Plasma Screens and delay

[robertk]

Has anyone had any problems with the delay of the video on plasma screens?

If we have external speakers connected the lip-sync is lost.

And some plasma screens that has speakers builtin also has a delay on the audio. So we atleast get lip-sync.

My concern in the later case is if this will interfer with the echo canceler in the VC system.

//Robert

[Morgan81]

I've never seen a delay on a plasma so large that the lip-sync is lost, what other components is the video traveling through?

[robertk]

I've never seen a delay on a plasma so large that the lip-sync is lost, what other components is the video traveling through?

An S-video Cable :^)

Try the cheaper models of Plasma screens, some of them have a looong delay.

//Robert

[GTC]

Robertk,

Are you sure its the screen that is the cause for the Lip sync delay?
I have seen numerous times and different sites the MXP platform video/audio decoders out of sync, up to about a second !!
I have seen this at various client sites (some ISDN and some IP), with different systems running early and later release codes.
Cannot find any real reason for it, however, I have found that if you select a different input then switch back to main cam and alls well again??

GTC

[robertk]

Robertk,

Are you sure its the screen that is the cause for the Lip sync delay?
I have seen numerous times and different sites the MXP platform video/audio decoders out of sync, up to about a second !!
I have seen this at various client sites (some ISDN and some IP), with different systems running early and later release codes.
Cannot find any real reason for it, however, I have found that if you select a different input then switch back to main cam and alls well again??

GTC

It's a known fact that Plasma screens has a built in delay. They even ahve a delay on the audio now so atleast TV and DVD's have a correct lipsync.

The question is if this audio delay cause any problems with the echo canceler.

//Robert

[AVGeek]

i had a client that wanted to do conferencing over their fiber video system but in the fiber processing it caused an audio delay much like you describe but more like a second off, i was using a polycom vortex and asked their tech support the same question and they said no matter what the delay the ec should work and it did, i guess that makes sense as the ec references the far end and does not allow it to be transmitted back via the vtc mics apparently no matter when it is received (i.e. the ec does not know it's delayed it just filters it out), hopefully all codecs would operate in a similar manner

[robertk]

i had a client that wanted to do conferencing over their fiber video system but in the fiber processing it caused an audio delay much like you describe but more like a second off, i was using a polycom vortex and asked their tech support the same question and they said no matter what the delay the ec should work and it did, i guess that makes sense as the ec references the far end and does not allow it to be transmitted back via the vtc mics apparently no matter when it is received (i.e. the ec does not know it's delayed it just filters it out), hopefully all codecs would operate in a similar manner

I like to think that delay work just like having the mics very far away from the codec, So in theory it should work. HOWEVER I don't belive the delay can be too long.

I figure a EC work as having a memory where it digitaly stores the receiving far-end audio for an amount of time (IE the time that the sound travels to the mic from the speakers) then apply that sound data as a filter to the sound that the mic picks up from the room.
So depending on the memory (or buffer size) the delay can't be too long.

This is the concern I have with the added delay in Plasma screens. If the accumulated delay is too large for normal Videoconference equipment. (Not the professional EC that is out there that probably has alot more features, like the Vortex for example.)

Any thoughts about this?

//Robert

[AVGeek]

well, i agree the theory gives the possibility of proper echo cancel a dim outlook, especially since the delay is happening after the audio hits the codec . . but how about this, throwing sense and reason out the window, i have done several small codec/plasma installs running the audio through the plasma with no echo problems, this is with both tandberg and polycom through maybe six different brands of plasmas . . real world trail and error > theory? hehe

[trapehzoid]

I like to think that delay work just like having the mics very far away from the codec, So in theory it should work. HOWEVER I don't belive the delay can be too long.

I figure a EC work as having a memory where it digitaly stores the receiving far-end audio for an amount of time (IE the time that the sound travels to the mic from the speakers) then apply that sound data as a filter to the sound that the mic picks up from the room.
So depending on the memory (or buffer size) the delay can't be too long.

This is called tail length. Its how long the system can 'wait' for acoustic echo and still be able to cancel it.

[robertk]

well, i agree the theory gives the possibility of proper echo cancel a dim outlook, especially since the delay is happening after the audio hits the codec . . but how about this, throwing sense and reason out the window, i have done several small codec/plasma installs running the audio through the plasma with no echo problems, this is with both tandberg and polycom through maybe six different brands of plasmas . . real world trail and error > theory? hehe

Well I have installed many plasmas, and used the builtin speakers/amplifiers with good result... but what I see today is that there are alot of CHEAP plasmas on the market (wich we are forced to use sometimes) and they have ALOT longer delay then the more proffesional models we use normally.

//Robert

[bgoez]

This delay can be quite annoying and yes the professional lines do have less of a delay. I have noticed that the LCD displays are a bit faster as well.

[Burt.Kloppers]

Hi

This is the first time I hear about a plasma with that much off a delay.

Is it a system strait onto the plasma with supplied cables and using the plasma audio or are you using something ells that is between the plasma and the unit like a control unit with Video/Audio switchers or boosters.

[djackson]

From the technical description of the Tandberg 6000 "classic" :
"-Echo cancelling tail length lowest frequency sub band 340 ms.
-Echo cancelling tail length highest frequency sub band 250 ms."

