Crop sensor Pro Body - Dp review tv

[MShot]

You know you can't get your groceries without a V8

 

[Skeeterbytes]

You know you can't get your groceries without a V8

I demand the spicy flavor, and keep that low-sodium version out of my house!

Recently counted up engine types of vehicles the spouse and I've owned and was..surprised. I-3 NA, I-4 NA, I-4 turbo, I-5 turbo, V-6 turbo, I-6 turbo, V-8 NA.

Looks like we're gonna need a V-12 to round out the set.

Agree with the above that 3:2 is arbitrary and less than ideal. YMMV.

Cheers,

Rick

 

[Mark Ransom]

Agree with the above that 3:2 is arbitrary and less than ideal. YMMV.

Not sure where anybody was discussing the 3:2 ratio, but OK.

For a long time I was biased against the 4:3 ratio, because that was my first digital camera experience and it was a pain. I had to crop different sides depending on whether I was doing 4x6 (3:2) snapshot prints or 8x10 (5:4) prints. If my framing put anything near the edges, one or the other wasn't going to work.

I got over it when I realized I hadn't done 4x6 in many years.

 

[Skeeterbytes]

Years and years printing from 135 negs and chromes to the bread-and-butter 8x10 and 11x14 papers had me composing to those formats when shooting, presuming one or more end of the x-axis would be cut out. Printing oneself accommodates fiddling to get the best compromise, including surrendering a little of the print's y-axis to get the whole frame. Sending the film in to a consumer lab meant they were making the decisions and often getting it wrong. Hrrmph!

Anyway, my eye thus trained I felt right at home with four-thirds, a little like when I got a 25mm prime and was transported back to my first SLR with a nifty fifty. Fits like a glove.

cheers,

Rick

 

[AiryDiscus]

Direct quote again below noise was only part of what he was declaring

https://www.dpreview.com/forums/post/61876146

"Oh, well of course Fuji's gonna win, even despite their crappy x-trans (3 or maybe it's 4 by now) and tempered RAW which includes noise reduction."

It does not matter what other people wrote. The thing is that you wrote that "m43 RAW files are amongst the most "tampered" with on the market if that concerns you". You fueled the flames claiming that this is the case although you were clearly wrong. Your original post is linked below:

https://www.dpreview.com/forums/post/61876156

If someone does not like X-Trans, then let it be. Opinions are one thing, but wrong facts are something that needs to be corrected, whether it's you or me who wrote them.

FWIW there were actually some fixed noise pattern tests in this forum to check for things like NR of raws. IIRC, B Claff was doing them, and the E-M1 II was definitely covered. I'm also fairly certain Jim is aware of those tests, and they just "slipped his mind".

(Whereby, "slipped his mind" I mean he intentionally neglected them, because that's nowhere near as 'big a story'). Always remember. Jim isn't actually interested in truth. Jim is interested in drama.

The post linked by Mr Stirling, above, is by a lens tester who uses an all-optical testing machine (OLAF machine at Lensrentals), so when a lens has half its corrections in the glass and half in the software, he has to 'guess/estimate' the corrections of the software and effectively his lens tests are invalid. He says it quintuples the time it takes to optically test them (with his machine). He hates that, so he won't have anything to do with m43.

Basically, he has a hammer (his OLAF machine), and he sees m43 lenses as really bad nails (not suited to his machine), so he criticizes the nail, instead of facing reality that the machine is effectively obsolete for the camera/lens systems of today.

The full extent of his rant can be seen here, https://www.dpreview.com/forums/post/61232509 . Note the balanced comment by Iliah Borg, also no fool, just above it. Even Mr Stirling made a balanced comment there.

cheers

My my, so many sentiments put in my mouth. Let me walk you through this step by step:

1. The machine isn't "OLAF," Olaf is the company. The machine is a Trioptics ImageMaster HR Max.

2. The machine uses an ISO standard method of MTF testing. This involves using a microscope to magnify a small image of a small target in order to satisfy some sampling requirements.

3. Microscopes have very small fields of view, in this case about 0.2 x 0.2 mm. The machine uses an idealized lens model (i.e. no distortion) to guesstimate where the next field point along the line will lie.

4. When there is a lot of distortion that changes quickly, the estimate becomes faulty and we have to use measured coordinates for the field points. This action quadruples the time it takes to test the lens.

5. We do not cook the data in any way to account for corrections done in software. That isn't science, we try to stick to science in our office.

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

 

[TN Args]

Direct quote again below noise was only part of what he was declaring

https://www.dpreview.com/forums/post/61876146

"Oh, well of course Fuji's gonna win, even despite their crappy x-trans (3 or maybe it's 4 by now) and tempered RAW which includes noise reduction."

It does not matter what other people wrote. The thing is that you wrote that "m43 RAW files are amongst the most "tampered" with on the market if that concerns you". You fueled the flames claiming that this is the case although you were clearly wrong. Your original post is linked below:

https://www.dpreview.com/forums/post/61876156

If someone does not like X-Trans, then let it be. Opinions are one thing, but wrong facts are something that needs to be corrected, whether it's you or me who wrote them.

