I am intrigued by all the FIR processing power and flexibility. But I'm a bit intimidated by all that's involved to get it right. Thus the appeal of a Dirac system.
And specifically for surround sound HT, the video to audio sync issues seem to be ongoing.
What I'd like is to have a killer 2 channel system that plays nicely with home theatre. I'm not married to Atmos but I figured I might as well build my room for it since I had the opportunity to.
Maybe my understanding of the pros and cons is wrong? Or my understanding of the potential solution is incomplete?
And specifically for surround sound HT, the video to audio sync issues seem to be ongoing.
What I'd like is to have a killer 2 channel system that plays nicely with home theatre. I'm not married to Atmos but I figured I might as well build my room for it since I had the opportunity to.
Maybe my understanding of the pros and cons is wrong? Or my understanding of the potential solution is incomplete?
Here's the way I would probably do it then.
I'd get a variant of a good receiver, but not the top of the line. They get obsolete and lose value so fast. Something of EQ built-in would be needed for the line arrays. Won't be perfect but maybe good enough for a movie experience.
Yamaha, Onkyo and Denon have variations of the same... with a some difference in sound processing. Cheaper than the Arcam, but not Dirac. Is Dirac that much better than those cheaper EQ systems? I couldn't tell you.
Then, get a HTPC, could be a small brick like the newer tiny computers that are pretty cheap. For DAC there would be a few different options, but I like the Pro offerings, so I'd get the Steinberg UR22MKII. It can decode up to 24-bit, 192kHz in pristine clarity for about $150.
Add JRiver ($50) and DRC Designer (free) and your line arrays would light up in 2-channel playback. For not much extra money.
Use one of the inputs on the receiver for the DAC and the receiver's own amps. playing in 2 channels mode. No switching cables, etc... just a press on the remote to change the input. Just making sure it's playing stereo direct, with no extra processing from the receiver to add to JRiver and DRC.
I'd get a variant of a good receiver, but not the top of the line. They get obsolete and lose value so fast. Something of EQ built-in would be needed for the line arrays. Won't be perfect but maybe good enough for a movie experience.
Yamaha, Onkyo and Denon have variations of the same... with a some difference in sound processing. Cheaper than the Arcam, but not Dirac. Is Dirac that much better than those cheaper EQ systems? I couldn't tell you.
Then, get a HTPC, could be a small brick like the newer tiny computers that are pretty cheap. For DAC there would be a few different options, but I like the Pro offerings, so I'd get the Steinberg UR22MKII. It can decode up to 24-bit, 192kHz in pristine clarity for about $150.
Add JRiver ($50) and DRC Designer (free) and your line arrays would light up in 2-channel playback. For not much extra money.
Use one of the inputs on the receiver for the DAC and the receiver's own amps. playing in 2 channels mode. No switching cables, etc... just a press on the remote to change the input. Just making sure it's playing stereo direct, with no extra processing from the receiver to add to JRiver and DRC.
Last edited:
Hmm, a few questions...
How would I incorporate my turntable? I have a DIY phono-preamp that would need to be part of the killer two channel system somehow. I was originally planning on feeding that to one of the line-inputs of the AVR. But that doesn't integrate JRiver. I'd have to have an ADC in the chain to feed into jRiver and then back to the AVR which seems less than ideal.
How do I manage the IB Sub for 2 channel? I currently have an Emotiva XDA-2 that I was planning on incorporating for two channel digital sources. So I can use that to plug in my PS3 for CD/SACD as well as computer audio. If I follow what you're saying, the DAC analog outs would feed the AVR and then I'd use the pre-outs to feed my nCore amps for two channel. But how to integrate the .1 channel? Would I rely on the AVR to manage that? Not sure how Jriver would do that on it's own? Or maybe I don't need to worry about it for 2-channel music. I'd just use my AVR for HT for the IB subs. Or... I also have a mini-DSP I can integrate. That could be what I use to split the pre-outs into 2.1 But then that makes switching the subs for HT more complicated, requiring a switch of cables of some sort. Likely would just ignore the subs for music given how low most have said the line arrays can go.
