Does 20-40mm f/2.8-4 cover full frame if baffle is removed?

Started Jun 5, 2014 | Discussions
viking79
viking79 Forum Pro • Posts: 14,137
Does 20-40mm f/2.8-4 cover full frame if baffle is removed?

Does anyone know if 20-40mm works with full frame with the rear baffle removed?  That rear baffle is very large.  Mine is a rental copy so I am not going to remove/cut it, but it appears to be the only spot causing vignetting.  I am reviewing the lens, was just curious.

Eric

 viking79's gear list:viking79's gear list
Sony Alpha 7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
Plakanina Contributing Member • Posts: 799
Re: Does 20-40mm f/2.8-4 cover full frame if baffle is removed?

I will be curious also. I have the 20-40 and raved about it at first. Still love it but in the studio shooting against a white background the vignetting is really obvious even shooting at f8 which is my standard studio/strobe f stop so I just use the FA43mm as usual, no vignetting whatsoever. I have never owned a lens that had so much vignetting.

 Plakanina's gear list:Plakanina's gear list
Ricoh GR Pentax K-5 II HD Pentax-DA 20-40mm F2.8-4 ED Limited DC WR
Confused of Malvern Senior Member • Posts: 1,197
Re: Does 20-40mm f/2.8-4 cover full frame if baffle is removed?

Plakanina wrote:

I have the 20-40 and raved about it at first. Still love it but in the studio shooting against a white background the vignetting is really obvious even shooting at f8 ..... I have never owned a lens that had so much vignetting.

Are you saying that it causes vignetting on a Pentax DSLR or on a full frame body?  I haven't seen anyone else mention vignetting on either a K5 or K3

-- hide signature --

Confused of Malvern
'The greatest fool can ask more than the wisest man can answer'

viking79
viking79 OP Forum Pro • Posts: 14,137
Re: Does 20-40mm f/2.8-4 cover full frame if baffle is removed?

viking79 wrote:

Does anyone know if 20-40mm works with full frame with the rear baffle removed? That rear baffle is very large. Mine is a rental copy so I am not going to remove/cut it, but it appears to be the only spot causing vignetting. I am reviewing the lens, was just curious.

Eric

PS, by the looks of it 24mm and up should cover the full frame image, but can't verify without taking the baffle off, and maybe cover slightly wider than 24mm with the baffle removed, but baffle doesn't block much at wider angles. The rear group is far from the sensor above 24mm, so the baffle will shade the sensor more.

Still, if it covers at 24 or 25mm completely, that makes it a nice APS-C 16mm field of view when used on full frame. However, with the baffle on, that is the only focal length which is free of vignetting, so it would basically be a 25mm prime on full frame with the baffle on. edit: or if used on full frame, you could get 16-40mm APS-C equivalent field of view, with reduced resolution away from 16mm APS-C field of view (i.e. set the lens at 24mm and zoom up to 40mm, but crop the image as you zoom out down to APS-C at 40mm).

Eric

 viking79's gear list:viking79's gear list
Sony Alpha 7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
Leandros S Senior Member • Posts: 1,970
Faulty maths?

viking79 wrote:

viking79 wrote:

Does anyone know if 20-40mm works with full frame with the rear baffle removed? That rear baffle is very large. Mine is a rental copy so I am not going to remove/cut it, but it appears to be the only spot causing vignetting. I am reviewing the lens, was just curious.

Eric

PS, by the looks of it 24mm and up should cover the full frame image, but can't verify without taking the baffle off, and maybe cover slightly wider than 24mm with the baffle removed, but baffle doesn't block much at wider angles. The rear group is far from the sensor above 24mm, so the baffle will shade the sensor more.

Still, if it covers at 24 or 25mm completely, that makes it a nice APS-C 16mm field of view when used on full frame. However, with the baffle on, that is the only focal length which is free of vignetting, so it would basically be a 25mm prime on full frame with the baffle on. edit: or if used on full frame, you could get 16-40mm APS-C equivalent field of view

That doesn't pan out - if you're going to convert 24 to 16, you have to convert 40 to 27(ish). Unless you're planning a more fundamental mod of the lens?

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

viking79
viking79 OP Forum Pro • Posts: 14,137
Re: Faulty maths?

Leandros S wrote:

viking79 wrote:

viking79 wrote:

Does anyone know if 20-40mm works with full frame with the rear baffle removed? That rear baffle is very large. Mine is a rental copy so I am not going to remove/cut it, but it appears to be the only spot causing vignetting. I am reviewing the lens, was just curious.

