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 :)

Trading 3K (4K) for 2K (3K) with better DR, less noise and other improvements?

Trading 3K (4K) for 2K (3K) with better DR, less noise and other improvements?

  • (in USA) I would trade some resolution for better DR, less noise and other +++

    Votes: 23 41.1%
  • (outside USA) I would trade some resolution for better DR, less noise and other +++

    Votes: 17 30.4%
  • (in USA) Nope - 4K is a must and I won't settle for anything less!

    Votes: 4 7.1%
  • (outside USA) Nope - 4K is a must and I won't settle for anything less!

    Votes: 4 7.1%
  • (in USA) I don't care...

    Votes: 6 10.7%
  • (outside USA) I don't care...

    Votes: 2 3.6%

  • Total voters
    56
In which case improving the signal to noise ratio is the way toward improving dynamic range.

You are right. But it does matter HOW You improve the SNR. The two methods are NOT the same, but they are complementary. In which case - according to Your logic - one would benefit by using both and there will be (I dare to say) great difference between gaining DR by "just" downscaling the image and gaining DR with Sensor that has greater DR in a first place and then downscaling the image...

Please, both of You, I am not saying gaining DR by image reduction is wrong. It is just NOT THE SAME THING as gaining DR with better sensor. What if You needed the full unscaled image? The better sensor (with higher DR / less noise) will make all the difference...
 
WTF is this inside-outside USA stuff?

No intention to discriminate. :innocent:

I only wanted to test my theory that people outside USA care little less about the actual resolution. I am glad I was proven wrong on this.

Maybe I should have mention "Hollywood" instead of "USA" :devil:
Don't kill me just yet...
 
(it doesn't matter how you get less noise)

This is where You are incorrect. It does matter a lot. What if you were shooting with Red One for 4K delivery and unable to "downscale" your footage? The better sensor will save the day...
 
No intention to discriminate. :innocent:

I only wanted to test my theory that people outside USA care little less about the actual resolution. I am glad I was proven wrong on this.

Maybe I should have mention "Hollywood" instead of "USA" :devil:
Don't kill me just yet...


No, I wasn't offended, just couldn't see what it was for.

Otherwise, yeah.

In EU our eyes perceive less resolution but more DR.

Fair.

:).

Everyone everywhere wants everything to be perfectly excellent. :sorcerer:
 
This is where You are incorrect. It does matter a lot. What if you were shooting with Red One for 4K delivery and unable to "downscale" your footage? The better sensor will save the day...

Um... what? Yes, of course if you have two sensors with photosites of the same size, and one has lower noise (e.g. more dynamic range) it will produce a better image. But that's not what we were talking about. What we were talking about, actually, was downscaled footage from a 3K sensor vs. non-downscaled footage from a 2K sensor of the same size (and therefore with larger photosites). You seemed to be of the opinion that somehow the dynamic range gained in the latter situation was different (and superior) in some meaningful way from the dynamic range gained in the former situation.

Both mechanisms provide increased dynamic range. The former provides more flexibility to choose between dynamic range and resolution. (And note that a 3K image scaled to 2K is still going to look sharper than an image shot at 2K.)
 
Um... what? Yes, of course if you have two sensors with photosites of the same size, and one has lower noise (e.g. more dynamic range) it will produce a better image. But that's not what we were talking about. What we were talking about, actually, was downscaled footage from a 3K sensor vs. non-downscaled footage from a 2K sensor of the same size (and therefore with larger photosites). You seemed to be of the opinion that somehow the dynamic range gained in the latter situation was different (and superior) in some meaningful way from the dynamic range gained in the former situation.

In any case I was talking about 3K 35mm (in relation to 4K 35mm Red One's).

Regardless, I dare to say that 2K from 2K 35mm sensor will look better (with higher DR) the 2K from 3K 2/3" sensor. Are You aware of the difference in surface area between 2/3" and 35mm sensors?
The later will most likely look sharper, but here we are talking latitude and noise...

I know from real-world experience that images taken on 8MPx Canon 30D using L-series Canon glass are far superior to 8MPx images downscaled from 12.1MPX Canon PowerShot SD950 IS with fixed lens. Coincidentally the later has almost 2/3" (1/1.7") sensor, whereby the 30D has APS-C sensor (not even 35mm).

You can expect similar results in video...
 
I voted in the "uncompromising" 4K-is-a-must category. The reason is simple enough... I think there's enough dynamic range in a 2/3" sensor to pull off excellent dynamic range at 3K, and I see no need to make sacrifices to specifications when I don't have to.

Implying there are huge sacrifices to be made is justifying the slow, boring trickle-ware policy that has been populating the market for ages. Go for the gusto RED!
 
Please, both of You, I am not saying gaining DR by image reduction is wrong. It is just NOT THE SAME THING as gaining DR with better sensor. What if You needed the full unscaled image? The better sensor (with higher DR / less noise) will make all the difference...

True but the reason I keep arguing this point is that I have seen a lot of posts on this forum that say "why 5k? give us more dynamic range! I only need 2k. The obsession with resolution is crazy" etc. Obviously Red is doing the best they can with the noise performance of the chip itself. At a certain point you run into heat issues and hit the limits of current technology. At that point moving up to a higher resolution can make sense as a way to gain some extra dynamic range.

I see what you're saying though. There are obviously some design trade offs to consider. Does the noise reduction you get by scaling down 4k footage from a 35mm sensor outweigh the inherent lower noise of a native 35mm 2k sensor? I have no idea but I trust that the smart engineers at Red have come up with what they feel is the best compromise between resolution, sensor size, and noise.
 
