I call Hot Rod Cameras Bluff with a $10,000 Bet (for Sharpest PL RF for Komodo)

[Adrian Jebef]

Focusing is something easily achieved with a part of the lens called the focus ring. I'm surprised I needed to explain it here. And no- it is not optical performance. It's is a mechanical adjustment.

lol

i thought we were discussing collimation to the sensor plane

what you talking bout? wut do you mean "the typical back-focus error"? does this has something to do wit ziess digiprimes?

 

[Jacek Zakowicz]

Colimation to the sensor plane has very little, if anything at all to do with the lens performance. Any collimation error is easily corrected using the focusing ring on the lens. Typical machining tolerance error of parts stack will not affect the lens performance at all unless the parts stack up too high and you can't reach infinity. But in that case you can still achieve the same performance from the lens at shorter distances just by turning the focus ring for the sharpest image. Sharpest image for a properly set lens and not properly set lens back focus will be exactly the same. I believe this is about 4th and 5th way of saying the same thing here just so some can understand. Lets see how many more ways we can come up with ;-)

 

[Adrian Jebef]

lol

JZ yer just being semantically silly. soooo much chatter about how a lens can totally resolve and perform just fine at a camera-to-subject distance of 10ft (or wutevs) even if collimation is out of spec. OK. sooooooo wut? wut happens when we need to put on the 18mm and do a nice clean in-focus wide shot? maybe it's in focus, maybe it's not? lol

this thread is clearly about PL Mount adapters for Komodo and how precise in terms of proper collimation they are (or are not). infinity focus is proper collimation spec. either the camera (and PL Mount adapter / lens) are properly collimated to spec and the lens hits infinity as it should or it doesn't. can the camera and the lens or maybe both be out of collimation and still produce a sharp image? sure. but why would anyone accept that as an acceptable way to work?! lol

so much agro as usual for RU.

i get it. the OP has been obtuse and a bit opaque and def a lotta cocky but reading thru his responses its clear as day to me he is simply saying his mount hits spec closer than any other mount on the market right out of the box and onto a right out of the box Komodo. that's it. nothing more nothing less.

 

[Michael Hastings]

This is probably the silliest thread I've seen on Reduser. :wink5: but what the hell - I'll throw in my two cents.

Everybody get out a piece of paper. Draw 2 lines one positive 20 degree angle the other negative 20 degree angle (so a 40 degree FOV). The point where the two lines (light rays) cross is the sharpest it can possibly be. If the crosspoint is at the position of the film or sensor that is the sharpest it can be. In a lens with a focus ring you can move the point where it crosses. So within limits, you can focus on a sensor that is not exactly where you (or the perfectly collimated lens) planned. The major effect will be that if the lens is supposed to be able to focus - for example - at 1 foot all the way to infinity either you won't be able to focus at infinity but you will be able to focus closer than 1 foot (this is the effect of macro extension tubes but much more dramatic) OR you will be able to focus at infinity but your closest focus will be greater than one foot. If you want to see why it's more critical with wide angle lenses - draw two lines at 45 degrees (90 degree for) and see how the gap is that much wider at the same distance from the cross point.

If you want to understand depth of field (or is it depth of focus - can never remember) look at the two lines just slightly to the left or right of the crosspoint (i.e. move your theoretical sensor to the left or right) - there's a tiny gap - if that gap is smaller than the distance between pixels then you are "in focus for practical purposes" i.e. the sensor can't tell the difference since it does not have enough resolving power. As you move farther from the crosspoint the gap gets larger than the distance between pixels and your OUT OF FOCUS.

But the point that "back focus" or sensor distance won't affect the ability to get the same level of "sharpness" is essentially correct as long as the sensor position isn't either so close to the lens or so far away that the focus ring can't get anything focused.

So could changes in Flange depth or collimation create some other optical problems? - probably. Modern high quality lenses tend to have numerous elements designed to correct for chromatic aberrations, sphericity etc. and those corrections may only be effective if both lens and sensor are at the design distances. But for lens/sensor settings that are within sort of "normal" i.e. correctable with the very thin shims that are normally used - I suspect these problems would be minimal i.e. slightly greater chromatic aberrations, distortion etc. And that is a completely different discussion than the basic initial argument that someones adapter comes out of the box that only needs two tiny shims rather than four tiny shims is going to be "sharper".

