I like this project idea, but I don't see why a PC is better necessarily than a DSP, aside from ease of development. I think you'll find that it's quite difficult to get multichannel digital audio in and out of a PC. For example, 2-channel input and 4-channel output is already beyond the capabilities of nearly any personal computer. You'll be needing ADAT or FireWire, and good luck prying the datasheets out of the stingy hands of Alesis and Texas Instruments.
Or, you can build your own interface hardware.
Or, you can build your own interface hardware.
I'm happy to see links to actual products, but all I've seen are cards with 2-channel PCM in and out, even giant breakout boxes like the Delta 1010 provide only stereo digital i/o. Of the card supporting multichannel i/o, they are all using AC3, which is obviously not appropriate for this application.
The multichannel stuff from RME uses ADAT, which is what I was pointing out earilier.
Links to actual products with multichannel S/PDIF outputs appreciated.
The multichannel stuff from RME uses ADAT, which is what I was pointing out earilier.
Links to actual products with multichannel S/PDIF outputs appreciated.
jwb said:
I agree that I have not seen any such products. There are, however, a multitude of firewire breakout boxes that provide many channels of analog output. If you are content using their DACs, this is a solution. I would expect these products to have very little, if any, relative latency between the channels.
Another solution would be to use multiple SPDIF outputs from your computer (many sound cards) and use multiple outboard DACs. This setup would certainly have more relative latency between the sound channels, but as long as you can get them to all have constant latency (negligible jitter), the difference would already be taken care of digitally.
You could also use a DSP or microcontroller to build your own firewire -> many DSP's breakout box, but that would be cost / time investment prohibitive for most people.
I admit that at least for me, all of these solutions are cost prohibitive at the moment. However, this is a function of current IC availability than of the basic idea.
I should note that for me, part of the problem is that I use a laptop and have to do everything externally. PCI sound cards are cheaper, and for many people adding a few of them might not be cost prohibitive. You get two channels per SPDIF, so for a pair of DIY two-way speakers you'd only need a pair of sound cards to have a foundation for your automatic room correction + automatic digital crossovers.
BTW, Has anyone here used simulink? It would be tempting to create a set of DSP application building blocks and use Jack to connect them all together in a similar paradigm. I suspect the CPU overhead for all of the extra signals would kill, and I don't know how portable Jack setups are, but it is tempting nonetheless. It also seems like Jack handles feedback loops fairly arbitrarily, which would not be ideal for this sort of thing.
jwb said:
Lynx AES-16, or any LynxTwo series card with the LS-AES addition. Technically, this is AES/EBU rather than spdif, but if you use a matching xformer it *should* work with most consumer gear. The AES-16 provides up to 16 channels (ie 8 spdif stereo signals), which is enough for any but the most esoteric application. Runs about $700 by the time you add the required breakout cabling.
www.lynxstudio.com
Creative's Live! and Audigy line of soundcards (those based on the emu10k* processors) have S/PDIF input and multiple S/PDIF outputs. All have four stereo S/PDIF outputs, but the number of them that are enabled depend on the model of the soundcard: the original Live! with 4-channel, so two are enabled. There is a firmware hack to enable a 3rd. The Live!5.1 has three enabled. The Audigy 7.1 has all four S/PDIF outputs enabled (8 channels).
Two caveats: the S/PDIF outputs are TTL level and need to be converted to either proper 75 Ohm 2 Vpp signals, or connected to optical TOSLINK transmitters. And, they are limited to 48 kHz only (specifically, not 44.1). Any digital input also goes through ASRC to 48 kHz.
ACXO
Although I don't think its exactly what you're looking for (and I'm not sure if it's been mentioned yet) here is the ACXO project. I've been wanting to try this package out (to replace my Behringer room EQ), it has room EQ and digital crossover in one package. If it works it would be really cool.
Although I don't think its exactly what you're looking for (and I'm not sure if it's been mentioned yet) here is the ACXO project. I've been wanting to try this package out (to replace my Behringer room EQ), it has room EQ and digital crossover in one package. If it works it would be really cool.
I'm confused too. SPDIF is constrained by standard, so you will never see more than stereo 192k over single SPDIF. I'm not sure how many channels of raw PCM are allowed at 96K, but at 48K all 8 are available. Whether they are used for AC3 or anything else is a matter of software..jwb said:
There are many cards that offer multiple SPDIF output ports that can be synced.
Take EMU 1820 - at the very least, 6 channels out at 96K each over 3 SPDIF ports. If you can live with 48K, you can output 8 channels over all 3 SPDIF output ports - 24 channels.
http://www.emu.com/products/product.asp?product=2210&category=754&maincategory=754&nav=spec
$400-500
It has AES/EBU support, ADAT, word clock io.
And 1820M has very decent analog part too. No need to use all external DACs. I don't imagine anyone using some exotic DAC for woofers.
wimms said:I'm confused too. SPDIF is constrained by standard, so you will never see more than stereo 192k over single SPDIF. I'm not sure how many channels of raw PCM are allowed at 96K, but at 48K all 8 are available. Whether they are used for AC3 or anything else is a matter of software..
Yes, you're confused. PCM over SPDIF is stereo. period, no exceptions. Things like DD/AC3 and DTS use the same physical transport layer, but are compressed and framed differently, so they aren't really the same thing. Two uncompressed pcm channels is all you'll ever get out of a single spdif link.
Not quite accurate. I have an 1820M, and there really are only 2 independent spdif outputs - the optical and coax ports on the card share the same signal, so you get one on the card and one on the dock. As above, changing the sample rate doesn't do anything to the spdif channel count.