FWIW there were actually some fixed noise pattern tests in this forum to check for things like NR of raws. IIRC, B Claff was doing them, and the E-M1 II was definitely covered. I'm also fairly certain Jim is aware of those tests, and they just "slipped his mind".

(Whereby, "slipped his mind" I mean he intentionally neglected them, because that's nowhere near as 'big a story'). Always remember. Jim isn't actually interested in truth. Jim is interested in drama.

The post linked by Mr Stirling, above, is by a lens tester who uses an all-optical testing machine (OLAF machine at Lensrentals), so when a lens has half its corrections in the glass and half in the software, he has to 'guess/estimate' the corrections of the software and effectively his lens tests are invalid. He says it quintuples the time it takes to optically test them (with his machine). He hates that, so he won't have anything to do with m43.

Basically, he has a hammer (his OLAF machine), and he sees m43 lenses as really bad nails (not suited to his machine), so he criticizes the nail, instead of facing reality that the machine is effectively obsolete for the camera/lens systems of today.

The full extent of his rant can be seen here, https://www.dpreview.com/forums/post/61232509 . Note the balanced comment by Iliah Borg, also no fool, just above it. Even Mr Stirling made a balanced comment there.

cheers

My my, so many sentiments put in my mouth.

I linked to your rant. Your words.

Let me walk you through this step by step:

1. The machine isn't "OLAF," Olaf is the company. The machine is a Trioptics ImageMaster HR Max.

OLAF isn't the company! I remember Roger's naming contest when the machine was new. As I recall, OLAF was the name that won. It's a nickname.

2. The machine uses an ISO standard method of MTF testing. This involves using a microscope to magnify a small image of a small target in order to satisfy some sampling requirements.

3. Microscopes have very small fields of view, in this case about 0.2 x 0.2 mm. The machine uses an idealized lens model (i.e. no distortion) to guesstimate where the next field point along the line will lie.

4. When there is a lot of distortion that changes quickly, the estimate becomes faulty and we have to use measured coordinates for the field points. This action quadruples the time it takes to test the lens.

5. We do not cook the data in any way to account for corrections done in software. That isn't science, we try to stick to science in our office.

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

OK -- you are saying M43 lenses don't get tested, because they aren't worth the effort. I assumed they were still getting tested, because your employer's lensauthority website says they do. Are you saying that your company lies about that? Or are you hiding the fact that they do get tested?

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

If you don't understand how hybrid design from the ground up can match or even exceed all-optical design.... then I understand your bias.

 

[AiryDiscus]

Direct quote again below noise was only part of what he was declaring

https://www.dpreview.com/forums/post/61876146

"Oh, well of course Fuji's gonna win, even despite their crappy x-trans (3 or maybe it's 4 by now) and tempered RAW which includes noise reduction."

It does not matter what other people wrote. The thing is that you wrote that "m43 RAW files are amongst the most "tampered" with on the market if that concerns you". You fueled the flames claiming that this is the case although you were clearly wrong. Your original post is linked below:

https://www.dpreview.com/forums/post/61876156

If someone does not like X-Trans, then let it be. Opinions are one thing, but wrong facts are something that needs to be corrected, whether it's you or me who wrote them.

FWIW there were actually some fixed noise pattern tests in this forum to check for things like NR of raws. IIRC, B Claff was doing them, and the E-M1 II was definitely covered. I'm also fairly certain Jim is aware of those tests, and they just "slipped his mind".

(Whereby, "slipped his mind" I mean he intentionally neglected them, because that's nowhere near as 'big a story'). Always remember. Jim isn't actually interested in truth. Jim is interested in drama.

The post linked by Mr Stirling, above, is by a lens tester who uses an all-optical testing machine (OLAF machine at Lensrentals), so when a lens has half its corrections in the glass and half in the software, he has to 'guess/estimate' the corrections of the software and effectively his lens tests are invalid. He says it quintuples the time it takes to optically test them (with his machine). He hates that, so he won't have anything to do with m43.

Basically, he has a hammer (his OLAF machine), and he sees m43 lenses as really bad nails (not suited to his machine), so he criticizes the nail, instead of facing reality that the machine is effectively obsolete for the camera/lens systems of today.

The full extent of his rant can be seen here, https://www.dpreview.com/forums/post/61232509 . Note the balanced comment by Iliah Borg, also no fool, just above it. Even Mr Stirling made a balanced comment there.

cheers

My my, so many sentiments put in my mouth.

I linked to your rant. Your words.

Let me walk you through this step by step:

1. The machine isn't "OLAF," Olaf is the company. The machine is a Trioptics ImageMaster HR Max.

OLAF isn't the company! I remember Roger's naming contest when the machine was new. As I recall, OLAF was the name that won. It's a nickname.

http://www.olafoptical.com/ It's an LLC. We do have a machine that the internet named OLAF, but it can't measure MTF.

2. The machine uses an ISO standard method of MTF testing. This involves using a microscope to magnify a small image of a small target in order to satisfy some sampling requirements.

