Welcome to our community

Be a part of something great, join today!

Register Log in
  • 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 :)

Big Resolution News...

Sweet Jesus. My engineer friends have been kicked is the head!
 
Damien Molineaux said:
Congrats Red Team ! Again !

Now about this new debayer, is it available to us yet, if not when ?

Cheers,
Damien
(aka. earthling)
Click to expand...

It is in Build 104 of REDCINE-X with a RED Rocket.

Jim
 
JD Holloway said:
I should think B4 posting. Theres no way the old OLPF could have ever let that amount of resolution though. Aliasing protection and the like.
Click to expand...

No, you were right. The pixel size didn't change, so there was no reason to change OLPF either. What changed was the size of the sensor, so only the size of the OLPF needed to be changed.

Having said that, it looks like OLPF has been improved over the original one. Not that the original was bad, but the results from the new one are simply spectacular.
 
Last edited:
new4-1k.jpg
 
“Leaders are born with an innate talent to question conventional wisdom.”
*~John G Agno

''Create a vision and never let the environment, other people's beliefs, or the limits of what has been done in the past shape your decisions.
Ignore conventional wisdom.”
~Anthony Robbins
 
Craig Ryan said:
So then...

The 2/3" Scarlet will theoretically resolve closer to 2.6k then right (assuming the new number is about 85%)? Giving us the capability of a solid .6k to move around, punch in, etc.. for 2k/1080p delivery.
Click to expand...

At the rate they are improving by the time the camera is released 3k will resolve 4k! ( I jest.) :001_cool:
 
so 4.5K is yielding 4.1K res. That's over 91 percent!
The other test is giving around 87.5%.

Then 5K S35 Epic has the potential to yield 4.55K ish

and 6K FF has the potential (if this trend continues) of 5.46K...4.8K using "the old" 80% rule.
Pretty remarkable. Somebody *cough* is going to talk about entry-level Imax (again) at some point lol.
Its 1.5K more than DigiIMAX at that point.

Just a thought about lens choices. I believe that the Salt test revealed that
"the 100mm was the weakest of the bunch". While the 100mm RPP may have some shortcomings in the testing, here we are not testing lenses but sensors.
Maybe the 100mm is ideal for this test as flare is not an issue etc.
Did you guys try any other RPPs?
 
Jannard said:
4.5K showing measured resolution of 4.1K. (85mm RPP)

1265973099.jpg
Click to expand...



OK the loud sycophantic cheering has drowned out my question about the graph on page 1 of the thread.




The graph shows things starting to get inaccurate at about 3500

Am I reading that wrong? If so, perhaps it's better not the post the graphs and just tell us that you've made the chip or whatever better... the 35 pages of responses all saying "well done red team" will still materialise... and you won't get bothersome people like me actually wondering what the graph is showing us.

Like I said on page 1, the graphs don't mean anything to me other than seeing some nice smooth gradient or whatever, and it seems to go to 4K but it has some spikes and such that break the curve severely from about 3500 onwards, and there's no explanation of how that's OK. Which is all I'm wondering about.
 
Michael Stanmore said:
OK the loud sycophantic cheering has drowned out my question about the graph on page 1 of the thread.




The graph shows things starting to get inaccurate at about 3500

Am I reading that wrong? If so, perhaps it's better not the post the graphs and just tell us that you've made the chip or whatever better... the 35 pages of responses all saying "well done red team" will still materialise... and you won't get bothersome people like me actually wondering what the graph is showing us.

Like I said on page 1, the graphs don't mean anything to me other than seeing some nice smooth gradient or whatever, and it seems to go to 4K but it has some spikes and such that break the curve severely from about 3500 onwards, and there's no explanation of how that's OK. Which is all I'm wondering about.
Click to expand...
Stop the thread! Back up to JBeale's post number 29. You weren't being ignored.
 
Michael Stanmore said:
OK the loud sycophantic cheering has drowned out my question about the graph on page 1 of the thread.




The graph shows things starting to get inaccurate at about 3500

Am I reading that wrong? If so, perhaps it's better not the post the graphs and just tell us that you've made the chip or whatever better... the 35 pages of responses all saying "well done red team" will still materialise... and you won't get bothersome people like me actually wondering what the graph is showing us.

Like I said on page 1, the graphs don't mean anything to me other than seeing some nice smooth gradient or whatever, and it seems to go to 4K but it has some spikes and such that break the curve severely from about 3500 onwards, and there's no explanation of how that's OK. Which is all I'm wondering about.
Click to expand...

screw the chart, look at your images and enjoy.
 
@MS

No doubt a statistical analysis for significance could be done for values at the end range of the curve.
This would determine if your getting more detail or junk information, and you're always going to get data that falls off the curve a little bit, at every part of the curve.
Imagine doing this test 24 times a second of a single chart, and for every frame, drawing a smooth line over the peaks and troughs of the ossillations.
Then overlaying the lines. Where does the graph become a useless blur?
The first test graph seems to have more errant data.
What if the test pattern is moving...

Just talk'n out my poop shoot.
 
new SDK with new debayer, when?

new SDK with new debayer, when?

Jannard said:
It is in Build 104 of REDCINE-X with a RED Rocket™.

Jim
Click to expand...

Thanks for the info Jim, but is that as in, only with a RedRocket?

And will the new debayer be included in a new SDK? I love RedCine X, and maybe that is all I will need but we are, for the time being, as are many others, using other software, for a variety of reasons. I know of course the integration of a SDK doesn't depend only on Red, but it must be available, before it can be integrated. When can we hope to see these evolutions become available for third parties?

Cheers,
Damien
 
JD Holloway said:
@MS

No doubt a statistical analysis for significance could be done for values at the end range of the curve.
This would determine if your getting more detail or junk information, and you're always going to get data that falls off the curve a little bit, at every part of the curve.
Imagine doing this test 24 times a second of a single chart, and for every frame, drawing a smooth line over the peaks and troughs of the ossillations.
Then overlaying the lines. Where does the graph become a useless blur?
The first test graph seems to have more errant data.
What if the test pattern is moving...

Just talk'n out my poop shoot.
Click to expand...

If image or camera is moving then the graph approximation of the curves across sliding windows of past 24 frames (1 sec) becomes more perfect, not blurred. Spatial movement on Bayer sensor, always increases perceived resolution not decreases.
But what blows me off here is the fact that it is end to end measurement test.

Test that includes the lenses, OLPF, sensor, Bayer pattern and its idiosyncrasies, debayering algorithm, REDRAW compression and uncompressing.
Full end to end test of optics, electronics, and software.

Jim, is this true that it is full end to end test?

So R1 is true 4K camera including lenses and software!

Andrew
 
Yeah, but the steppin fetzer valve is not in juxtaposition to the Z axis, which dictates recalculation of the hyperducinator... seriously folks, what's with all the nit-pickin?
Comparative charts look pretty obvious to me... I see nothing at 3100. WTF?

Nice work RED TEAM.

Our R1 just got back from rental, and will be enroute to RED HQ today for M-X upgrade.. cant wait!
 
You must log in or register to reply here.
Back
Top