Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Saturday 11 February 2006 19:36, Steve Underwood wrote:
Matthew Fredrickson wrote:
On Feb 10, 2006, at 10:25 PM, Steve Underwood wrote:
Matthew Fredrickson wrote:
On Feb 10, 2006, at 1:21 PM, Gerard Saraber wrote:
DC in the signal through the echo canceller represents a signal the
canceller's adaption can never eliminate. It fights; it fails; it many
get very upset trying. DC needs to be eliminated before cancellation.
A-law/u-law ports are not supposed to give you any DC, but some do. The
following will estimate and remove DC from the signal. Prime
dc_estimate with zero.
int 16_t dc_removal(int32_t dc_estimate, int16_t sample)
{
dc_estimate += int32_t) sample 15) - dc_bias) 9);
sample -= (dc_estimate 15);
return sample;
}
Shouldn't it be *dc_estimate ?
Paul
--
Paul Hewlett - CottonPickinMinds - www.cottonpickinminds.co.za
Tel: +27 21 852 8812 Cel: +27 84 420 9282 Fax: +27 86 672 0563
--
___
--Bandwidth and Colocation provided by Easynews.com --
Asterisk-Users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Fri, 2006-02-10 at 16:05 -0600, Matthew Fredrickson wrote: On Feb 10, 2006, at 1:21 PM, Gerard Saraber wrote: Found it, going to go test it right now :) thanks! So far reports have been positive on the echo, but its a slow day ;) We're using cisco 7960 phones, they're pricy, but they work great and sound good, if it wasn't for the echo issue, I would have been able to roll the whole setup out already. Actually that's not quite true, I still have to make the 7914 addon module work with the 7960 phone, but that's not a show stopper. Either way, so far big thumbs up for the MG2 echo can, and if any developers read this, feel free to add a compile flag to make it more cpu intensive ;) and do more canceling. Does latest MG2 behave better than KB1 on your analog lines? I heard in the past that in some cases (primarily with analog lines) that KB1 worked better. Also, have you tried the echotraining=800 (in zapata.conf) tweak as well? --- Matthew Fredrickson In my case, MG2 blows KB1 away, the trunk version is a huge improvement, in the past, echotraining=some number was always worse compared to echotraining=yes so I didn't change it. I'll definately try that if I get any echo complaints, so far, so good though. -- Regards, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Feb 10, 2006, at 10:25 PM, Steve Underwood wrote: Matthew Fredrickson wrote: On Feb 10, 2006, at 1:21 PM, Gerard Saraber wrote: Found it, going to go test it right now :) thanks! So far reports have been positive on the echo, but its a slow day ;) We're using cisco 7960 phones, they're pricy, but they work great and sound good, if it wasn't for the echo issue, I would have been able to roll the whole setup out already. Actually that's not quite true, I still have to make the 7914 addon module work with the 7960 phone, but that's not a show stopper. Either way, so far big thumbs up for the MG2 echo can, and if any developers read this, feel free to add a compile flag to make it more cpu intensive ;) and do more canceling. Does latest MG2 behave better than KB1 on your analog lines? I heard in the past that in some cases (primarily with analog lines) that KB1 worked better. Also, have you tried the echotraining=800 (in zapata.conf) tweak as well? A lot of the variability is probably due to thr lack of a DC blocker at the front of the echo canceller. As far as I remember, none of the cancellers in * has a DC blocker. Where can one find out more information on writing a DC blocker? I google'd around a bit, but couldn't find a definitive overview of what one was, and how to write one. Thanks! --- Matthew Fredrickson ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
Matthew Fredrickson wrote:
On Feb 10, 2006, at 10:25 PM, Steve Underwood wrote:
Matthew Fredrickson wrote:
On Feb 10, 2006, at 1:21 PM, Gerard Saraber wrote:
Found it, going to go test it right now :) thanks!
So far reports have been positive on the echo, but its a slow day ;)
We're using cisco 7960 phones, they're pricy, but they work great and
sound good, if it wasn't for the echo issue, I would have been able to
roll the whole setup out already.