How do I have my fronts driven both by the nCore mono blocks fed by the AVR pre-outs and then by the AVR directly when in HT mode? I'd need to swap cables I think. Or I suppose AVRs can send the signal to both the pre-outs as well as the speaker outs simultaneously? And with no lag issues?
How would I incorporate my turntable? I have a DIY phono-preamp that would need to be part of the killer two channel system somehow. I was originally planning on feeding that to one of the line-inputs of the AVR. But that doesn't integrate JRiver. I'd have to have an ADC in the chain to feed into jRiver and then back to the AVR which seems less than ideal.
How do I manage the IB Sub for 2 channel? I currently have an Emotiva XDA-2 that I was planning on incorporating for two channel digital sources. So I can use that to plug in my PS3 for CD/SACD as well as computer audio. If I follow what you're saying, the DAC analog outs would feed the AVR and then I'd use the pre-outs to feed my nCore amps for two channel. But how to integrate the .1 channel? Would I rely on the AVR to manage that? Not sure how Jriver would do that on it's own? Or maybe I don't need to worry about it for 2-channel music. I'd just use my AVR for HT for the IB subs. Or... I also have a mini-DSP I can integrate. That could be what I use to split the pre-outs into 2.1 But then that makes switching the subs for HT more complicated, requiring a switch of cables of some sort. Likely would just ignore the subs for music given how low most have said the line arrays can go.
How do I have my fronts driven both by the nCore mono blocks fed by the AVR pre-outs and then by the AVR directly when in HT mode? I'd need to swap cables I think. Or I suppose AVRs can send the signal to both the pre-outs as well as the speaker outs simultaneously? And with no lag issues?
Last edited:
Here are my thoughts on the whole thing.
Line arrays are a different beast. They got bad rep because people built them, plugged them into their clean setup and got disappointed.
Line arrays need a LOT of work to get them sounding great, as Wesayso, and many others like myself discovered. Once they are tuned though, they are simply amazing. The image and the ambience is mesmerizing.
But, they are not for a purist system like phono-Pre-amp. That's a recipe for disappointment and disaster.
If you want to use your setup as you described, then I would suggest the next best thing after line arrays.... Horns!
A nice setup of your subs, with mid and high horns would sound great for both music and movies. The EQ, etheir Dirac or other would probably be enough for that kind of a system.
Ultimately the decision is yours, but I would hate to see someone build arrays and then feel they do not sound great.
If you are dead set on line arrays, something in your setup will have to give.
If you want to keep your setup, I'd suggest you look at other designs for your audio/movie room.
Line arrays are a different beast. They got bad rep because people built them, plugged them into their clean setup and got disappointed.
Line arrays need a LOT of work to get them sounding great, as Wesayso, and many others like myself discovered. Once they are tuned though, they are simply amazing. The image and the ambience is mesmerizing.
But, they are not for a purist system like phono-Pre-amp. That's a recipe for disappointment and disaster.
If you want to use your setup as you described, then I would suggest the next best thing after line arrays.... Horns!
A nice setup of your subs, with mid and high horns would sound great for both music and movies. The EQ, etheir Dirac or other would probably be enough for that kind of a system.
Ultimately the decision is yours, but I would hate to see someone build arrays and then feel they do not sound great.
If you are dead set on line arrays, something in your setup will have to give.
If you want to keep your setup, I'd suggest you look at other designs for your audio/movie room.
Here's an idea, setup a bass array of 12" drivers in the corners with a Synergy horn in the middle of it. That would do it 
I would build it in a heartbeat if I could sell that system to my girl. Less processing needed for excellent results. You can still improve on it with FIR though.
For movies use an extra Synergy horn in the center. That sounds like a recipe that would work with your proposed receivers, at least to me. Add all the other stuff you want with simpler direct radiators or dipoles, whatever suits your taste.
But I agree with Perceval, line arrays do need the extra help. And they can be more than awesome for home theatre if you include the needed steps. But please don't take short cuts.
I would build it in a heartbeat if I could sell that system to my girl. Less processing needed for excellent results. You can still improve on it with FIR though.
For movies use an extra Synergy horn in the center. That sounds like a recipe that would work with your proposed receivers, at least to me. Add all the other stuff you want with simpler direct radiators or dipoles, whatever suits your taste.
