So basically sub would be on one of the channels and center would be on the other? I'll have to see if there's room to have two output stages. If someone writes a kernel mod before I can throw together a circuit please let me know to which channel the sub and center are routed.

Stu

I may have read your post wrong but I thought the centre channel comprised of stripping out the differences and only allowing through what's common to both channels. Otherwise you'll get all those nicely separated signals coming from the centre of the soundstage and stuff like vocals (which are usually placed in the centre) removed completely. This is how it works isn't it?

In reply to:

---

I thought that, in a car stereo, the proper configuration for a center channel speaker is to only play the difference between the left and the right channels, not the sum of the left and right channels.

---

I don't think that's what he means. I believe he is saying that one of the channels on the I2S carries the DSP generated center and the other, the DSP generated sub.

Stu

With a center channel, is true 5.1 channel audio possible? Or would we need more than just software to decode it? Of course we would probably have the same issues as getting OGG support, unless the empeg gods smile upon us, and the player software becomes open source... I know I know... I'll never live to see the day. But would the 5.1 audio be possible?

I may have read your post wrong but I thought the centre channel comprised of stripping out the differences and only allowing through what's common to both channels.

Okay, I think that's what I meant.

If you summed the two channels, you'd get everything as if it were a mono signal (including things only on the left channel for instance), so that's not what you want.

If you subtracted the two channels, you'd get only the stuff that was unique to the outside channels, which is also not what you want.

I don't know the exact mathematical operation that would result in "only the center channel sounds appear in the center", but simply summing the two channels would suck. Does the DSP simply sum the two channels to get the center channel, or does it Do The Right Thing?

With a center channel, is true 5.1 channel audio possible?

Negative.

5.1 audio requires a special kind of digital datastream which can't be read or decoded by the stereo-CD-player-targeted DSP. I don't even think you could do it with the digital output board because all the 5.1 decoders expect a different samplerate than what our player delivers (48 instead of 44 if I recall correctly).

Not even matrixed Dolby Surround (4 channel: L, R, Center, RearSurround) could be done, because there's no Surround decoder on the DSP chip, and its only input is stereo, so you couldn't do Dolby decoding in software either because there would be no place to send the other two channels.

In reply to:

---

I don't know the exact mathematical operation that...

---

Hmmm... I've been trying to get my head around that one too .

Thats what I was afraid of. Thanks for the quick reply!

Hmmm... I've been trying to get my head around that one too.

Well, let's see. The way I'd do it (I think) is...

Phase-invert one of the channels and subtract it from the other channel. This will give you a track that's only the outside two channels. Then take THAT and subtract it from a mono-summed version of the original signal. The result is just the stuff that's common to both channels.

Note that the first half of the equation (phase-invert and subtract) is how most of the "vocal eliminators" work. They assume that the voice is going to be mono center, and the instruments will be panned L and R.

5.1 audio requires a special kind of digital datastream which can't be read or decoded by the stereo-CD-player-targeted DSP. ...

Not to mention that mp3 and wav only support two channels. Ogg Vorbis can support more, but no current implementation does.

Page 108 of the 7705 manual (as it is noted in the PDF itself) covers the issue of how the center and sub signals are derived. There is some sort of illustration on that page but it won't load it properly, so I can't really find the answer, but it does say that the "the center signal is derived from the sum signals of the front channels just in front of the Soft Audio Mute coefficient." The sub signal is the sum of all four channels and is output on the left portion of the I2S secondary, while the center is output on the right portion of the secondary I2S.

Stu

I'm not sure that would work. If you phase-invert one of the channels and then subtract it from the other one aren't you just going to double the amplitude of the bit you're trying to get rid of? ie you've got a point on the waveform which has a value of 0.3 (we won't worry about units they're irrelevant) if you invert this you get -0.3. If you take this away from the corresponding 0.3 on the other channel you'll get 0.3 - -0.3 = 0.6
It's getting too late for all this brainwork (it's 02:25 here) I'm off to bed. Hopefully someone will have figured it out by the time I wake up. Then we'll all say how obvious it is .

Yeah, I have seen Dolby Digital sound files however.. some came on my CD with my Sound Blaster Live! So there is the format out there. I don't remember off the top of my head their extension, but I had to use the Creative PlayCenter to listen to them.

The Mathematical term is (I believe) The "Intersection" of the two channels. The symbol is an upside-down "U".

Marius (Escort Cab + Mark II)

My point was that the empeg doesn't support any formats that have more than two channels, so it's a rather moot point.