Eric

PS, by the looks of it 24mm and up should cover the full frame image, but can't verify without taking the baffle off, and maybe cover slightly wider than 24mm with the baffle removed, but baffle doesn't block much at wider angles. The rear group is far from the sensor above 24mm, so the baffle will shade the sensor more.

Still, if it covers at 24 or 25mm completely, that makes it a nice APS-C 16mm field of view when used on full frame. However, with the baffle on, that is the only focal length which is free of vignetting, so it would basically be a 25mm prime on full frame with the baffle on. edit: or if used on full frame, you could get 16-40mm APS-C equivalent field of view

That doesn't pan out - if you're going to convert 24 to 16, you have to convert 40 to 27(ish). Unless you're planning a more fundamental mod of the lens?

No, this is weird so bear with me. I mean if you leave it uncropped at 24mm, but crop it to APS-C size at 40mm, it would share field of view as a 16-40mm APS-C lens used on an APS-C camera. A full frame 24-70mm used on a full frame is just as flexible as an APS-C size 16-70mm used on APS-C. This is because the wide angle field of view is the same (24mm on full frame and 16mm on APS-C), and you can always crop the 70mm end of the full frame to APS-C (effectively what the APS-C sensor is doing, why they call them crop sensors) and get the same field of view as the APS-C camera.  You will lose some resolution doing this.  If you had a 36 MP full frame it would only be about 16 MP cropped to APS-C.

Using the 20-40mm some tonight and I can tell the corrections are meant for APS-C (field curvature, etc gets out of hand near edge of full frame image), but could work well for centered subjects. So I wouldn't recommend it for full frame, but it could still be sort of useful (typical of many APS-C lenses on full frame, some of Pentax lenses are actually well corrected on full frame, like DA 55mm f/1.4).

Eric

 viking79's gear list:viking79's gear list
Sony Alpha 7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
Plakanina Contributing Member • Posts: 799
Re: Does 20-40mm f/2.8-4 cover full frame if baffle is removed?

Confused of Malvern wrote:

Plakanina wrote:

I have the 20-40 and raved about it at first. Still love it but in the studio shooting against a white background the vignetting is really obvious even shooting at f8 ..... I have never owned a lens that had so much vignetting.

Are you saying that it causes vignetting on a Pentax DSLR or on a full frame body? I haven't seen anyone else mention vignetting on either a K5 or K3

-- hide signature --

Confused of Malvern
'The greatest fool can ask more than the wisest man can answer'

K5 ll

 Plakanina's gear list:Plakanina's gear list
Ricoh GR Pentax K-5 II HD Pentax-DA 20-40mm F2.8-4 ED Limited DC WR
Leandros S Senior Member • Posts: 1,970
Re: Faulty maths?

viking79 wrote:

Leandros S wrote:

viking79 wrote:

viking79 wrote:

Does anyone know if 20-40mm works with full frame with the rear baffle removed? That rear baffle is very large. Mine is a rental copy so I am not going to remove/cut it, but it appears to be the only spot causing vignetting. I am reviewing the lens, was just curious.

Eric

PS, by the looks of it 24mm and up should cover the full frame image, but can't verify without taking the baffle off, and maybe cover slightly wider than 24mm with the baffle removed, but baffle doesn't block much at wider angles. The rear group is far from the sensor above 24mm, so the baffle will shade the sensor more.

Still, if it covers at 24 or 25mm completely, that makes it a nice APS-C 16mm field of view when used on full frame. However, with the baffle on, that is the only focal length which is free of vignetting, so it would basically be a 25mm prime on full frame with the baffle on. edit: or if used on full frame, you could get 16-40mm APS-C equivalent field of view

That doesn't pan out - if you're going to convert 24 to 16, you have to convert 40 to 27(ish). Unless you're planning a more fundamental mod of the lens?

No, this is weird so bear with me. I mean if you leave it uncropped at 24mm, but crop it to APS-C size at 40mm, it would share field of view as a 16-40mm APS-C lens used on an APS-C camera. A full frame 24-70mm used on a full frame is just as flexible as an APS-C size 16-70mm used on APS-C. This is because the wide angle field of view is the same (24mm on full frame and 16mm on APS-C), and you can always crop the 70mm end of the full frame to APS-C (effectively what the APS-C sensor is doing, why they call them crop sensors) and get the same field of view as the APS-C camera. You will lose some resolution doing this. If you had a 36 MP full frame it would only be about 16 MP cropped to APS-C.

