Breakthrough Photography Promises The Best Circular Polarizer

[grahamclarkphoto]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

 

[jeffa4444]

To a point raised by Graham Clark about adding the CPL first...

[truncated]

...Critical sharpness on 36MP camera or higher will likely be more affected by technique (mirror lock-up, tripod, cable release, point of focus etc) than the filters and most colour shifts are correctable in Lightroom as long as they are not severe but its always better to get in right in-camera. As for polarizers they are here to stay regardless of MP.

I have the Lee system and I also own the Canon 5DS. Used the landscape polarizer along with the little stopper & a warming filter on Dartmoor shooting waterfalls a couple of weeks back never had any issues with sharpness. In cinematography Matte Boxes (filter holders) are mounted on bars not the lens so its easier to fit the polarizer at the rear. Unless your making large prints for exhibition then most people will struggle to see the difference a filter will make bad technique will do far more damage but the old maxim remains only use filters when you need to and try & limit the use to two the more you stack the more your degrade the image and create image shift.

Thanks, Jeffa. So this opens up a few questions:

1) Why is Lee Filters -- one of (if not the) biggest names in professional landscape work -- not espousing a similar Lens > CPL > Adaptor Ring > Holder > Slot-in Filter setup? 36MP is not a recent development, and Lee continue to market a system which places the CPL beyond the slot-in filters (i.e. as the furthest forward piece of glass). Do they think vignetting is a bigger deal than sharpness, or is the threat to sharpness not that great?

2) Have any stills shooters run a head to head with the two different ways we've discussed? I looked at Graham's samples, but I didn't see any full res shots to compare.

Again, I don't mean to doubt Graham's posting -- I find it noteworthy and want to learn more. I just need to see the kind of sharpness hit we're talking about just from the order of stacking everything. Does anyone have that?

- A

I used the Lee system with the Canon 5DS on Dartmoor shooting waterfalls three weeks ago with a polarizer, warming filter (to enhance autumn leaves) and a little stopper. Shot everything on a tripod, with mirror lock-up release. All the shots were sharp and I got a good number of keepers. Cinematography matte boxes (filter holders) are mounted on bars not on the lens so its easy to rear mount polarizers. The old maxim of filters is only use them when you have to and try and limit yourself to stacking two the more you stack the bigger the chance of image shift and softness.

 

[jeffa4444]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

 

[ahsanford]

Go down to the Lee vs. X3 ND comparison. Notice how the sharpness sucks?

The reason for this is not that the glass has no MRC coatings, or that the glass is bad, both of which could be true, but rather the glass is thick and it sits much further away from the front lens element.

Doh. I missed the crops, Graham -- apologies. That's a clear sharpness difference, but again, we've changed two variables -- the filters and the order of setup of those filters.

Again -- does anyone have head to head shots where only the order of stacking has changed? I don't have a perfect comparison setup or I'd do it myself.

- A

 

[jeffa4444]

To a point raised by Graham Clark about adding the CPL first...

[truncated]

...Critical sharpness on 36MP camera or higher will likely be more affected by technique (mirror lock-up, tripod, cable release, point of focus etc) than the filters and most colour shifts are correctable in Lightroom as long as they are not severe but its always better to get in right in-camera. As for polarizers they are here to stay regardless of MP.

As far as Im aware all Lee Filters filters are 2mm thick regardless of what they are. In cinematography the standard thickness is 3mm as I stated in all our comprehensive tests down to pixel level the difference between 1mm, 2mm & 3mm is almost zero if they are optically flat and have a high clarity / transmittance value.

Thanks, Jeffa. So this opens up a few questions:

1) Why is Lee Filters -- one of (if not the) biggest names in professional landscape work -- not espousing a similar Lens > CPL > Adaptor Ring > Holder > Slot-in Filter setup? 36MP is not a recent development, and Lee continue to market a system which places the CPL beyond the slot-in filters (i.e. as the furthest forward piece of glass). Do they think vignetting is a bigger deal than sharpness, or is the threat to sharpness not that great?

