The Lynx TWO B also doesn't offer any internal loopback so no go for windows users who want full channel count with hosts such as Console, Audiomulch etc.
Will work fine with Linux and ALSA though.
Will work fine with Linux and ALSA though.
It does, but only for stereo loopback. So this may be a showw stopper for quite a few.
Possibly 4 channels of loopback with two in parallell, but of course that isn't enough for 5.1 material.
Possibly 4 channels of loopback with two in parallell, but of course that isn't enough for 5.1 material.
BerntR said:
Yep you don't get full channel count without physical routing and wastage of other I/O's on the card such as ADAT or AES/ABU. RME, MAudio and EMU cards are amongst the best for the windows IMO since these allow full internal loopback and all channels to be used.
Obviuosly I can't read this whole thread, and I don't want to. What I read at the begining was enough - I was impressed. Sounds like you know this material well.
So here is my question:
Could this techology strip off the LFE and LF signals from a movie or CD bit stream, process them and output them to just the subs while passing through an inprocessed bit stream to SPDIF? It seems that stripping off the LF from the L,R and C from a movie track would not be too hard and getting the LFE, but can this be acomplished easily enough while still passing the unprocessed signal through to the SPDIF output?
Clearly what I want to do is to process only the sub signals and pass all the rest unchanged.
So here is my question:
Could this techology strip off the LFE and LF signals from a movie or CD bit stream, process them and output them to just the subs while passing through an inprocessed bit stream to SPDIF? It seems that stripping off the LF from the L,R and C from a movie track would not be too hard and getting the LFE, but can this be acomplished easily enough while still passing the unprocessed signal through to the SPDIF output?
Clearly what I want to do is to process only the sub signals and pass all the rest unchanged.
I see no reason why you couldn't do that with a decoder plugin hosted in something like Console.
You'd have a software player which output to SPDIF, the sound driver and routing software loops this back ready to be picked up as an input for console. From there you load the plugin such as:
ffdshow Audio Decoder
Nvidia Audio Decoder
DScaler Audio Decoder
AC3 Filter
etc. etc.
All of these are capable of decoding AC3 and DTS but not the HD variants.
After you load a suitable plugin you'd split the bitstream signal creating a duplicate. One would go straight back to a suitable digital output in its original and unaltered form ready to be sent to the AV amp and the other to the decoder plugin where its split into multichannel. Since your only interested in the LFE you'd take that from the decoder and into whatever processing plugins you wanted.
Now I haven't tried what your asking and what I'm proposing so can't say how successful it is but straight away I'd imagine processing lag might be a problem - your decoding the audio in two separate boxes, your AV amp and the PC along with a bunch of processing on the LFE. If could be that you need a delay on the AV amp should you find the PC is lagging behind. This could in turn put the sound out of sync with the video and you begin to have a small nightmare on your hands. Its all solvable but its not very elegant.
A better solution is to have the PC as source for both audio and video with all decoding done within. You can setup a similar but less complicated system as above then route all the decoded/processed audio through analogue outputs to amps. The advantage here is that there's a fixed delay through just one box rather than two and everything is easily manipulated from all pass delays to FIR filtering to basic PEQ for each channel.
You'd have a software player which output to SPDIF, the sound driver and routing software loops this back ready to be picked up as an input for console. From there you load the plugin such as:
ffdshow Audio Decoder
Nvidia Audio Decoder
DScaler Audio Decoder
AC3 Filter
etc. etc.
All of these are capable of decoding AC3 and DTS but not the HD variants.
After you load a suitable plugin you'd split the bitstream signal creating a duplicate. One would go straight back to a suitable digital output in its original and unaltered form ready to be sent to the AV amp and the other to the decoder plugin where its split into multichannel. Since your only interested in the LFE you'd take that from the decoder and into whatever processing plugins you wanted.
Now I haven't tried what your asking and what I'm proposing so can't say how successful it is but straight away I'd imagine processing lag might be a problem - your decoding the audio in two separate boxes, your AV amp and the PC along with a bunch of processing on the LFE. If could be that you need a delay on the AV amp should you find the PC is lagging behind. This could in turn put the sound out of sync with the video and you begin to have a small nightmare on your hands. Its all solvable but its not very elegant.
A better solution is to have the PC as source for both audio and video with all decoding done within. You can setup a similar but less complicated system as above then route all the decoded/processed audio through analogue outputs to amps. The advantage here is that there's a fixed delay through just one box rather than two and everything is easily manipulated from all pass delays to FIR filtering to basic PEQ for each channel.