Of course each product will have a different tail length. A Gentner or Vortex should have a longer buffer, which is one reason you want to use these devices in large auditoriums.

The speed of sound at sea level = 340.29 m/s. For sound to bounce from the front to back of a 10 meter (33ft) long room it would take approx 2*10m/(340.29m/s)= 0.059 seconds or 59 ms, assuming a straight front to back bounce and that my math is correct, which are both slightly faulty assumptions.

Depending on the size of the room, I'm guessing as long as the delay introduced by the plasma is less than 180-200ms, you should be OK. The best way is to test it in the room you are using.

You can measure the delay on the plasma by hooking up a PC that runs a program with timer, hopefully with ms and try taking a picture with a camera of both the PC and the Plasma. Please share with us the delay encountered. If you are looking for a cheap Plasma, you could probably go to a place like BestBuy with a laptop and measure the delay of individual systems. The sales guy will love you, as long as you buy something in the end.

[robertk]

From the technical description of the Tandberg 6000 "classic" :
"-Echo cancelling tail length lowest frequency sub band 340 ms.
-Echo cancelling tail length highest frequency sub band 250 ms."

Of course each product will have a different tail length. A Gentner or Vortex should have a longer buffer, which is one reason you want to use these devices in large auditoriums.

The speed of sound at sea level = 340.29 m/s. For sound to bounce from the front to back of a 10 meter (33ft) long room it would take approx 2*10m/(340.29m/s)= 0.059 seconds or 59 ms, assuming a straight front to back bounce and that my math is correct, which are both slightly faulty assumptions.

Depending on the size of the room, I'm guessing as long as the delay introduced by the plasma is less than 180-200ms, you should be OK. The best way is to test it in the room you are using.


The plasma had LONG delay :)

And it worked quite well when connected pt-pt, but connected over a MPS multisite just was terrible.. lots of echo.

We solved the problem by adding a external speaker to each system... now it works as expected.


You can measure the delay on the plasma by hooking up a PC that runs a program with timer, hopefully with ms and try taking a picture with a camera of both the PC and the Plasma. Please share with us the delay encountered. If you are looking for a cheap Plasma, you could probably go to a place like BestBuy with a laptop and measure the delay of individual systems. The sales guy will love you, as long as you buy something in the end.

I Measured it by sending a mono sound (drums) trough both a speaker and the builtin speakers of the plasma... it sounded like two beats instead of just one :) IE Long delay.


//Robert - Who will be extremly careful choosing plasma screen in the future.

[tom9933]

I wonder if the problem is related to the quality of the scalar. I know in the HT world a lot of people run the audio through the scalar so that it will automatically delay the audio. I guess it makes sense that a plasma might have the same issue.

[mIstabone]

I have also run into echoing problems using LCD (syntax/olevia 32" LCD) speaker system. The echoing would occur when we had 3 or more sites never with Point-to-Point. Just curious what external speaker systems are people using with VSX7000/8000 series?

[robertk]

I have also run into echoing problems using LCD (syntax/olevia 32" LCD) speaker system. The echoing would occur when we had 3 or more sites never with Point-to-Point. Just curious what external speaker systems are people using with VSX7000/8000 series?

We use as simple speakers as possible. Prefferable ACTIVE speakers with it's own (fixed volume) amplifier.

We have used TANNOY speakers (i5 MP) and Bose speakers latly with good results.

Just keep away from anything that distorts the sound, like home-theatre systems and you should be fine.

//Robert

[tom9933]

For us it really depends on the room. In permanent rooms we typically use JBL or now Extron ceiling speakers. For smaller systems we normally use the speakers in the TV. I suppose once we start using LCD we might see more problems however I’ve not see any at this point. In our portable systems we use whatever PA system we are using for the event. In the long run an educated user/operator can go a long way. Where I’ve seen big echo problem is with the user who puts the mic in front of the speaker, then sits at the other end of the table and cranks up the speakers to the limit.
On the positive side though, with current echo canceller technology echo problems are becoming few and far between. Unless of course you do something like put the audio form the codec back into the codec :)

[bgoez]

better to keep the mics away from the speakers. Echo cancellation works much better when you eliminate problems before you start. speaker recommendation = soundtube very good sounding speakers work well with mic equipment

[AVGeek]

a little off-topic but a handy calculator for mics/speakers placement . . mainly for feedback but i've found it applies for echo cancellation limits as well:
http://www.lectrosonics.com/dl/gainc15.zip

[dnshbedi]

Has anyone had any problems with the delay of the video on plasma screens?

If we have external speakers connected the lip-sync is lost.

And some plasma screens that has speakers builtin also has a delay on the audio. So we atleast get lip-sync.

My concern in the later case is if this will interfer with the echo canceler in the VC system.

//Robert

As such we have not found this problem if speaker is connected whether they are external or internal u will not find this prob hey better u check other hard ware unit or connections.

rgds,
Dinesh

[robertk]

As such we have not found this problem if speaker is connected whether they are external or internal u will not find this prob hey better u check other hard ware unit or connections.

rgds,
Dinesh

I don't understand a word of what you are writing!!!

I suggest you read the complete thread first and try to understand what it's all about.

Regards

//Robert