 

[Jacek Zakowicz]

lol

JZ yer just being semantically silly. soooo much chatter about how a lens can totally resolve and perform just fine at a camera-to-subject distance of 10ft (or wutevs) even if collimation is out of spec. OK. sooooooo wut? wut happens when we need to put on the 18mm and do a nice clean in-focus wide shot? maybe it's in focus, maybe it's not? lol

this thread is clearly about PL Mount adapters for Komodo and how precise in terms of proper collimation they are (or are not). infinity focus is proper collimation spec. either the camera (and PL Mount adapter / lens) are properly collimated to spec and the lens hits infinity as it should or it doesn't. can the camera and the lens or maybe both be out of collimation and still produce a sharp image? sure. but why would anyone accept that as an acceptable way to work?! lol

so much agro as usual for RU.

i get it. the OP has been obtuse and a bit opaque and def a lotta cocky but reading thru his responses its clear as day to me he is simply saying his mount hits spec closer than any other mount on the market right out of the box and onto a right out of the box Komodo. that's it. nothing more nothing less.

OK i found another way:
All still camera lenses are setup short from the factory- i.e the back focus is incorrect. The bodies are also short- i.e. the flange is incorrect. Since the are both short they actually compound into a bigger error. It's clearly visible by the multiple index marks on still camera zoom lenses. The reason behind it is strictly practical one- infinity focus must be assured no matter what temperature, environment these lenses are used in. Do you think that lens manufacturers would compromise optical performance for that. Do you think that still camera lenses perform worse because of that. With 50-100mpix sensors? Here -we found 6 ways to explain the same thing now...

 

[Adrian Jebef]

i like this thread. it's classic RU

so much posturing that collimation isn't an issue. that any lens can focus sharp. hell that lenses are built to not be in tolerance. that all lens designs are the same. all lenses can focus so any type of mount is suitable. back-focus errors are something to embrace. ok wuteva.

so why is there even a 52mm PL spec? Why not just make it all short like cheap still lenses? why not just make the focus marks stickers so i can move them around at will during prep?

i guess you guys are right: a PL Mount is a PL Mount and any PL Mount that is a PL Mount works exactly like a PL Mount should work as long as it's a PL Mount which simply means as long as the lens doesn't fall off while I'm shooting it's mos def doing it job.

 

[Björn Benckert]

Listen to Jacek, he is a paid professional when it comes to fiddle wiht lenses and he has done it more than most. And yes, as he writes for still work backfocus is no much of an importance, as focus marks are not really used and zoom lenses does not really need to be parfocal. However Canon, Nikon and the others still care more than a little about lens performance and lens sharpness.

 

[Björn Benckert]

lol

JZ yer just being semantically silly. soooo much chatter about how a lens can totally resolve and perform just fine at a camera-to-subject distance of 10ft (or wutevs) even if collimation is out of spec. OK. sooooooo wut? wut happens when we need to put on the 18mm and do a nice clean in-focus wide shot? maybe it's in focus, maybe it's not? lol

this thread is clearly about PL Mount adapters for Komodo and how precise in terms of proper collimation they are (or are not). infinity focus is proper collimation spec. either the camera (and PL Mount adapter / lens) are properly collimated to spec and the lens hits infinity as it should or it doesn't. can the camera and the lens or maybe both be out of collimation and still produce a sharp image? sure. but why would anyone accept that as an acceptable way to work?! lol

so much agro as usual for RU.

i get it. the OP has been obtuse and a bit opaque and def a lotta cocky but reading thru his responses its clear as day to me he is simply saying his mount hits spec closer than any other mount on the market right out of the box and onto a right out of the box Komodo. that's it. nothing more nothing less.

You dont get it.

TS has made claims that trees shot with his mount looked sharper than trees shot with other manufactured mounts. Like his mount was more tack sharp then other mounts. Well it´s not. If he said your focus marks will line up better with his mount if you use a perfectly colimated lens. Sure that could be true. But no, thats likely not true either as all komodos are coming out different even striaght out of the box, temperature is in play, wear and tear and how solid the mount is seated etc. All parameters that is way more in play than TS claimed tolerances of 0.00001mm or what ever the distance he mention was set when milling the mount.

 

[Jacek Zakowicz]

Listen to Jacek, he is a paid professional when it comes to fiddle wiht lenses and he has done it more than most. And yes, as he writes for still work backfocus is no much of an importance, as focus marks are not really used and zoom lenses does not really need to be parfocal. However Canon, Nikon and the others still care more than a little about lens performance and lens sharpness.

Thank you Bjorn for your kind words.
If anyone here cares this is my opinion based on selling (high end electronic movie camera) adapters for 12 years:
1.Adapters should be set up with close tolerance with the infinity focus bias at all times. Shimming provisions are needed but only to bring the adapter to specs within the tolerance. No individual shimming to a specific camera body is acceptable.
2.Lenses should be checked and corrected to specs.
3.Camera flanges should be checked and corrected to specs.
Notes:
Adjustable adapters will introduce more problems than they solve. Not to mention the size/expense penalty to the user.
Both camera flange corrections and lens back focus corrections to specs assure the lens camera interchangeability across different brands with the same mounting system. This is absolutely crucial IMO
THE END

 

[Bruce Allen]

Looks like I need to post some examples. "as far as film makers are concerned", most interesting statement that made me actually angry today.

