Canon R5 Announced with 8k

I went back the other week to watch Yedlin's test as I haven't seen it since 2016 or so (whenever it was) and then realized there was no 6K version, or even 4K version of the files (at least anywhere I could find), so didn't even bother as it's a useless test in that regard as the higher resolution cameras are dumbed-down.

But if you're mastering in 2K and watching on a 2K screen, results make sense, I'm with it. (I'd love to see the 6K+ version(s) though on a 8K screen.)

Like I've said dozens of times, resolution is useless in the visual sense if you aren't properly watching it.

It's like combining 720p and 4K in a 720p timeline and then watching the result on a SD TV and saying, "Look, there's no difference!"
 
What about the benefits of oversampling that come through at lower resolution? Brawley claimed that the Phil Holland AppleTV 16K (or whatever) screensaver captures had extra pop in HD.

Personally, I just dont care as much about minute details coming across as I do about overall prettiness of the frame, and sometimes the extra detail is uglier.

Obviously, there are sweet spots of detail and there's a time and a place for everything. Detail in nature/sports/war/noir - sure, bring it on. But by and large, I prefer a degree of gloss, especially when rendering faces
 
Just to add something...I honestly think that point above about monitors and mastering is still relatively new to many people.

For some reason, I've had a feeling these past years that some still aren't grasping the advantages of higher-resolutions with proper workflows and screens.

It is a mystery, for real.
 
ahalpert said:
What about the benefits of oversampling that come through at lower resolution? Brawley claimed that the Phil Holland AppleTV 16K (or whatever) screensaver captures had extra pop in HD.

Personally, I just dont care as much about minute details coming across as I do about overall prettiness of the frame, and sometimes the extra detail is uglier.

Obviously, there are sweet spots of detail and there's a time and a place for everything. Detail in nature/sports/war/noir - sure, bring it on. But by and large, I prefer a degree of gloss, especially when rendering faces
Click to expand...

That's different, and that is totally something we need more of. This was a big conversation back in the BM days on BMCuser.

I will take great 4K from 8K+.

Honestly, the 8K was a PITA to work with from the R5. Not just because it's H.265 (what I used), but the resolution made transcoding take longer.
 
I'm 6.5 feet from a 55 inch 4k (UHDTV) barely close enough to see 4k. According to this, to see a benefit from an 8K (SUHDTV) I'd have to either sit 3 ft closer or even a 100 inch tv would be too small from my present 6.5 ft viewing distance.

optimal-viewing-distance-television-graph-size.png
 
NorBro said:
But if you're mastering in 2K and watching on a 2K screen, results make sense, I'm with it. (I'd love to see the 6K+ version(s) though on a 8K screen.)

Like I've said dozens of times, resolution is useless in the visual sense if you aren't properly watching it.

It's like combining 720p and 4K in a 720p timeline and then watching the result on a SD TV and saying, "Look, there's no difference!"
Click to expand...

NorBro said:
That's different, and that is totally something we need more of. This was a big conversation back in the BM days on BMCuser.

I will take great 4K from 8K+.
Click to expand...

Sounds like contradicting statements. On the one hand, you say that you can't appreciate high res without adequate playback. On the other, you say that oversampling improves lower resolution video
 
Focus...

You ready?

They are two different applications.

(1) To see the difference in 4K and higher resolutions, you need to export the highest resolution files you can. And you need to play them back on the highest resolution displays. And the distance. And your eyes need to work well.

(2) Yes, 4K improves lower resolutions. How long have you been doing this? You haven't seen the difference in HD cameras get better over the years since 2010? In this case resolution is used to improve lower-resolution hardware but there is still a ceiling. Pixels.

You good?
 
I'm still not seeing how - if there is a perceptible difference between 4K and higher resolutions - that difference might not carry over into detail in an HD export, as implied by Brawley with the 16K screensavers viewed in HD. There may be diminishing returns from greater and greater downsample or a point where the increased resolution is totally wasted in the downsample...

Either way, the point is moot if people are happy with binned HD when compared to 6K->4K downsample, viewed together in a 4K timeline
 
I think you're over-analyzing it (that's a surprise, lol) - but I actually think this is a decent over-analyzation.

I would agree with you about eventual diminishing returns for super-sampled HD, but the fact of the matter is we do not know what the diminishing returns really are, don't you agree?

We just listen to other people. We have no way of testing them.

We do not have a variety of 8K or 12K cameras that also provide the best-quality HD those systems are able to offer (from cameramakers we trust) to test ourselves.

We also don't have those same 12K and 16K and 20K cameras we can use in their highest resolutions and watch that 20K footage on a 20K display to see what diminishing returns there are.

So don't you (anyone) think that visual IQ diminishing returns is mostly theory?
 
If Brawley is saying that, I would have to disagree. For oversampling, the ceiling is not pixels, the ceiling is when the oversampling is enough that 444 chroma sub-sampling has been obtained or the optical limits have been reached, whichever occurs first. At exactly what level of oversampling that is could be subject to some debate, but not 8k or 12k to HD, because 444 has already been obtained before that.
 
What I just said is not right, YUV chroma sub-sampling is is not the same as debayered GRGB, you can get full 444 (RGB) 4k from as little as oversampled 5k raw, so anything more than that is likely overkill.
 
NorBro said:
So don't you (anyone) think that visual IQ diminishing returns is mostly theory?
Click to expand...

Yes, I'm speaking purely theoretically. I have no idea what the specific thresholds are, and it's not that important to me. Perhaps it can be worked out in the math.

I do agree that fine detail in an image - be it a close-up of something textured or a wide shot of a complex landscape - can be breathtakingly beautiful for having all that detail.

It's just that in my applications, by and large, it's not as important as having nice-looking facial skin or just an overall pretty rendering. Especially at the screen sizes and resolutions my work is looked at.

Incidentally, here were 2 shots I liked that I came across in recent lens research. Do people like them? I'll be amazed if someone could guess what the lenses are just by looking, although there's no reason you should be able to.

02-Optics.jpg

12-Close-Up-Work.jpg
 
Bokeh is nice in the dog shot.

My first guess would be something like a Zeiss Milvus with that creamy look (although I don't know if compression here is affecting the image a bit or the lens is showing some artifacts).

But also by the sound of your post I wouldn't be surprised if you told me it was some new 2020 50mm f/1.8 for $125.

On a side note, shallow DOF is another mind-blowing topic that tricks everyone's brains.
 
ahalpert said:
Yes, I'm speaking purely theoretically. I have no idea what the specific thresholds are, and it's not that important to me. Perhaps it can be worked out in the math.

I do agree that fine detail in an image - be it a close-up of something textured or a wide shot of a complex landscape - can be breathtakingly beautiful for having all that detail.

It's just that in my applications, by and large, it's not as important as having nice-looking facial skin or just an overall pretty rendering. Especially at the screen sizes and resolutions my work is looked at.

Incidentally, here were 2 shots I liked that I came across in recent lens research. Do people like them? I'll be amazed if someone could guess what the lenses are just by looking, although there's no reason you should be able to.

View attachment 140303

View attachment 140304
Click to expand...

I recognise the Samyang AF 45mm F1.8 FE and tamron 28-200mm F2.8-5.6 RXD anywhere.
 
You must log in or register to reply here.
Back
Top