ShinOBIWAN said:
..This is exactly what I want to do. It'll be a while though. Looks like I have a lot of reading to do anyway.
edit:
Without doing much reading except from the first page like 2 years ago.. how well would something like the M-Audio delta1010 (if I'm remembering the model correctly) work? DAC is external.
BHTX said:
According to Shin, the delta1010 is a good card. However, the topic that we always return to seems is the tricky is the matter of routing. If you are fine with using a mediaplayer where your filters can live then M-Audio is OK. If you want to send all your system sounds to the card for filtering then you need the go with E-MU, RME, Lynx or Audiotrak.
I use Windows Media Center so I need to use one of the above with Console.
Re: Re: Thanks a lot for the quick response Shin!
Many thanks for that Shin! I am also curious to hear you input wether you advise using a dedicated DRC computer or do you prefer to do it in the same machine? I am currently just using one box, but could 12 channels with XO and DRC cause a problem?
ShinOBIWAN said:
Many thanks for that Shin! I am also curious to hear you input wether you advise using a dedicated DRC computer or do you prefer to do it in the same machine? I am currently just using one box, but could 12 channels with XO and DRC cause a problem?
Ninfendo said:
Thanks for the info! Kinda disappoints me tho, as I have 19" rack mount equipment, and the Delta1010 would have been a nice piece of gear to add to it. Plus, having the DAC away from the PC is definitely a plus I'm sure, but I'm also sure there are plenty of others like that most likely. Anyway, what is it that allows these big multichannel cards to be controlled by the Console software, and why can't the Delta1010 do it?
I guess I just need to read. 🙁
ShinOBIWAN said:
Thanks Sin (oops I mean Shin!)
I do want more than just the LFE, I would also need the signals below about 200 Hz from each channel.
I thought about the processing delay, but here is what I think could work. If all signals are summed and then resampled at say 1 kHz. they would automatically be LP filtered. At 1 kHz sampling, delay is insignificant and since we are talking about the subs the delay is very unlikely to be audible anyways.
JBL sells their Multi-Subwoofer processing unit for $1200. I know that this could easily be done in a PC (and better) and since I already use a PC for DVD and Blue-Ray it seems a natural. BUT, with Blue-Ray the CPU is darn near peaked out on a lot of disks, so I can't afford much processing time. Thats why I don't want to do everything in the PC. I just want to do what cannot be done in the receiver.
Also, since these signals would all be very low sample rate and subs only, the sound card could be the one in the PC - if this is possible. In the end a simple solution is a highly valuable thing, but a complex one defeats the purpose - just buy the JBL unit.
Any more help that you could give would be appreciated.
Volume control EMU-1820M + console
Hi,
I have working setup with a EMU-1820M and console for PC XO. This works great, but I would like to find a better way of regulating the volume of playback. Currently I use a surround receiver(Primare SPA20) for volume control, but I figure that it would be better to just be able to reduce the output from the EMU directly (yes, I know that this involves throwing bits away...) I would then make some cool Ucd400 multichannel amps and adjust the gain in each UCD400 to a reasonable level.
Is it possible to control some sort of attenuator in the EMU mixer application or in console via remote(IR) control? It would also be nice to have a big nice knob on the pc to turn the volume down.
regards,
Øyvin Eikeland
Hi,
I have working setup with a EMU-1820M and console for PC XO. This works great, but I would like to find a better way of regulating the volume of playback. Currently I use a surround receiver(Primare SPA20) for volume control, but I figure that it would be better to just be able to reduce the output from the EMU directly (yes, I know that this involves throwing bits away...) I would then make some cool Ucd400 multichannel amps and adjust the gain in each UCD400 to a reasonable level.
Is it possible to control some sort of attenuator in the EMU mixer application or in console via remote(IR) control? It would also be nice to have a big nice knob on the pc to turn the volume down.
regards,
Øyvin Eikeland
gedlee said:
No problem, any combination of signal routing, LP or HP is possible once the audio is within Console and decoded.
Delay can be minimised by using IIR filtering and avoiding FIR. There's also a sample delay caused by the ASIO buffer and the processing delay from any plugins you use. Some sound cards have delays as they route the audio through drivers to outputs too.