Be back in a few days when I'm done shooting I guess.

Phil do you have time to post some images? I would love to see images showing the actual difference between:

a) lens on a mount with perfect backfocus

b) lens on a mount with slightly off backfocus (let's say missing a few shims) but focus adjusted to compensate (eg we are trying to focus at infinity but find that best results are found just a hair short of infinity, so we set focus there)

All we have seen so far are the Breakthrough Filters' test where I think they didn't bother to focus the lens? That seems kinda deliberately misleading... but hey, I have done those ads for Best Buy trying to sell HDTVs where we deliberately softened the SD image, so I guess I'm guilty of this too

So yeah, Phil what you are saying is that there is a real difference (in some cases I would guess) visible when viewed at HD or maybe UHD, between a mount that's shimmed correctly and a mount that's a few shims off, even if in both cases the lenses are actually focused on the object perfectly? I am legit super curious! Thank you.

Bruce Allen
www.bruceallen.tv

 

[Phil Holland]

Phil do you have time to post some images?

When I get my next batch of free time I'll showcase some interesting things across some lenses. Bare in mind I joyfully use extension tubes and understand the general perspective here on why in theory this shouldn't matter much, but it was also there where subtle things could be observed. For instance a lens that could not achieve focus with a rather small tube attached required investigation, which also led to some interesting discoveries.

Specifically when it comes to certain lenses (some primes, zooms, most anamorphic lenses) there's many reasons to nail things outside of merely getting to infinity or hitting scales. Key variable is what's going on in the optical path on various cameras as well. Not to mention any hope on achieving parfocal focus on a zoom that can do such things, which is actually the heartbreaker to many post houses when things aren't carefully tended to or looked at.

Outside of this, for what I shoot professionally I'm often dealing with temperature variations that can lead to problems occasionally that need to be dealt with. Which is why I travel with an FDC and a bunch of tools.

 

[Robert Hofmeyr]

Hey Phil. I think your backfocus was out - the image you posted looks pixellated

I'm also really interested to hear how backfocus can influence sharpness (other than not being in focus). On a recent shoot with an EF mount telephoto zoom on an Ursa, I noticed the zoom was way less parfocal than normal. It is not normally perfect, but it is close, but now it was totally out. The next time I mounted the lens, it was back to normal. This has never happened to me before, and I am not sure exactly what caused it, but I guess the lens must have been mounted badly? Anyhow, footage still looks sharp (except for the one zoom I attempted, that looks like crap).

 

[Bruce Allen]

When I get my next batch of free time I'll showcase some interesting things across some lenses. Bare in mind I joyfully use extension tubes and understand the general perspective here on why in theory this shouldn't matter much, but it was also there where subtle things could be observed. For instance a lens that could not achieve focus with a rather small tube attached required investigation, which also led to some interesting discoveries.

Specifically when it comes to certain lenses (some primes, zooms, most anamorphic lenses) there's many reasons to nail things outside of merely getting to infinity or hitting scales. Key variable is what's going on in the optical path on various cameras as well. Not to mention any hope on achieving parfocal focus on a zoom that can do such things, which is actually the heartbreaker to many post houses when things aren't carefully tended to or looked at.

Outside of this, for what I shoot professionally I'm often dealing with temperature variations that can lead to problems occasionally that need to be dealt with. Which is why I travel with an FDC and a bunch of tools.

OK thank you Phil. Leaving extension tubes out of it, can you post an example from a spherical prime on a Komodo for example where this difference is visible? Sorry I know it's a lot of work!

Jacek posted this:
"I think we need to clear up one thing- the typical back focus error has zero effect on a spherical lens optical performance, i.e. resolving power, contrast, field curvature, lateral dispersion, longitudinal dispersion, coma, spherical aberration, field curvature, astigmatism, etc. None whatsoever s far as the film makers are concerned."

And you replied it made you "actually angry".

So that's my question - in what normal filmmaking case with a normal spherical prime is his statement incorrect?

Sorry I'm not an expert, I'm just trying to learn. What Jacek stated is my understanding too, so I want to know if I'm wrong!

And also, that "comparison" Graham was using is not a proper test, right? Like he seemed to be stating that his lens mount was "sharper" but actually he didn't bother to focus the one that needed shimming, right?

Just want to know I'm not crazy. Thank you!

Bruce Allen
www.bruceallen.tv

 

[Robert Hofmeyr]

And also, that "comparison" Graham was using is not a proper test, right? Like he seemed to be stating that his lens mount was "sharper" but actually he didn't bother to focus the one that needed shimming, right?www.bruceallen.tv

I was also confused about this for a while. Graham did say

Objective: shoot hard stop infinity focus between two mounts, compare which is more critically sharp.