But I agree with Perceval, line arrays do need the extra help. And they can be more than awesome for home theatre if you include the needed steps. But please don't take short cuts.
We're going to make this work, damnit! Thanks for all the help so far and all the help that's to come in the future I'm sure!
Can I ask why a phono source wouldn't be good? Are you saying that it's not good directly feeding the line array? Or even with the convolution in between? My plan would be for some sort of EQ (IR/FIR/Dirac or otherwise) so I was assuming that would be fine as it would be for any other music source.
I'm committed to the line arrays (the backordered drivers finally arrived yesterday) so as much as I'd like to try horns, for now, I'm set. Perhaps I could try one for the center channel per wesayso's suggestion.
As for the original problem statement, my understanding is that Dirac incorporates FIR into it's processing and also has a fair amount of flexibility built into it so you can adjust to get close to JRiver convolution. Marketing-speak here. Also, it has the benefit of measuring in different locations to optimize more broadly. In any case, I'm wondering if there's direct experience out there that says Dirac wouldn't do a good job with line arrays. I tried a google search for someone with direct experience one way or the other but came up empty.
And finally, as for solutions, another idea I came up is this:
I would have the sources all feed into the AVR. And then the AVR would have pre-outs that would feed a mini-DSP DDRC-88A (It's a Dirac live unit with 8 analog inputs and 8 analog outputs. I would need separate amplification for all the channels... but I already have four channels plus the sub covered so I would just need the center channel and one of the surround pairs. It's not a cheap solution... that unit is almost $1000. But I could then buy a pretty nice AVR and still come in less than the Arcam 550.
In addition, I can still play with JRiver and DRC designer and just feed that output separately into the AVR and have the signal directly pass through the DDRC-88A. Then I could actually directly compare results!
Thoughts?
Thanks for all the help so far and all the help that's to come in the future I'm sure!Can I ask why a phono source wouldn't be good? Are you saying that it's not good directly feeding the line array? Or even with the convolution in between? My plan would be for some sort of EQ (IR/FIR/Dirac or otherwise) so I was assuming that would be fine as it would be for any other music source.
I'm committed to the line arrays (the backordered drivers finally arrived yesterday) so as much as I'd like to try horns, for now, I'm set. Perhaps I could try one for the center channel per wesayso's suggestion.
As for the original problem statement, my understanding is that Dirac incorporates FIR into it's processing and also has a fair amount of flexibility built into it so you can adjust to get close to JRiver convolution. Marketing-speak here. Also, it has the benefit of measuring in different locations to optimize more broadly. In any case, I'm wondering if there's direct experience out there that says Dirac wouldn't do a good job with line arrays. I tried a google search for someone with direct experience one way or the other but came up empty.
And finally, as for solutions, another idea I came up is this:
I would have the sources all feed into the AVR. And then the AVR would have pre-outs that would feed a mini-DSP DDRC-88A (It's a Dirac live unit with 8 analog inputs and 8 analog outputs. I would need separate amplification for all the channels... but I already have four channels plus the sub covered so I would just need the center channel and one of the surround pairs. It's not a cheap solution... that unit is almost $1000. But I could then buy a pretty nice AVR and still come in less than the Arcam 550.
In addition, I can still play with JRiver and DRC designer and just feed that output separately into the AVR and have the signal directly pass through the DDRC-88A. Then I could actually directly compare results!
Thoughts?
I know the marketing speak of Dirac. The trouble I see with it is the closed box solution it brings. All the ones I mentioned, Audiolense, Acourate and DRC-FIR have user adjustable (and frequency dependent) windows to get the optimization right.
I see no problem hooking up a record player, aside from adding the AD/DA step. You could do that and route it trough JRiver. Minimising AD/DA conversions would always be a wise move i.m.h.o.
Now why would you want user adjustability? For one because line arrays have high frequencies arriving from all of the drivers in the array, very different from a point source.
Due to that I have a longer correction window up top (high frequencies) and way less at mid and low frequencies. Dirac is a "one size fits all" kind of solution compared to the ones I mentioned. MiniDSP has some FIR capable devices though. I'd still prefer the flexibility of a PC with a good DAC/soundcard. Due to using large amounts of boost my preferred solution was a DAC hooked up with optical.
