Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

RED 8K S35

so what is the ideal pixel pitch for 8K to get no diffraction effects before say f8?
I have sony a7rii which is 8k across for photos and I don't see diffraction effects below f10
There is NO OLPF and I believe it has 4.5 micron pixel pitch
 
so what is the ideal pixel pitch for 8K to get no diffraction effects before say f8?
I have sony a7rii which is 8k across for photos and I don't see diffraction effects below f10
There is NO OLPF and I believe it has 4.5 micron pixel pitch

as mentioned earlier, 5 micron (MX|Dragon) has diffraction kick in at T7.9 in the red channel - not f-stop, t-stop.

That is the extinction limit. Diffraction may become visible already earlier.

The problem is - and ur statement is a great example - that once u see diffraction it's way too late.

Because u are monitoring on lower than native res, and the downscaling algorithm cleans some of that up.

A true comparison would be to shoot the exact same frame, one below and one above the diffraction limit and then compare in native res. Now u see the image detail that was lost.
 
Example:

Shoot 2 images with the same sensor and same lens. Focus dead on in both images. Shoot one @ 6K and then one @ 8K. Then look at the image in full native resolution, so look at one in 6K and at one in 8K - zoom in to pixel level. They will both have the exact same sharpness. One image just has less image detail than the other.

This doesn't hold water. Substitute different resolutions into the argument- 0.5k versus 50k for example.

People are are overthinkng a simple situation here. Greater sensor resolution allows us a better view of what the lens is doing. If it's doing bad things, we'll see them sooner, if it's doing good things we'll see them better. For any given output more data only improves your final result.

Now if if you feel that going to 8k s35 is like Monet having a cataract removed, and you preferred the world a little muddier, that's a creative choice, but anyone arguing that higher res capture is going to make their images worse is misunderstanding the situation.
 
This doesn't hold water. Substitute different resolutions into the argument- 0.5k versus 50k for example.

People are are overthinkng a simple situation here. Greater sensor resolution allows us a better view of what the lens is doing. If it's doing bad things, we'll see them sooner, if it's doing good things we'll see them better. For any given output more data only improves your final result.

Now if if you feel that going to 8k s35 is like Monet having a cataract removed, and you preferred the world a little muddier, that's a creative choice, but anyone arguing that higher res capture is going to make their images worse is misunderstanding the situation.

sharpness is a lens attribute. Like I said, u need to look at and compare the 2 example frames in their native resolution - NOT scaled down to your display resolution as that adds perceived sharpness and cleans up the image.

I think nobody here is opposed to a higher res capture.

The point is to get the highest possible IQ out of the capture. Resolution is merely one parameter in that equation.
 
So does that mean a 10K VistaVision cut from the same wafer that makes the 8K S35 sensor?
I certanly hope so. And ready in time for my VV allocation.
Please please please.
 
I didn't get a response from J when I asked, but I'm wondering if Helium and its various cuts will be the only new DSMC2 sensor.
Curious if a sensor with new color science and more DR will be available for these bodies in this "body -> sensor" cycle.

I'm happy with my Epic Dragon and I can live without 8K for now. Just wondering if there's anything coming that will tempt me within the next couple years, since I told myself that I won't spend on more gear and the Epic Dragon upgrade path won't last forever.
 
I didn't get a response from J when I asked, but I'm wondering if Helium and its various cuts will be the only new DSMC2 sensor.
Curious if a sensor with new color science and more DR will be available for these bodies in this "body -> sensor" cycle.

I'm happy with my Epic Dragon and I can live without 8K for now. Just wondering if there's anything coming that will tempt me within the next couple years, since I told myself that I won't spend on more gear and the Epic Dragon upgrade path won't last forever.

nothing is impossible in the RED world, but why would u think there's another sensor coming ?

Helium is the new sensor. It has new color science, as all sensors have it. Which means u can also use the currently existing color science on it, but a dedicated HeliumColor will squeeze more out of it.

IMO, RED has their hands full already with one sensor in the tick-tock cycle, so not sure if a second disruption would benefit anybody ;-)
 
If anyone is interested in the Post Workflow I had to work with for the Underdog. I just posted a blog about it :)

http://www.abandonvisuals.com/blog/underdog

Hey Edward -

Great work on Underdog.

I have a question - only from point of view of curiosity of why people choose different post paths / workflows - not at all to challenge your process. There are endless combinations of workflow steps that can be used to get from A to B.

I'm curious why you chose to transcode vs recording ProRes proxy at the same time as 8K RAW - and just offload and cut the proxy and re-conform directly to the 8K R3D RAW files and then color and export to master?
 
Hey Edward -

Great work on Underdog.

I have a question - only from point of view of curiosity of why people choose different post paths / workflows - not at all to challenge your process. There are endless combinations of workflow steps that can be used to get from A to B.

I'm curious why you chose to transcode vs recording ProRes proxy at the same time as 8K RAW - and just offload and cut the proxy and re-conform directly to the 8K R3D RAW files and then color and export to master?

Recording prores and R3D at the same time, it seams people still have not really understood is possible. I see DIT´s doing it all the time. Same thing with lut´s they still use their lut boxes and struggle to get luts onto client screens and multiple output. When it´s just to feed it into camera. I think the rapid progress for new firmwares and camera models, people tend to be a few steps behind. Others are worried that the prores files in camera are eating to much bandwidth, which I really don't think is the case if using 422 2k on the side of the r3d.

So far the 8k R3D´s you can only view in RCXP, difficult to finish an edit in there. I guess next update of Premiere and the other NLE´s will have the SDK implemented in their next update.
 
So far the 8k R3D´s you can only view in RCXP, difficult to finish an edit in there. I guess next update of Premiere and the other NLE´s will have the SDK implemented in their next update.

That is just a momentary blip - I already have pre-release beta builds of 3rd party software with the new SDK working with 8K.
 
Our tests so far with Helium has surpassed what I was hoping for. Imagine going from HD to 4k for the first time. This is whats happening here.
What monitor resolution are you viewing this on to compare?

Are you talking about comparing 6K Dragon downscaled to 4K on a 4K monitor vs downscaling 8K Helium to 4K on a 4K monitor?
 
Would love to see some motion R3D for low end comparisons.

I'm still trying to understand how smaller photosites gives a cleaner image. From Epic to Dragon, the image was noisier, Weapon cleared that up a bit, then we're now at an even smaller pixel in the same area. What am I missing?

Does anyone have a side by side by side yet with a dimly lit scene?
 
Back
Top