SALT III- High Speed Lens Shootout "WFO" Results

[Matt Uhry]

Yes, and look at the Cine lenses section of this (http://www.lensrentals.com/blog/2012/01/the-great-50mm-shootout) to see why.

The premium people pay for longer focus throw and correct distance marks with optical performance the same (at best) as 1/10th the price stills lenses...boy I don't know. I've gotten shouted down pretty bad for saying that on this forum, but there are the numbers in black and white.

HI Lee,

No Shouting down, I own or co-own about 20 PL lenses and about as many in Stills mounts and use whichever is appropriate for the job at hand. Stills lenses are a great value when compared to Cine glass, it's just way harder to put their resolution to use by keeping good focus in typical cine situations where the camera and / or subject are moving. This limitation has created it's own aesthetic - small budget, big sensor, just look on vimeo, there are countless examples.

In the future there will be nifty and hopefully reliable ways to use the built in AF motors with a remote focus knob, this will be pretty cool, and it will probably make still lenses a more viable choice, but you'll still need a good focus puller and they are not easy to come by.

It's a big mistake in my opinion to pick a lens on maximum resolution numbers alone. The camera cannot use the excess resolution beyond the photo site size - on the Epic I think this works out to be about 70lp/mm. Once you've achieved that # low of flare and good microcontrast are much more relavant to sharp looking images, and that's poorly expressed in the MTF #'s from the LensRentals 50mm test.



Matt Uhry
www.mattuhry.com

 

[Bill Anderson]

Now that sensor sizes threaten to push beyond the industry standard lens coverage and possibly way beyond, it's going to be interesting to see what lens manufacturers come up with. On the cine camera front, I wonder who will be the first brave soul to push sensors into that uncharted territory. Or will we always hover around what is best known?

 

[Jiri Vrozina]

Surely not to come in Jarred's defense as He sure needs none, but I was in fact surprise that no one form RED had yet to respond and put you in your place for your third comment, which was completely out of place!

You can't expect responses from the RED team at your own time, you should ask a question, or two, and then patiently wait to see whom and when they can answered if some one does in fact fill that it is a question they wish to answer.


But it is also offensive to the rest of us when you say things like "unless if you are "Raving Fan-Boy" , your tech questions are Ignored".

Some moNths ago i respectfully asked RED if there would be RED pro 17-50 zoom available For Sale in Future. No answer. Last week i asked twice regarding coverage of certain lenses for new sensor. No answer. What is a point of RED team not responding to very importat question like that? After all, many people(companies) have invested is some decent glass? Like i said before, simple answer " I do not know, i can not tell, i will let you know in due time" would be fairly intelligent think to say in 2013...

Finally, I must admit that 3rd question was out of place and for that i owe an apology...

 

[Will Keir]

It's good to hear people are watching the entire 45 minute video! hahahaha You must really love your lenses.

Give Matt Hayslett all the credit on the music. Somewhere in the madness of four solid days of editing, Muse must have saved his sanity.

Thanks for finding the time and resources to do this guys! Nice music selection as well, made the 45 minutes enjoyable.

 

[Matt Ryan]

It's good to hear people are watching the entire 45 minute video! hahahaha You must really love your lenses.

Give Matt Hayslett all the credit on the music. Somewhere in the madness of four solid days of editing, Muse must have saved his sanity.

haha, they are my employers and I like their music so it felt right.

 

[Timur Civan]

Yes, and look at the Cine lenses section of this (http://www.lensrentals.com/blog/2012/01/the-great-50mm-shootout) to see why.

The premium people pay for longer focus throw and correct distance marks with optical performance the same (at best) as 1/10th the price stills lenses...boy I don't know. I've gotten shouted down pretty bad for saying that on this forum, but there are the numbers in black and white.

its as simple as time savings. my 90mm vivitar is far sharper and has better boken than any PL lens ive used, but its a nightmare to work with.

the few seconds of time saved with each lens change with physically matched PL lenses adds up. over the course of a day, can mean a few minutes, and thats the difference between getting one more take or not. all high end gear is designed to save those few seconds. collectively it means more shooting time perday. when you are spending 50-300K per day, those minutes are very valuable. its really just about time.

 

[Les Dittert]

Like Phil said, thats what the projection test was for. We didn't want to do the standard resolution chart shots, that gets really old. That's not what we wanted for this test. So we gave you both a practical shot, a beauty shot, and results from the projector tests.

The projector test results in the center seemed to only find two results from all the lens tests: 140 and 200 lp/mm . How is that possible ?

To de-emphasize everything but the center is like testing a sports car only on a straight road. Tacking a couple of charts in the near corners would have been nice.