Actually that's not quite true, I still have to make the 7914 addon
module work with the 7960 phone, but that's not a show stopper.
Either way, so far big thumbs up for the MG2 echo can, and if any
developers read this, feel free to add a compile flag to make it more
cpu intensive ;) and do more canceling.
Does latest MG2 behave better than KB1 on your analog lines? I
heard in the past that in some cases (primarily with analog lines)
that KB1 worked better. Also, have you tried the echotraining=800
(in zapata.conf) tweak as well?
A lot of the variability is probably due to thr lack of a DC blocker
at the front of the echo canceller. As far as I remember, none of the
cancellers in * has a DC blocker.
Where can one find out more information on writing a DC blocker? I
google'd around a bit, but couldn't find a definitive overview of what
one was, and how to write one. Thanks!
DC in the signal through the echo canceller represents a signal the
canceller's adaption can never eliminate. It fights; it fails; it many
get very upset trying. DC needs to be eliminated before cancellation.
A-law/u-law ports are not supposed to give you any DC, but some do. The
following will estimate and remove DC from the signal. Prime
dc_estimate with zero.
int 16_t dc_removal(int32_t dc_estimate, int16_t sample)
{
dc_estimate += int32_t) sample 15) - dc_bias) 9);
sample -= (dc_estimate 15);
return sample;
}
Its a first order noise shaped single pole IIR. '9' is the damping
factor. If you make it bigger, the low frequency response will improve,
but the estimate will take longer to settle after step changes. This may
affect initial convergence if a DC hiccup occurs as the line is picked
up. 9 should be a good starting point to try.
Regards,
Steve
___
--Bandwidth and Colocation provided by Easynews.com --
Asterisk-Users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
Gerard Saraber wrote: Thanks! testing it now, on my test calls it appears to start out with less echo then the Mark3 canceler, but it trains slower, seems like it took a long time for the echo to completely disappear, the real test will be seeing what the people at my company have to say. Feb 9 14:47:51 [kernel] Zapata Telephony Interface Registered on major 196 Feb 9 14:47:51 [kernel] Zaptel Version: SVN-trunk-r934M Echo Canceller: MG2 I've had really good luck with the echocan preload patch that was posted on the asterisk dev list a while back as well, and I've been recommending it to people as well. This has really helped minimize the echo problems to a minimal level, although I don't know about recommending this system to our customers. I still think a lot of my audio quality problems are being caused by my phones (not echo, but clicks and pops and various overmodulation problems). We're getting there, but I'm still nervous with trying to sell an * system to someone who is used to the quality of a traditional PBX or key system. Clint ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Fri, 2006-02-10 at 11:01 -0600, Clint Sharp wrote: Gerard Saraber wrote: Thanks! testing it now, on my test calls it appears to start out with less echo then the Mark3 canceler, but it trains slower, seems like it took a long time for the echo to completely disappear, the real test will be seeing what the people at my company have to say. Feb 9 14:47:51 [kernel] Zapata Telephony Interface Registered on major 196 Feb 9 14:47:51 [kernel] Zaptel Version: SVN-trunk-r934M Echo Canceller: MG2 I've had really good luck with the echocan preload patch that was posted on the asterisk dev list a while back as well, and I've been recommending it to people as well. This has really helped minimize the echo problems to a minimal level, although I don't know about recommending this system to our customers. I still think a lot of my audio quality problems are being caused by my phones (not echo, but clicks and pops and various overmodulation problems). We're getting there, but I'm still nervous with trying to sell an * system to someone who is used to the quality of a traditional PBX or key system. Clint Found it, going to go test it right now :) thanks! So far reports have been positive on the echo, but its a slow day ;) We're using cisco 7960 phones, they're pricy, but they work great and sound good, if it wasn't for the echo issue, I would have been able to roll the whole setup out already. Actually that's not quite true, I still have to make the 