2) Have any stills shooters run a head to head with the two different ways we've discussed? I looked at Graham's samples, but I didn't see any full res shots to compare.

Again, I don't mean to doubt Graham's posting -- I find it noteworthy and want to learn more. I just need to see the kind of sharpness hit we're talking about just from the order of stacking everything. Does anyone have that?

- A

Talking about results is a lot less effective than viewing results, click here: http://breakthrough.photography/performance-gallery

Go down to the Lee vs. X3 ND comparison. Notice how the sharpness sucks?

The reason for this is not that the glass has no MRC coatings, or that the glass is bad, both of which could be true, but rather the glass is thick and it sits much further away from the front lens element.

Graham

 

[jeffa4444]

To a point raised by Graham Clark about adding the CPL first...

[truncated]

...Critical sharpness on 36MP camera or higher will likely be more affected by technique (mirror lock-up, tripod, cable release, point of focus etc) than the filters and most colour shifts are correctable in Lightroom as long as they are not severe but its always better to get in right in-camera. As for polarizers they are here to stay regardless of MP.

As far as Im aware all Lee Filters filters are 2mm thick regardless of what they are. In cinematography the standard thickness is 3mm as I stated in all our comprehensive tests down to pixel level the difference between 1mm, 2mm & 3mm is almost zero if they are optically flat and have a high clarity / transmittance value.

Thanks, Jeffa. So this opens up a few questions:

1) Why is Lee Filters -- one of (if not the) biggest names in professional landscape work -- not espousing a similar Lens > CPL > Adaptor Ring > Holder > Slot-in Filter setup? 36MP is not a recent development, and Lee continue to market a system which places the CPL beyond the slot-in filters (i.e. as the furthest forward piece of glass). Do they think vignetting is a bigger deal than sharpness, or is the threat to sharpness not that great?

2) Have any stills shooters run a head to head with the two different ways we've discussed? I looked at Graham's samples, but I didn't see any full res shots to compare.

Again, I don't mean to doubt Graham's posting -- I find it noteworthy and want to learn more. I just need to see the kind of sharpness hit we're talking about just from the order of stacking everything. Does anyone have that?

- A

Talking about results is a lot less effective than viewing results, click here: http://breakthrough.photography/performance-gallery

Go down to the Lee vs. X3 ND comparison. Notice how the sharpness sucks?

The reason for this is not that the glass has no MRC coatings, or that the glass is bad, both of which could be true, but rather the glass is thick and it sits much further away from the front lens element.

Graham

As far as Im aware all Lee Filters filters are 2mm regardless of filter. In cinematography they are 3mm and as long as the glass has a high clarity value and is optically flat the difference between 1mm, 2mm & 3mm is almost zero in terms of sharpness. I for one would not like to use 1mm glass in a 4x4 (10cm x 10cm) or 6x4 (10cm x 150cm) holders your just asking for trouble Ive broken 2mm Big Stoppers.
Their resin filters use optical grade ADC polymers of the type Zeiss use for high end specticles

 

[privatebydesign]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

 

[grahamclarkphoto]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

No one ever said diffraction becomes worse with higher MP, we only said that if you enlarge already inherent diffraction it magnifies the detail.

Graham

 

[RobertG.]

For professional setups at 36MP or higher, you can't use CPLs in front of slot-loading filters. It kills sharpness.

The optical thickness must be thin, and the distance to front lens element needs to be close for resolving power to be high.

The correct order to easily achieve critical sharpness:

LENS > CPL > ADAPTER RING > HOLDER > GND

Graham

So Joe Cornish has been doing it wrong all these years?

He uses the Lee 100mm and SW150 setups with CPLs in front of his slot loaders: https://www.youtube.com/watch?v=QN1Q31jSfQ8

There are two vital reasons to do it this way:

• It principally de-couples CPL rotation from ND grad rotation. This is super handy you don't want to have to reposition your ND grad every time you change your polarization (or vice versa).
  
