Sign in with
Sign up | Sign in
Your question

AUDIGY 2 ZS bad CMSS in 7.1

Last response: in Components
Share
February 17, 2006 2:27:18 PM

Hi There,

1) I have the Audigy 2 ZS soundcard accompanied with Creative Gigaworks S750 speakers (btw. they ROCK!!!). All drivers for audigy are up to date. I have concerns regarding the Audigy CMSS function, that should make the mp3´s sound from ALL of the 7 speakers. The problem is that all the speakers sound OK except the rears, that have only about 10% of the volume the other speakers have (the games work absolutely properly in surround mode, in EAX for ex). And the problem is not in the surround mixer, all the speakers are set to 100%, the rears too... The CMSS is set to the SECOND option (makes the best results i think) and the speaker configuration is set to 7.1. And there is a slider in the right part, by which you can adjust the FADER, in other words you can adjust how much volume the front speakers will have and how much rear. And here is the problem, when i move the slider down, naturally the REAR speakers should go louder, but they remain the same and the SIDE speakers go louder instead, so there is NO way to increase the volume of rear speakers when listening the music (i use winamp 5, if that matters). All this changes (and works properly) only when i change the speakers set to 5.1 in the THX console, but that has no use for me (the sides are completely OFF in that case), as i have 7.1 speakers and so, naturally, i want 7.1 setup. And when you dont hear the rear speakers, you simply absolutely dont have the feel of surround CMSS sound and thats what worries me.

So id like to ask whether you have the same problem or you know a solution that could possibly help me or other ppl with such problem. Thanks in advance!

2) Second question, do you know a program that uses a microphone to setup your speakers? That program should create a sound from each speaker and, using the input from the microphone, set up the dB values for each channel i think (for example make rear speakers louder if they are too far behind). Because as i know, the manual setup console that creative labs uses doesnt work as it resets the values after each restart.

3) Maybe a little off topic, but worth trying: Do you know a program that can normalize your mp3 songs (it would be the best if it had some sort of batch processing, where you could select a directory and the program would normalize all the mp3´s in that directory to a certain level)? Because i want to get rid of the well known problem: when you copy some mp3´s to your mp3 player and play them, some of the songs play a lot louder, some have relatively normal volume and some play a bit silent on the other hand... and its impossible to adjust (normalize) EACH song in some sound editor...

Really thanks in advance for ANY help!!!

More about : audigy bad cmss

February 18, 2006 9:04:05 AM

1) CMSS is a matrixing system for "pseudo" surround. This involves taking the transients and putting them into the rear channels for "ambience", which are very quiet signals. CMSS chases Pro Logic's tail in an attempt to bring "matrixed" surround, and there's not much you can put into the rears from the front soundstage that can convincingly sound "maybe" as room reflections, and not just play the front soundstage from all channels, and create nulls from distructive interferrence.

Anway, it seems like this just isn't working properly. For you, you might want to consider swapping the surrounds with the rears, if the wire lengths permit?

2) http://homepage.ntlworld.com/john.mulcahy/roomeq/index....

Room EQ Wizard, as per the link, or TrueRTA. You probably need a microphone like the Radioshack microphone.

If you do use that, the correction values are:

10hz +20db
12.5hz +16.5db
16hz +11.5db
20hz +7.5db
25hz +5.0db
31.5hz +3.0db
40hz +2.5db
50hz +1.5db
63hz +1.5db
80hz +1.5db
100hz +2.0db
125hz +0.5db
60hz -0.5db
1KHz +0.0db

3) I think Foobar can normalize mp3s.
February 27, 2006 7:24:12 AM

Thank you very much for help astrallite! I have some other questions:

1) What exactly you meant by those correction values? Are they somehow connected with the Room EQ Wizard or TrueRTA?