But he also said

The result is the camera resolves sharper than you thought possible.

And other images posted in this thread just showed sharp images next to blurry ones with very little info so I got confused. I think it is clear now that the blurry images could have been sharp if the lens had been focussed, but I could be wrong.

We also don't know what lens was used, at what f-stop, or if the images are crops from the originals. In my limited understanding, without this information, the images are quite meaningless: the subjects look like they are all quite close to the camera. If both cameras are focussed at infinity, why wouldn't close objects be out of focus? (Of course this depends on the focal length of the lens, but this was not provided as far as I remember).

Then after many pages of back and forth, it turns out some people believe lenses are less sharp if the back focus is not set correctly, even if they are perfectly in focus. This came as a surprise to me, and others disagree. I'm not sure if Graham meant to claim this, but it certainly was implied. It does seem like a pretty simple thing to test for someone with the right equipment. I am no expert, but I too would love to see some tests that prove this either way.

 

[Jacek Zakowicz]

Of course anamorphic lenses are different and need proper back focus and flange. I've said it about 5 times in this thread.

 

[emil_h]

I gather this bet has gone with the wind lol

 

[J. Orlandi]

I can’t believe this thread is still going... the bet should just be burned and forgotten about at this point.

 

[Bruce Allen]

I actually learned a bunch on this thread listening to everyone!

The one thing that hasn't been cleared up is for Phil to post the images of how a small backfocus error can result in soft images on a normal spherical lens for normal narrative filmmaking purposes?

Phil you're an amazing resource but you're kinda the only expert here defending Graham's claims - I dunno man, don't you think him posting that really soft image from another adapter and that really sharp image from his (but not bothering to focus the one on the other adapter) should be called out rather than boosted by someone like you? The way it is, it seems like you're kinda just covering for him because you're his buddy, or you're getting stuff from him, or he's RED's buddy or something. Which is cool, just, I mean, tell us. I think experts have a moral duty to stand for the truth and the whole truth.

Like the real way to prove it is show:

a) adapter with proper backfocus with lens properly focused
b) adapter with slightly off backfocus not bothering to adjust lens focus
c) adapter with slightly off with lens focus adjusted so it's sharp

Everyone except Phil and Graham are saying that for most cine purposes, a) and c) will be identical, on a normal spherical lens for normal narrative filmmaking.

Graham and Phil are saying a) and c) are not?!

Again I would love to see tests. I'm not an expert, I'd love to learn.

Bruce Allen
www.bruceallen.tv

 

[Blair S. Paulsen]

Back focus backer

Back focus backer

It is my understanding that the difference between compensating for back focus discrepancy with the focus ring and having the back focus dead on is seen in focus falloff characteristics. So in Bruce's example a), you would get the full DoF available for a particular lens at a particular stop vs example c) where your DoF is truncated or shifted vs spec.

I also believe that the corrective elements of some lens designs are compromised when the back focus is not at spec.

As someone who believes in science and math (yes, I know, I'm old school , if I have this wrong, I would welcome being educated on the subject. Until then, I'm going to stick to my belief that accurate back focus is foundational to consistent optical performance.

Cheers - #19

 

[Bruce Allen]

Agree, Blair! Any practical tests you have to share? Again, we're talking where it's a little off, not vastly off.

In terms of actual practical results where backfocus is a little off, I'd love to see an example.

A lot of knowledgeable people said it wouldn't matter for practical cine terms.

Phil said this "made him angry" and said he would provide an example.

He has been supporting Graham, who has only so far posted a very false and misleading comparison which people called him out on. Phil has (I think) defended him. Why, Phil? Like, dude, why are you taking the side of some guy posting a misleading test? Why not take a nuanced position and call out Graham for the misleading marketing but also make your point if you want...

Anyway, let's see Phil or Graham post a proper comparison, right?

BTW... I actually got inspired by this thread to put my money where my mouth is and buy the Hotrod Cameras PL to E mount mk2 (the one with shims you can remove etc). Playing with it on Sony cameras - and the Sigma MC-31 mount on my Panasonic which is also shimmable.

I haven seen no practical difference so far between perfect backfocus and also removing or adding a shim and refocusing, but need more time to test.

If this thread stays open, I'll post my results when I have time. But again, I do hope Graham (since he's into "calling bluffs") or Phil (the only knowledgeable guy who seems to be standing by Graham's really misleading "test" and who is "angry" about people calling Graham out) step up and surprise us because we could learn a lot then.

It's 2021, I'm frikkin' tired of marketing hype and lots of words. I spent a lot of the last decade reading too much Reduser and believing a bunch of hype about RED vs Arri because I believed marketing BS and quasi facts and didn't have the money and time to test enough myself. Now I have the resources, I'm not spending another decade doing that. Let's see some proper tests and proof, right?

Bruce Allen
www.bruceallen.tv