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 :)

Epic noisy at 2K

I'm really not sure why it does it, or did it. I have seen it thought when we first bought the camera, but I haven't seen it since. We even did a low light test.. and it never showed up.
It's a weird one.

http://www.zoomfilmtv.com.au/ftp/Dynamic_range/300fps_1600ASA.mov

Mark,
i notice some flickering on the shot you posted there....just to the right of the glass---and behind--i think a window? but is that what you are talking about with the strangeness you are or were seeing?
 
I have yet to bring this up with RED, AND would be surprised they don't know about it already

Send it to them. Don't assume they have seen it just because you have. This happens all the time in the software industry, people don't report issues because they assume others have already reported it. :)
 
no... thats just a funny hertz range from the monitors in the back ground. 50 htz, and 300 fps does not sync so you end up with flicker. If I shot 200fps.. no worries, Shoot 400 fps.. no flicker, but 300 is of sync to 5o htz.

got it....i thought it was a window
 
You are not alone. 2K 300 fps is especially in the realm of what your mentioning. It's not as bad as a snowstorm but it's certainly not useable.


Has anyone else experienced this problem? My epic is extremely noisy at 2K. It is beyond unusable. Just a snowstorm.

I have tried lodging an enquiry with red about a weak ago, very little response from the tech department.:furious3:

Has anyone else had/ solved this problem?
 
trees vs. test charts

trees vs. test charts

Some time ago, RED ONE was having "codec error" when pointed up toward the sky in winter at tree branches without leaves with a wide angle lens.

One problem in testing and developing a camera that reacts differently to what is being photographed, as REDCODE does (rather than a True RAW camera that records everything the same), is that you cannot simulate all possible subjects in the "lab" to know if something the camera can be pointed at will "freak out" the compressor, maybe by making a variable overflow or something.

I was in the Sony store here and walked up to one of their camcorders with my coat that had a random color texture to it, at a certain distance the image went into wild fits of noise, while at other distances the cameras image processing look normal. I'm sure the engineers at Sony never knew that their image processing would "freak out" with just normal texture in a subject, but testing can never prevent such issues as they only come up once the device is in the field shooting unknown subjects.

In the lab, shooting test charts, "it works", but in the real world you are not shooting a test chart a f/4.

When an issue like this came up before, I suggested using a 75mm lens and keeping it open to reduce the bandwidth needed in the background, Having the background black can also reduce bandwidth used so more is there for the subject.

At the lower compression rates, it is probably more a issue of the processing speed needed (less time to compress each frame) or increased noise in the pre-compression signal, so the optimizations of the compressor do not relate well to the detail vs. noise threshold which can be difficult to predict.

Hopefully the camera has enough processing power to be able to do the compression for all subjects at the top fps, given the history of dealing with the codec errors before, they will probably be able to adjust the compressor to have it more stable for complex subjects.
 
Just skimmed this thread, so don't know if it's been mentioned... You guys should do a black-shading with the camera set for the high frame rate and/or fast exposure time for best results. Drastic changes in exposure time are often a good reason to do a shading calibration.
Along with temperature or other significant environmental changes. Black shade at the intended project specs is always part of my camera prep. It's not just something I do only after upgrading firmware.
 
I have black-shaded w/ the camera set to the high frame rate (2K 300fps) but it still is not crisp. I've done numerous tests indoors and out set at 2K 300fps but it's just not useable for a job. It's been that way for me on more than one Epic also, so I never thought it was just an issue w/ a specific camera. The image at 2K 300fps (regardless of what project settings I'm at besides that frame rate) just looks to compressed. Maybe I'm missing something...?

Just skimmed this thread, so don't know if it's been mentioned... You guys should do a black-shading with the camera set for the high frame rate and/or fast exposure time for best results. Drastic changes in exposure time are often a good reason to do a shading calibration.
Along with temperature or other significant environmental changes. Black shade at the intended project specs is always part of my camera prep. It's not just something I do only after upgrading firmware.
 
I am on a shoot in Spain wit 30 people on set, I have that issue and the shots are full of noise.

HELP HELP needet! RED where are you!!

What is to do? Bright daylight, 300 fps, 180°, ISO 800 and lots of noise in the shot!!!
I did a black calibrating and not any better!
 
I am on a shoot in Spain wit 30 people on set, I have that issue and the shots are full of noise.

HELP HELP needet! RED where are you!!

What is to do? Bright daylight, 300 fps, 180°, ISO 800 and lots of noise in the shot!!!
I did a black calibrating and not any better!

also redcode 6:1 or 7:1?

post a still
 
7:1
263279_224817710897980_100001092570224_608826_7790556_n.jpg
 
exposure right in the middle

Just did a Black shading, stopped with Calibration falied!!!

1. I did black shading calibration set at 300 fps / 180° and the calibration falied
2. I did a black shading with 24 fps and it went ok
 
Last edited:
Mike we have the same issues as well. Same type of noise no matter what exposure and amount of light available. Even in Bright daylight there is still noise in 2k 300fps at iso 800. Tired different exposure and it doesn't help.... But then again RED one MX was also always noisy in 2k 120fps...so maybe this is how it's supposed to be....?
 
Set it to the highest redcode you can for now.
We're working on improving the high speed compression more.
 
As M owners we're reaping the many benefits of an incredible camera at an early stage in its life, but there are still reminders that this system is as yet unfinished. I've every faith that Red will come through, sadly there are a few elements of pain to go with our pleasure. I've said it before, but its essential to remember that Red do still have their disclaimer next to the EPIC-M on the store: 'for those who can provide testing and constructive feedback'
 
I discovered something: We were perfect all day with 300 fps until we did one shot at 5K HDRx.
Then we went back to 2K / 300fps and we had a lot of noise. I did a downgrade to 1.6.20 and a
upgrade to 1.6.24 again and now its a lot better again.

And John, what do you want to say with your statement? I do know what I paid 58.000 for and yes
it is still beta. I know.....

Thank's Deanan
 
Last edited:
No offense meant Michael, you're clearly smart enough to do your own tests and draw your own conclusions. Maybe I'm just spilling over a touch from the fact that I've heard a lot of moans recently off-line from people who expected perfect functionality and to be able to rent their EPIC-M unattended from day one... Touches a nerve as I'm at the opposite end of the spectrum, an inveterate tester and tweaker and almost expecting Red to want more from me on that front!
 
I discovered something: We were perfect all day with 300 fps until we did one shot at 5K HDRx.
Then we went back to 2K / 300fps and we had a lot of noise. I did a downgrade to 1.6.20 and a
upgrade to 1.6.24 again and now its a lot better again.

And John, what do you want to say with your statement? I do know what I paid 58.000 for and yes
it is still beta. I know.....

Thank's Deanan

Can you send me an r3d snapshot from the 5k HDRx, the 2k shot right after, and the 2k shot from later on? Use RCX 487 to do the trims.
 
I agree, 300FPS compression is a little below par. Just did a Hot wheels commercial, hope they can use it. But looks like the root of the problem is we need less compression, which means faster SSD's. When does Red anticipate moving to 2.5"?
 
I agree, 300FPS compression is a little below par. Just did a Hot wheels commercial, hope they can use it. But looks like the root of the problem is we need less compression, which means faster SSD's. When does Red anticipate moving to 2.5"?

Alex, 2.5" what? not a spinning drive ?? or is there a new 2.5" SSD that is faster?
 
Back
Top