PL or stills glass - my 10 cents

[Ola Christoffersson]

I have had my F3 for little over a week now and I've spent some serious time trying to find a strategy when selecting glass. I have also done some tests comparing PL-glass with still lenses.

If this is bugging you to, you might want to read about my tests and findings here: www.kamrat.tv

[Chuck Fishbein]

Thanks for the comment Ola. I have to agree with you that many of us are coming from different and unique production backgrounds and see the need for top glass on the F3s very differently. For many of us, a set of nikon lenses will be more than sufficient and for those on a commercial set, perhaps unbearable without a set of pl primes.

The great thing is that you don't have to wait. You can get started with what you have, what you're used to or what you can afford until you need and/or can afford something better.

All of the still camera lens shortcomings such as breathing, focus direction, short rotation, and difficulty working with a pro matte box will mean nothing to the next generation wanting to make their first feature, doc or commercial, using a high quality sensor and recording to a decent medium.

Exciting times!

[Matt Davis]

Thanks for publishing your tests, Ola.

I'm assuming your Nikkors are all full-frame 35mm stills glass.

Have you tried any APS-C lenses?

Just thinking of the multitude of Canon 7D video shooters moving their glass to F3 (if/when the Birger Canon to PL adaptor appears).

What I'm getting at is that I have heard that the F3's sensor is a smidgen larger than the Canon APS-C size, so Canon glass for 7D et al would vignette. However - and please correct me if I'm wrong - isn't Nikon APS-C is slightly larger than Canon APS-C too?

Perhaps I should hold onto my Nikkor 17-55/2.8 after all...

[Ola Christoffersson]

You definately should. It will work very well. I have a Sigma 18-50 f2.8 that covers the sensor with no trace of vignetting. The other lenses I have are full frame Nikkors but I think you'll be fine with any APS-C-glass.

[Steve Kalle]

Yes, Canon APS-C sensors are slightly smaller than Nikon/Sony (Sony makes most Nikon's sensors). The Canon APS-C crop factor is 1.6 whereas Nikon/Sony is 1.5.

If you choose to use still lenses, it is best to use Full Frame lenses because you will get the very best performance from the lens. They are always sharpest in the middle and get softer towards the outside; so, with an APS-C sensor, you only use the sharpest part of the lens. Also, CA and distortion tend to be less in the middle.

The two best sites I have found for lens testing are SLRGear.com and photozone.de . Both sites test many full frame lenses with both APS-C and FF sensor cameras, thus, you can see the difference in performance when using a FF lens on a smaller sensor.

[Brian Lai]

Matt, Will the foreseeable Birger F3 to Canon adaptor works for EF lens and being able to control aperture? I have a lot of L series glass from my 5D days that would love to pair up with the F3 - am so desperate that I ordered the present MTF Canon adaptor with intention to shoot those lens wide open and control the aperture with ND's and ISO adjustments on camera.

[Brian Lai]

Ola, thanks for publishing the test. I looked at footages shot last week with a Cooke Panchro 100mm Prime from a TV logic 32" monitor paired up to a FilmMaster color correction suite, and was surprised that a Canon 5D shooting side by side with an EF 70-200 L series 2 lens matches the Alexa's footage quite well, under ideal lighting conditions. The subject was skin and hair, notoriously hard for HD cameras to get pleasing textures.

[Matt Davis]

Originally Posted by Brian Lai

Matt, Will the foreseeable Birger F3 to Canon adaptor works for EF lens and being able to control aperture?

From the sources I have, yes - aperture and focus. Similar to the Red Scarlet demos. You're paying a lot of cash for it, but for that you get to use and accurately control the lens's built in servo motors for iris and focus (not zoom, of course).

I'm still willing to sell my lovely Nikkor to fund more Canon L-series (I think...).

[Steve Kalle]

Originally Posted by Brian Lai

Ola, thanks for publishing the test. I looked at footages shot last week with a Cooke Panchro 100mm Prime from a TV logic 32" monitor paired up to a FilmMaster color correction suite, and was surprised that a Canon 5D shooting side by side with an EF 70-200 L series 2 lens matches the Alexa's footage quite well, under ideal lighting conditions. The subject was skin and hair, notoriously hard for HD cameras to get pleasing textures.

Brian, I have a VERY hard time believing 5D footage would come anywhere close to Alexa footage because the 5D can only capture about a third of the resolution of an Alexa and a 1/4 of the chroma resolution. Somewhere in a recent thread, someone posted a comparison of a 5D and Alexa using a standard color chart to see how far each could be pushed in post. Right from the start, the 5D's color was very different from the Alexa's. For what you saw, the only explanation I can think of is that the Alexa's settings were adjusted to look like a 5D.

Furthermore, a recent class over at FXPHD.com showed the variation in exposures and white balance between a 5D, 7D and Red One MX by shooting a gray card. The 5D looked nothing like the 7D and neither Canons looked anything like the Red. The 5D had more red in its image, the 7D had too much blue and the Red1 was spot on. This test was designed to show how much variation there is in the DSLRs because each camera can see the same image and expose differently.

