MiniDSP Flex HTx

[Fredygump]

The gain is listed in the ICEpower datasheet (see attached). One thing to watch out for is the ASX series does not have balanced inputs, which may or may not be OK depending on your system.

If you can get away with unbalanced, the 50ASXS2 SE only has 20.5 dB gain and would be great with high efficiency drivers that do not need tons of power.

Michael

Ah yes, there it is on page 11! I was noting it's absence from the "key specifications" section, thinking that gain was one of the more important specs.

But yes, that amp, the icepower 50asxs2 was the obvious choice, since it is pretty quiet and because the horns do not need much power. Previously I noticed that it doesn't have the absolute lowest residual noise at 25µV, and until last night I didn't fully understand how lowering the amplifier gain can have a greater impact on reducing total system noise than just reducing the amp's residual noise. It's sitting in my cart over at parts-express....

The 50asx amp and the minidsp processor together would get me under 50µV, so that is going to be my plan.

 

[Balthazar B]

In the end... get the Apple TV 4K. Everything else will be a small nightmare and rather problematic.

Looking for the most straightforward way to combine elements of a new, simple 5.1 or 5.2 HT/audio setup, using a MiniDSP device for optimizing room correction, etc., in lieu of a conventional AV processor.

The screen will be a Sony A75L OLED, which supports a variety of audio formats, including LPCM Channels (bitstream) and LPCM Dolby/MAT over eARC, and I don't believe the device does decoding itself. We stream a lot of Google TV content today, and the Sony, having a Google TV OS, is a natural fit in that regard. We're also mulling over an Apple TV 4K as a content source. Is there some way to leverage the single eARC interface on the TV, the ATV, and the MiniDSP such that the ATV can decode and pass through decoded multichannel audio from the TV via LPCM to the MiniDSP, and alternatively its own its own content, and sending the video to the screen simultaneously? Intuitively, I suspect not due to device and eARC protocol capabilities, so in such a situation, what would be the preferred way to support a scenario where content streamed from the Internet via either the TV or ATV can have its audio processed through the MiniDSP, and thence to the transducers? I'm making the assumption that steams from an Apple TV app running on the Sony, or Google TV screens running from an app on the ATV, will be compromised in terms of their audio component.

I should add that we'll also be streaming 2/2.1 channel music content (Qobuz, Apple Music, CD quality Internet radio, etc.) from a dedicated streamer through the MiniDSP as well.

Thanks for any thoughts or advice!

 

[Fredygump]

Looking for the most straightforward way to combine elements of a new, simple 5.1 or 5.2 HT/audio setup, using a MiniDSP device for optimizing room correction, etc., in lieu of a conventional AV processor.

The screen will be a Sony A75L OLED, which supports a variety of audio formats, including LPCM Channels (bitstream) and LPCM Dolby/MAT over eARC, and I don't believe the device does decoding itself. We stream a lot of Google TV content today, and the Sony, having a Google TV OS, is a natural fit in that regard. We're also mulling over an Apple TV 4K as a content source. Is there some way to leverage the single eARC interface on the TV, the ATV, and the MiniDSP such that the ATV can decode and pass through decoded multichannel audio from the TV via LPCM to the MiniDSP, and alternatively its own its own content, and sending the video to the screen simultaneously? Intuitively, I suspect not due to device and eARC protocol capabilities, so in such a situation, what would be the preferred way to support a scenario where content streamed from the Internet via either the TV or ATV can have its audio processed through the MiniDSP, and thence to the transducers? I'm making the assumption that steams from an Apple TV app running on the Sony, or Google TV screens running from an app on the ATV, will be compromised in terms of their audio component.

I should add that we'll also be streaming 2/2.1 channel music content (Qobuz, Apple Music, CD quality Internet radio, etc.) from a dedicated streamer through the MiniDSP as well.

Thanks for any thoughts or advice!

As you said, the MiniDSP does not decode multichannel signals. But I believe the setup you want is pretty simple with E-arc

Please correct me if I'm wrong, but I believe the E-Arc HDMI just functions as an audio output from the TV. So run an HDMI cable from the E-Arc HDMI port to the MIniDSP, and then plug your other devices into the remaining HDMI inputs on your TV. Then in your TV audio setup menu enable EArc mode and make sure it is PCM. Sony has instructions, and I just found the settings on my TV.

