PC Crossover virtual Playback Device driver?

[rcohen]

Does anyone know about PC software that can create a Windows virtual Playback Device with 7.1 channels, then apply flexible crossovers, delay, and routing to a 16 channel DAC?

 

I'm currently using JRiver and Dirac Live. JRiver has great crossover and routing features, but Dirac limits me to 7.1 channels. Ideally, this stuff would go after Dirac in the processing chain.

 

7.1ch PC Audio -> 7.1ch Dirac Live Playback Device -> 7.1-16ch Crossover/Routing -> 16 ch DAC

 

This would be useful to use the PC for stuff like active crossovers, bass management to all subs, mid-bass management to closest subs, etc.

[dallasjustice]

Crossover must happen after DIRAC. I don't think any such virtual crossover or WDM exists. I think it would be cool if Jriver ASIO driver could be selected as output in DAP and then route the signal through a second zone which applies all crossovers and other DSP routing. This would solve the problem requiring a crossover after DIRAC without an external crossover box. My understanding is that Jriver is working on some stuff like WDM but I have never heard of them offering multiple Jriver ASIO zones. It would work though.

Michael.

Does anyone know about PC software that can create a Windows virtual Playback Device with 7.1 channels, then apply flexible crossovers, delay, and routing to a 16 channel DAC?

 

I'm currently using JRiver and Dirac Live. JRiver has great crossover and routing features, but Dirac limits me to 7.1 channels. Ideally, this stuff would go after Dirac in the processing chain.

 

7.1ch PC Audio -> 7.1ch Dirac Live Playback Device -> 7.1-16ch Crossover/Routing -> 16 ch DAC

 

This would be useful to use the PC for stuff like active crossovers, bass management to all subs, mid-bass management to closest subs, etc.

[rcohen]

Crossover must happen after DIRAC. I don't think any such virtual crossover or WDM exists. I think it would be cool if Jriver ASIO driver could be selected as output in DAP and then route the signal through a second zone which applies all crossovers and other DSP routing. This would solve the problem requiring a crossover after DIRAC without an external crossover box. My understanding is that Jriver is working on some stuff like WDM but I have never heard of them offering multiple Jriver ASIO zones. It would work though.

Michael.

I'd really only need one zone. The trick would be to do this with JRiver as the source:

JRiver 7.1 source -> Dirac Live 7.1 -> JRiver ASIO 7.1 -> JRiver Processing 7.1-16ch -> 16 ch DAC

[dallasjustice]

You can't without an external crossover. I mentioned a second zone only as a way Jriver could do what you want. But that doesn't exist.

 

I'd really only need one zone. The trick would be to do this with JRiver as the source:

JRiver 7.1 source -> Dirac Live 7.1 -> JRiver ASIO 7.1 -> JRiver Processing 7.1-16ch -> 16 ch DAC

[rcohen]

You can't without an external crossover. I mentioned a second zone only as a way Jriver could do what you want. But that doesn't exist.

Okay, thanks.

 

I just posted a feature request at the JRiver forums:

Two DSP zones with WDM feature request

[dallasjustice]

Thanks. I hope they do something to expand audio processing and routing. I honestly don't know whether a WDM would work. It may not since DIRAC has it's own WDM that needs to be selected. The only idea that makes sense to me would be a second Jriver ASIO driver or a configurable Jriver ASIO driver that would let the user select a zone for playback. I don't even know whether this is possible.

 

Okay, thanks.

 

I just posted a feature request at the JRiver forums:

Two DSP zones with WDM feature request

[rcohen]

Thanks. I hope they do something to expand audio processing and routing. I honestly don't know whether a WDM would work. It may not since DIRAC has it's own WDM that needs to be selected. The only idea that makes sense to me would be a second Jriver ASIO driver or a configurable Jriver ASIO driver that would let the user select a zone for playback. I don't even know whether this is possible.

Dirac can select JRiver's ASIO driver. I think the problem is that JRiver can only select one output driver, rather than Dirac before processing and the DAC after processing.

[dallasjustice]

Dirac can select JRiver's ASIO driver. I think the problem is that JRiver can only select one output driver, rather than Dirac before processing and the DAC after processing.

Jriver can output to a second soundcard but you need to use another zone for that.

[retro]

Since I'm also very interrested in this topic, I'm taking the liberty to post here what I just posted in the thread rcohen started at jrivers forum:

 

 

Jriver user mojave wrote:

Once you perform the crossovers/routing, the frequency response and phase change. Therefore, room correction such as Dirac always has to be last in the signal chain.

 

Sounds logical, but afaik, the measurements in Dirac are always done with full-frequency sweeps, even in the mega-expensive Datasat unit. Basically, I think it's the same Dirac Live Correction Suite that's available for our HTPC's..?!?

 

And here's what an actual owner of the Datasat, djnickuk on avsforum, say about the correct workflow to get it running:

 

A key tip for setting up Dirac.

 

Set up your delays and spl levels first. I find the best way to set up delays is to use REW with a loop back to get the exact time of flight in ms from speaker to mic. I then use these measurements to work out the exact delays required.

 

Then ensure bass management is turned off. Speakers set to large and full range to subs.

 

Then start the Dirac process.

 

Once you have completed Dirac you need to hard reboot the machine.

 

Then you need to enable bass management and use the setting you require, the settings that you will have ideally learned from using REW or similar tools in learning xovers / slopes etc.