I think you're getting confused between spdif and ADAT. ADAT was originaly an 8 channel/48k interface. However, various 'bit splitting' approaches have been layered on top to allow trading off channels for higher sample rates. So, you can get 4 channels at 96k, or in some cases 2 channels at 192 - all of these have the same 'bit rate' but different interpretations, and not all the approaches are compatible with each other (eg EMU's 192k ADAT mode seems to be proprietary).
If you're so sure. Original SPDIF had no specification for 192k PCM either. So it has been extended. I relied on spdif frame format description, as it has specific bit to signal 4-channel PCM data, sometimes called DO-4 iirc.dwk123 said:
http://www.epanorama.net/documents/audio/spdif.html
I can't seem to find the damned files where 192k extensions were covered. I just remember that spdif was basically constrained by max bitrate, and 48k subchannel, not much else. But memory may fail me.
though http://www.ebu.ch/CMSimages/en/tec_doc_t3250-2004_tcm6-12767.pdf describes multichannel pretty clearly. AES/EBU or spdif with pro bit. So I assume from your assertion that this is basically not implemented or easy to use.
Btw, check this: http://www.bionicfx.com/
I've always thought how hard it is to utilise Graphics cards for DSP purposes. 40Gflops DSP with 256M gddr3 ram - how's that?
I've always thought how hard it is to utilise Graphics cards for DSP purposes. 40Gflops DSP with 256M gddr3 ram - how's that?
WOW ! GREAT GUYS !!
This is it !!
This is the thing, which I have been searching for.
Its really nice to meet like minded people like me. Great efforts.
In my view:
* Sounds will be stored in PC – could be compressed or uncompressed.
* The new PCs come up with Multi channel, flexible Hi-definition Audio Cards (Sound Cards). Yes, its flexible, Input can function as out put & vice versa.
Refer Intel’s Mother board 915GVBW(L) which I had earlier & Currently I have ASUS M2NPV-VM.
http://www.newegg.com/Product/Product.asp?Item=N82E16813131014
Have Logitech Speakers
http://www.logitech.com/index.cfm/products/details/IN/EN,CRID=2173,CONTENTID=11459
Some how, the Audio reproduction Clarity doesn’t satisfy me.
Every sub-system has its own frequency response curve & apart from this, the room acoustics too adds up to the worst.
Please note even though my PC’s primary purpose is not listening to music, but I play music daily in it.
I don’t Know how many of you are aware of SonicFocus
http://www.sonicfocus.com/
My earlier Mother Board used it to enhance the DEPTH & improved sonic stage from MP3 files. But not all motherboards support it. (You may give it a try)
Also I use XMPlay (Xmplay from http://www.un4seen.com/ ) to play the MP3 files, which is a light weight, No installation, Low system Resources player.
Do explore it with the Equalizer set it to a Smile (I like it that way) & Reverb ON in the DSP settings.
Good thing about XMPlay is you can use all the WinAmp’s DSPs.
Well currently, what I did is:
1: used DSSF3 from http://www.ymec.com/ (Free Trial) to find out My Room’s response for my Logitech speaker hooked up to the PC.
2. In trial & error method adjusted a 120Band Equalizer plug-in for winamp & used the same in Xmplay.
3. Now, if I generate a sweep sinwave from PC & its hooked up speaker, I nearly get a flat FFT curve. But it took me 4 weeks time to adjust it.
Hence I strongly felt the need of Auto Room Tuner / MultiEQ /DRC.
Let me know if any of you have located such program
Greetings,
Swami.
This is it !!
This is the thing, which I have been searching for.
Its really nice to meet like minded people like me. Great efforts.
In my view:
* Sounds will be stored in PC – could be compressed or uncompressed.
* The new PCs come up with Multi channel, flexible Hi-definition Audio Cards (Sound Cards). Yes, its flexible, Input can function as out put & vice versa.
Refer Intel’s Mother board 915GVBW(L) which I had earlier & Currently I have ASUS M2NPV-VM.
http://www.newegg.com/Product/Product.asp?Item=N82E16813131014
Have Logitech Speakers
http://www.logitech.com/index.cfm/products/details/IN/EN,CRID=2173,CONTENTID=11459
Some how, the Audio reproduction Clarity doesn’t satisfy me.
Every sub-system has its own frequency response curve & apart from this, the room acoustics too adds up to the worst.
Please note even though my PC’s primary purpose is not listening to music, but I play music daily in it.
I don’t Know how many of you are aware of SonicFocus
http://www.sonicfocus.com/
My earlier Mother Board used it to enhance the DEPTH & improved sonic stage from MP3 files. But not all motherboards support it. (You may give it a try)
Also I use XMPlay (Xmplay from http://www.un4seen.com/ ) to play the MP3 files, which is a light weight, No installation, Low system Resources player.
Do explore it with the Equalizer set it to a Smile (I like it that way) & Reverb ON in the DSP settings.
Good thing about XMPlay is you can use all the WinAmp’s DSPs.
Well currently, what I did is:
1: used DSSF3 from http://www.ymec.com/ (Free Trial) to find out My Room’s response for my Logitech speaker hooked up to the PC.
2. In trial & error method adjusted a 120Band Equalizer plug-in for winamp & used the same in Xmplay.
3. Now, if I generate a sweep sinwave from PC & its hooked up speaker, I nearly get a flat FFT curve. But it took me 4 weeks time to adjust it.
Hence I strongly felt the need of Auto Room Tuner / MultiEQ /DRC.
Let me know if any of you have located such program
Greetings,
Swami.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.