7914 addon module work with the 7960 phone, but that's not a show stopper. Either way, so far big thumbs up for the MG2 echo can, and if any developers read this, feel free to add a compile flag to make it more cpu intensive ;) and do more canceling. -- Regards, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Feb 10, 2006, at 1:21 PM, Gerard Saraber wrote: Found it, going to go test it right now :) thanks! So far reports have been positive on the echo, but its a slow day ;) We're using cisco 7960 phones, they're pricy, but they work great and sound good, if it wasn't for the echo issue, I would have been able to roll the whole setup out already. Actually that's not quite true, I still have to make the 7914 addon module work with the 7960 phone, but that's not a show stopper. Either way, so far big thumbs up for the MG2 echo can, and if any developers read this, feel free to add a compile flag to make it more cpu intensive ;) and do more canceling. Does latest MG2 behave better than KB1 on your analog lines? I heard in the past that in some cases (primarily with analog lines) that KB1 worked better. Also, have you tried the echotraining=800 (in zapata.conf) tweak as well? --- Matthew Fredrickson ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
Matthew Fredrickson wrote: On Feb 10, 2006, at 1:21 PM, Gerard Saraber wrote: Found it, going to go test it right now :) thanks! So far reports have been positive on the echo, but its a slow day ;) We're using cisco 7960 phones, they're pricy, but they work great and sound good, if it wasn't for the echo issue, I would have been able to roll the whole setup out already. Actually that's not quite true, I still have to make the 7914 addon module work with the 7960 phone, but that's not a show stopper. Either way, so far big thumbs up for the MG2 echo can, and if any developers read this, feel free to add a compile flag to make it more cpu intensive ;) and do more canceling. Does latest MG2 behave better than KB1 on your analog lines? I heard in the past that in some cases (primarily with analog lines) that KB1 worked better. Also, have you tried the echotraining=800 (in zapata.conf) tweak as well? A lot of the variability is probably due to thr lack of a DC blocker at the front of the echo canceller. As far as I remember, none of the cancellers in * has a DC blocker. Steve ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] more cpu intensive echo cancellers ?
I appreciate the input, but after doing a little research on that card, it looks like I'll still need the channel bank, I think with some carefull ebaying, I should be able to do the hardware canceling for about $1000 less then what i saw the 104d card for. not to mention it seems total overkill, I've got a wimpy 10 pstn phone lines, a quad T1 card seems a little excessive to me. If there was a single T1 with the G.168 echo canceler card for say $800, I'd be all over that (still researching). I've got all this extra cpu power, and nothing to use it on ;) in the mean time, I'll put the 104d card on the list of possibilities, Thanks, Gerard Saraber [EMAIL PROTECTED] On Wed, 2006-02-08 at 17:26 -0800, Canuck15 wrote: Gerard, Just get yourself a Sangoma card with hardware echo can and be done with it. It is worth every penny just for the headaches it will save you. It's a better solution for most situations compared to a channel bank. Cheaper, simpler and works just as good IMHO. -Original Message- From: Gerard Saraber [mailto:[EMAIL PROTECTED] Sent: Wednesday, February 08, 2006 12:20 PM To: Asterisk Users Mailing List - Non-Commercial Discussion Subject: RE: [Asterisk-Users] more cpu intensive echo cancellers ? Thanks for the quick reply :) It happens when we call a lot of different people, and obviously doesn't happen with our old analog phone system, so even if its caused by someone else, *we* still have to fix it. we're kind of weighing our options, I'm hoping to take care of this with some fancy software, but if not we'll be going the hardware canceller route. Thanks, Gerard Saraber [EMAIL PROTECTED] On Wed, 2006-02-08 at 11:45 -0800, Michael Collins wrote: Gerard, I'll bet your side is working great for echo cancellation. It sounds like the equipment at the other end of the call might need some help. You know the old rule if you and I are talking on the phone: If I hear echo, you've got a problem; if you hear echo, I've got a problem. If only all echo