.  If you use 50 mm on both cameras from the same distance and crop the full frame to match fov of m4/3 then the blur should be the same. (That is as long as they are stopped down f1.4 m4/3 and f2.8 for ff.  This does make things really difficult.

-- hide signature --

new to technology,always learning

 labe's gear list:labe's gear list
Fujifilm FinePix IS-1 Fujifilm X-S1 Fujifilm X30 Nikon Coolpix P900 Fujifilm FinePix S5 Pro +3 more
Reply   Reply with quote   Complain
David Kieltyka
Veteran MemberPosts: 4,498Gear list
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to Steen Bay, Feb 16, 2014

Steen Bay wrote:

According to DxO the t-stop of the 25/1.4 on E-M1 is 1.7, but not quite clear whether that affects the DoF too.

It doesn't. DOF is affected by the actual aperture diaphragm diameter rather than the light transmission characteristics of the lens. I've observed this many times in actual use with fast lenses.

-Dave-

 David Kieltyka's gear list:David Kieltyka's gear list
Leica M8.2 Pentax 645D Olympus OM-D E-M5 Panasonic Lumix DMC-GX7 Olympus E-M1 +7 more
Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,792
Like?
Re: MFT 25mm f/1.4 fails to blur background like Full Frame 50mm lens @ f/2.8
In reply to David Kieltyka, Feb 16, 2014

David Kieltyka wrote:

Steen Bay wrote:

According to DxO the t-stop of the 25/1.4 on E-M1 is 1.7, but not quite clear whether that affects the DoF too.

It doesn't. DOF is affected by the actual aperture diaphragm diameter rather than the light transmission characteristics of the lens. I've observed this many times in actual use with fast lenses.

-Dave-

But what if the pixels can't 'see' the outer parts of a large aperture (because of 'pixel vignetting')?

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads