Time to start testing Canon lenses?

[Vicente Nogueira]

Now that we will have the EOS Birger mount, will anyone start a consistent Canon Lens testing? I understand that Evin prefers and owns Nikon lenses (and I certainly respect that), and that before the Birger lens mount it was a problem to use modern Canon lenses with Red because of the lack of aperture ring, however this is not the case anymore since we will have the Birger Mount.

There are several great Canon Lenses, such as the 16-35 2.8L II, 24-70 2.8L, and 70-200 2.8L IS, besides several prime lenses that seem like great options for Red. It would be nice if someone would perform consistent Canon tests, prior or during LART, similar to those that Evin performed with Nikon. There are Canon lens tests and rankings available, but I have not seen anything consistent, like Evin did with Nikon lenses.

I understand the issue of Canon CMOS sensor size vs. megapixel being different from Red, but it would be nice to have a clear comparison of the best available Canon lenses. Perhaps the tests could use the 12mp crop of the 16.7mp 1Ds MK II sensor image, would that be similar to the Red sensor size?

[Evan Owen]

I understand the issue of Canon CMOS sensor size vs. megapixel being different from Red, but it would be nice to have a clear comparison of the best available Canon lenses. Perhaps the tests could use the 12mp crop of the 16.7mp 1Ds MK II sensor image, would that be similar to the Red sensor size?

Actually, I believe the 1.6x FOV (compared to 35mm) of the EOS 30D (8MP) will be almost exactly the same as 4K footage from RED, with no cropping in post. Should make for easy comparisons.

Anyone else want to confirm? This is just what I remember from seeing comparisons of frame sizes and field of view...

[Vicente Nogueira]

However, the resolution would be much lower at 8mp. The Canon 400D offers 10mp at 1.6x FOV and the 1DIII an excellent 10mp at 1.3x FOV. In order to get a cropped 12mp frame I think the only option is the 1DsII, but I don't know the crop factor for that.

[Evan Owen]

However, the resolution would be much lower at 8mp. The Canon 400D offers 10mp at 1.6x FOV and the 1DIII an excellent 10mp at 1.3x FOV. In order to get a cropped 12mp frame I think the only option is the 1DsII, but I don't know the crop factor for that.

4K 16x9 from RED is 4096px x 2304px, or about 9.5MP–not that much greater than 8.2MP.

I'm wondering which would be more important to have similar to the RED when testing these lenses, field of view or resolution... ?

[EDIT]
A 16x9 crop from the 30D would yield a 3,504px x 1,972px image or about 6.9MP, so it is quite a bit lower resolution, but the same FOV–I think. You may be on to something with the crop from a 1Ds Mark II, but I haven't done the math.

[Evin Grant]

I'm totally planning on testing EOS lenses but not until my Red arrives in October. There just isn't a DSLR from Canon that has a pixel density in the right amount in an APS-C format chip. The Rebel 400D comes closest but it's still not going to do justice in a lens review. Maybe the new 40D will be 12MP or higher but by the time that ships I'll probably be cradelling #199.

[Vicente Nogueira]

Great! I am very happy to hear that you will test EOS lenses.

[Eddie]

I'm totally planning on testing EOS lenses but not until my Red arrives in October. There just isn't a DSLR from Canon that has a pixel density in the right amount in an APS-C format chip. The Rebel 400D comes closest but it's still not going to do justice in a lens review. Maybe the new 40D will be 12MP or higher but by the time that ships I'll probably be cradelling #199.

Evin you have used this argument about the sensor size of the d2x before to kind of dissuade people from posting there own testings of very different lenses. And I dont really get it... Sensor size is just one spec, and you are cropping to 16:9, and using a completely different codec than the redcode, anyway, so what is it, that make the d2x sooo important? The fuji s5 has a dynamic range that fits Red one much better than a d2x, so why not make that the criteria? or something else, like noise at high iso, that should put the canons on the map?
Questions like: Does it breath? how is the bokeh?, how does IS or VR affect the still camera when it is recording in video mode. focus throw etc. are all relevant and can be tested in spite of minor differences in sensor size.
In a way I can understand, that you want a kind of consistency, but then again, why hold off?
As much as I have enjoyed your reviews, I still don´t see any of the lenses which people are most likely to buy. Namely 17-55mm DX, 70-200mm VR dx zeiss zf canon, 17-55mm IS USM the L series, Leica summicron etc.
So why not use the whole community, and let us have it all out in the open and see what happens... :biggrin:

[Evan Owen]

... Sensor size is just one spec, and you are cropping to 16:9, and using a completely different codec than the redcode, anyway, so what is it, that make the d2x sooo important? The fuji s5 has a dynamic range that fits Red one much better than a d2x, so why not make that the criteria? or something else, like noise at high iso, that should put the canons on the map? ...

Because sensor size is a very important spec. The cropping of smaller sensors effect things like field of view, edge sharpness and vignetting–all important when picking a lens. Dynamic range and noise have nothing to do with lenses... he picked the D2x for a reason.

[EDIT] I agree that now that we have a way to use electronic lenses such as the 70-200 f/2.8 VR (with the Birger mount), it might be nice to see some tests. Like Evin mentioned though, we might as well wait to try Canon lenses until we have RED and the mount. It won't be long now, and it only makes sense... to me anyway.

[Patrick Tresch]

Canon lenses are also interesting for their optically stabilized technology, as on XL-H1(base lens) and still photography, this device can be a great help...

I know that stabilizer are not used in 35mm cinematography, but after shooting hand held a documentry for 1 year with this type of optical tabilizer, I can apreciate the advantage depending on the situation.

[Eddie]

I am sorry if I got it all wrong... But if you use a camera with a bigger sensor than red, wouldn´t that be an even better indication of problems with vignetting and edge sharpness? In a way dynamic range is also telling something about the lens, namely how its going to look, when you put it on a Red One. I mean if you took a camera with 5 stops of DR, then it would be harder to imagine what kind of image it will deliver on red, contrastwise etc. wouldn´t it?
Now, about FOV, i don´t know how far away evin is from his camera, so it is a bit hard for me to judge the fov?
Btw, I am not trying to upset anyone at all... I have been very grateful for the lense database. It would just be sad if it kept a sort of "nikonian" feel throughout.

all the best eddie

[Evan Owen]

I am sorry if I got it all wrong... But if you use a camera with a bigger sensor than red, wouldn´t that be an even better indication of problems with vignetting and edge sharpness? ...

I understand exactly what you're saying and the point is well taken.

The way I see it is that we're looking for lenses for a specific camera. Every lens vignettes, every lens loses sharpness towards the edge. But not every lens do these things the same way, or to the same degree. A lens that exhibits unacceptable vignetting on a full-frame sensor camera might work fine on a APS-C sensor camera like the EOS 20D – or RED. Yes, the color will be a little different, the dynamic range might not be exactly the same, but those are pretty minor issues that can usually be fixed in post. Field of view, sharpness and vignetting are a little harder to correct.

I think expanding the database to include Canon lenses is a great idea. Until recently it didn't make sense because of the lack of good quality manual Canon lenses. Now that we can use electronic lenses, the options double! Maybe someone with access to some of these lenses would like to do some tests?

[EDIT] Just to clarify, although I don't own any glass right now, historically I've been a Canon shooter. To hear that we can use L glass is extremely exciting to me...

[Evin Grant]

I am sorry if I got it all wrong... But if you use a camera with a bigger sensor than red, wouldn´t that be an even better indication of problems with vignetting and edge sharpness?
No, because they would not reflect what you could expect from your Red.


In a way dynamic range is also telling something about the lens, namely how its going to look, when you put it on a Red One. I mean if you took a camera with 5 stops of DR, then it would be harder to imagine what kind of image it will deliver on red, contrastwise etc. wouldn´t it?

That is not the point of the review, but more precicely it's not relevant because the D2x does have about 10 stops of DR so it's geting pretty close.