Figure out the time differences of arriving frequencies from this drawing:
Yet with DRC-FIR I get a clean impulse out of it:
The slight downwards peak you see in this graph is from my downwards sloped Room Target. A straight line target would only show a single peak upwards, followed by the wiggly shapes caused by my room and up to some part the array behavior.
I see no problem hooking up a record player, aside from adding the AD/DA step. You could do that and route it trough JRiver. Minimising AD/DA conversions would always be a wise move i.m.h.o.
Now why would you want user adjustability? For one because line arrays have high frequencies arriving from all of the drivers in the array, very different from a point source.
Due to that I have a longer correction window up top (high frequencies) and way less at mid and low frequencies. Dirac is a "one size fits all" kind of solution compared to the ones I mentioned. MiniDSP has some FIR capable devices though. I'd still prefer the flexibility of a PC with a good DAC/soundcard. Due to using large amounts of boost my preferred solution was a DAC hooked up with optical.
Figure out the time differences of arriving frequencies from this drawing:
Yet with DRC-FIR I get a clean impulse out of it:
The slight downwards peak you see in this graph is from my downwards sloped Room Target. A straight line target would only show a single peak upwards, followed by the wiggly shapes caused by my room and up to some part the array behavior.
Last edited:
Thanks for the explanation. So how many frequency windows do you have setup? Is it based on how much of a delay is noticeable? It would be nice to have something that automatically adjusts based on actual frequency vs. a freq window.
As for my idea, here's a schematic of what I was thinking:
I don't know that there is such a thing as a receiver that can take an input and pass through to something external while simultaneously taking a second input and outputting that through another set of outputs. My thinking is that I can use a zone 2 capability but this may not be a capability that exists. It's a little bit like an effects loop for guitar amps.
But if this doesn't exist, the fix would be fairly simple... just add a switch between the DDCR-88A and the nCore amps that the HTPC would feed instead of looping back into the AVR.
I could also just skip the DDCR-88A for the time being and just use the AVR and the Audyssey or whatever comes with it for home theatre and perhaps the analog inputs (unless I do A/D for the HTPC). That would save me $1000 and more as I wouldn't need another three channels of amplification.
As for my idea, here's a schematic of what I was thinking:
I don't know that there is such a thing as a receiver that can take an input and pass through to something external while simultaneously taking a second input and outputting that through another set of outputs. My thinking is that I can use a zone 2 capability but this may not be a capability that exists. It's a little bit like an effects loop for guitar amps.
But if this doesn't exist, the fix would be fairly simple... just add a switch between the DDCR-88A and the nCore amps that the HTPC would feed instead of looping back into the AVR.
I could also just skip the DDCR-88A for the time being and just use the AVR and the Audyssey or whatever comes with it for home theatre and perhaps the analog inputs (unless I do A/D for the HTPC). That would save me $1000 and more as I wouldn't need another three channels of amplification.
Last edited:
It's a sliding frequency dependent window which you setup in DRC, having a different length at the top end compared to the mids. I've done a lot of experimentation on various window lengths until I narrowed down something that worked for me in my room.
I do use other tricks as well, most all of them I describe in that huge tread I started.
I have no clue if this would work, I can run and process 8 channels from my HTPC, I only use 4 right now. 4 for plain Stereo and 4 for HT with slight differences in processing.
For stereo I use 2 for the arrays and 2 for my ambient channels. In HT mode the ambient speakers play the SL and SR signals. I might play with more along the way, but for now this keeps me busy enough with experiments.
I would highly recommend the ambient channels for Stereo reproduction though. One of the most fun projects I've done so far!
I do use other tricks as well, most all of them I describe in that huge tread I started.
I have no clue if this would work, I can run and process 8 channels from my HTPC, I only use 4 right now. 4 for plain Stereo and 4 for HT with slight differences in processing.
For stereo I use 2 for the arrays and 2 for my ambient channels. In HT mode the ambient speakers play the SL and SR signals. I might play with more along the way, but for now this keeps me busy enough with experiments.
I would highly recommend the ambient channels for Stereo reproduction though. One of the most fun projects I've done so far!
