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 :)

EPIC FAN NOISE - HEARTBREAKING DURING INTERVIEW

Scott Brown said:
Have the team at RED acknowledged the fan noise issue at all?

Just wondering what can be done technically to improve this issue - apart from the fan noise we love the Epic but for long interviews which can often be about personal and sensitive issues, the fan noise during and between takes kills the moment and for now is making us consider alternative cameras for interviews.

Scott
Click to expand...

To me a One MX with hard drives is still the interview king.

That said a steady noise is extremely easy to remove in post. Just record a room-tone.
 
Hello,

I did a non scientific test...

At 21 Celsius room temp

Fan @ 25% = 8-9 min
Fan @ 28% = 9-10 min
Fan @ 30% = 12 min

recording before 100% fan kiks in.


Pat
 
Patrick Tresch said:
Hello,

I did a non scientific test...

At 21 Celsius room temp

Fan @ 25% = 8-9 min
Fan @ 28% = 9-10 min
Fan @ 30% = 12 min

recording before 100% fan kiks in.


Pat
Click to expand...

What power source were you using?

What recording setting were you in (18:1 > 6:1

Cheers
Antonio
 
Patrick Tresch said:
Wich Sr. number do you have? Is it an X?


Pat

(Snr. Epic-X2013 delivered 02.2012 is subjectively ok)
Click to expand...

An X, #1646, 01/2012, no more than 10 min in single shoot, really unusable with direct sound.

sorry for the late.
 
To be a little more scientific, here are a couple of scenarios.

35% fan speed at room temperature. It just can't quite avoid the temps peaking. I often use 38% on long takes - I'll graph this when I get more time

epicm35fan.jpg



And at 25% (on a scarlet, as our EPIC-M is limited to 35%) things get ugly fast. 25% is for short takes only, without a doubt.

scarletx25fan.jpg



What's worth noting is that reducing fan speed to reduce noise is marginally a game of diminishing returns. There's a smaller noise difference between 35% and 25% than there is between 60% and 50% for example. Compromising to 30% not 25% could double your safe record time for very little harm to the sound dept.

fan%20loudness.jpg



Just as significantly we should all bear in mind that sound dissipates on an inverse square rule.
ie. its possible in an intimate scenario that moving the camera position only a very small amount relative to mics will make a disproportionate improvement to the intrusion. Getting the mic 15 inches away rather than 18 and the camera 7 feet away not 6 could just make all the difference.


On another note, I learned today that 'auto' doesn't ride the fan setting up and down during record. I thought it did, but it appears only to 'thermostat' the camera in standby mode. Is that right? When I've another spare second I need to rtfm on that one.
 
Last edited:
@John Merchant
would love for RED engineering to provide us this quality of data on all things Epic. Start with Black Shading: Signal to Noise at different temps and hard science on the topic - not fuzzy guidelines. Move onto a Compression matrix: data on a standardized/repeatable test scenario running from 3:1 up to 18:1 so we can see where the point of diminishing returns is: data rate/file size vs image quality. Heck, even a list of Tested & Approved 3rd Party Products would be a bit helpful. Someone has it at HQ . . . why not post it??

There's always time for this type of support since it dramatically reduces user guess work, on-set failures and warranty issues. "We're too busy inventing cool new stuff" doesn't cut it. I'm sure John and the OP are busy too.

That's the camera tests/results I would like to see - not some contrived Zacuto death match
 
Last edited:
Can you plot Signal to Noise from steady state temp to the maximum before the fan kicks on at 100%? Curious to see where the sensor temp/noise tipping point is.
 
Antonio Ribeiro said:
What power source were you using?

What recording setting were you in (18:1 > 6:1

Cheers
Antonio
Click to expand...


RC 8:1 25p 5k 2:1 Vlock 180wh bat.
I actually think (find out but not tested extenvively) that compression rate, in contrary of HDRx use, doesn't change a lot the temp.

The problem is not only the sensor temp but the "CORE" temp (wich I think is the CPU/ASIC of the cam).
The core temp is shown in the second setting 32/55C. This temp increase incredibly fast and gets the cam to speed up the fan.

Pat
 
John Marchant said:
Getting the mic 15 inches away rather than 18 and the camera 7 feet away not 6 could just make all the difference.
Click to expand...