2) Yes, the fact that rears sound silent makes sense as the CMSS plays only small amount of frequencies there. But here, a new question comes up. Is there any way to play the mp3 songs MONO from ALL the 7 speakers without any surround effect, but without the problems with silent rears, while that way all the speakers would play at 100% (you have mentioned it too: "play the front soundstage from all channels")? Because i think that would sound MUCH more "surroundish" than using the CMSS, that on one hand tries to divide the frequencies among all the speakers, but the only result is that there is no surround feeling because of those silent rears...and the rears are crucial for creating a surround feeling while those are the only speakers that are behind you.

And another similar question, when you use the aux input (located on the control unit of the speakers), why the Gigaworks play 2.1 only and cant get rid of the STEREO 2.1 sound and distribute the signal to ALL 7 speakers MONO instead? Is it because the circuitry cant do that or the amplifier has problems distributing the sound? But firstly, WHY should there be such trivial problems distributing the sound which is in fact just a current??? As i know there are for ex. the Y shaped hubs that make two 3,5 jacks out of one (good when for ex. two people want to listen to one mp3 player at the same time and they have headphones only). So why shouldn´t the complex circuitry and amplifier do the job that can be done by some trivial piece of plastic and metal (i mean the Y hub)?
Related resources
February 27, 2006 8:48:21 AM

1) The meter isn't the most accurate, and more accurate ones aren't particularly affordable

2) A recording that was made in stereo (2 mics) is meant to be played back in 2-channel for accurate soundstage and instrumental/vocal location. Playing that (or a mono source) out in multiple channels, I'm not even going to speculate what the true effect is, but not only will you get lobing (destructive interferrence) but the effect of having the same band play from multiple locations around you has a dubious effect at best, IMO.

Most people with decent equipment don't even consider mirroring stereo; it's stereo or buy real surround recordings.

The Gigaworks I suspect internally is just a device with no real sound processing. It's wired so input = output. On a media player like Foobar you can use DSPs like (downmix stereo into mono) which effectively just plays the left (mono) channel. There are cable splitters you can get at Radioshack that will do effectively the same thing--connect a stereo mini to RCA cable to your sound card, then connect it to a RCA to stereo mini adaptor to the speakers, but attaching only the white (mono) cable instead of both (the red, right channel) for example. Anyway, mono sounds horrible to me, you get no sense of space at all, and you miss alot of frequencies because (no surprise here)--destructive interferrence when multiple sources play identical frequencies.

I played around with stuff like stereo mirroring, CMSS, and PL2 back when I had computer speakers, and honestly they just sounded horrible in my opinion. But your preferences might be different from mine.
February 27, 2006 6:38:10 PM

Quote:
Hi There,

1) [..] I have concerns regarding the Audigy CMSS function, that should make the mp3´s sound from ALL of the 7 speakers. The problem is that all the speakers sound OK except the rears, that have only about 10% of the volume the other speakers have (the games work absolutely properly in surround mode, in EAX for ex).[..]
So id like to ask whether you have the same problem or you know a solution that could possibly help me or other ppl with such problem. Thanks in advance!

This behavior is by design. Creative designed it to behave this way, therefore, nothing's "wrong" and nothing can be done to "fix" it, as it's not a "problem". However, why exactly Creative designed it this way is beyond me. If you really want to have both rear and surround speakers in use, you can set your speaker output to 5.1, and set the console on the Gigaworks themselves to upmix 5.1 into 7.1. This will play the rear channel sthrough both the rear and side speakers simultaneously.

Quote:
2) Second question, do you know a program that uses a microphone to setup your speakers? [..] Because as i know, the manual setup console that creative labs uses doesnt work as it resets the values after each restart.

No, they don't, at least they're not supposed to. If they do, then something is wrong.

Quote:

3) Maybe a little off topic, but worth trying: Do you know a program that can normalize your mp3 songs (it would be the best if it had some sort of batch processing, where you could select a directory and the program would normalize all the mp3´s in that directory to a certain level)? [..]

Various plugins for Winamp can normalize, but if you want to normalize your MP3's themselves you will have to re-encode them all. This takes a lot of time, and causes losses in quality as well. If you want to do this, virtually any common MP3 encoder has the function(s) built in. If you want a standalone program to do it, try on Google - http://www.google.co.uk/search?hl=en&q=Normalize+MP3+Batch&btnG=Google+Search&meta=.

