Welcome to our community

Be a part of something great, join today!

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 Change...

Adobe is great. But isn't FCP support the big news everyone's waiting for?
 
Johan Malmsten said:
After reading the statement from Jim about Canons "parlor trick" (skipping pixels to get 1080p recording from their "full" sensor area)... I've started thinking...

Actually... Even though it isn't exactly in the RED-way. Why not use that same parlor-trick and better Canons result. The Canon sollution does have it's merits, though it's hampered by lack of controll of exposure (as I have been told at least. I don't know if it can be worked around like with the semi-auto HV20), crippling compression, limited FPS control and no raw.
Click to expand...

The R3D file is described as being a 'wavelet'. I'm no expert but I believe this means that the analog waveform is stored as a waveform, albeit digitally. It only turns into pixel information when you convert it to another video or still image format.

If you want to skip or combine pixels you would have to first convert the analog waveform to the 'pixel defined' result format, then combine the pixels and store on the medium. If I understand it correctly the Red method has the camera's CPU and A to D subsystem encapsulate the waveform only, and your computer's CPU does the rest in post. We all know how slow this process is and how much CPU power it takes. For this reason the only way the camera can reduce the rate of storage is to reduce the frequency of the digital storage (Redcode28 or Redcode36) or to only store a portion of the analog signal (window).

Again, this all depends on if I understand it correctly...

Bob
 
Wavelet compression is a mathematical transform applied to the raw digital data. The most common type, classic JPEG, is a discrete cosine transform. Both are mathematical algorithms for reducing image file size but have a different effect on how the compressed image looks.

Here is interesting site that compares the two types on photographic images at different quality levels.

http://www.cs.sfu.ca/CC/365/li/material/cgi-bin/wavelet.cgi
 
Stop! I need to think!

Stop! I need to think!

WTF??? new setups, againnnnn????
OK! great !!! :greedy:

Tell me Jim... in the world of post production? .R3D fles new features???
-It will be possibe to do first light correction redcine style on the original .R3D files inside Adobe(after, premiere), Discreet(lustre, smoke) , Digital vision(nucoda, film master) and Quantel(pablo, EQ)
-Redray HDMI output and Media Station
- AND REDPHONE:bleh: it's like the IPHONE but you also can record/render 3k files heheheh

Bring it on :gun:
 
Boris_Epix said:
...People that ask Red to stop constantly changing the specs shall get happy with Canon/Sony ...

Let's be grateful for the enthusiasm and hard effort Jim and his people are putting into Red.
Click to expand...

That is exactly why we dumped them and are loyal customers and rally for Red Co.!!! I am VERY grateful for all people at Red; I included them in my thanks on TG last Thursday.

Cheers!!!

Luis


PS

I just open the door (I live in Iowa) and get chilled in a second...:cold:
 
kidrobot said:
I hope the change includes a mirror housing for a proper optical VF (and quick AF). That would put the S back into DSMC.
Click to expand...

The traditional moving mirror reflex viewfinder is an anachronism on a digital camera and unusable for motion picture acquisition. RED DSMC and Panasonic's Lumix G1 are the way of the future: High resolution electronic viewfinders that give an accurate visual reference of what the sensor is actually seeing.
 
David Rasberry said:
The traditional moving mirror reflex viewfinder is an anachronism on a digital camera and unusable for motion picture acquisition. RED DSMC and Panasonic's Lumix G1 are the way of the future: High resolution electronic viewfinders that give an accurate visual reference of what the sensor is actually seeing.
Click to expand...

Couldn't agree more. The mirror causes vibration, the optical viewfinder gives less than 100% view, can't be used for infrared photography, is difficult in low light etc. and an EVF gives you exposure information you'll never get from optical.

Nobody in the stills world focusses manually anymore. The viewfinder screens don't support it. Of course, you often focus 'manually' but you use the electronics (focus lock) to do it. And I doubt anybody truly misses their split image/microprism either. I sure as hell don't.

As for auto focus, I believe (Jim, tell me if I'm wrong) that the focus matrix can read off the front of the sensor just like light meters do after many years living in the pentaprism.
 
The big issue for me with an OVF in a DSLR is that they lie about DOF - they make it appear greater than it really is, which can be a shock when you look at the result. Focus is tricky too - things that appear in focus in the OVF really are not. I often have to use the angle finder with 2x zoom to see if things are in focus.

An EVF might not tell the whole truth, but it does not lie about these things.

Graeme
 
Reddec?

Reddec?

Hey what if the RED team, in their spare time (LOL), developed an external 4k debayering box connected via PCI-e that would give all post applications the ability to debayer on the fly. Sorta like a 4k RAW Multibridge chocka-block with dedicated CPUs. Now that would make the 13th insigificant.

Of course, AJA could do it (the connections are there) but they haven't yet. BMD could but won't (they certainly have the brains but I think this is all a bit 'pro' for them nowadays). Naw, RED are the perfect guys to do it. They wouldn't have to release the codec to anyone and could update firmware everytime they changed the R3D codec so nobody would be let down.

Two more guys on staff Jim. One hardware, one software, working on the REDDEC. Prototype at NAB. C'mon, whatdya say?
 
Actually the likely best way to go, as well as the most modular/open, is to write a special version of REDcine/REDAlert to run on the CUDA platform. You could hook up a whole room full of Tesla cards. And it could run on lower-end nVidia graphics cards as well.
 
autofocus?.. make it and i will be loyal to you forever.. scout's word.. well.. i never was a boy scout.. ehehehehehe
 
David Rasberry said:
Wavelet compression is a mathematical transform applied to the raw digital data. The most common type, classic JPEG, is a discrete cosine transform. Both are mathematical algorithms for reducing image file size but have a different effect on how the compressed image looks.

Here is interesting site that compares the two types on photographic images at different quality levels.

http://www.cs.sfu.ca/CC/365/li/material/cgi-bin/wavelet.cgi
Click to expand...

Thanks for the great link. Very interesting to see how much better wavelet is.
 
For me, the release #1 for both, RedCine and RedAlert, fully functional (real time playback with sound, etc.) will make my life so easy...:biggrin:
 
bettsy said:
Nobody in the stills world focusses manually anymore.
Click to expand...

I need to agree with this.. a couple months ago, I started to get comfortable with a mamiya RZ67 with a digital back, and although the images where great, I just couldn't handle all the soft shots via the manual focus... and even when they were in focus the time it took to get there was debilitating for anything but studio tabletop work.

That said, one thing though that I do enjoy is Manual focus but with Intelligent focus Confirmation...
 
If Red could pull off autofocus in the cine world too... that would be awesome.

I know someone said manual still was needed in because focus should be proactive versus reactive, but if the autofocus system could be fast and responsive enough for at least normal human movement, that still would be great.
 
You must log in or register to reply here.
Back
Top