Now, about FOV, i don´t know how far away evin is from his camera, so it is a bit hard for me to judge the fov?
Btw, I am not trying to upset anyone at all... I have been very grateful for the lense database. It would just be sad if it kept a sort of "nikonian" feel throughout. all the best eddie

I have set the shots up to reflect the most often used angle of the lens tested, if it's long I do a close up if it's medium I do a medium, I don't really have the space for a true wide so if I have a wide lens I do cowboys.
The point being is that the tests are supposed to give you an reasonable expectation of the performance a lens can give you in the situation that you are most likely to use it in. As for the Nikonian feel, well I am a Nikonian and I've never tried to hide that. Also bear in mind that no one pays me for these tests, no one sends me free lenses and even if there were an EOS body capable of substituting for the Red no one has provided me a Canon body to do testing with. I am not against Canon but I won't be buying an entire new camera system just to test before my Red arrives. As to why I've discouraged others from doing testing, I haven't. I just won't be making stickies about reviews that I can't verify the testing techniques.
This sub forum does have my name on it you know. Once LART hits we will be testing a ton of glass and since there is no Birger Nikon mount yet, there will undoubtedly be quite alot of EOS glass at that test.

[PaulClements]

There just isn't a DSLR from Canon that has a pixel density in the right amount in an APS-C format chip. The Rebel 400D comes closest but it's still not going to do justice in a lens review.

The 400D has exactly the same width sensor as Red's at 35mm 4k (22.2mm). It is slightly higher at 14.8 compared to Red's 12.49, which means that an image shot on the 300D requires a vertical crop of 405px.

Whilst the resolution would be slightly less the sensor would actually give a very good indication of what the red will see from the lens.

If you resize the images to 2048 x 1152 having cropped vertically it would give pretty much exactly what you would see from Red when 2k.

[Michael Hastings]

I think you may be forgetting that Evin does this for free as a sharing thing with the RED community. I think if we were to all get together and buy an XTi, a dozen EOS lenses, and the others you mentioned plus pay his bills for a couple months, then maybe he could accelerate testing. (edit: I see evin just checked in while I was writing so he covered some of this himself.)

And his point about the sensor is that he owns a camera with a sensor directly comparable. I am scheduled for RED 206 delivery in Oct. I will be buying a 10.2MP XTi and for my purposes it will be close enough (I will use it as a directors finder and light meter, as well as location stills) but I won't be representing myself as an objective tester. There have been other people on Reduser posting their reviews of EOS lenses, but they are of limited usefulness because we don't know the testing parameters.

Evin has been trying to keep things as objective as possible and I salute that. He has also provided a valuable preliminary look at the subject, opened many of our eyes to the possibilities with still lenses.

Also remember that he initially started out with the idea of using Nikon manual lenses with the RED Nikon F mount as the only real low cost alternative to expensive PL lenses. Jim changed things with his announcement of a low cost set of RED cine primes, the new red zooms, etc. Then the birger mount for EOS became a reality - ie. announced as an official RED association - (and since birger has been making an eos mount for several years it was pretty clear that it was for real), followed by the announcement that a birger Nikon mount was in the works.

So the test universe grew exponentially. At this point it makes sense to start looking at testing on RED since deliveries are now with 2 weeks. At that point we can movies showing breathing, vignetting etc. as well as sharpness.


Evin you have used this argument about the sensor size of the d2x before to kind of dissuade people from posting there own testings of very different lenses. And I dont really get it... Sensor size is just one spec, and you are cropping to 16:9, and using a completely different codec than the redcode, anyway, so what is it, that make the d2x sooo important? The fuji s5 has a dynamic range that fits Red one much better than a d2x, so why not make that the criteria? or something else, like noise at high iso, that should put the canons on the map? :biggrin:

[Eddie]

Everybody loves Evin... me too :love: That is not the issue.
All I am saying is, that with all the lens options now available, people should not be discouraged from posting their homemade testings just because the sensor size isn´t an exact match.
If you are testing on a bigger sensor, without signs of vignetting or poor sharpness at the edges, that should suffice.
But you are right aquavideo, that the lens options are exploding, that is one reason it would be better to encourage people to post their findings, even though they are testing on a bigger sensor. Lets say someone is sitting on a bunch of old Zeiss contax glass and an EOS adapter, he makes a trustworthy and thourough review, and advise people on the pros and cons... Why should this review be inferior to a review made on the d2x? And if it is a great review, showing a way to get cheap zeiss primes, why not make it a sticky for everyone to see? now all they see is the "why not canon EOS"- sticky... When Evin began his testings, he stated that he would welcome criticisms and suggestions from the community, and it is only in that spirit I have made my points. I would likely defend Evin on many other points just not on this one