problems were so easy to diagnose! In any case, is it possible that some of the echo you're hearing is being caused by poor echo handling on the other end of the line? Just a thought. -MC -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gerard Saraber Sent: Wednesday, February 08, 2006 11:03 AM To: Asterisk Users Mailing List Subject: [Asterisk-Users] more cpu intensive echo cancellers ? Hi, I've had some decent luck with the mark3 echo canceller from the zaptel driver, echos on about 20% of the calls, people I've called say I sound great now, but our side hears echos. I was wondering if there was any way to tweak the current software cancelers into using more CPU (and hopefully doing a better job, close to a hardware canceler), I only have 10 lines, and a single call takes 0.5% cpu, I would have no problem if it went up to 5-10% if they would work better. Or should I just give up now and buy the channel bank, tellabs hardware echo canceler and a T1 pci card? (hope TDM400P cards have decent resale value ;) -- Regards, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] -- Regards, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Feb 8, 2006, at 1:03 PM, Gerard Saraber wrote: Hi, I've had some decent luck with the mark3 echo canceller from the zaptel driver, echos on about 20% of the calls, people I've called say I sound great now, but our side hears echos. I was wondering if there was any way to tweak the current software cancelers into using more CPU (and hopefully doing a better job, close to a hardware canceler), I only have 10 lines, and a single call takes 0.5% cpu, I would have no problem if it went up to 5-10% if they would work better. Or should I just give up now and buy the channel bank, tellabs hardware echo canceler and a T1 pci card? (hope TDM400P cards have decent resale value ;) Yeah there is, upgrade to trunk and use the new echo canceller there (MG2). It's supposed to rock, at least from what I've heard. All the MEC cancellers are _OLD_. At least switch to 1.2 and the KB1 echo canceler before giving up. --- Matthew Fredrickson ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Thu, 2006-02-09 at 10:05 -0600, Matthew Fredrickson wrote: Yeah there is, upgrade to trunk and use the new echo canceller there (MG2). It's supposed to rock, at least from what I've heard. All the MEC cancellers are _OLD_. At least switch to 1.2 and the KB1 echo canceler before giving up. --- Matthew Fredrickson Aha! good to know, I am running asterisk 1.2.4 and zaptel-1.2.3, should I switch to CVS ? I've tried the MG2 canceler with the above versions, each time I tried it, I had a constant echo, where with the mark3 it went away after a second or two at the beginning of the call. (which I can live with, but some of the calls are completely unusable due to continuous or returning echos) I'll go play with the mg2 and kb1 again and see what happens -- Thanks, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Feb 9, 2006, at 10:50 AM, Gerard Saraber wrote: Aha! good to know, I am running asterisk 1.2.4 and zaptel-1.2.3, should I switch to CVS ? I've tried the MG2 canceler with the above versions, each time I tried it, I had a constant echo, where with the mark3 it went away after a second or two at the beginning of the call. (which I can live with, but some of the calls are completely unusable due to continuous or returning echos) I'll go play with the mg2 and kb1 again and see what happens Try MG2 with trunk and KB1 with 1.2. KB1 is supposed to be fairly reliable in 1.2, and MG2 in trunk has a good possibility of outperforming KB1 from 1.2. Matthew Fredrickson ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Thu, 2006-02-09 at 14:16 -0600, Matthew Fredrickson wrote: Try MG2 with trunk and KB1 with 1.2. KB1 is supposed to be fairly reliable in 1.2, and MG2 in trunk has a good possibility of outperforming KB1 from 1.2. Matthew Fredrickson Thanks! testing it now, on my test calls it appears to start out with less echo then the Mark3 canceler, but it trains slower, seems like it took a long time for the echo to completely disappear, the real test will be seeing what the people at my company have to say. Feb 9 14:47:51 [kernel] Zapata Telephony Interface Registered on major 196 Feb 9 14:47:51 [kernel] Zaptel Version: SVN-trunk-r934M Echo Canceller: MG2 -- Thanks again, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] more cpu intensive echo cancellers ?