  
• Stacking the CPL behind the adapter ring effectively makes the filter ring nearest the front element 'taller', which will cause you to (mechanically) vignette 'sooner' on UWA focal lengths. That's why most folks I've seen (myself included) tend to get the CPL out of that thickness stack by puttting a massive (105mm or greater) CPL in front of the filter stack.

But sharpness is obviously important so I find your claim fascinating. Has anyone published anything on this, run a head to head comparison, etc.? I'd like to read more about that, thx.

- A

If he shot on 36MP+ sensors with the best glass, then yes, but he hasn't because the advent of ultra-high resolution sensors is a recent advance.

36MP+ with best glass, never use thick filters that sit far away from front lens element, otherwise images will suck.

For an example of this click the below link and view the Lee Big Stopper vs X3 ND, look for sharpness, ignore blue color cast and 2-3 stops darker exposure:

http://breakthrough.photography/performance-gallery

Graham

Hi Graham,

I shoot not only with the 5D II but also with the 5Ds R. "images will suck" is a pretty bold statement. I did just that recently: I shot with my 5Ds R + TS-E 24mm with a grad ND and my old Sigma CPL in front of it. The sharpness is OK and more than sufficient for my landscape and nature shots. In a 100% view of the CR2 file I can see that the picture is a tiny bit softer than without the filters. But this advantage of filterless taken photos is lost in post processing and on print-outs you could not see it. With my TS-E 17mm & 24mm and also the EF 24-70mm L II sharpness is no concern for me. These 3 lenses are used for more than 80% of my pictures at the moment. I own another 15 lenses.... but just a handful of CPL filters.

Sharpness won't be the reason for me to buy a new CPL filter. Accurate colors are much more important for me. That is the reason why I bought a Canon 5Ds R instead of a Sony camera.

 

[privatebydesign]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

No one ever said diffraction becomes worse with higher MP, we only said that if you enlarge already inherent diffraction it magnifies the detail.

Graham

Where does anybody point that out in this thread? I don't see the word 'magnification' used anywhere.

The implication in comment after comment entirely sidesteps the fact that diffraction is a constant related to aperture and magnification only. Not lens, not sensor, and not MP.

 

[grahamclarkphoto]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

No one ever said diffraction becomes worse with higher MP, we only said that if you enlarge already inherent diffraction it magnifies the detail.

Graham

Where does anybody point that out in this thread? I don't see the word 'magnification' used anywhere.

The implication in comment after comment entirely sidesteps the fact that diffraction is a constant related to aperture and magnification only. Not lens, not sensor, and not MP.

It's implied.

Don't feed the trolls.

Graham

 

[grahamclarkphoto]

For professional setups at 36MP or higher, you can't use CPLs in front of slot-loading filters. It kills sharpness.

The optical thickness must be thin, and the distance to front lens element needs to be close for resolving power to be high.

The correct order to easily achieve critical sharpness:

LENS > CPL > ADAPTER RING > HOLDER > GND

Graham

So Joe Cornish has been doing it wrong all these years?

He uses the Lee 100mm and SW150 setups with CPLs in front of his slot loaders: https://www.youtube.com/watch?v=QN1Q31jSfQ8

There are two vital reasons to do it this way:

• It principally de-couples CPL rotation from ND grad rotation. This is super handy you don't want to have to reposition your ND grad every time you change your polarization (or vice versa).
  
  
• Stacking the CPL behind the adapter ring effectively makes the filter ring nearest the front element 'taller', which will cause you to (mechanically) vignette 'sooner' on UWA focal lengths. That's why most folks I've seen (myself included) tend to get the CPL out of that thickness stack by puttting a massive (105mm or greater) CPL in front of the filter stack.

But sharpness is obviously important so I find your claim fascinating. Has anyone published anything on this, run a head to head comparison, etc.? I'd like to read more about that, thx.

- A

If he shot on 36MP+ sensors with the best glass, then yes, but he hasn't because the advent of ultra-high resolution sensors is a recent advance.

36MP+ with best glass, never use thick filters that sit far away from front lens element, otherwise images will suck.