Sorry, this may be a stupid question... but we've mostly been talking about netflix and video sources that start with the computer. Can you get other sources into your HTPC and then out to your speakers and display? Like, is there an HDMI input in an HTPC? For me, working on a new 13" Macbook Pro, I can only see something like a HDMI - Thunderbolt adapter for the input perhaps? And I have an HDMI port out that could go to a receiver. I could theoretically find a good USB ADC for my phono setup.
And does the HTPC decode Dolby and DTS and such?
I think I essentially am describing a PC based AVR! =)
And does the HTPC decode Dolby and DTS and such?
I think I essentially am describing a PC based AVR! =)
Yeah.. I've looked into this myself. Not easily realized. I think you will be happier sticking with a "typical" setup (AVR) apart from running pre outs into a MiniDSP for the line arrays. You can delay the rest of the speakers to match the MiniDSP latency on the AVR. Not sure if any AVR supports delaying the video signal to match...
Not too bad... I used to run this way:
On a Mac, there is a way to route the audio to do the processing, then out to an interface.
I did it for a while. You need VLC, SoundForge and AULab. Lots of how-to online, including Youtube or just step-by-step websites.
VLC will decode all the channels on a DTS DVD or BR, output to Soundforge.
Opening AULab, you open the number of channels you need, add AU (Audio Units) like convolver, EQ, etc...
Output to an interface supporting the number of channels you need, and connecting your amps to their dedicated channel.
If you have Video and Audio sync issues, you can adjust delay within VLC to match the delay of the audio processing.
But now, I have pretty much ripped all my DVDs and BR to a hard drive with MakeMKV. Does a great job, and it keeps the HD audio streams, like DTS, etc...
Then using JRiver and its processing power.
Multi TB drives are pretty cheap now, and it's so much easier than shuffling through a stack of disks when looking for a movie!
The interface could also be used to hook up your turntable phono outputs. This time using AULab in stereo mode only. Changing the input to the interface. You can save different setups, and just open the one you will use.
But again, with an interface, you could digitize your vinyl collection. Don't use a phono/USB product. It is very limited.
On a Mac, there is a way to route the audio to do the processing, then out to an interface.
I did it for a while. You need VLC, SoundForge and AULab. Lots of how-to online, including Youtube or just step-by-step websites.
VLC will decode all the channels on a DTS DVD or BR, output to Soundforge.
Opening AULab, you open the number of channels you need, add AU (Audio Units) like convolver, EQ, etc...
Output to an interface supporting the number of channels you need, and connecting your amps to their dedicated channel.
If you have Video and Audio sync issues, you can adjust delay within VLC to match the delay of the audio processing.
But now, I have pretty much ripped all my DVDs and BR to a hard drive with MakeMKV. Does a great job, and it keeps the HD audio streams, like DTS, etc...
Then using JRiver and its processing power.
Multi TB drives are pretty cheap now, and it's so much easier than shuffling through a stack of disks when looking for a movie!
The interface could also be used to hook up your turntable phono outputs. This time using AULab in stereo mode only. Changing the input to the interface. You can save different setups, and just open the one you will use.
But again, with an interface, you could digitize your vinyl collection. Don't use a phono/USB product. It is very limited.
What do you guys think about the Open DRC products from miniDSP? It's open source so I can use the Acourate or whatever to generate the filters themselves. I could use the analog version and just put it between the AVR pre-outs and the nCore mono blocks. Or if my AVR can do digital out, I can use the digital to analog version or even the digital to digital version with my DAC in between which may be best from a fidelity POV.
I would rely on the AVR to do bass management for HT and drive the other channels.
Man, so many options! So much to learn! So much fun! Good thing I have so much time!
I would rely on the AVR to do bass management for HT and drive the other channels.
Man, so many options! So much to learn! So much fun! Good thing I have so much time!
Me again... two questions:
ONE:
One thing that may be an issue... the openDRC product maxes out at 6144 taps/ch @48kHz for stereo. Does that seem like enough for the line arrays? My research mostly showed that folks weren't using that many but then I found some using powerful computer based convolution (a professional sound mixing engineer) was using something like 65000 taps/ch. There's also the issue of whether my AVR would be able to time delay the other channels enough to sync up so more taps wouldn't necessarily be "free" either.