[Brian Lai]

Steve, I wasn't suggesting the 5D recording a compressed codec can ever match the Alexa recording on 4.4.4 Log C on a real world basis. Since I was shooting with both cameras side by side (my F3 didn't arrive till after the shoot or it would have been an even more interesting comparison), I did lit the scenes to avoid the pitfalls of the 5D compression - very soft light, avoiding bright sharp colors and filling in the blacks, and using different filtration on both cameras in an effort to match the footage - to make the images less digital and more like film. I did not purposely degrade the Alexa's image to match the 5D.

When the Alexa first came out, I did a similar comparison shooting side by side with an Arri 535 on ultra primes and came up with a filter combo for the Alexa which I felt matched the film texture of Kodak 5219 transfer with a Spirit telecine locally, all in the effort to convince die hard film clients to let us go digital.

I was also echoing Ola's test on certain still glasses performing equal to or better than the Zeiss master primes when I saw the Canon zooms' shots looking very admirably next to the Cooke Panchros.

[Brian Lai]

[QUOTE=Steve Kalle "Furthermore, a recent class over at FXPHD.com showed the variation in exposures and white balance between a 5D, 7D and Red One MX by shooting a gray card. The 5D looked nothing like the 7D and neither Canons looked anything like the Red. The 5D had more red in its image, the 7D had too much blue and the Red1 was spot on. This test was designed to show how much variation there is in the DSLRs because each camera can see the same image and expose differently"

I'd be curious to find out whether the test was shot with identical lens as different glass have their inherent coloration and contrast differences.

"The Red1 was spot on" meaning what? Were they judging by RED COLOR which was just a viewable representation of the raw recorded image in RED CODE and by no means definitive to what is being recorded on the sensor? And if you color correct RED CODE, you can make anything perfectly gray including images shot by the 5D & 7D.

The price points differences between a $2K DSLR, a $25K RED and a $60K Alexa make comparing their merits and shortcomings a mood point. At the end of the day, cameras and lenses provide a function to acquire images to your liking. Like a toothbrush that has a function, I'd be more concerned about how I brush my teeth to get the desired results rather than counting the bells and bristles of the toothbrush.

[Jim Tittle]

Brush. Amen!

[Steve Kalle]

Originally Posted by Brian Lai

The price points differences between a $2K DSLR, a $25K RED and a $60K Alexa make comparing their merits and shortcomings a mood point.

Producers, Directors and DoP's need to know the difference so they CAN select the best tool for the job. The people making the choice on what tools are used must know the shortcomings of each tool, and the Canons have some serious issues.

For the color difference, each camera was white balanced; thus, theoretically speaking, there shouldn't be much of a difference. The Red's image had a Rec 709 LUT applied to match the Canons. The class also showed how much more noise is in the image between the first second of recording and 30mins later. In addition, he showed how Canon uses Auto Gain at certain ISO levels to reduce noise.

In the attached image, each camera was set to ISO 320, 1/50, F11 except the 5D was f13. This was shooting a 50% grey card. Another test showed how the 5D's metering can be way off from f1.2 to 5.6 to f16 with nothing else changing but the aperture.

[Brian Lai]

Steve, Point taken, and thanks for posting the test.

Not being anally argumentative here, if this was a controlled experiment, then the lens and F stops used had to be identical. It wouldn't make sense for the RED to use a Zeiss ultra prime lens calibrated in T stops for true light transmission against the Canons using a cheap zoom kit lens, with different colored glasses, and calibrated in mathematical F stops, (not accounting for light loss from its glass elements).

And I wonder why the 5D used F13 instead of F11 for the other cameras? From the picture, the grey cards shot with both the Canons are overexposed against the RED's, so instead of the color shifts, there is also an exposure difference, or differences in sensors sensitivity. During my recent Alexa shoot, our data tech did remind me my 5D's signal was 1/3 stop hotter, observing both cameras through a Leader wave form monitor, so I had to stop the 5D down by one ISO increment to match the footage. I wonder if this was why they used F13 and F11 for the other cameras.

In the real world, production budgets and logistics pretty much dictate what tools we can shoot with. But since we're in a F3 forum, I'm going to convince everyone to shoot with the F3 even if they can afford the Alexa. For the Canons, I'm just happy that a camera costing so little have accomplished so much and changed our industry forever. Cheers.

[Thierry Humeau]

I purchased those great, light and fast SLR lenses to use on the F3...

Sigma 17-50 F2.8
Sigma 30mm F1.4
Sigma 50mm F1.4
Sigma 85mm F1.4
Tokina 11-16 F2.8

I am on a hunt for a great long lenses. Unlike the wides, it does not have to be fast but I need a long throw and a good focus ring. Tried the Nikon AF-S 55-300, focus ring no good. I now have a Sigma 70-300 APO, focus ring is not great but it is usable. I am now looking at Tokina's 80-400. It seems to have a good focus ring.... Anyone has experience with that lens?

Thanks.