But I am coming to an understanding of the multi leveled desirability of fancy 'classic' glass. To use my car metaphor again, you don't take a Bently on the race track either.
It's all about pedigree, build quality, durability, creamy and organic optical qualities. Sure some of these qualities are a side effect of less that stellar scientific optical performance.
Young filmmakers especially seek to mimic classical masters. Enter shallow DOF , LOMO , forced vignetting, introducing anamorphic flares etc.
The dilemma here on a 4K camera site is that many of the looks created also come with much of the frame imaged at 2K or worse.
I'm glad that Red users can attain images either which way!

 

[Ryan Patrick O'Hara]

Almost all lenses are optimized for F5.6, this is known as a standard bell curve. The only exception are the Leica's which are designed for best performance between f1.4-2.8.

That is a generalization of how all photography lens optics generally work, and that last part of the statement in my experience is not true as Leica's are not an exception. The summilux C's at 1.4 are not optimal over 5.6. Not sure who told you that, but yeah. If clinical sharpness is the goal, along with even field illumination and other attributes, like CA... you'll find Leica's will do better at T/5.6 versus T/1.4, at least that is my finding. Most nice cinema lenses hit their stride early, and you'll typically see 2.8-8 be very close in performance, although that bell curve does exist. But really I only notice the small improvements past 2.8/4 on projectors, with of course a decrease in performance on very closed stops. My .02$

 

[Ryan Patrick O'Hara]

Yes, and look at the Cine lenses section of this (http://www.lensrentals.com/blog/2012/01/the-great-50mm-shootout) to see why.

The premium people pay for longer focus throw and correct distance marks with optical performance the same (at best) as 1/10th the price stills lenses...boy I don't know. I've gotten shouted down pretty bad for saying that on this forum, but there are the numbers in black and white.

\\

There are more advantages to Cinema lenses ( when using them for cinema purposes) than you quickly mentioned: http://www.dvxuser.com/V6/showthread.php?273326-Article-Why-We-Need-Cinema-Lenses

I also must add, when it comes to comparing a nice photography lens to a cinema lens, the only rather consistent category the photography lens will win (besides size) is center sharpness. That's it. So this test, only taking center sharpness into account, was a very strong proponent for photography lenses. I do give the writer of the test credit for being very even keeled. I did not feel a big bias, and I think it was well done.

That said, I think sharpness is a conquered frontier, at least for now. I mean, the center sharpness they were measuring is beyond necessary on digital cinema cameras, so who cares *that* much?! Sure we want sharp, but there is a limit a sensor can detect, and cinema sensors are nowhere near reaching the limits of most cine lenses and I don't think will be for some time. As of now and the foreseeable future, digital cinema isn't crossing over 25-30-35mp just yet and I think we're ok. I'd rather have a car that had good handling, comfort, brakes, shocks, air conditioning, safety, and a decent horsepower, than a car that was worse in all categories but had killer horsepower. When digital cinema cameras reach 20, 30, 35+mp, maybe this will matter, and considering the screen size to viewing distance ratio needed for a 30mp image to take full hold in theater form and considering people are thinking lenses are too sharp now (BPM's, vintage glass, and other things being done today) makes me think out resolving a digital cinema sensor chip by many fold now is ok.

I think photography lenses have a very good and useful place in photography, and some uses in cinematography, but I also think rating or judging any lens based off of center sharpness alone is really missing the whole picture.... literally.

Not saying you were, but just saying it's just one of the many things that make an image. Just as resolution is just one attribute of a sensor.

 

[Clayton Burkhart]

That is a generalization of how all photography lens optics generally work, and that last part of the statement in my experience is not true as Leica's are not an exception. The summilux C's at 1.4 are not optimal over 5.6. Not sure who told you that, but yeah. If clinical sharpness is the goal, along with even field illumination and other attributes, like CA... you'll find Leica's will do better at T/5.6 versus T/1.4, at least that is my finding. Most nice cinema lenses hit their stride early, and you'll typically see 2.8-8 be very close in performance, although that bell curve does exist. But really I only notice the small improvements past 2.8/4 on projectors, with of course a decrease in performance on very closed stops. My .02$

I have seen the charts and their performance on the bench (Leica C) and the difference is minimal between 2.8 and 5.6. Their curve really does not look like the rest. In addition they are sharp across the whole field even close to wide open, with less CA. I have never seen any other lens set perform like they do. In addition a rep also told me, that their objective was to maximize the range between 1.4 and 2.8. He actually said to me that if one was shooting beyond 2.8 they might as well get something else! However, I want to emphasize, that this does not mean they are sharper at 1.4 than 5.6, it means that the curve starts higher and moves very steeply upwards to 2.8 with only a gentle gain to 5.6. Most lenses move much more gently to 5.6 and descend again, hence the bell metaphor.

