This is topic CP650 Headroom Config in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.


To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=006310

---

Posted by Antonio Marcheselli (Member # 373) on 06-27-2004, 03:14 PM:
 
Hi everyone

I upgraded all of my Cp650 recently to 2.2.5.0. I found a new menu item called "Headroom config" that can be set as "Typical" or "noise floor optimization".
I read the software help, but I didn't understand completely what it means. Is it a sort of sound limiter or dynamic compressor?

If I left the setting on "Typical" I disable this feature, isn't it?

Thanks for your time

Bye
Antonio
 

---

Posted by Ferdinando Innocenti (Member # 2477) on 06-27-2004, 04:32 PM:
 
Hi, Antonio!

Well, noise floor optimization was present in previous CP650 software version. It’s a feature which permits to lower noise floor.

From a Dolby bulletin:
Headroom is the difference (in dB) between the highest peaks of the signal and the highest peak that can be passed through the system without distorting.

It tries to use all available headroom in the B-chain, boosting the levels to do it, and reducing output levels by the same amount. That will be the optimum noise floor.

Dolby says to be aware that you can have distortion if your fader is set above 7.0

Bye
Nando
 

---

Posted by Antonio Marcheselli (Member # 373) on 06-27-2004, 04:54 PM:
 
Hi Nando

It's my first time that I speak in english with an Italian people!

I understand what you say. So if I left "typical" there is no variation.

But why should I have improvements boosting levels and reducing outputs? Where CP boosts the levels? On the 31 band equalizer? And where it decrease the level?

But since all the processing is in digital domain shouldn't I expect no noise problems? I believe that we are talking about few dBs, is the improvement so big?

Ciao
A
 

---

Posted by Steve Guttag (Member # 268) on 06-27-2004, 09:29 PM:
 
One of the big misconceptions is that audio behaves differently in the digital domain than in the analog domain. The rules on noise floor still apply in the digital domain.

So if you have a digital processor that has a particular noise floor but you do not maximize the available headroom within the digital processor, you do not get that noise floor in the specification, you get the noise floor of the unit plus the wasted headroom. In short, it is a noisier system.

If you do screening room work with the CP-650...it is too noisy (as are most analog or digital cinema processors) using conventional set up proceedures.

In both analog or digital processors you want the unit to be putting out the strongest signal it can without distortion. This will keep your signal as far above its noise floor as possible.

A well designed processor, like the CP-650, will not have the end user worrying about turning things up too much. The design criteria would be such that with a 0dBFS signal, gain trims as max (127?) and the volume on 10 and the EQs turned up...the system shouldn't go into digital clipping.

However, with the headroom for the EQ and the volume increase from 7 to 10...you have wasted quite a bit of the potential noise floor of the unit. If you know that you will not raise your volume above 7.0 and you have not boosted most of your EQ bands to the max, one can relaim that digital storage for improved noise characteristics.

So, if you tune up your room properly (in terms of EQ and level) and you find that your theatres don't ever raise the volume above 7.0 on typical films...then you can safely use the noise optimization of the CP-650 and pick up a noticable noise floor improvement.

If your theatre is relatively noisey or plays intermission music, it is unlikely you'd notice or need the change. Again, in a screening room where the stage speakers are less than 10-feet from the listener and there is no intermission music, one will hear a hiss and other possible noise with most stock processors. Panastereo is a notable exception...if the output stage is set up properly, it will run quite quiet.

Now there are other factors as to noise optimization with the equipment after the processor, including the crossovers and amplifiers that will sometimes limit what one can do.

Steve
 

---

Posted by Michael Schaffer (Member # 1204) on 06-27-2004, 09:48 PM:
 
If I understand this correctly, the function uses the available ca 10dB headroom between 7 and 10, boosts the signal internally by a fixed ratio and reduces the signal again at the output to achieve the desired output level, but with better internatal s/n ratio? In other words, the CP does NOT "see" where the fader is set, and calculates the remaining headroom? For instance, if the fader was at 8, it would see that there is still ca 6dB left, if the fader was at 6, it would see 12dB left etc. The warning not to turn up the fader suggests that there is a fixed boost.

EDIT- I just read the installation manual addendum. Apparently the system does measure the actual signal level. Still, why would there be a danger of clipping over 7 if it does so and relate it to the output level?
 

---

Posted by Christopher Seo (Member # 7) on 06-28-2004, 09:37 AM:
 
Are there any standard procedures for optimizing the noise floor in analog processors? Is it possible to do so?
 

---

Posted by Mark Gulbrandsen (Member # 72) on 06-28-2004, 06:57 PM:
 
When setting up a b chain its best to optimise the output level of an analog processor to create the optimum noise floor. dolby used to reccomend that you run CP-65's at about 750mv level average and then adjust the input gain on the power amps a bit lower to bring them into that range while keeping output levels standardized to what you like them set to. Many techs run power amp gains wide open and you generally will not acheive optimum noise floor(with either a digital or analog processor for that matter) by running them wide open. Either consult the manufacturer of your processor to see what optimum output level they reccomend, or experiment to find the best output level vs. audible noise with the processor in SR film playback.

Mark
 

---

Posted by Antonio Marcheselli (Member # 373) on 06-28-2004, 06:57 PM:
 
Thank you Steve for your reply.

I'm interest too for answers for Michael's question, very interesting.

Steve
About noise floor in digital domain: I understand that have the strongest singnal out of a CP650 (keeping the amps as low as possible) will give the best S/N ratio. However using the "noise floor optimization" of the Cp650 will boost the signal of x dB in the CP and lower the same signal of the same x dB in the unit again. The signal has the same level at the exit.

Why should I have better S/N ratio this way? I understand that a CP with levels set to 100 has better S/N ratio than a CP with levels set to 20 but I miss the meaning of the other way...

Bye!
A
 

---

Posted by Mark Gulbrandsen (Member # 72) on 06-28-2004, 06:58 PM:
 
When setting up a b chain its best to optimise the output level of an analog processor to create the optimum noise floor. dolby used to reccomend that you run CP-65's at about 700mv level average and then adjust the input gain on the power amps a bit lower to bring them into that range while keeping output levels standardized to what you like them set to. Many techs run power amp gains wide open and you generally will not acheive optimum noise floor(with either a digital or analog processor for that matter) by running the amps wide open. Either consult the manufacturer of your processor to see what optimum output level they reccomend, or experiment to find the best output level vs. audible noise with the processor in SR film playback.

Mark
 

---

Posted by Antonio Marcheselli (Member # 373) on 07-14-2004, 05:03 AM:
 

quote: Steve Guttag

---

A well designed processor, like the CP-650, will not have the end user worrying about turning things up too much. The design criteria would be such that with a 0dBFS signal, gain trims as max (127?) and the volume on 10 and the EQs turned up...the system shouldn't go into digital clipping.

---