What matters is how many ms delay you have at the output of the PC compared to your AV amp. I think it will be either undetectable or very small if using IIR. FIR is where you need to employ video delay because of a ~500ms processing time.
I also have Blueray drive and PowerDVD for playback. You can now get GPU assisted video decoding of H.264 codec. It lowers CPU usage tremendously on my system from around 80% to 20%. I use an ATI 4870 graphics card with 3Ghz Intel quad core.
You'd be surprised just how little processing time is needed for the audio processing. I see CPU usage between less than 1% upto 15% whilst playing back audio only. That's with 8 channels of FIR filters and delays needed to build up the crossovers for a 3.5way system. If you use IIR and less channels then its less than that. Multicore processors are a good thing too for obvious reasons.
The PC is more complicated than the rigid dedicated units created to serve one purpose such as JBL, DCX, DEQX etc. Its the nature of the beast but given a little time the PC , even when seemingly complicated at first, is quite easy and certainly more flexible.
BHTX said:
Its all so long ago but didn't someone in this thread mention that the 1010 could do internal loopback? Maybe I'm imagining it because I had a quick look and couldn't find anything.
Its all decided by if a card has a feature called internal loopback, allow me to explain...
With these pro audio cards there's a bunch of internal playback and record channels within the drivers and then there's the physical I/O on the card itself.
All the software being talked about here uses ASIO to shuffle audio around between playback application, drivers and so on. Problem with ASIO is that each secures a channel and doesn't let any other application touch it eg. Winamp is assigned a playback channel and you want to take that audio and process it in another application. This is impossible because ASIO won't allow it.
You could output the sound on a physical output and then bring it straight back in using an input. From there you could do whatever you like with it. BUT your wasting an input and output doing so.
This is where internal loopback comes in real handy. What it does is take an ASIO playback channel such as the winamp example above and create a duplicate in a new ASIO record channel so that any software can now access it and all without conflict or using up the physical I/O on the card.
Hope that makes sense.
There are multi client ASIO cards/drivers. My RME Hammerfall is multiclient and so is my Presonus Firebox. You get errors if you try to jam different sampling rate streams onto the card simultaniously, but once they all march to the same beat, I can send audio to the same output from two different apps.
Recently I have been playing with Jack for Windows.
It's an internal audio patching application (originally for Linux) that lets you route audio between many sources and apps. It's been fairly solid for me on Windows XP.
Once I have a couple of days to write it up I will do a page on it.
It's a good solution for cards that can't route streams internally.
Recently I have been playing with Jack for Windows.
It's an internal audio patching application (originally for Linux) that lets you route audio between many sources and apps. It's been fairly solid for me on Windows XP.
Once I have a couple of days to write it up I will do a page on it.
It's a good solution for cards that can't route streams internally.
Thunau said:
Man, scooped again - when I saw the discussion on routing, I though I could add something new 🙂. I use jack on linux, and didn't realize jackmp/jack2 had progressed as far as it had. I grabbed the new 1.90 release last night but haven't tried it out yet.
IMHO this looks like it may be a huge advance in routing on Windows. It works by creating a virtual ASIO driver, so asio enabled apps simply connect into the jack fabric, and then you route everything within jack. I don't know how MME/KS inputs would be handled in this case, though.
Jan - have you tried netjack2 between boxes yet? I'm really hoping this works OK, so I can bridge my mpd/brutefir setup over to an ASIO playback on Windows, as my Emu 1820m isn't fully supported under Linux.
Thunau said:
Looks good if your card doesn't already have loopback. Looks like there's no reason to use Linux at all now, unless your an MS hater 😀
Shin
One question that I didn't get answered. Do these PC applications do down-sampling? Or do you have a fixed sample rate that you have to stay with?
One question that I didn't get answered. Do these PC applications do down-sampling? Or do you have a fixed sample rate that you have to stay with?
gedlee said:
Fixed sample rate in ASIO sessions but you can have a sample rate converter ie. run a 96Khz session and upsample 44.1/48Khz material to 96Khz. Or if you don't fancy that idea then you can have separate sessions for differing rates.
Ok, lets be specific. What would be required to sum all the channels and then down-convert to 4 kHz sample rate for the processing. One session for the summing and another for the further processing on that monaural channel? I've scanned the web information and it seems the learning curve is quite high. It all seems to be geared for pro audio sound mixing and recording and seems to assume a high level of proficiency at those tasks. Is there a "For dummies" on getting started in this?
- Status
- Not open for further replies.