For an example of this click the below link and view the Lee Big Stopper vs X3 ND, look for sharpness, ignore blue color cast and 2-3 stops darker exposure:

http://breakthrough.photography/performance-gallery

Graham

Hi Graham,

I shoot not only with the 5D II but also with the 5Ds R. "images will suck" is a pretty bold statement. I did just that recently: I shot with my 5Ds R + TS-E 24mm with a grad ND and my old Sigma CPL in front of it. The sharpness is OK and more than sufficient for my landscape and nature shots. In a 100% view of the CR2 file I can see that the picture is a tiny bit softer than without the filters. But this advantage of filterless taken photos is lost in post processing and on print-outs you could not see it. With my TS-E 17mm & 24mm and also the EF 24-70mm L II sharpness is no concern for me. These 3 lenses are used for more than 80% of my pictures at the moment. I own another 15 lenses.... but just a handful of CPL filters.

Sharpness won't be the reason for me to buy a new CPL filter. Accurate colors are much more important for me. That is the reason why I bought a Canon 5Ds R instead of a Sony camera.

Did you not click the link I sent you? Seems we're going around in circles on this one...

Yea, Canon 5Ds R is great, but it's utterly useless for landscape photography with it's ISO 12k limitation, unfortunately.

But the A7R2 is great, and although the color doesn't seem as natural out of camera, Sony seems to have nailed a number of other things.

Graham

 

[privatebydesign]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

No one ever said diffraction becomes worse with higher MP, we only said that if you enlarge already inherent diffraction it magnifies the detail.

Graham

Where does anybody point that out in this thread? I don't see the word 'magnification' used anywhere.

The implication in comment after comment entirely sidesteps the fact that diffraction is a constant related to aperture and magnification only. Not lens, not sensor, and not MP.

It's implied.

Don't feed the trolls.

Graham

Hmm, So we have gone from 'nobody said', to 'it is implied'. Sorry, you guys are talking self indulgent crap.

Yea, Canon 5Ds R is great, but it's utterly useless for landscape photography with it's ISO 12k limitation, unfortunately.

Graham

As for trolling, that has to be one of the best here for a while. How many landscape shooters are there? How many of them shoot at 12,000 iso for serious images?

 

[grahamclarkphoto]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

No one ever said diffraction becomes worse with higher MP, we only said that if you enlarge already inherent diffraction it magnifies the detail.

Graham

Where does anybody point that out in this thread? I don't see the word 'magnification' used anywhere.

The implication in comment after comment entirely sidesteps the fact that diffraction is a constant related to aperture and magnification only. Not lens, not sensor, and not MP.

It's implied.

Don't feed the trolls.

Graham

Hmm, So we have gone from 'nobody said', to 'it is implied'. Sorry, you guys are talking self indulgent crap.

Yea, Canon 5Ds R is great, but it's utterly useless for landscape photography with it's ISO 12k limitation, unfortunately.

Graham

As for trolling, that has to be one of the best here for a while. How many landscape shooters are there? How many of them shoot at 12,000 iso for serious images?

ISO 12k for Live View, not shooting. lol.

Graham

 

[Stu_bert]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

No one ever said diffraction becomes worse with higher MP, we only said that if you enlarge already inherent diffraction it magnifies the detail.

Graham

Where does anybody point that out in this thread? I don't see the word 'magnification' used anywhere.

The implication in comment after comment entirely sidesteps the fact that diffraction is a constant related to aperture and magnification only. Not lens, not sensor, and not MP.

It's implied.

Don't feed the trolls.

Graham

Hmm, So we have gone from 'nobody said', to 'it is implied'. Sorry, you guys are talking self indulgent crap.

Yea, Canon 5Ds R is great, but it's utterly useless for landscape photography with it's ISO 12k limitation, unfortunately.

Graham

As for trolling, that has to be one of the best here for a while. How many landscape shooters are there? How many of them shoot at 12,000 iso for serious images?

ISO 12k for Live View, not shooting. lol.

Graham

Sorry for going off thread, but what's the problem with ISO & live view on the 5Ds please ?

 

[grahamclarkphoto]

Should have added defraction will become the biggest enemy the higher we climb the MP ladder.