3. Microscopes have very small fields of view, in this case about 0.2 x 0.2 mm. The machine uses an idealized lens model (i.e. no distortion) to guesstimate where the next field point along the line will lie.

4. When there is a lot of distortion that changes quickly, the estimate becomes faulty and we have to use measured coordinates for the field points. This action quadruples the time it takes to test the lens.

5. We do not cook the data in any way to account for corrections done in software. That isn't science, we try to stick to science in our office.

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

OK -- you are saying M43 lenses don't get tested, because they aren't worth the effort. I assumed they were still getting tested,

faulty assumption. We don't test Pentax, either.

because your employer's lensauthority website says they do.

LA isn't my employer, nor is LR, nor does my employer belong to the same conglomerate.

Are you saying that your company lies about that? Or are you hiding the fact that they do get tested?

Bowling alleys with custom test charts at the end, not MTF tested.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

If you don't understand how hybrid design from the ground up can match or even exceed all-optical design.... then I understand your bias.

Ah, you assume they can. That is not a given.

 

[Adielle]

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

It definitely defies logic, and it also defies logic to use a machine that's unsuitable for testing lenses that are not super over-engineered for the application of general photography, considering the fact that most lenses on the market are not such.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

When it comes to purists, they'll often ignore practicality and focus solely on specs, and that's exactly what you do. Lenses with superb specs are usually far too expensive or large or heavy, and compromises have to be made to produce a product that suits the market. By having full control over both the optical and the software aspects, the companies are able to produce products that are good for the vast majority of the market, and not just purists. The truth is that the optical side of superb lenses also usually includes a serious amount of optical corrections within the lens, with elements that correct all kinds of things. Moving a part of the correction to the digital side is a very sensible and smart design decision, which purists disapprove of and act as if corrections done on the optical side are always acceptable while corrections done on the software side are never acceptable. Purism is funny like that.

 

[AiryDiscus]

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

It definitely defies logic, and it also defies logic to use a machine that's unsuitable for testing lenses that are not super over-engineered for the application of general photography, considering the fact that most lenses on the market are not such.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

When it comes to purists, they'll often ignore practicality and focus solely on specs, and that's exactly what you do. Lenses with superb specs are usually far too expensive or large or heavy, and compromises have to be made to produce a product that suits the market. By having full control over both the optical and the software aspects, the companies are able to produce products that are good for the vast majority of the market, and not just purists.

-

The truth is that the optical side of superb lenses also usually includes a serious amount of optical corrections within the lens, with elements that correct all kinds of things.

What is this technobabble supposed to mean? That well corrected lenses are well corrected?

Moving a part of the correction to the digital side is a very sensible and smart design decision, which purists disapprove of and act as if corrections done on the optical side are always acceptable while corrections done on the software side are never acceptable. Purism is funny like that.

It's not about acceptable, it's about preferrable. If the M4/3 manufactures accepted lower margins to put another asphere or two in their wide angle lenses, they wouldn't have 10, or even 20% distortion. They would't be selling fisheyes as rectilinear.

 

[Adielle]

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

It definitely defies logic, and it also defies logic to use a machine that's unsuitable for testing lenses that are not super over-engineered for the application of general photography, considering the fact that most lenses on the market are not such.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

When it comes to purists, they'll often ignore practicality and focus solely on specs, and that's exactly what you do. Lenses with superb specs are usually far too expensive or large or heavy, and compromises have to be made to produce a product that suits the market. By having full control over both the optical and the software aspects, the companies are able to produce products that are good for the vast majority of the market, and not just purists.

-

The truth is that the optical side of superb lenses also usually includes a serious amount of optical corrections within the lens, with elements that correct all kinds of things.

What is this technobabble supposed to mean? That well corrected lenses are well corrected?

Again with the purism. You think that lenses that were designed to give a result that should be software-corrected are not "well corrected", but lenses that were designed with more or different optical correction are "well corrected". Despite this common "technobabble", the fact is that optical corrections are CORRECTIONS due to deficiencies in the design of the lens, and they have side-effects. Same with digital corrections. To purists like you, the optical corrections are "well" while digital are "not". Don't try to pass it off as objectivity, it's just your opinion.

Moving a part of the correction to the digital side is a very sensible and smart design decision, which purists disapprove of and act as if corrections done on the optical side are always acceptable while corrections done on the software side are never acceptable. Purism is funny like that.

It's not about acceptable, it's about preferrable. If the M4/3 manufactures accepted lower margins to put another asphere or two in their wide angle lenses, they wouldn't have 10, or even 20% distortion. They would't be selling fisheyes as rectilinear.

Whatever. I bet they know a lot more than you about how to design good lenses that are suitable for their market. At least the results I and many people are getting from those lenses are superb, despite the company's choice to exploit the fact that in the modern world, software can be used to correct some things well enough and in some cases even better than optical elements (which add even worse artifacts in some cases).