Is anyone able to properly view the entire 7705 manual? I keep getting an error on page 108 (among others) regarding MSDRAW which is preventing me from seeing the figures explaining the sub and center channel functionality.

Stu

The error I get on that page is not related to MSDRAW.

It says it cannot find the FONT named "MS Line Draw". I suspect you only need to locate a copy of that TrueType font on the net somewhere and install it on your system.

Yeah I just checked, it's MSLINEDRAW. That cleared up the problem. Thanks

Stu

For others missing the MS LIneDraw font... I have attached a copy to this post...

Stu:
I assume you are currently using IISOUT1 (pin 34) for your digital card. Do you know if IISOUT2 (pin 35) connects to anything external on the DSP?

It does, it connects to the IIC output header for just this sort of reason.

The only design flaw does appear to be that we forgot to put the MCLK (256fs) on the same header. Oops.

Hugo

So Hugo...

Stu does connect IISOUT1 to his digital out board

OR

There is a connection for IISOUT2 on the header

OR

both?

Shonky, could you please post another copy of this file? When the BBS move happened, a couple of the most recent file attachments did not make the move, I assume yours is one of them.

Unless the resulting ZIP file was over 100k, in which case, the attachment will never work.

Here it is.

Both IISOUT1 and IISOUT2 pins on the DSP are connected by traces leading to an area marked as "IIS" on the Empeg motherboard. Besides IIS1 and IIS2 (actually SDATA signals), this output header contains a pad with LRCLK, SCLK, and ground. I believe Hugo meant that the MCK pad (the 11.2896 MHz oscillator signal which is 256 times the 44.1 KHz sampling rate of the Empeg hardware) was inadvertently omitted. from this "IIS" header. That header is not labeled "IIC" despite what Hugo says. The digital out board has a ribbon cable which connects to IISOUT1 as well as LRCLK, SCLK (we get ground from a screw rather than the ground pad because it was tough to solder to), and directly to one side of the 11.2896 MHz crystal. The sub out board would use IISOUT2 rather than IISOUT1, but all other connections would be the same.

I have attached a picture from our soon to be completed manual to illustrate.

Stu

Sorry, brain fade. I did mean IIS not IIC...

Hugo

OK, Well I've done the code for subwoofer and centre channel output.

Subwoofer has a 40, 60, 80, 100 or 120 Hz 4th order filter with gain from -20 to +24db. Centre channel has a gain of -20 to +24db.

So Stu, I've done the software, have you got any hardware? ( just jokes). I am going to build that hardware we discussed to test it hopefully this weekend.

I've just been looking at the empeg CPU card. I hope your pictures aren't as cropped in the install manuals. The part showing the crystal can be misleading since the crystal can be soldered to the board either way around. Only if you look very very carefully at the vias can you make out which way is up. My board happens to have the crystal the other way around. A less cropped view will obviously show which side is which with the DSP and the 6MHz crystal on either side.

Also are you sure that side of the crystal you are connecting to is the OSC_OUT (pin 64) side? i.e. is it the side driven by the internal oscillator. I'm not saying you're wrong, just check you have it right.

I hope your pictures aren't as cropped in the install manuals.

No, what I posted was about 3% of the frame (I told you they were high-res ) The manual will show the full frame to make it easy for people to get their bearings.

Also are you sure that side of the crystal you are connecting to is the OSC_OUT (pin 64) side?

I never traced it out. I tried both sides actually, in an effort to avoid using a buffer, but I got the same results from either side, so I didn't really worry about it. After adding the buffer circuitry, the output from the digital out board is rock solid, so I don't think there is anything wrong with my approach.

Stu

Don't they perform something similar to create a center channel in the "old" Dolby Pro Logic : (L+R)-(L-R) ? I beleive this is what Jane is referring to with the mathmatical intersection (sum minus the difference).

Hi there.

I really like the idea of adjusting the sub output from the empeg - I'm sick of diving behind the back seats of my car to tweak the sub amp whilst waiting at traffic lights and stuff.

I think I'd like to buy one of these (probably as a kit, coz I like a challenge) but I'm wondering about the acutal connections from the empeg to the outside world....

Are there are spare pins on the car sled connector to allow an analog bass output to be passed through? Any kind of trailing leads flowing from the back of the empeg are going to be a real pain to connect and would probably foul when trying to dock the empeg...

Anybody have any thoughts on this? Are there any relatively unimportant pins that could be reused, or would it involve fitting a new custom connector to the empeg & the sled?

Looking forward to hearing more news on this project....

Cheers,

Andrew.