 

(From Official DataSat RS20i thread. (Setup Tips, Questions,General Info, etc) - Page 52)

 

This kind of contradicts what you write above, mojave..? Not meant as critique, but I'm also trying to get the hang of this to properly DRC my system. Running Dirac now in my HTPC and it sound's fabulous, although I haven't corrected my new subs yet.

Also have a license of Audiolense, may have to give that a try again...

 

Link to jriver thread: Two DSP zones with WDM feature request

 

Mojave, are you also a member on this forum..?

[dallasjustice]

Since I'm also very interrested in this topic, I'm taking the liberty to post here what I just posted in the thread rcohen started at jrivers forum:

 

 

 

 

Link to jriver thread: Two DSP zones with WDM feature request

 

Mojave, are you also a member on this forum..?

I jumped in on your jriver thread. Hopefully, something useful will come of it.

[sshd]

I'm currently using JRiver and Dirac Live. JRiver has great crossover and routing features, but Dirac limits me to 7.1 channels. Ideally, this stuff would go after Dirac in the processing chain.

 

Ideally Dirac Live would be your crossover, but it lacks this feature.

 

JRiver has a convolution engine that works well with alernative software like Accourate and Audiolense. Mitchco wrote an advanced guide to Accourate XO. Audiolense is much simpler to setup.

[rcohen]

Ideally Dirac Live would be your crossover, but it lacks this feature.

 

JRiver has a convolution engine that works well with alernative software like Accourate and Audiolense. Mitchco wrote an advanced guide to Accourate XO. Audiolense is much simpler to setup.

How does Audiolense compare to Dirac? I really like how Dirac makes it so fast to tweak and compare target curves. It makes it really practical to fine tune the sound. Does Audiolense have that?

[sshd]

How does Audiolense compare to Dirac?

 

Like comparing trains and aeroplanes. Quite difficult.

 

Dirac Live does all correction before the crossovers. It can only correct the speakers and crossovers as one unit and not say time align the drivers.

 

Audiolense combines crossovering and correction in one process. There is no support for standard crossovers like Linkwitz-Riley or Butterworth - only the Audiolense XO. You define crossover frequencies and crossover-width (not level). Then proceed to meassure each driver individually. Then set up a filter procedure, design a target and finally generate a FIR filter.

 

The FIR filter can be used with JRiver's convolution engine (and others like convolver.sf.net and foobar2000 gapless convolver). You can use the virtual ASIO device with JRiver or console.jp+convolver.sf.net to capure the output of your media player.

 

As you do surround you will have more features like bass routing and multiple targets for say rear and side channels. I don't have surround myself, so no hands-on experience. But a friend has a 5.1/14 channel active setup with Audiolense XO which sounds fantastic.

 

Audiolense can combine multiple meassurements (multi-seat) if you like, but it is not required.

 

Since I am unable to test drive Dirac Live, I have no idea how it sounds and how it compares.

But I have been using Audiolense XO with an active 6 channel stereo setup for years and I am perfectly happy.

[sshd]

I really like how Dirac makes it so fast to tweak and compare target curves. It makes it really practical to fine tune the sound. Does Audiolense have that?

 

I can't really answer this, as I have zero experience with Dirac Live.

 

Audiolense offers nothing for comparing filters, but there are practical approaches:

 

Audiolense displays simulated results on frequency and impulse response graphs similar to Dirac Live. There is no way to make a visual comparison of two targets/graphs inside Audiolense. However you can open two Audiolense windows and then do <alt><tab> to make a visual comparison.

 

As for listening to the difference between two filters:

- Use foobar2000 with foo_gapless_convolver and foo_mm

- Save DSP settings for your FIR filters A and B

- Setup short cut keys for loading A and B

- Listen to music and switch between them with the short cuts.

- Overwrite B to try something new

 

Maybe you can do the same in JRiver...

[rcohen]

I can't really answer this, as I have zero experience with Dirac Live.

 

Audiolense offers nothing for comparing filters, but there are practical approaches:

 

Audiolense displays simulated results on frequency and impulse response graphs similar to Dirac Live. There is no way to make a visual comparison of two targets/graphs inside Audiolense. However you can open two Audiolense windows and then do <alt><tab> to make a visual comparison.

 

As for listening to the difference between two filters:

- Use foobar2000 with foo_gapless_convolver and foo_mm

- Save DSP settings for your FIR filters A and B

- Setup short cut keys for loading A and B

- Listen to music and switch between them with the short cuts.

- Overwrite B to try something new

 

Maybe you can do the same in JRiver...

 

Thanks. I mean AB comparing the audio, not the visual. I'll look closer at Audiolense XO and JRiver.

 

I'm actually extremely happy with the sound I'm getting with Dirac Live. I'm just sad that I can't do more channels, active crossovers, or experiment with routing mid-bass to the closest subs.

 

Regarding the sound quality, aside from the room correction that seems to work great, Dirac Live's ability to experiment and fine tune & AB test target curves contributed a ton to getting the sound dialed in. That wasn't practical with Audyssey Pro, which is the only similar product I've compared it with. On the other hand, Audyssey Pro supports 11.2 speakers.

[LocutusEstBorg]

X-Fi MB3 is a proof of concept that it can be done, but unfortunately it doesn't support the Dirac Processor play back device. It only lists my physical devices.