 

[AiryDiscus]

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

It definitely defies logic, and it also defies logic to use a machine that's unsuitable for testing lenses that are not super over-engineered for the application of general photography, considering the fact that most lenses on the market are not such.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

When it comes to purists, they'll often ignore practicality and focus solely on specs, and that's exactly what you do. Lenses with superb specs are usually far too expensive or large or heavy, and compromises have to be made to produce a product that suits the market. By having full control over both the optical and the software aspects, the companies are able to produce products that are good for the vast majority of the market, and not just purists.

-

The truth is that the optical side of superb lenses also usually includes a serious amount of optical corrections within the lens, with elements that correct all kinds of things.

What is this technobabble supposed to mean? That well corrected lenses are well corrected?

Again with the purism. You think that lenses that were designed to give a result that should be software-corrected are not "well corrected",

Well, if you have a lens and you add aberrations to it, it is logically less well corrected.

but lenses that were designed with more or different optical correction are "well corrected".

That would imply the lenses I have in mind are more corrected, but they aren't as sharp as their FF counterparts and have about an order of magnitude more distortion. I wouldn't call that differently corrected, only less.

Despite this common "technobabble", the fact is that optical corrections are CORRECTIONS due to deficiencies in the design of the lens,

No, the use of 'corrected' is just grammar. A single lens element cannot create a diffraction-limited image. You need a harmonious ensemble of them to form a good (diffraction limited) image. An optical correction is not a deficiency of the design of a lens, it is exactly the opposite; optical correction is a feature while residual aberrations are defects or deficiencies.

and they have side-effects.

Perhaps size weight or cost, but nothing more.

Same with digital corrections.

Your smartphone lens has << 2% distortion and they aim for those cameras to go through a whole lot of post processing. If undercorrection going in was a good thing, your smartphone lens would be terrible, but it is actually remarkably good.

To purists like you, the optical corrections are "well" while digital are "not". Don't try to pass it off as objectivity, it's just your opinion.

Garbage in, garbage out as they say.

Moving a part of the correction to the digital side is a very sensible and smart design decision, which purists disapprove of and act as if corrections done on the optical side are always acceptable while corrections done on the software side are never acceptable. Purism is funny like that.

It's not about acceptable, it's about preferrable. If the M4/3 manufactures accepted lower margins to put another asphere or two in their wide angle lenses, they wouldn't have 10, or even 20% distortion. They would't be selling fisheyes as rectilinear.

Whatever. I bet they know a lot more than you about how to design good lenses that are suitable for their market.

The failure of M4/3 to capture a larger market share seems to indicate they do not know how to design products the market wants more than others. You also see much more chatter online speaking against their bigger, more expensive, larger aperture lenses than you do comments in support of them.

At least the results I and many people are getting from those lenses are superb,

And the paints and brushes the mona lisa was made with are worse than those available today. That someone is successful with a tool does not indicate that it is a good tool, only that it is not unusable.

despite the company's choice to exploit the fact that in the modern world, software can be used to correct some things well enough and in some cases even better than optical elements (which add even worse artifacts in some cases).

Digital correction can never be better than optical with the exception of synthesized information from a neural net under ideal conditions (i.e. true superresolution).

That's why, for example, deconvolution, the most sophisticated form of digital correction, produces ringing artifacts.

 

[Adielle]

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

It definitely defies logic, and it also defies logic to use a machine that's unsuitable for testing lenses that are not super over-engineered for the application of general photography, considering the fact that most lenses on the market are not such.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

When it comes to purists, they'll often ignore practicality and focus solely on specs, and that's exactly what you do. Lenses with superb specs are usually far too expensive or large or heavy, and compromises have to be made to produce a product that suits the market. By having full control over both the optical and the software aspects, the companies are able to produce products that are good for the vast majority of the market, and not just purists.

-

The truth is that the optical side of superb lenses also usually includes a serious amount of optical corrections within the lens, with elements that correct all kinds of things.

What is this technobabble supposed to mean? That well corrected lenses are well corrected?

Again with the purism. You think that lenses that were designed to give a result that should be software-corrected are not "well corrected",

Well, if you have a lens and you add aberrations to it, it is logically less well corrected.

but lenses that were designed with more or different optical correction are "well corrected".

That would imply the lenses I have in mind are more corrected, but they aren't as sharp as their FF counterparts and have about an order of magnitude more distortion. I wouldn't call that differently corrected, only less.

Despite this common "technobabble", the fact is that optical corrections are CORRECTIONS due to deficiencies in the design of the lens,

No, the use of 'corrected' is just grammar. A single lens element cannot create a diffraction-limited image. You need a harmonious ensemble of them to form a good (diffraction limited) image. An optical correction is not a deficiency of the design of a lens, it is exactly the opposite; optical correction is a feature while residual aberrations are defects or deficiencies.

An optical correction is not a "deficiency" by itself, the design of the lens itself is deficient enough that it requires elements that correct distortions of the image by another element or group of elements. The lens's design, which requires such corrective elements in the lens, which introduce artifacts, is deficient. Therefore, correction is needed. You gonna do a part of it digitally? Well, that's the designer's choice.

and they have side-effects.

Perhaps size weight or cost, but nothing more.