 

[Nick Morrison]

its as simple as time savings. my 90mm vivitar is far sharper and has better boken than any PL lens ive used, but its a nightmare to work with.

the few seconds of time saved with each lens change with physically matched PL lenses adds up. over the course of a day, can mean a few minutes, and thats the difference between getting one more take or not. all high end gear is designed to save those few seconds. collectively it means more shooting time perday. when you are spending 50-300K per day, those minutes are very valuable. its really just about time.

The mechanics of cine lenses do save time, that is true. However if you plan ahead, it's possible to make stills lenses as cine-friendly as possible, and limit some of the time wasting you're talking about.

For example, I'm having my main personal Contax set MODDED at RPLENS, as we speak. Not only is Peter adding 80mm fronts and declicking every lens, he is also MAKING SURE (at my request) that all the FOCUS GEARS are in the same place (so they work as a true SET). This way my AC's are saving time, and can swap lenses faster.

Just sayin'.

That being said, they don't have accurate focus marks (etc), so I'm not suggesting they are true replacement for cine lenses, they ain't.


SPEAKING OF FOCUS MARKS: I have a question for you all.

Have you guys noticed that your marks for S35 lenses are "off" when you shoot 5K on Epic? Because the FOV gets wider than standard S35? Just occurred to me this may be an interesting quirk that comes w/ shooting on sensors that extend beyond S35. Will the marks still work?

 

[Ryan Patrick O'Hara]

The mechanics of cine lenses do save time, that is true. However if you plan ahead, it's possible to make stills lenses as cine-friendly as possible, and limit some of the time wasting you're talking about.

For example, I'm having my main personal Contax set MODDED at RPLENS, as we speak. Not only is Peter adding 80mm fronts and declicking every lens, he is also MAKING SURE (at my request) that all the FOCUS GEARS are in the same place (so they work as a true SET). This way my AC's are saving time, and can swap lenses faster.

Just sayin'.

That being said, they don't have accurate focus marks (etc), so I'm not suggesting they are true replacement for cine lenses, they ain't.


SPEAKING OF FOCUS MARKS: I have a question for you all.

Have you guys noticed that your marks for S35 lenses are "off" when you shoot 5K on Epic? Because the FOV gets wider than standard S35? Just occurred to me this may be an interesting quirk that comes w/ shooting on sensors that extend beyond S35. Will the marks still work?

What in the what? Nothing changes at all.

 

[Nick Morrison]

What in the what? Nothing changes at all.

Haha. OK. I figured I was probably wrong. Thanks!

 

[Will Keir]

Les,

The projector test we used center and s35 edge. That means every test used the center then outside on the edges out to the s35 boundaries. Hope that helps.

The projector test results in the center seemed to only find two results from all the lens tests: 140 and 200 lp/mm . How is that possible ?

 

[Charles Pickel]

Effect of Horizontal Angle of View

Effect of Horizontal Angle of View

Haha. OK. I figured I was probably wrong. Thanks!

You do bring up a sometimes overlooked point. Although sensor size will not have an effect on the accuracy of focus scales, it will have an effect on the relative nature of focal length. That is, the rule of thumb for a "normal" focal length is that it is approximately equivalent to the diagonal of the format. Hence the "normal" lens for ff35 (24 x 36) is a considered to be 50mm (ff35 diagonal is actually 43.266mm). As you increase sensor size and/or horizontal angle of view, the "normal" focal length will increase proportionally. So that classic human-eye POV 40mm lens you used in S35/4k starts to look a bit wide-ish in 5K 15x30. As sensor sizes continue to bounce around (exactly HOW big is that Black Magic sensor again ?), the cinematographer has to make adjustments.

The new standard is there are no standards.

 

[Lee Saxon]

its as simple as time savings. my 90mm vivitar is far sharper and has better boken than any PL lens ive used, but its a nightmare to work with.

the few seconds of time saved with each lens change with physically matched PL lenses adds up. over the course of a day, can mean a few minutes, and thats the difference between getting one more take or not. all high end gear is designed to save those few seconds. collectively it means more shooting time perday. when you are spending 50-300K per day, those minutes are very valuable. its really just about time.

This argument absolutely drives me crazy.

First, let me clarify that I'm not talking about real handmade cine lenses like Summliux-C's. Now:

If you're willing to pay 10-15 times more for some convenience (convenience which even you make sound minor), that's your business.

