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...

but i want to know plain and simple that does the red one mx can yield 3.5k res at 4k and true 4k or 4.1k at 4.5k resolution?????
 
and i am looking for this answer for a really long time,
 
What lens are you using? I guess you could always do a test yourself.
 
well i havent really done a test myself on resolution,but i would shoot a movie soon on red one mx a 4k 2.1 at redcode 42 most probably on zeiss ultra primes
 
still waiting for an answer.................
pls reply
 
I'm curious about this too. I'd never read those notes before, I just always heard about the 3.2-3.5k the RED One could resolve. Not holding my breath as I haven't heard of any of this for a long time, just think it would be cool to know one way or the other, and if this would apply to the newer cameras, especially the SCARLET. Would love to see the SCARLET get up a little to around 2.7k as opposed to 2.4, roughly 91% seems like an awesome leap over 80%. Though, kudos to RED and Graeme for getting 80% over the "theoretical" 75%....91% though...WANT :-P
 
has anyone who has done some resolution test on the production red one mx's has the answer to my questions.......??? pls still waiting for a proper answer,can red one mx yield 3.5k at 4k and 4.1k at 4.5k,plssssssssss someone give me the answer.......................................
 
Rohan... Search is your friend. It's been tested and confirmed by others and has been posted here several times. Results vary from one individual test to another, but generally speaking the 85% nominal or measured resolution is pretty accurate. You can figure that 4K on the RED One is going to measure about 3.4~3.5K+ if your'e using good lenses and other aspects of your shoot being ideal, you hit focus properly, etc.. 4.5K or 4880 pixels wide, should give you right about 4K+ nominal.
 
Rohan B said:
but i want to know plain and simple that does the red one mx can yield 3.5k res at 4k and true 4k or 4.1k at 4.5k resolution?????
Click to expand...

Why does it matter? A difference in resolution of less than about 0.3K or 0.4K is probably undetectable by the human eye, so it's not like the difference between 4K and 4.1K matters. At that resolution, the quality of your lens probably makes the biggest difference.
 
Stephen Gentle said:
At that resolution, the quality of your lens probably makes the biggest difference.
Click to expand...

Definitely true - and your tripod! I tried to help someone who was getting extremely variable results while trying to resolution test his stills camera. His images with quite varied results led to a lot of head scratching, but seeing his testing methodology revealed the problem immediately: he was shooting test charts in an office while other people were walking around, shaking the floor his tripod was on, and depending on the exact instant he triggered the camera, it was seriously dropping the resolution visible on the chart.

That was the day it finally dawned on him that just because he lots of megapixels in his camera, he wasn't getting anywhere near the full resolution possible unless everything else - lens, tripod, and even the ground the tripod was resting on - enabled that sort of resolution to be delivered to the sensor.
 
Jeff Kilgroe said:
Rohan... Search is your friend. It's been tested and confirmed by others and has been posted here several times. Results vary from one individual test to another, but generally speaking the 85% nominal or measured resolution is pretty accurate. You can figure that 4K on the RED One is going to measure about 3.4~3.5K+ if your'e using good lenses and other aspects of your shoot being ideal, you hit focus properly, etc.. 4.5K or 4880 pixels wide, should give you right about 4K+ nominal.
Click to expand...

Thanks Jeff for the answer.........
 
I'd use 80% for a conservative number. Remember that this varies depending on which OLPF is used and who does the debayer code. I haven't checked the other DSLRs recently but if they hold true to form as they have in the past, they are subject to the same general numbers.

Captured resolution for all cameras is a package size... like the size of a drawer. Measured resolution is how much stuff fits in that drawer... and the drawer is never full from any camera.

1080P is the drawer size... or 1.92K horizontal. The best 1080P camera so far for resolution is the F35 which measures 1.7K inside the 1.92K package. The Alexa measure 1.6K from its 2.8K RAW, primarily due to their OLPF. EPIC 5K measures 4K in a 5K package, just under 4K in a down-scaled 4K package.

Jim
 
Jim, I thought I heard Jarred mention in one of the NAB 2011 videos that EPIC will have interchangeable internal ND filters.

Is there any chance of having interchangeable OLPFs and tuning them for certain resolutions, say 3K or 2K so if you need to use that mode, for example to get increased frame rates, that the image would be more dialed in as a result?

Is this overcomplicating things and not a good cost/benefit ratio?

Or is the OLPF not interchangeable at all from a practical standpoint?
 
And if you try to stuff too much in the drawer, that's when you get unacceptable levels of aliasing or moire artifacts. Same as if you start with a nearly full drawer and downsize to a smaller one - if you aim to stuff all the resolution of the big drawer into the smaller one, you're heading for trouble.

Also, although the OLPF and demosaic algorithm are big factors, the lens plays a part too, as does compression if you're testing the full system (as you should). We always test with REDCode R3D files as that's how the camera is used. For other cameras we try to capture the HD-SDI uncompressed, giving them all the resolution advantage we can, but still we get the figures Jim mentions above. If you're capturing to any kind of compressed file, the numbers could be worse.

Graeme
 
Tuning an OLPF to 2k is only needed if you're doing something like line skipping to extract the lower (2k for instance or 1080p) resolution. Whenever we downsample we use proper electronic downsample filters (which are the electronic equivalent of an OLPF). Also, an OLPF finds it hard to cope in the line skipping scenario too because now the effective fill-factor of the sensor has been reduced which makes the MTF of the sensor higher, giving the OLPF a much harder job to do. That is why we crop or electronically downsample properly - not skip.

Graeme
 
There are a lot of factors that go in to Bayer resolution numbers. My main issue is that I shoot primarily in low-light conditions (sunrise, sunset, magic hour), where the camera is constantly struggling for light. In a well-lit, studio environment, this is not a major issue.
 
Of course, such low light factors will impinge upon the measured resolution of any type of camera system - so all you can do there is start with as much good resolution as you can to begin with.

Graeme
 
Thanks Graeme. I'm learning!

On another note... I find it the level of involvement here by RED Team members truly amazing. The only other company I found where questions are answered by the people who actually develop and engineer the products is Dr. B.J. Buchalter, the co-owner/founder (along with his brother) of Metric-Halo.
 
You must log in or register to reply here.
Back
Top