Nothing more? No effect on aberrations, color, flare, etc? Give me a break. This "harmonious" design which includes corrective elements always has deficiencies, there's never a free lunch.

Same with digital corrections.

Your smartphone lens has << 2% distortion and they aim for those cameras to go through a whole lot of post processing. If undercorrection going in was a good thing, your smartphone lens would be terrible, but it is actually remarkably good.

"Undercorrection" is yet again just your opinion. In the modern world where the final image ends up in a digital format, companies can design a lens that performs generally better with less optical correction than it could with more optical correction, by design, and choose to rely on digital correction as the remainder. This is the modern world, we're not in an analog-only world anymore.

To purists like you, the optical corrections are "well" while digital are "not". Don't try to pass it off as objectivity, it's just your opinion.

Garbage in, garbage out as they say.

"Garbage" is abundant in optics, and the corrections you so love, with all their deficiencies are not inherently better. A design that relies on a combination of optical and digital corrections can have as good or better quality than a design that relies solely on optical correction. It's that simple.

Moving a part of the correction to the digital side is a very sensible and smart design decision, which purists disapprove of and act as if corrections done on the optical side are always acceptable while corrections done on the software side are never acceptable. Purism is funny like that.

It's not about acceptable, it's about preferrable. If the M4/3 manufactures accepted lower margins to put another asphere or two in their wide angle lenses, they wouldn't have 10, or even 20% distortion. They would't be selling fisheyes as rectilinear.

Whatever. I bet they know a lot more than you about how to design good lenses that are suitable for their market.

The failure of M4/3 to capture a larger market share seems to indicate they do not know how to design products the market wants more than others. You also see much more chatter online speaking against their bigger, more expensive, larger aperture lenses than you do comments in support of them.

You have nothing to support your claim that MFT is failing. Market share by itself means nothing, and MFT hasn't had any significant change in market share.

At least the results I and many people are getting from those lenses are superb,

And the paints and brushes the mona lisa was made with are worse than those available today. That someone is successful with a tool does not indicate that it is a good tool, only that it is not unusable.

Yeah, you're gonna keep claiming that excellent lenses that enable excellent images are bad, no matter what anyone says, I know. Unless it's a purist-approved design, it's not gonna be good enough, just "usable" at best, as far as you're concerned.

despite the company's choice to exploit the fact that in the modern world, software can be used to correct some things well enough and in some cases even better than optical elements (which add even worse artifacts in some cases).

Digital correction can never be better than optical

That's nonsense, and you know it. A simple case is avoidance of aberration or flare that an optical element would have caused. A digital implementation doesn't have that. Another simple case is effect that additional elements have on the contrast, or tinting color. Avoidable with a digital implementation. I have no idea how you can dare to claim what you did.

with the exception of synthesized information from a neural net under ideal conditions (i.e. true superresolution).

It's not really "synthesized".

That's why, for example, deconvolution, the most sophisticated form of digital correction, produces ringing artifacts.

Deconvolution has nothing to do with correction, it's just a processing method, and in the context of correction, it is far from the most sophisticated form of correction, it's mostly just much more brute force and processing cycle intensive than classical methods, and most implementations are pretty terrible.

 

[AiryDiscus]

Micro 4/3 is about 6% of the rental market. Devoting 5x the effort to 6% of the market defies logic.

It definitely defies logic, and it also defies logic to use a machine that's unsuitable for testing lenses that are not super over-engineered for the application of general photography, considering the fact that most lenses on the market are not such.

When it comes to opinions, I'm a purist. I'll buy a well corrected lens at a premium. But an under-engineered effort that only gets across the finish line with a fancy DSP? I'll pass.

When it comes to purists, they'll often ignore practicality and focus solely on specs, and that's exactly what you do. Lenses with superb specs are usually far too expensive or large or heavy, and compromises have to be made to produce a product that suits the market. By having full control over both the optical and the software aspects, the companies are able to produce products that are good for the vast majority of the market, and not just purists.

-

The truth is that the optical side of superb lenses also usually includes a serious amount of optical corrections within the lens, with elements that correct all kinds of things.

What is this technobabble supposed to mean? That well corrected lenses are well corrected?

Again with the purism. You think that lenses that were designed to give a result that should be software-corrected are not "well corrected",

Well, if you have a lens and you add aberrations to it, it is logically less well corrected.

but lenses that were designed with more or different optical correction are "well corrected".

That would imply the lenses I have in mind are more corrected, but they aren't as sharp as their FF counterparts and have about an order of magnitude more distortion. I wouldn't call that differently corrected, only less.

Despite this common "technobabble", the fact is that optical corrections are CORRECTIONS due to deficiencies in the design of the lens,

No, the use of 'corrected' is just grammar. A single lens element cannot create a diffraction-limited image. You need a harmonious ensemble of them to form a good (diffraction limited) image. An optical correction is not a deficiency of the design of a lens, it is exactly the opposite; optical correction is a feature while residual aberrations are defects or deficiencies.

An optical correction is not a "deficiency" by itself, the design of the lens itself is deficient enough that it requires elements that correct distortions of the image by another element or group of elements.