Quote:

[..]
2) Yes, the fact that rears sound silent makes sense as the CMSS plays only small amount of frequencies there. But here, a new question comes up. Is there any way to play the mp3 songs MONO from ALL the 7 speakers without any surround effect, but without the problems with silent rears, while that way all the speakers would play at 100% (you have mentioned it too: "play the front soundstage from all channels")?

Yes. Set CMSS to "Stereo Surround". That's the third option in the CMSS modes list.
"In multi speaker setups, the stereo signal is played over each speaker pair (stereo x2 or stereo x3)"

Quote:

Because i think that would sound MUCH more "surroundish"

Actually I think that it sounds rubbish, but that's just my personal opinion.

Quote:
And another similar question, when you use the aux input (located on the control unit of the speakers), why the Gigaworks play 2.1 only and cant get rid of the STEREO 2.1 sound and distribute the signal to ALL 7 speakers MONO instead? Is it because the circuitry cant do that or the amplifier has problems distributing the sound? But firstly, WHY should there be such trivial problems distributing the sound which is in fact just a current??? As i know there are for ex. the Y shaped hubs that make two 3,5 jacks out of one (good when for ex. two people want to listen to one mp3 player at the same time and they have headphones only). So why shouldn´t the complex circuitry and amplifier do the job that can be done by some trivial piece of plastic and metal (i mean the Y hub)?

There are several reasons for this.
1) They don't want to. As you can already tell from the lack of decoder in the speakers, they were designed to minimize the amount of logic circuitry incorporated.
2) It would add costs to the speakers for something that most people couldn't care less about.
3) Creative prefer to promote (and assume that you have access to) their "better" proprietary CMSS technology.
4) It can be done, as you should already know from the 5.1 and 6.1 => 7.1 upmix function on the speakers.
5) As I said, I, and most people I know think that duplicating the stereo signal across all speakers sounds rubbish
6) One would normally assume that by using a stereo input, you're inputting stereo sound. As astrallite said, stereo sound was meant to be heard in stereo.
March 1, 2006 11:58:06 AM

Hello. Thank you again for your answers. Here is my reply (sorry for big amount of brackets):

1) qasdfdsaq surely had the point, setting the speakers set to 5.1 in the console and then upmixing that sound by gigaworks to 7.1 does the job quite well.

2) Yes, i think you are absolutely right that MONO output from 7 channels together can sound rubbish, bud i couldnt know that as i never managed to play my songs that way on my gigaworks 8D. BUT...then another question comes up: you said that stereo sound is meant to be played as stereo, therefore from two speakers only (and subwoofer voluntarily). But then i recalled listening to the music in my car. There, i have the factory standard radio set (not the basic one but one level higher) that consists of 10 Harman Kardon speakers (2x treble front speakers, 2x mid-range speakers in the front doors, 2x mid + treble REAR speakers and i think 2x bass speakers, but here i am not sure if they are bass, as i havent found them yet 8D) and an aux input. And when i connect my portable mp3 player into the AUX with simple 3,5 jack (or play a CD or radio), the sound is SUPERB surround (that means the sound goes from ALL the speakers) just from stereo signal and nothing sounds rubbish. You can even adjust the FADER, so move the sound from the front to the back of the car interior and vice versa (and therefore setting the soundstage perfectly into the center of the car interior, when you can´t recognize whether that sound goes from the front or from behind and thats the way its meant to be i think) and everything works perfectly (in contrast to the CMSS slider which i mentioned before). And whats important, also the rears play at normal volume (and they shouldnt as astrallite said that the rears get almost unhearable frequencies, but this seems to confute that argument), not just 10% like the gigaworks ones in CMSS mode. So my question is, how is it possible that my car´s radio (and not just my car´s) can engage even 10 speakers just with stereo source (what is important with the rears sounding perfectly OK and overall no rubbish sound made) and my high-end PC speakers with high-end sound card (surely more complex than circuitry in my car´s radio which even doesnt have the advantage of some software that it can exploit) made just for that set of speakers can´t engage 7 speakers only (compared to 10 in my car).
March 1, 2006 3:03:33 PM

Basically, 5.1 computer audio is commonly found, and was intended to exist in this way, profesionnaly mastered stereo audio was intended to be reproduced in stereo. Therefore, it's an issue of "political correctness" in audio reproduction, the CMSS function to upmix is basically a shoddy extra which wasn't given much thought, designed for people bent on using all speakers to do so.