I know from real-world experience that images taken on 8MPx Canon 30D using L-series Canon glass are far superior to 8MPx images downscaled from 12.1MPX Canon PowerShot SD950 IS with fixed lens. Coincidentally the later has almost 2/3" (1/1.7") sensor, whereby the 30D has APS-C sensor (not even 35mm).

Far superior how? Noise? Sharpness? Compression issues? CA? There are a lot of variables between those two cameras. But yes, I'm sure that there are more expensive 2k cameras that will make a nicer image than the 3k Scarlet. I don't think anyone here is saying that resolution is everything.
 
I think this is a nonsense poll. It suggests that trading off resolution in the sensor will improve its DR more than it would when you downscale it in post.
Both methods will produce the same DR, but doing it in post gives you a choice, so would be the better choice period.
 
I think this is a nonsense poll. It suggests that trading off resolution in the sensor will improve its DR more than it would when you downscale it in post.
Both methods will produce the same DR, but doing it in post gives you a choice, so would be the better choice period.

Back to ground zero - don't even get me started...

I have made my point as clear as possible - go and read this thread through-out before You make such statement...
 
I think this is a nonsense poll. It suggests that trading off resolution in the sensor will improve its DR more than it would when you downscale it in post.
Both methods will produce the same DR, but doing it in post gives you a choice, so would be the better choice period.

LOL, now I have to back Peter up. Both methods can increase the dynamic range but not necessarily by the same amount. Which I think is what Peter has been trying to get at all along. Apart from the Red engineers, the rest of us here simply can't know which approach squeezes the maximum possible dynamic range for the buck.
 
LOL, now I have to back Peter up. Both methods can increase the dynamic range but not necessarily by the same amount. Which I think is what Peter has been trying to get at all along. Apart from the Red engineers, the rest of us here simply can't know which approach squeezes the maximum possible dynamic range for the buck.

Thanks mate...
 
We should all consider another (often overlooked) benefit of having less resolution:

Given the limitation of recording bandwidth for the new sensor (actually the recording media) at 100MBps, reducing the captured resolution from 4K to 3K saves app. 43% from this bandwidth which can be used to either increase the FPS, or more importantly (at least for me) - reduce the level of compression!

Now 43% saving in compression is huge! Ever noticed difference between quality 12 (MAX) and 10 when saving JPEGs? Almost impossible to notice. But have a look at the difference between quality 10 and 6 - now that is a different story.

I am sure RED-CODE-RAW is more efficient then JPEG (otherwise they would use JPEG), but being able to gain 43% less compression in Your images should be very high on the list of priorities.

I stand here to correct myself.

I've said before that the higher DR of film is the "last mountain" we have to climb to bring the digital cinematography up to (or over) the level of analogue film...

Well - there is one more - analogue film captures all frames "without compression" (but with grain). Now that is something to think about as well...
 
"trading"

"trading"

Some interesting comments. I see the poll results and I respect the passion on this subject.

But before this thread takes any potential turn towards "RED isn't listening to its customers" comments, let me just say we disagree that this is the best way to improve performance given the mix of technologies we have available,
 
Some interesting comments. I see the poll results and I respect the passion on this subject.

But before this thread takes any potential turn towards "RED isn't listening to its customers" comments, let me just say we disagree that this is the best way to improve performance given the mix of technologies we have available,

What exactly comes to the equation when calculation "performance" and how much does the compression/image quality influence this equation?

With RED ONE You have managed to squeeze 4K stream to 36MBps. Now uncompressed SD is little over 20MBps. I don't have RED camera (yet), but I would like to know what are the most common compression artifacts in RED footage, because at this compression level something's gotta give...
 
BTW Stuart, thanks for dropping in.

I was proposing the 3K 35mm sensor in the "new" camera (http://www.reduser.net/forum/showthread.php?t=12235) in order to fit it between Scarlet and Red One and to be able to "reuse" all the internal electronics from Scarlet while giving to all that want it the interchangeable lens mounts instead of 35mm DOF adapters...
 
what "new" camera? IMO 3 cameras is more than enough of a workload for RED's R&D team. Let them concentrate on their current designs, make a ton of money off Scarlet then make the camera to end all cameras. Putting anything more on the plate right now is just going to tax resources and slow things down.
 
Average R&D process

Average R&D process

what "new" camera? IMO 3 cameras is more than enough of a workload for RED's R&D team. Let them concentrate on their current designs, make a ton of money off Scarlet then make the camera to end all cameras. Putting anything more on the plate right now is just going to tax resources and slow things down.

Before NAB no one had clue about EPIC and similar comments (by me amongst others) were made that Scarlet would distract RED from RED ONE...

The point being is that RED can't fall asleep on roses (or Scarlets :blush: ). The industry WILL respond and (as You know) it takes time from idea to real product in hand. I'll bet anything that RED is already playing around ideas for next cameras.

Average R&D process:

IDEA --> Research & Proof of Concept --> Development of Prototype --> Tuning and Advanced Engineering --> Release Product --> Updates & Tweaks

Red One is now in the last stage of "Updates & Tweaks"
Scarlet & EPIC are somewhere between "Development of Prototype" and "Tuning and Advanced Engineering" stage...
So for sure someone at RED has to be playing around with the "IDEA" and "Research & Proof of Concept"
 
Back
Top