Using the 20-40mm some tonight and I can tell the corrections are meant for APS-C (field curvature, etc gets out of hand near edge of full frame image), but could work well for centered subjects. So I wouldn't recommend it for full frame, but it could still be sort of useful (typical of many APS-C lenses on full frame, some of Pentax lenses are actually well corrected on full frame, like DA 55mm f/1.4).

Eric

Once you get into that kind of logic, the whole exercise kind of loses its point. I mean, why assume a 1.5 crop factor when you could assume 1.6 or even 2 or 3? Just because Nikon cameras have a crop mode that happens to default to 1.5? Seems arbitrary.

When I get tempted to think that kind of math, I try to think of something I call resolving power, i.e. focal length * crop factor * megapixels, or EFL * megapixels. Obviously other things factor in, such as the actual sharpness of the lens and noise levels of the sensor. But I think it gives a better approximation than assuming arbitrary crop factors.

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

viking79
viking79 OP Forum Pro • Posts: 14,137
Re: Faulty maths?

Leandros S wrote:

Once you get into that kind of logic, the whole exercise kind of loses its point. I mean, why assume a 1.5 crop factor when you could assume 1.6 or even 2 or 3? Just because Nikon cameras have a crop mode that happens to default to 1.5? Seems arbitrary.

When I get tempted to think that kind of math, I try to think of something I call resolving power, i.e. focal length * crop factor * megapixels, or EFL * megapixels. Obviously other things factor in, such as the actual sharpness of the lens and noise levels of the sensor. But I think it gives a better approximation than assuming arbitrary crop factors.

It isn't arbitrary, but the same as APS-C. This goes back to the larger sensor being more flexible and focal length is a fixed property of the lens. A 40mm lens is always a 40mm lens, it just depends on how much your sensor crops the image circle on the field of view you get.

A 36 MP full frame cropped to APS-C is near identical to the K-5IIs or something. The K-3 or a 24 MP will give you slightly more resolving power if your lenses can handle it.

I don't really have any point here other than a larger sensor is more flexible since you can always choose to crop it to the smaller sensor size. You could argue the same with APS-C vs 4/3" etc.

My original example of the 20-40mm on full frame is not a good one since it really isn't designed for full frame, but it is valid to say 24-70mm on full frame is just as flexible as 16-70mm on APS-C. Depending on the resolution of either your full frame or APS-C camera, one might have higher effective resolution at 70mm, but I would argue that at 70mm on any kit lens is going to be lens limited at APS-C and not sensor limited (i.e. most lenses are weaker at 70mm).

Eric

 viking79's gear list:viking79's gear list
Sony Alpha 7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
Leandros S Senior Member • Posts: 1,970
Re: Faulty maths?

viking79 wrote:

Leandros S wrote:

Once you get into that kind of logic, the whole exercise kind of loses its point. I mean, why assume a 1.5 crop factor when you could assume 1.6 or even 2 or 3? Just because Nikon cameras have a crop mode that happens to default to 1.5? Seems arbitrary.

When I get tempted to think that kind of math, I try to think of something I call resolving power, i.e. focal length * crop factor * megapixels, or EFL * megapixels. Obviously other things factor in, such as the actual sharpness of the lens and noise levels of the sensor. But I think it gives a better approximation than assuming arbitrary crop factors.

It isn't arbitrary, but the same as APS-C. This goes back to the larger sensor being more flexible and focal length is a fixed property of the lens. A 40mm lens is always a 40mm lens, it just depends on how much your sensor crops the image circle on the field of view you get.

And yet you say that a 20mm lens is also a 16mm lens. You're keeping the cake and eating it at the same time. See my comment below.

A 36 MP full frame cropped to APS-C is near identical to the K-5IIs or something. The K-3 or a 24 MP will give you slightly more resolving power if your lenses can handle it.

I don't really have any point here other than a larger sensor is more flexible since you can always choose to crop it to the smaller sensor size. You could argue the same with APS-C vs 4/3" etc.

It seems you do agree that it's arbitrary. Why not extrapolate from FF to 4/3? Or why, indeed, not specify it the other way around and say it is a 20-60mm lens? Or a 20-80mm? I'm giving these examples simply to show how arbitrary and confusing your abuse of the equivalency system is.

My original example of the 20-40mm on full frame is not a good one since it really isn't designed for full frame, but it is valid to say 24-70mm on full frame is just as flexible as 16-70mm on APS-C. Depending on the resolution of either your full frame or APS-C camera, one might have higher effective resolution at 70mm, but I would argue that at 70mm on any kit lens is going to be lens limited at APS-C and not sensor limited (i.e. most lenses are weaker at 70mm).