So... are you on theangryphotographer's team "more elements more bad" ? A single element is not good enough to form a good image. Every other element is added to make the image better and correct more of the aberrations left over from that original 1 element.

So where does the buck stop? Should we just make singlets and throw all the computational power in the world at the image? Doublets? Triplets?

The lens's design, which requires such corrective elements in the lens, which introduce artifacts, is deficient.

A more highly corrected design is not a deficient one. I would also like you to precisely define artifact in this context.

Therefore, correction is needed. You gonna do a part of it digitally? Well, that's the designer's choice.

More likely marketing/sales, but ok.

and they have side-effects.

Perhaps size weight or cost, but nothing more.

Nothing more? No effect on aberrations, color, flare, etc? Give me a break. This "harmonious" design which includes corrective elements always has deficiencies, there's never a free lunch.

Reducing aberrations (i.e., "correction") definitely doesn't add them. Color is essentially entirely to do with coatings, which are essentially independent to the base optical design. Flare is also largely to do with coatings.

A lens which is diffraction limited over its full field of view has no deficiencies left. That isn't a free lunch, it's the result of hard work. A free lunch would be a pinhole.

Same with digital corrections.

Your smartphone lens has << 2% distortion and they aim for those cameras to go through a whole lot of post processing. If undercorrection going in was a good thing, your smartphone lens would be terrible, but it is actually remarkably good.

"Undercorrection" is yet again just your opinion.

No, it is a technical term from the lens design community. If there are residual aberrations, they are undercorrected.

In the modern world where the final image ends up in a digital format, companies can design a lens that performs generally better with less optical correction than it could with more optical correction, by design, and choose to rely on digital correction as the remainder. This is the modern world, we're not in an analog-only world anymore.

Ah. Ok, I'll bite. The people underdesigning these lenses are essentially all in M4/3, or some niches of the industrial sector. The statement that these lenses are generally better must mean they are generally better than the alternative, i.e. APS-C, full-frame, or larger formats.

If I ignore APS-C for a moment and concentrate on FF, you have a system which:

• within the same generation, always offers a higher resolution in MP
• offers lenses with higher optical quality
  • less flare
  • higher resolution
  • less distortion
• offers greater control over depth of field
• offers, generally, a better quality bokeh
• offers cheaper lenses when compared at the "excellent" tier of both systems.

To purists like you, the optical corrections are "well" while digital are "not". Don't try to pass it off as objectivity, it's just your opinion.

Garbage in, garbage out as they say.

"Garbage" is abundant in optics, and the corrections you so love, with all their deficiencies are not inherently better.

The vast majority of the optics community would disagree with you there.

A design that relies on a combination of optical and digital corrections can have as good or better quality than a design that relies solely on optical correction. It's that simple.

Ok. Show me two equivalent designs and demonstrate the superiority of the lens that uses software corrections.

Moving a part of the correction to the digital side is a very sensible and smart design decision, which purists disapprove of and act as if corrections done on the optical side are always acceptable while corrections done on the software side are never acceptable. Purism is funny like that.

It's not about acceptable, it's about preferrable. If the M4/3 manufactures accepted lower margins to put another asphere or two in their wide angle lenses, they wouldn't have 10, or even 20% distortion. They would't be selling fisheyes as rectilinear.

Whatever. I bet they know a lot more than you about how to design good lenses that are suitable for their market.

The failure of M4/3 to capture a larger market share seems to indicate they do not know how to design products the market wants more than others. You also see much more chatter online speaking against their bigger, more expensive, larger aperture lenses than you do comments in support of them.

You have nothing to support your claim that MFT is failing. Market share by itself means nothing, and MFT hasn't had any significant change in market share.

I didn't say it was failing. I said it wasn't growing in any significant fashion. If the product development and marketing teams for M4/3 had a better ability to address what the market wants than their competitors, their share would grow. Plain and simple.

At least the results I and many people are getting from those lenses are superb,

And the paints and brushes the mona lisa was made with are worse than those available today. That someone is successful with a tool does not indicate that it is a good tool, only that it is not unusable.

Yeah, you're gonna keep claiming that excellent lenses that enable excellent images are bad, no matter what anyone says, I know.

Well, who gets to say excellent is what matters here. I would say that there are exceptionally few "excellent" lenses for the M4/3 platform. But the 75/1.8 is one of them, if you want to know where my head is.

Unless it's a purist-approved design, it's not gonna be good enough, just "usable" at best, as far as you're concerned.

It doesn't have to be purist approved, just well designed.

despite the company's choice to exploit the fact that in the modern world, software can be used to correct some things well enough and in some cases even better than optical elements (which add even worse artifacts in some cases).

Digital correction can never be better than optical

That's nonsense, and you know it. A simple case is avoidance of aberration or flare that an optical element would have caused.

Putting in elements to reduce aberrations won't add them. Flare will only rise in significant fashion from a dramatic change of design form.

A digital implementation doesn't have that.