Well, your examples reduce the noise only by about 1.5dB. Not very much.
 
Patrick Tresch said:
Hello,

I did a non scientific test...

At 21 Celsius room temp

Fan @ 25% = 8-9 min
Fan @ 28% = 9-10 min
Fan @ 30% = 12 min

recording before 100% fan kiks in.


Pat
Click to expand...

Great test and amazing graphical information. Thanks to all of you.
 
Mike Kushner said:
Wondering if anyone else has had a similar experience or has any advice on how to keep the fan from starting spontaneously during an intimate interview in a quiet room that may require long clips rolling continuously.
Click to expand...
This is only a suggestion, but note that there are ways to cancel and/or filter out some (I emphasize some) of the fan noise in post.

The problem with fan noise like this -- which I've heard referred to as "Hair Dryer mode" -- is that it's broadband, meaning that many frequencies are affected simultaneously. I can recall when we were working on Steven Soderberg's Che, he had stories about icing down one of the early Red Ones with frozen peas to keep it quiet (and cool). Red has come a long way towards improving the fan's efficiency and the camera's noise since then.

The only solutions I can think of are: don't get the camera too close. Use lots of sound blankets in the general area to absorb any fan noise. Use very directional microphones to minimize the ambience of the room. Hire a good sound person who knows how to minimize noise problems. Shoot in a quiet location conducive to controlling noise issues. Always bring a backup camera body for a critical shoot. (I think these steps apply to all cameras, not just Red's.) Get a good post sound guy who can use CEDAR, iZotope RX, and other tools to cancel out the noise in multiple passes.

Ironically, many of these problems also happened in the very early days of sound (1927-1930), when they had to put massive blimps around the mechanical film cameras to silence them. This no longer works in the digital age, since modern cameras have to be ventilated.

BTW, I just worked on a Scarlet project this week where we were in a factory setting with an average noise level of 87dB (measured on a SPL meter). By using Countryman B2 lavs, about a foot from the actor's mouths, we were able to knock out about half the noise. Directional mikes are a little problematic, but for a situation like this, there was no other practical solution.

Rob Ruffo said:
That said a steady noise is extremely easy to remove in post. Just record a room-tone.
Click to expand...
"Extremely" is not a word I would throw around. Everything you remove creates artifacts. There are no precise surgical solutions that are perfect and artifact-free. It's possible to take bad sound and make it barely tolerable, but it's not possible to take bad sound and make it good. We can take OK sound and make it much better. Good sound does not happen at the push of a button. I wish it was that easy.
 
The sound dispertion of the Epic is very directional :


- horizontal front to the subject (fan position)
- vertical above the camera (heat sink position)

I think the solution is to have a fan running 30-35% and do a BLIMP.

I'll try to make my blimp.

Pat
 
First off, as a sound mixer, I can appreciate a thread that deals with the impact of this camera on sound. Secondly, set life is a cooperative collaboration between different departments and armed with some of the information I've learned here, it will certainly allow me to approach problem solving with some additional 2nd hand experience. Although I would prefer to work with others who share an open and respectful view of each others' trades, at the end of the day I'm coming in as a professional to ply my trade and not have a happy hand holding moment with those around me. I'm going to tell it like it is, if it is too loud, it is too loud. I don't care how much the camera has improved or will be improving. My yardstick is not graduated for firmware versions. Of course I'll apply the tools at my disposal to minimize issues or find creative solutions, but one of the things you won't find me doing is to making suggestions to the camera department on set as how to operate their equipment. That is not my responsibility nor my place.

So, here is the thought that has occurred to me in helping to fix this issue. The challenge as I see it is to keep the required velocity of air moving over the camera's electronics to prevent the camera's logic from kicking the fan up to unworkable speeds, probably in the 35 to 50% range, depending on the distance of the camera to subject, framing, room, etc... The main issue or failure is depending on the small circumference built in fan to provide this air. I've seen installations in projection booths or equipment closets that have to deal with a similar excavation of hot air to keep the equipment running. Would anyone be up for trying a remotely placed larger capacity fan with cooling tubes? Here is a company that makes a product specifically designed to meet certain acoustic targets:
http://www.activethermal.com/page24.html

Please be more respectful to monkeys, they are beautiful creatures. Aloha
 
You must log in or register to reply here.
Back
Top