I don't see what that has to do with your argument. With the same pixel density and lens, the lens limitations are the same no matter how large your sensor.

And you know and I know that there were times when sensors weren't outperforming lenses, and that those cameras are still in use. A paradigm that applies only to SOME cameras is not a good one.

Also, pixel density is one thing, but file size and time spent doing the actual cropping is another. Not to mention that the D800 or A7r are not perfect cameras - they each work for some kinds of shooting much better than for others. If your argument only works when the highest resolving cameras are considered, and ignores the opportunity costs involved in cropping, it strikes me as limited.

There is a school of thought that says you should avoid cropping if you can, and that a real teleconverter is better than a digital one.

I have no problem, btw, with you shooting a 20mm on full frame that was originally designed for APS-C. But stating you now have a 16-40mm lens breaks the established framework in ways that aren't useful, imo. Cropping, after all, was invented before your time.

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

viking79
viking79 OP Forum Pro • Posts: 14,137
Sample at 24mm f/4 (approximately)
1

viking79 wrote:

Does anyone know if 20-40mm works with full frame with the rear baffle removed? That rear baffle is very large. Mine is a rental copy so I am not going to remove/cut it, but it appears to be the only spot causing vignetting. I am reviewing the lens, was just curious.

Eric

20-40mm Sample at 24mm f/4 (approximately) on A7R .

LR: Punch preset, +34 shadows, Sharpen Scenic. I corrected some lens vignetting, but added vignetting 2 preset. It changes the rolloff a bit. The extreme corners are a bit dark on full frame.

This lens is virtually immune to flare. There is some on the purple leaf, and my cheap adapter will sometimes add a ring around it (and it might be causing that tiny spot dead center, don't know if that is the lens or the adapter, most likely the adapter since it is in the same spot in shots it shows in.  If you are hyper critical you can even see the maligned Sony 11+7 delta encoding errors on the veins of the leaf (they get jagged horizontal lines of pixels of incorrect value where they veins are bright).

A7R 20-40mm f/2.8-4 at approximately 24mm f/4 (adapter is limited control).

APS-C crop at 20mm f/2.8 (I think)

 viking79's gear list:viking79's gear list
Sony Alpha 7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
viking79
viking79 OP Forum Pro • Posts: 14,137
and a couple landscapes
1

These pretty much all have sharpen scenic, remove CA (it is one button click and removes lateral CA very effectively, the lens has quite a bit), and punch done in Lightroom.

Here is one shot at 24mm and left full frame. Note how much wider the usable field of view is on full frame. A Pentax K full frame would be nice

24mm on full frame, focus center, left uncropped, some light vignetting correction. f/8 or f/11, or 16mm APS-C equivalent.

24mm on full frame, focus corner, left uncropped, some light vignetting correction. f/8 or f/11, or 16mm APS-C equivalent.

And another shot at 20mm and cropped to APS-C. This lens looks pretty decent in the corners APS-C if you stop down, although another shot I had looked a little worse but it was also at f/4 or f/5.6.

APS-C crop of 20mm shot (so 30mm full frame equivalent) at f/8 or f/11.

uncropped 20mm view, same image as above.

 viking79's gear list:viking79's gear list
Sony Alpha 7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
Leandros S Senior Member • Posts: 1,970
Re: Sample at 24mm f/4 (approximately)

Thanks for these. Maybe you could upload a sample image to lensfun - if you haven't already.

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

viking79
viking79 OP Forum Pro • Posts: 14,137
Re: Sample at 24mm f/4 (approximately)

Leandros S wrote:

Thanks for these. Maybe you could upload a sample image to lensfun - if you haven't already.

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

Yes, I can do that, but what is lensfun?

I am not familiar with that one.

Eric

 viking79's gear list:viking79's gear list
Sony Alpha 7R Samsung NX1 Samsung NX 30mm F2 Pancake Samsung NX 85mm F1.4 ED SSA Samsung NX 60mm F2.8 Macro ED OIS SSA +5 more
Leandros S Senior Member • Posts: 1,970
Re: Sample at 24mm f/4 (approximately)

viking79 wrote:

Leandros S wrote:

Thanks for these. Maybe you could upload a sample image to lensfun - if you haven't already.

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

Yes, I can do that, but what is lensfun?

I am not familiar with that one.

Eric

Open source lens correction:

http://wilson.bronger.org/calibration

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

Keyboard shortcuts:
FForum MMy threads