Then you just need to make sure you configure the MiniDSP to work as you want it to. It has a "channel 9" which is an automatic summed low frequency + LFE channel that you can select for your subwoofer.

 

[Balthazar B]

As you said, the MiniDSP does not decode multichannel signals. But I believe the setup you want is pretty simple with E-arc

Please correct me if I'm wrong, but I believe the E-Arc HDMI just functions as an audio output from the TV. So run an HDMI cable from the E-Arc HDMI port to the MIniDSP, and then plug your other devices into the remaining HDMI inputs on your TV. Then in your TV audio setup menu enable EArc mode and make sure it is PCM. Sony has instructions, and I just found the settings on my TV.

Then you just need to make sure you configure the MiniDSP to work as you want it to. It has a "channel 9" which is an automatic summed low frequency + LFE channel that you can select for your subwoofer.

If the Sony doesn't decode the multichannel bitstream, though, how can the MiniDSP process the audio (since it doesn't either)? An Apple TV 4K can decode bitstreams, I believe, but I don't know:

• If it can do so with streams from the Sony;
• How it can send those decoded streams via LPCM to the MiniDSP if its only eARC port is connected to the Sony's only eARC port;
• If it can send decoded streams to the MiniDSP and the video content to the Sony simultaneously (and if so, can the video on the Sony be in synch with the audio emanating from the collection of tranducers).

Obviously I'm a bit out of my depth here, but I think the high level question I'm asking is how to make everything work nicely if there are two content sources (Google/Sony TV and Apple TV), one decoder (Apple TV), one audio processor (MiniDSP) downstream from the two previous devices, and one video display (the Sony). I think the multichannel analog inputs on the MiniDSP may need to be leveraged somehow, but I can't picture what working dataflows are feasible with just those components.

FWIW, here's what the Sony A75L supports with eARC (got this from RTINGS, which is usually quite accurate, as I don't have the device yet):

 

[bungle]

Then in your TV audio setup menu enable EArc mode and make sure it is PCM.

It may not be that. Setting TV to PCM, may make source to believe that you have 2.0 or it can make TV to only send 2.0 to eARC.

Sometimes you need a HDMI switch in between to ”lie” about capabilities to source. And then you need your source to send stuff as LPCM or perhaps in some cases bitstream, if your TV can decode. Same with CEC. It may work or it may not, and then a switch may complicate it.

For example with my setup of AppleTV, HTx and LG UST, I need to first have a HDMI switch to connect Apple TV that sends the right EDID info to Apple TV (that my LG does not seem to), so that I can enable Atmos in Apple TV. Then I connect LG UST to switch, and in LG I need to set the HDMI to bitstream, and passthrough so that it does not do anything to Apple TV source signal (yes, it is not bitstream). Then I need to enable eARC in LG and connect HTx to there. I tried to change the switch to eARC mode and connect HTx to there, but this caused so bad lipsync issue that it was not solvable. Now I still have some problems. It seems my setup may at times just shutdown without me doing anything. This has to do with CEC that I am chasing, which may lead to disabling CEC everywhere, but that then leads to usability issues. Also it seems the LG UST has some weird bug when connected to switch, which may make it reboot out of nothing. Or perhaps it is the switch that causes it.

Codecs are cancer of hifi (and video) and have always been. We would be so much better without them. Or at least have a royalty free ones. Same with HDMI. Everything you just need to try, and most of the time you will likely get just some of the stuff working. Needless to say that in my system everything I use analog inputs in HTx works flawless. The S/PDIF, Toslink and USB works as well. It is just this HDMI/CEC/Codecs world that makes everyone suffer needlessly. Only way to know is to really test with your equipment.

 

[bungle]

It has a "channel 9" which is an automatic summed low frequency + LFE channel that you can select for your subwoofer.

Care must be taken how Channel 9 is mixed. Basically if the LFE does not have +10db boost already added by source, which is likely, you need to make all input channels other than LFE to -10db in bass management. Also on down mixing you may need to -3db channels that you are downmixing to get proper LoRo. With Dirac there is a clitch with this too and when running Dirac you may need to change output of LFE to -10 and then after dirac, set it back to 0. Otherwise dirac may try correct it.