Hi, I've had some decent luck with the mark3 echo canceller from the zaptel driver, echos on about 20% of the calls, people I've called say I sound great now, but our side hears echos. I was wondering if there was any way to tweak the current software cancelers into using more CPU (and hopefully doing a better job, close to a hardware canceler), I only have 10 lines, and a single call takes 0.5% cpu, I would have no problem if it went up to 5-10% if they would work better. Or should I just give up now and buy the channel bank, tellabs hardware echo canceler and a T1 pci card? (hope TDM400P cards have decent resale value ;) -- Thanks, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] more cpu intensive echo cancellers ?
Gerard, I'll bet your side is working great for echo cancellation. It sounds like the equipment at the other end of the call might need some help. You know the old rule if you and I are talking on the phone: If I hear echo, you've got a problem; if you hear echo, I've got a problem. If only all echo problems were so easy to diagnose! In any case, is it possible that some of the echo you're hearing is being caused by poor echo handling on the other end of the line? Just a thought. -MC -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gerard Saraber Sent: Wednesday, February 08, 2006 11:03 AM To: Asterisk Users Mailing List Subject: [Asterisk-Users] more cpu intensive echo cancellers ? Hi, I've had some decent luck with the mark3 echo canceller from the zaptel driver, echos on about 20% of the calls, people I've called say I sound great now, but our side hears echos. I was wondering if there was any way to tweak the current software cancelers into using more CPU (and hopefully doing a better job, close to a hardware canceler), I only have 10 lines, and a single call takes 0.5% cpu, I would have no problem if it went up to 5-10% if they would work better. Or should I just give up now and buy the channel bank, tellabs hardware echo canceler and a T1 pci card? (hope TDM400P cards have decent resale value ;) -- Thanks, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] more cpu intensive echo cancellers ?
Thanks for the quick reply :) It happens when we call a lot of different people, and obviously doesn't happen with our old analog phone system, so even if its caused by someone else, *we* still have to fix it. we're kind of weighing our options, I'm hoping to take care of this with some fancy software, but if not we'll be going the hardware canceller route. Thanks, Gerard Saraber [EMAIL PROTECTED] On Wed, 2006-02-08 at 11:45 -0800, Michael Collins wrote: Gerard, I'll bet your side is working great for echo cancellation. It sounds like the equipment at the other end of the call might need some help. You know the old rule if you and I are talking on the phone: If I hear echo, you've got a problem; if you hear echo, I've got a problem. If only all echo problems were so easy to diagnose! In any case, is it possible that some of the echo you're hearing is being caused by poor echo handling on the other end of the line? Just a thought. -MC -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gerard Saraber Sent: Wednesday, February 08, 2006 11:03 AM To: Asterisk Users Mailing List Subject: [Asterisk-Users] more cpu intensive echo cancellers ? Hi, I've had some decent luck with the mark3 echo canceller from the zaptel driver, echos on about 20% of the calls, people I've called say I sound great now, but our side hears echos. I was wondering if there was any way to tweak the current software cancelers into using more CPU (and hopefully doing a better job, close to a hardware canceler), I only have 10 lines, and a single call takes 0.5% cpu, I would have no problem if it went up to 5-10% if they would work better. Or should I just give up now and buy the channel bank, tellabs hardware echo canceler and a T1 pci card? (hope TDM400P cards have decent resale value ;) -- Regards, Gerard Saraber Network Admin, Rarcoa, Inc. (630) 654-2580 x11 [EMAIL PROTECTED] ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] more cpu intensive echo cancellers ?
On Wednesday 08 February 2006 14:45, Michael Collins wrote: I'll bet your side is working great for echo cancellation. It sounds like the equipment at the other end of the call might need some help. Kind of. Please keep in mind that even if the far end is reflecting more energy back than it should be, it only becomes a problem when the combined delay makes it apparent. And anytime you're bringing telephone audio into a PC you are adding a significant delay to the audio path. -A. ___ --Bandwidth and Colocation provided by Easynews.com -- Asterisk-Users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