Yes, but not for the lenses that are being designed for high resolution sensors, like the Canon 16-35 F4 IS.

This lens was designed for 50.2MP+, and the results are shockingly impressive in the CA department.

Graham

Nyquist pure & simple. I have the lens.

Diffraction is exactly the same for a 20MP sensor as it is for a 120MP sensor. If you enlarge the images the same the diffraction is the same.

MP have nothing to do with diffraction.

No one ever said diffraction becomes worse with higher MP, we only said that if you enlarge already inherent diffraction it magnifies the detail.

Graham

Where does anybody point that out in this thread? I don't see the word 'magnification' used anywhere.

The implication in comment after comment entirely sidesteps the fact that diffraction is a constant related to aperture and magnification only. Not lens, not sensor, and not MP.

It's implied.

Don't feed the trolls.

Graham

Hmm, So we have gone from 'nobody said', to 'it is implied'. Sorry, you guys are talking self indulgent crap.

Yea, Canon 5Ds R is great, but it's utterly useless for landscape photography with it's ISO 12k limitation, unfortunately.

Graham

As for trolling, that has to be one of the best here for a while. How many landscape shooters are there? How many of them shoot at 12,000 iso for serious images?

ISO 12k for Live View, not shooting. lol.

Graham

Sorry for going off thread, but what's the problem with ISO & live view on the 5Ds please ?

The Canon 5Ds R uses dual DIGIC 6 processors, with a hardware capability of going to ISO 200k+.

Understandably, due to high shadow detail ISO noise, in the software Canon limited the shooting ISO to ISO 12k. Check out the images below and you'll see that this is actually may be perceived by some as a reasonable move.

However for Live View they messed up and by accident, also in the software, limited the Live View to ISO 12k as well.

That means the Canon 5D2 with a Live View of ISO 25k, released in 2008, outperforms the 5Ds R by a factor of two when composing images with Exposure Simulation turned off (as it should be) with say ND filters, or during sunset where light is low.

For noise performance of the 5Ds R vs. A7R2 vs. 6D check out my in-progress review: http://www.grahamclarkphoto.com/canon-5ds-r-review-by-a-landscape-photographer/

Graham

 

[AvTvM]

I still don't understand the 12k ISO problem. You mean, that in low light there is more visible noise on the small 3" rear LCD of the 5Ds R compared to a 5D II? And is this really a (serious) problem? ???

 

[HighLowISO]

Sounds like the 5Ds live view could be improved easy enough with software update.

Just a little closer to on topic; now that you are working with polarizers, does than mean we can expect a variable ND filter in the not to distant future? If so, would that still give up most of the quality gains of the X3 fixed ND filters?

 

[lholmes549]

Great to see the interaction from Graham!

I have question before purchasing the 105mm filter. What is the thickness of the filter and how is the vignetting at wide angles e.g. 16mm FF?
This is important to me in order to reduce chances of vignetting and have been looking at the Formatt Hitech 105mm UltraSlim CPL for this reason.

Thanks

 

[ahsanford]

I have question before purchasing the 105mm filter. What is the thickness of the filter and how is the vignetting at wide angles e.g. 16mm FF?

Graham can make some more or less vignetting statements comparatively to other filters he may have measured. But you can't state a hard 'it will vignette at Xmm' without more information:

• He'll need to know the lens in question. Every lens has its own filter thread to front element distance that is slightly different.
  
  
• If you are putting this polarizer in front of a slot-in filter holder with the 105 ring*, he'll need to know more about that Lee setup -- are you using a standard or UWA adaptor ring, how many slots are between the lens and the CPL, etc.

* You have to be, don't you? Not aware of any UWA lenses with a 105mm filter ring.

To truly know your vignetting situation, there are enough variables that it's often best to just build that apparatus and test it, tell everyone how it went, and let the interwebs absorb that information for future people who need that answer.

Here's my hyperspecific offering on that front:
http://www.canonrumors.com/forum/index.php?topic=21554.msg409701#msg409701

- A