But that doesn't mean it's ethical for them to charge that. The expense of providing those conveniences is not that large. Larger, matched-diameter housings? A longer focus throw? Actually measuring where the distance marks are supposed to go, which they ought to be doing anyway? I'll eat my hat if a CP.2 costs more than 5-10% more to build than ZF/ZE (and I bet 90% of that is the different diaphram). Letting Zeiss pass that expense on to you at an absolutely outrageous mark-up just because you can afford it screws over everyone who isn't so lucky.

 

[Ryan Patrick O'Hara]

You should learn how a cinema lens is made and how an average assembly line photography lens is made. Of course this varies depending on what you compare. It's not about mechanics , as it is about precision, tolerances, materials and quality. It also has a little to do with supply, demand and overhead.


If you don't like it, don't buy it. But you should educate yourself before ranting. It's one thing to understand and still disagree, it's another to be ignorant. I don't know about cp.2's, but an assembly line generic photo lens and a Cooke cinema lens go through a very different birth canal, and the materials, time, and service costs reflect that.

Costs are exponential in a way, and that leads to frustrating costs for many. As you increase up the chain, so very often many people find the increase in quality is not worth the increase in price... However some productions do and for very good reasons. Just because it's not you, doesn't make them ludicrous. For reasons you may not know about or dismiss as trivial, these lenses provide the appropriate productions with a service they are very willing to pay for and have paid for throughout the history of cinema.

I like to remind everyone that cinema lenses are not always the best choice for every video production. In fact, depending on the circumstances, they can be quite wrong. And we should acknowledge that fact and how many types of videography can benefit from other types of lenses. It's a per production basis.

It's not a reason to hate nor is it an OPEC style price gauge. People pay premium for anything hand built with utmost precision by a specialist, especially when these specialist workers are typically from Germany, France, Japan, England, etc.... Where labour rates are higher, than say a low-wage non specialist Chinese assembly line worker turning out masses of photography lenses everyday. Now this is not to say those lenses are crap, it's more about extreme QC, tolerances and precision, along with the premi quality in design, features, coatings and more.

So again, the price to performance ratio does get very aggressive toward the top, and this certainly does not lend itself to be affordable to many and certainly not all, however just because you are not one of those productions who value paying for such lenses does not negate their validation as a worth while tool.

This is why there are options, and I guess it is too bad that the most worthwhile cinema lenses are the ones with complicated sophisticated designs, built out of the highest quality materials, hand built/tested/rebuilt/tested by specialist lens makers and in low numbers. Supply and demand is sometimes a cruel mistress. But that is a capitalist market.

If you'd like to know more about the production of cinema lenses, you may go to Jon fauer's Film and Digital Times magazine and look upa Cooke book factory tour, Zeiss factory tour or Leica factory tour. They discuss the process and you can see the difference. If that difference is not worth the price to you, than by all means there are options from a 25,000 prime, to a $30 prime. Find what suits you.

 

[Evin Grant]

Sorry for the tardy post guys, have a newborn in the house and time is not in great supply.
Here is my take on the test...

http://www.reduser.net/forum/showth...quot-Results&p=1147835&viewfull=1#post1147835

 

[PatrickFaith]

Have been spending hours upon hours, trying to correct some lens distortions on a heavy vfx shoot (using the LensDistortion node in nukex), will probably need to actually get the lenses and do test shots with the lens on a checkboard grid(I'm lucky that I can actually get the lens back to do the lens distortion charts).

What I noticed from the SALT III test, is that in all cases the higher cost lenses had less lens distortion problems. I could easily see someone spending $100k in vfx hours on a movie trying to undistort lenses for match merges that need camera solves(I'm finding this very important for 3d mesh generation and deep pixel operations ... the meshes turn out curvey on flat surfaces without running the lens distortion nodes). Just another thing to think about if your doing a lens test for a film and there will be a decent amount of vfx.

 

[Phil Holland]

Have been spending hours upon hours, trying to correct some lens distortions on a heavy vfx shoot (using the LensDistortion node in nukex), will probably need to actually get the lenses and do test shots with the lens on a checkboard grid(I'm lucky that I can actually get the lens back to do the lens distortion charts).

Interesting. I rarely do that at the tail of a production. Usually either before production starts or during the first couple weeks of filming we shoot checker boards and create our distortion geometry for either correction or addition. It's a big time saver getting this done at the beginning of a project as CG elements are created during the production these days. As approved elements come in and surveying/tracking data is culled or created we can get right to work often before the production even wraps.

I've certainly seen variation across primes from different sets, usually slight though. The worst if the set data collector labels a lens as "zoom" instead of marking the actual focal length used on that zoom. iPads and tablets have greatly enhanced that on set workflow because they now have a "checklist" for each shot to go through.