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 :)

MEET FELIX....

Or that might be because of the bayer pattern.
I asked last week about it and Jim said that we keep about 80% of the pixels so that make about 2,4K roughly.
I assume that the 3K image uses a complicated math technic to be upscaled from that.

That being said, I didn't notice the problem mentionned on my iMac.
 
Or that might be because of the bayer pattern.
I asked last week about it and Jim said that we keep about 80% of the pixels so that make about 2,4K roughly.
I assume that the 3K image uses a complicated math technic to be upscaled from that.

That being said, I didn't notice the problem mentionned on my iMac.



I am certainly not one to speak for Jim, but I think you may have misinterpreted what he was saying. What I believe he probably told you is that the 3k "resolves" to about 2.4K. That doesn't mean that any pixels are thrown away, just that when the picture is evaluated with the use of a chart built for measuring resolution, it measures at about 2.4K. I don't know how this explains or doesn't explain what was being asked but "upscaling" would not be part of it.

- or maybe I am not understanding what you are saying here?
 
Saw also that : maybe it's because the interpolation from our computers (not running on 3K native displays) is less effective than the 1/2 interpolation from Redcine-X ?

I've compared frames from HDV clips that are downscaled to 50% displayed at 100%, versus unscaled frames displayed at 50%. Unscaled viewed at 50% tends to have a slightly softer look than the frames that were actually downscaled, so I'm sure you're correct. There's nothing to worry about. :)
 
Or that might be because of the bayer pattern.
I asked last week about it and Jim said that we keep about 80% of the pixels so that make about 2,4K roughly.
I assume that the 3K image uses a complicated math technic to be upscaled from that.

That being said, I didn't notice the problem mentionned on my iMac.

The Felix clip was recorded as 3072 x 1536 pixels. When I deBayer a frame and export it as a 16bit RGB tiff still frame, it is still 3072 x 1536 pixels unless I crop or scale it to a different format. No pixels are discarded.
All digital cameras, including 3-chip 1080p video cameras, optically resolve less than their sensor pixel resolution, but the frame size in pixels and the format are not changed. The very best 1080p cameras will resolve between 800 and 900 lines vertical, a Scarlet should resolve around 1200 lines vertical, or around 30% more, if the lens is up to the task.
 
I've compared frames from HDV clips that are downscaled to 50% displayed at 100%, versus unscaled frames displayed at 50%. Unscaled viewed at 50% tends to have a slightly softer look than the frames that were actually downscaled, so I'm sure you're correct. There's nothing to worry about. :)

so it shouldn't be there if we do it the other way around?
Full resolution at 100% - http://weareglitter.com/stuff/felix_v2_2.jpg
Half resolution at 200% - http://weareglitter.com/stuff/felix_v2_1.jpg

anyways, thanks for checking it! (I'm a newbie, this was the first RED footage I opened, also hoping to own a Scarlet later on:)
 
I am certainly not one to speak for Jim, but I think you may have misinterpreted what he was saying. What I believe he probably told you is that the 3k "resolves" to about 2.4K. That doesn't mean that any pixels are thrown away, just that when the picture is evaluated with the use of a chart built for measuring resolution, it measures at about 2.4K. I don't know how this explains or doesn't explain what was being asked but "upscaling" would not be part of it.

- or maybe I am not understanding what you are saying here?

Maybe... I'm not sure of what I understood now...
I've been told in classes that the Bayer pattern causes "loss" and that when RED says "Red One is 4K", they are toalking photosites, but the Bayer pattern need 4 photosites to make a pixel (Red Green Green Blue or in another order can't remember that one).
I'm confused now...:emote_head_explode:
Please note that I'm not complaining in anyway or whatever. Even if there's a loss, those cameras are still badass and the images out them are still beautiful and so on, and if I'm on Reduser it's because I am really pleased by what Red is doing. I was just trying to put some technical stuff to clarify some... stuff.
 
Bayer pattern needs 4 photo sites to define full color pixels, 2 green pixels and a red and blue pixel in a block. But these are processed by debayer algorithms to produce 4 full RGB pixels in the image file, one for every pixel in the original block.
 
so it shouldn't be there if we do it the other way around?
Full resolution at 100% - http://weareglitter.com/stuff/felix_v2_2.jpg
Half resolution at 200% - http://weareglitter.com/stuff/felix_v2_1.jpg

anyways, thanks for checking it! (I'm a newbie, this was the first RED footage I opened, also hoping to own a Scarlet later on:)

We need real names here...

Jim


And in case you are wondering how to change it... http://reduser.net/forum/showthread.php?t=32432
 
Yeah, i know i'm a little late with that.... but hanks for posting the Scarlet-footage. While i'm one of those who would buy a camera even without seeing footage/samples (i totally trust you guys, never trusted a company this much before, but with RED it's so different) i'm a little amused how much fun i had with just a few frames.

It was my first time with RC-X, and i really like that tool, very clean, very straight forward, short learning curve. I came back to it alot of times to just play around and try different things, the possibilities of RED-RAW are just awesome. It will be a great addition to my workflow in the future, can't wait to play with my own RED-footage.

BTW, i wonder what kind of pet RED will use for future monstro-test-shoots.
Maybe a T-Rex or Godzilla? You know, it has to be something worth shot on a MONSTRO. ;-)
 
Actually, a test shoot with one of the Godzilla suits would be pretty damn cool. It could certainly be arranged if they wanted to do it, though I think it might look hilariously bad at such a high resolution. Perhaps they could shoot a faux trailer for "Godzilla Vs. Jarred Land" or something. At any rate, it's great to see that the quality just keeps on rising with these cameras and the sensor tech is definitely headed in the right direction.
 
Whoa. This is my first time looking at any kind of raw RED footage.

DSLR what...? I mean, I still love my 7D for what it is, as a stepping stone for a broke filmmaker but... Holy cow.

For a while, as a DSLR user, I was kind of offended when I read I wasn't using a pro enough tool for filmmaking in a press release a while back.

...

DSLR what...?

Time to start saving up my pretty pennies. $6,000 isn't impossible. Might be a while, but not impossible. :smile5:
 
Hi Jarred - First of, the Scarlet is really finally coming together nicely! Fantastic job you've done. Would it be possible to show any normal outdoor / indoor footage? Felix is one beautiful beast but he is not that common in what I have in mind. And if the footage could be slightly longer - say around a minute that would be nice. :)
 
Hi Jarred - First of, the Scarlet is really finally coming together nicely! Fantastic job you've done. Would it be possible to show any normal outdoor / indoor footage? Felix is one beautiful beast but he is not that common in what I have in mind. And if the footage could be slightly longer - say around a minute that would be nice. :)

i been barking up that tree for a minute now!!! LOL
 
Back
Top