There is no technical reason to concede that optical correction invites flare, but I will give you that a digital correction can't add flare. But it also can't fix softness, makes the image softer when you correct distortion, alters the FoV when you correct distortion, damages edges when you fix chromatic aberrations, and has a host of other problems.

Another simple case is effect that additional elements have on the contrast, or tinting color. Avoidable with a digital implementation. I have no idea how you can dare to claim what you did.

Ok. Show me (that is, with data) the correlation that adding elements lowers contrast and tints color.

with the exception of synthesized information from a neural net under ideal conditions (i.e. true superresolution).

It's not really "synthesized".

The function of a neural net is exactly to form a collection of hypothesis about what a pixel should be, and pick the one it is most confident in. That is synthesis.

That's why, for example, deconvolution, the most sophisticated form of digital correction, produces ringing artifacts.

Deconvolution has nothing to do with correction,

It is the only DSP technique to reverse the point spread function and "un-aberrate" an image.

it's just a processing method, and in the context of correction, it is far from the most sophisticated form of correction,

Things like Sobel filters, unsharp masks, etc, are vastly less sophistocated than something like richardson-lucy.

it's mostly just much more brute force and processing cycle intensive than classical methods, and most implementations are pretty terrible.

What is "classical?" Blind deconvolution generally sucks. Deconvolution with an estimate of the PSF can work quite well - most microscopy works that way. You trade SNR and some artifacts (edge ringing, etc), both of which are problems you wouldn't have with an optical correction.

 

[Raist3d]

thanks for sticking to science Some people here have the *fantasy* that software corrections for a lens are better than optical for image quality they are better to make cheaper perhaps or smaller perhaps but not image quality

how people think that inventing information where there’s no original information is something that can yield the same result i don’t know

This is a key reason why 4/3 lenses are better overall than m43 lenses optically

 

[TN Args]

thanks for sticking to science Some people here have the *fantasy* that software corrections for a lens are better than optical for image quality they are better to make cheaper perhaps or smaller perhaps but not image quality

how people think that inventing information where there’s no original information is something that can yield the same result i don’t know

This is a key reason why 4/3 lenses are better overall than m43 lenses optically

Clearly you don't understand the concept. I particularly love the way you manage to insult everyone who does. See if you can find a better way to express your self in future.

This is not a case of reinstating lost information. What makes you think it is??

 

[Martin.au]

thanks for sticking to science Some people here have the *fantasy* that software corrections for a lens are better than optical for image quality they are better to make cheaper perhaps or smaller perhaps but not image quality

how people think that inventing information where there’s no original information is something that can yield the same result i don’t know

This is a key reason why 4/3 lenses are better overall than m43 lenses optically

Are they always? I swapped my 12-60 for my 12-40 because it gave better results. Mustache distortion is a pita.

 

[glassoholic]

thanks for sticking to science Some people here have the *fantasy* that software corrections for a lens are better than optical for image quality they are better to make cheaper perhaps or smaller perhaps but not image quality

how people think that inventing information where there’s no original information is something that can yield the same result i don’t know

This is a key reason why 4/3 lenses are better overall than m43 lenses optically

Are they always? I swapped my 12-60 for my 12-40 because it gave better results.

SHG FT lenses are very good but the m43 Pro's are excellent and if software correction allows cheaper and smaller outcomes then I am good with that.

 

[Raist3d]

thanks for sticking to science Some people here have the *fantasy* that software corrections for a lens are better than optical for image quality they are better to make cheaper perhaps or smaller perhaps but not image quality

how people think that inventing information where there’s no original information is something that can yield the same result i don’t know

This is a key reason why 4/3 lenses are better overall than m43 lenses optically

Clearly you don't understand the concept. I particularly love the way you manage to insult everyone who does. See if you can find a better way to express your self in future.

I believe i didn’t call any names here I am saying it’s difficult to understand how people canconclude that software corrections somehow can put information accurately where it’s not there

This is not a case of reinstating lost information. What makes you think it is??

You are using the data you got to stretch and interpolate to pixels While thisnis way better than generating random data it is still not the same as having a lens optically give you that deficinitoonnin those areas from the get go

so when you software correct for distortion because you have to warp/unwrap/ stretch you necessarily lose somewhere some resolution whether thisnis joiceaboe or not will vary per individual, how strong the correction is etc

what i can say- again- this is one reason why the older 4/3 system lenses were better overall than current most m43 lenses

Of course older 4/3 lenses were bigger which is a drawback the best m43 lenses avoid usually some or most of this

 

[Mark Ransom]

thanks for sticking to science Some people here have the *fantasy* that software corrections for a lens are better than optical for image quality they are better to make cheaper perhaps or smaller perhaps but not image quality

how people think that inventing information where there’s no original information is something that can yield the same result i don’t know

This is a key reason why 4/3 lenses are better overall than m43 lenses optically

Clearly you don't understand the concept. I particularly love the way you manage to insult everyone who does. See if you can find a better way to express your self in future.

I believe i didn’t call any names here I am saying it’s difficult to understand how people canconclude that software corrections somehow can put information accurately where it’s not there

This is not a case of reinstating lost information. What makes you think it is??