[planet e]

i dunno. i've been very appreciative of evin's testing, mostly because i am ignorant about nikon glass. so it has been a nice education. i even gave some thought to switching. but not since the birger mount was announced. canon glass i know quite well. i have a 70-200mm f. 2.8, the 16-35 f2.8, the 70-200mm f2.8, the 200mm 1.8, and a 400mm 5.6. also a really amazing sigma 300-800mm f.5.6, which is a damn fine piece of glass, looking forward to the moon in 4K....

and my opinion is that, if you're hungry for lens tests and reviews, check out fredmiranda.com...a lot of the outcomes there mirror what evin has done here, so if you're just looking for general opinions about lenses worth owning, which is all that we can really formulate until we have a specific lens mounted on a RED, then that's a good place to look.

my big concerns, having mounted all of these lenses at one time or another onto other video cameras, are CA and breathing. and we can't really adequately test for those issues until we have the camera in hand.

if anyone with a RED reservation earlier than mine wants to borrow my lenses to test them, i'm willing to lend them out. i offered them to LART but nothing came of it, and now LART is pushed out until october, and i would really like to obtain some first-hand information on how these lenses will work on a RED in advance of my own purchase. it would be worth the exchange, and we could all learn a lot.

[Michael Hastings]

Interestingly CA and breathing are on the way to being handled. The birger mount is capable of precise feedback as to where the zoom and focus are so they are working on breathing compensation in conjunction with the zoom motors. CA is already pretty well handled in the still world with post software and all it will take is a Graeme Nattress-type genius to adapt that technology to motion pictures - maybe a little time consuming but capable of near perfection.

I'm red #206 with delivery scheduled for Oct 10th. I am very interested in testing your lenses. We make underwater housings so will definitely be setting up for the birger mount and the 10-22mm EF-s but I would like to try the others before finalizing my other purchases. We will have an early birger mount and I may have access to an even earlier camera for some testing as well. BTW breathing could also be handled via software as well.

Feel free to PM with your phone and address.


i dunno. i've been very appreciative of evin's testing, mostly because i am ignorant about nikon glass. so it has been a nice education. i even gave some thought to switching. but not since the birger mount was announced. canon glass i know quite well. i have a 70-200mm f. 2.8, the 16-35 f2.8, the 70-200mm f2.8, the 200mm 1.8, and a 400mm 5.6. also a really amazing sigma 300-800mm f.5.6, which is a damn fine piece of glass, looking forward to the moon in 4K....

and my opinion is that, if you're hungry for lens tests and reviews, check out fredmiranda.com...a lot of the outcomes there mirror what evin has done here, so if you're just looking for general opinions about lenses worth owning, which is all that we can really formulate until we have a specific lens mounted on a RED, then that's a good place to look.

my big concerns, having mounted all of these lenses at one time or another onto other video cameras, are CA and breathing. and we can't really adequately test for those issues until we have the camera in hand.

if anyone with a RED reservation earlier than mine wants to borrow my lenses to test them, i'm willing to lend them out. i offered them to LART but nothing came of it, and now LART is pushed out until october, and i would really like to obtain some first-hand information on how these lenses will work on a RED in advance of my own purchase. it would be worth the exchange, and we could all learn a lot.

[Vicente Nogueira]

Canon just announced their 1Ds Mk.III full frame camera with 21.1MP and 14-bit RAW. They also announced a new 14mm f2.8L mk.II. Considering the new 21.1MP and that they recently announced a 16-35mm f2.8L mk.II, they seem to be upgrading their top L lenses to catch up with the higher MP of their professonal cameras. This is very good news in my oppinion for those with Canon systems.