TWO:
Also, seems like there are a lot of FIR filter generators out there. Free ones, expensive ones, etc. Do the free ones mostly do what the non-free ones do? The ones with the nice GUI are appealing to me as a noob. But I don't want to leave too much on the table either. RePhase good enough? List of others on the miniDSP site here.
ONE:
One thing that may be an issue... the openDRC product maxes out at 6144 taps/ch @48kHz for stereo. Does that seem like enough for the line arrays? My research mostly showed that folks weren't using that many but then I found some using powerful computer based convolution (a professional sound mixing engineer) was using something like 65000 taps/ch. There's also the issue of whether my AVR would be able to time delay the other channels enough to sync up so more taps wouldn't necessarily be "free" either.
TWO:
Also, seems like there are a lot of FIR filter generators out there. Free ones, expensive ones, etc. Do the free ones mostly do what the non-free ones do? The ones with the nice GUI are appealing to me as a noob. But I don't want to leave too much on the table either. RePhase good enough? List of others on the miniDSP site here.
I have DRCDesigner already loaded up via Wine on my Mac based on folks' input here. So that's ready to go. I might try rePhase as well based on the mini-DSP site having recommended it quite strongly.
wesayso/RA7 (or anyone else with similar line arrays) - how many taps are you guys using in your filters?
wesayso/RA7 (or anyone else with similar line arrays) - how many taps are you guys using in your filters?
You could have guessed as much, I'm using 65536.
RePhase isn't really comparable to DRC-FIR. DRC-FIR is an automated tool to create a correction impulse based and adjustable by multiple parameters. In RePhase you'd have to do it all by hand and wouldn't have the frequency dependent windowing available etc.
I use RePhase to make some fine adjustments and things like the phase shuffler in the http://www.diyaudio.com/forums/multi-way/277519-fixing-stereo-phantom-center-55.html#post4717755 thread. It wouldn't be easy to do with RePhase what DRC-FIR is capable of. Not saying it is impossible, but you'd need some extra tools for proper frequency dependent windowing as REW's implementation is smoothing a bit too much for my taste.
Last edited:
Hrm... this integrating HT and stereo system with the added complexity of an analog source is turning out to be quite the nut to crack. I wonder if I might be better off with a switch box to drive the speakers from different amps and splitting off the stereo digital and analog sources upstream of that. Then there are no compromises. The additional cost really isn't too bad actually... especially if I de-prioritize the HT part of it. Just need to find a speaker switch box. Or really, it's the opposite... one pair of speakers, two amps. Haven't seen one of those before.
Or I'll just keep it simple and use a wall plate and swap the amps at the wall when going from HT to stereo:
The more I think about it, the more I think separating the systems is the "simple" solution.
For HT:
I'm not sure what kind of result to expect with the Yamaha room correction (YPAO) built into my existing low budget receiver, but if it's terrible, I'll deal with it then. Perhaps just with some simple miniDSP IIR solution.
For analog I have two options:
1. I can use an existing set of 3.5 way speakers (Zaph ZDT3.5) as the rear surrounds for HT and then plug in the turn table signal chain when needed. They're a big pair of speakers and my couch would be facing the wrong direction so not ideal. But an option nonetheless.
2. I can find a good ADC to convert the signal to digital and then through the convolver as with the other sources.
Theoretically, I could have both at my disposal... now to start researching ADCs. Any suggestions?
An externally hosted image should be here but it was not working when we last tested it.
The more I think about it, the more I think separating the systems is the "simple" solution.
For HT:
I'm not sure what kind of result to expect with the Yamaha room correction (YPAO) built into my existing low budget receiver, but if it's terrible, I'll deal with it then. Perhaps just with some simple miniDSP IIR solution.
For analog I have two options:
1. I can use an existing set of 3.5 way speakers (Zaph ZDT3.5) as the rear surrounds for HT and then plug in the turn table signal chain when needed. They're a big pair of speakers and my couch would be facing the wrong direction so not ideal. But an option nonetheless.
2. I can find a good ADC to convert the signal to digital and then through the convolver as with the other sources.
Theoretically, I could have both at my disposal... now to start researching ADCs. Any suggestions?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.