You are using the data you got to stretch and interpolate to pixels While thisnis way better than generating random data it is still not the same as having a lens optically give you that deficinitoonnin those areas from the get go

so when you software correct for distortion because you have to warp/unwrap/ stretch you necessarily lose somewhere some resolution whether thisnis joiceaboe or not will vary per individual, how strong the correction is etc

what i can say- again- this is one reason why the older 4/3 system lenses were better overall than current most m43 lenses

Of course older 4/3 lenses were bigger which is a drawback the best m43 lenses avoid usually some or most of this

Since we're sticking to science...

There's no need to make up any missing information to warp an image, as long as the stretching is minimal. Information theory says that the true value of any point can be known as long as the proper filter function is used and the input didn't contain any frequencies above the Nyquist limit. In our case the Bayer filter on the sensor puts an upper limit on the frequencies, since not every pixel has a sample for every color.

The perfect filter function is called Sinc, and it can't be practically implemented since the terms go out to infinity. But you can get some pretty good approximations. I'd expect desktop software to do a better job of this than the camera, since the camera processor is less powerful and has less time to do the job.

As for whether it's better to add a glass element to do the same thing, that's debatable. Every element will add some degradation because glass isn't perfect. Although we expect the degradation to be minimal there's no denying that it will happen. Certainly the lens designer wouldn't add it if the expected benefit didn't grossly outweigh the disadvantage. But if you could use software to completely eliminate an element, would the resulting image be better or worse? I don't think you can make a definitive statement either way.

 

[AiryDiscus]

thanks for sticking to science Some people here have the *fantasy* that software corrections for a lens are better than optical for image quality they are better to make cheaper perhaps or smaller perhaps but not image quality

how people think that inventing information where there’s no original information is something that can yield the same result i don’t know

This is a key reason why 4/3 lenses are better overall than m43 lenses optically

Clearly you don't understand the concept. I particularly love the way you manage to insult everyone who does. See if you can find a better way to express your self in future.

I believe i didn’t call any names here I am saying it’s difficult to understand how people canconclude that software corrections somehow can put information accurately where it’s not there

This is not a case of reinstating lost information. What makes you think it is??

You are using the data you got to stretch and interpolate to pixels While thisnis way better than generating random data it is still not the same as having a lens optically give you that deficinitoonnin those areas from the get go

so when you software correct for distortion because you have to warp/unwrap/ stretch you necessarily lose somewhere some resolution whether thisnis joiceaboe or not will vary per individual, how strong the correction is etc

what i can say- again- this is one reason why the older 4/3 system lenses were better overall than current most m43 lenses

Of course older 4/3 lenses were bigger which is a drawback the best m43 lenses avoid usually some or most of this

Since we're sticking to science...

There's no need to make up any missing information to warp an image, as long as the stretching is minimal. Information theory says that the true value of any point can be known as long as the proper filter function is used and the input didn't contain any frequencies above the Nyquist limit.

No - sampling theory states that if a signal is Nyquist sampled, its sampled representation contains enough data to recreate exactly the original signal. This has to do only with reconstruction. Correcting distortion is a non-affine, non-reversible transformation and is not a filter. Because it is non-reversible, it is lossy.

In our case the Bayer filter on the sensor puts an upper limit on the frequencies, since not every pixel has a sample for every color.

This is absolutely not true. The bayer filter makes the chromatic signal sparse. Sparcity is not equivalent to filtered. The presence of aliasing and things like moire are proof positive that Bayer images can be less than Nyquist sampled. In fact, consumer preferences are for something around Q=1.5 being "ok" sharpness and Q=0.5 being "critical" sharpness. You need Q>2 for Nyquist sampling!

The perfect filter function is called Sinc, and it can't be practically implemented since the terms go out to infinity. But you can get some pretty good approximations. I'd expect desktop software to do a better job of this than the camera, since the camera processor is less powerful and has less time to do the job.

Sinc can be practically implemented - exactly even. The FT of sinc is just a rect function, which has finite support. The camera doesn't calculate sinc, but even if it did it's only about 5 flops. To process a 20MP image with one sinc per pixel that's 100Mflops - a measly 800Mhz processor could do it in about 100 milliseconds.

As for whether it's better to add a glass element to do the same thing, that's debatable. Every element will add some degradation because glass isn't perfect.

In what way is the glass imperfect and how does that impact the image?

Do you mean homogeneity and things like striae? No substantive impact on an imaging system far from the diffraction limit like a camera lens. Surface roughness? Matters in UV, not VIS. Surface irregularity? The elements in a camera lens are good enough for that to not matter.

Although we expect the degradation to be minimal there's no denying that it will happen. Certainly the lens designer wouldn't add it if the expected benefit didn't grossly outweigh the disadvantage. But if you could use software to completely eliminate an element, would the resulting image be better or worse? I don't think you can make a definitive statement either way.

If software were king, lithography lenses wouldn't have 40-80 elements in them, they'd have 6 and software thrown on top. Here's one that's a mild derivative of a 90s design, good enough for a 100 nm or so process:

All that fused silica is definitely because software is superior!