Canon L Series Glass at 2K

[Peter Cote]

Has anyone tested Canon L series glass at 2K? My Canon mount will be here (hopefully) in a few weeks. I am going to be shooting a lot of 300 FPS footage for a project. I did a test using a Duclos 11-16 and felt the lens could be a little sharper at 2K and I wanted to know how Canon L series optics might hold up at 2K. I will primarily be using the 24-70 and 70-200.

 

[Terry VerHaar]

I keep seeing these questions about sharpness at 2K and I am wondering if I am missing something...

Since 2K is windowed on the sensor, why would sharpness be any different than at any other (higher) resolution? You are using the same section of glass that you would be using if shooting a full sensor at 5K or at 4K or whatever. Help me out here.

 

[IAN SUN]

I keep seeing these questions about sharpness at 2K and I am wondering if I am missing something...

Since 2K is windowed on the sensor, why would sharpness be any different than at any other (higher) resolution? You are using the same section of glass that you would be using if shooting a full sensor at 5K or at 4K or whatever. Help me out here.

5K or 4K acquisition output at 1080 is going to be sharper than 2K acquired and output to 1080 (all other things being the same) due to the massive over sample. The concern then, is that if 2K acquisition is relatively softer, you want to maximize sharpness in the other parts of your imaging chain. If you are shooting at 2K you want to make sure your lens is nice and sharp. It is a bigger concern than when you are shooting 4/5K.

 

[kevin holmgren]

I know this seems like an obvious comment, but in Photoshop, if you have a high rez image and down rez, SOMETIMES it actually looks worse than just starting with an lower rez image. Does this not happen with red code?

 

[Peter Cote]

I know this seems like an obvious comment, but in Photoshop, if you have a high rez image and down rez, SOMETIMES it actually looks worse than just starting with an lower rez image. Does this not happen with red code?

no.

 

[Mark Andersen]

The smaller the capture resolution the better the lenses need to be. Its true , over sampling can hid the sins of a less than perfect lens. When you shoot 2K you aren't getting much data for the over sample so the lens needs to be spot on, as well as focus. I recently shot some R1 2K with the RPP lenses and it looked quite good in 1080. The RPPs a really sharp, so is the RPZ 17-50

 

[Mark Andersen]

To answer, the 70-200 mark 2 is really good. I haven't had much luck with my 24-70 but it a older model, maybe they are better now. I actually like my 24-105 better, but it's only F4. The primes are pretty good 35 1.4 is awesome, 24 1.4 is really good too. All the long L glass is really nice, love my 300 2.8 and 600 4.

 

[Peter Cote]

To answer, the 70-200 mark 2 is really good. I haven't had much luck with my 24-70 but it a older model, maybe they are better now. I actually like my 24-105 better, but it's only F4. The primes are pretty good 35 1.4 is awesome, 24 1.4 is really good too. All the long L glass is really nice, love my 300 2.8 and 600 4.

Thanks for sharing. Just to verify, these were shots at 300 FPS? I would consider the 24-105 but since I am shooting high frame rate I need all the light that I can get. F/4 might get me in trouble.

 

[Dan Kanes]

Canon 10-22 or Tokina 11-16 for 2k man.

One word of personal advice: debayer your stuff to 2048x854 (native res) when outputting - then upscale/crop/smoosh whatever you need to do in the NLE or FX program of your choosing.

 

[Peter Cote]

Canon 10-22 or Tokina 11-16 for 2k man.

One word of personal advice: debayer your stuff to 2048x854 (native res) when outputting - then upscale/crop/smoosh whatever you need to do in the NLE or FX program of your choosing.

No the Tokina 11-16 is the same lens at the Duclos 11-16 except modded for cinema and that lens is even sharper than the Canon 10-22. EF-S optics are not built to be pushed to those limits. For really wide shots at 2K I am using super 16 lenses as they are built to a higher standard, will cover the full sensor at 2K, and can be much wider than 35 lenses.

 

[Dan Kanes]

I've used the 70-200 and the 24-70 at 2k. I've also used the 11-16 at 2k. All of them were "Great" for what I was trying to do.

I dunno - I'd use the 11-16 in a heartbeat if I was looking for a 2k shot with a 24-35mm field of view...

 

[paulherrin]

I know this seems like an obvious comment, but in Photoshop, if you have a high rez image and down rez, SOMETIMES it actually looks worse than just starting with an lower rez image. Does this not happen with red code?

that does partially depend on how you resize it (which filter: gaussian, bicubic, etc). otherwise, perceived sharpness is a combination of many factors including resolution and local contrast in the image.

 

[Terry VerHaar]

5K or 4K acquisition output at 1080 is going to be sharper than 2K acquired and output to 1080 (all other things being the same) due to the massive over sample. The concern then, is that if 2K acquisition is relatively softer, you want to maximize sharpness in the other parts of your imaging chain. If you are shooting at 2K you want to make sure your lens is nice and sharp. It is a bigger concern than when you are shooting 4/5K.

But these folks are talking about the effect of the glass on sharpness when acquiring 2K - that isn't going to change if you are acquiring 2K or 4K - it's the same sharpness as acquiring 4K since the sensor is simply windowed. Downsampling is another matter but the glass has no effect on that.

 

[Peter Cote]

But these folks are talking about the effect of the glass on sharpness when acquiring 2K - that isn't going to change if you are acquiring 2K or 4K - it's the same sharpness as acquiring 4K since the sensor is simply windowed. Downsampling is another matter but the glass has no effect on that.

No that's not true.