A large number of your cars speakers are likely to be linked, so you're not likely to have 10 individual channels. Nevertheless, they were designed to be fed a stereo signal as 5.1 car audio is rare. Also, accurate reproduction is usually not the major concern in an automobile. Thus, your car system is designed to "sound good" in a car, while your PC system is designed to "sound exactly the way it was produced".

Lastly, again, the CMSS setting was by choice by Creative. They probably thought it sounded better compared to using all seven speakers equally. Also, arguably, your car does not have 7.1 surround. It's a 4-point system with 2 bass units, which could well be linked as previously mentioned. This would be equivalent to using your gigaworks as a 4.1 system, or, at best, 5.1. Again, there's nothing "wrong" with what CMSS can or can't do, it does everything it's told to and was designed to do.

I'm not advocating CMSS is perfect, and I'm not saying your car's audio is bad in any way. I'm saying CMSS was designed this way, for reasons that you appear not to agree with, but there's nothing "wrong" with it and it's not for reasons like "it can't" do 7.1. Would you say there's something "wrong" with your car if it couldn't do 100km/h in reverse?
March 3, 2006 10:28:56 AM

Hi

Thanks for reply.

Yes, it seems that gigaworks werent designed to produce just stereo sound, but they surely could count with that as mp3´s are widely used...

Anyway, qasdfdsaq, thanks for your help regarding the mp3 normalization. I found a program called mp3doctor and thats just the thing ive been looking for! Perfectly simple yet effective. But you mentioned that reencoding the mp3s causes losses in quality.

Is that true also when you normalize those mp3s with standalone program like the mp3 doctor?

Because i just tried that and the normalized mp3s seem to have absolutely the same quality as original and i couldnt hear any distortion, losses in quality or sound damage caused by the normalization. Maybe you meant the distortion which can occur when you increase the volume of that song by normalization and the peaks will be out of range, but thats not the only way normalization can be done and for example mp3 doctor can use a method that prevents that type of distortion.
March 3, 2006 10:57:45 AM

It quite often depends on how the program goes about normalizing. Some for example, might just read the file and set a ReplayGain or similar tag somewhere to tell the output program to adjust the volume when playing it.

Others, will (and IMO this is the "proper" way of doing it) read the file, and adjust the waveform to compensate for changes in amplitude, and then re-write it out to the MP3. This involves re-encoding the file, and thus a loss of quality. Some people argue that it's very difficult to notice, some people claim it is obvious, your mileage may vary. Put simply, the loss for each encoding cycle is slight, but it most definately is there. MP3 itself is a lossy format so quality is lost simply by using MP3 in the first place. If you start off with a poor quality file, it will only get poorer, possibly making it easier to notice.

In my personal opinion, if your original file was encoded at above 192kbps then the quality of the normalized file should be OK for general use. It'll be difficult for most people to notice the difference without listening carefully and through good audio equipment.
March 5, 2006 7:46:42 AM

If you normalize you are reducing the dynamic range of the song. Most music (not pop obviously) will have a certain volume range, from say 40db to 100db within the song. Classical music shows this pretty well.

If you normalize, you are raising the minimum volume level. However, since there is a limitation to the peak volume level (depends on the bitdepth, but 144db is a common one), the effect of raising the minimum volume is the same as "compressing" the song. Increase the minimum volume "too much" and you'll get a song that sounds like a continuous monotone (which poorly recorded heavy metal songs will often sound like).
!