One thing I also found with dirac and bass management that mains play together with sub. I think you need to mute the channels in bass management to not do that before running dirac and unmuting after). Still I am not sure which is more correct (measure mains in dirac with/without sub, the lfe runs independently of course).

 

[Fredygump]

The gain is listed in the ICEpower datasheet (see attached). One thing to watch out for is the ASX series does not have balanced inputs, which may or may not be OK depending on your system.

If you can get away with unbalanced, the 50ASXS2 SE only has 20.5 dB gain and would be great with high efficiency drivers that do not need tons of power.

Michael

I received the MiniDSP HTx yesterday and have it temporarily set up with unbalanced cables. It is indeed much quieter. If I put my ear up to the horn I can hear a soft hum/buzzz, which I suspect is due to the messy temporary wiring. It is much quieter than the noise before, and is inaudible from ~3ft away. I am going to be modifying my short XLRs to have TRS connectors, which I expect will reduce this even more.

I probably would not benefit from a quieter amp right now. With the L-Pads I am using on the sensitive drivers, I have zero gain on all channels on the DSP. If I removed the L-Pad on the drivers, I'd have to reduce the gain by that amount. So I'll leave it until I find a compelling reason to change things.

Interestingly, my initial setup requires almost none of the EQs needed before. I'm not sure why this is different on the MiniDSP. The biggest difference in the setup is previously I had to use a PC for programming, but with MiniDSP I can run both REW and the MiniDSPs program from my Mac Mini that I use as an HTPC.

 

[Fredygump]

If the Sony doesn't decode the multichannel bitstream, though, how can the MiniDSP process the audio (since it doesn't either)? An Apple TV 4K can decode bitstreams, I believe, but I don't know:

• If it can do so with streams from the Sony;
• How it can send those decoded streams via LPCM to the MiniDSP if its only eARC port is connected to the Sony's only eARC port;
• If it can send decoded streams to the MiniDSP and the video content to the Sony simultaneously (and if so, can the video on the Sony be in synch with the audio emanating from the collection of tranducers).

Obviously I'm a bit out of my depth here, but I think the high level question I'm asking is how to make everything work nicely if there are two content sources (Google/Sony TV and Apple TV), one decoder (Apple TV), one audio processor (MiniDSP) downstream from the two previous devices, and one video display (the Sony). I think the multichannel analog inputs on the MiniDSP may need to be leveraged somehow, but I can't picture what working dataflows are feasible with just those components.

FWIW, here's what the Sony A75L supports with eARC (got this from RTINGS, which is usually quite accurate, as I don't have the device yet):

View attachment 355401

I feel like there is a misunderstanding in how E-Arc works. The E-Arc is functionally just an external audio output from your TV. Any audio that would play through the TV's built in speakers is re-directed through the E-Arc HDMI cable. It does not matter how many different sources you have plugged into the TV, or what app on the TV that you are using.

The decoding of multichannel audio part of your question is the hard part. This makes me ask why you are choosing this MiniDSP HTx in the first place? If you want an affordable processor that will play multichannel content, maybe you should buy a unit designed for that purpose, like the Emotiva BasX MC1? I don't know if this is the best one, but it is a similar price, it is designed to do multi-channel decoding, it has a room correction feature, and it has more pre-outs for more channels than the MiniDSP HTx.

For perspective, the reason I bought the MiniDSP is because I need the DSP processing for my unique 4-way speaker design. I don't need E-Arc, and I don't need multichannel. I'm interested in trying to make multi-channel work, just to see if it contains anything different than the stereo mix, but at the end of the day I don't care about those features. If all of the digital inputs were removed, I'd still pay the same amount for this device. It is a great piece of hardware, unique at this price point, but it may not be the best device for you?

 

[ARNiTECT]

Care must be taken how Channel 9 is mixed. Basically if the LFE does not have +10db boost already added by source, which is likely, you need to make all input channels other than LFE to -10db in bass management. Also on down mixing you may need to -3db channels that you are downmixing to get proper LoRo. With Dirac there is a clitch with this too and when running Dirac you may need to change output of LFE to -10 and then after dirac, set it back to 0. Otherwise dirac may try correct it.

One thing I also found with dirac and bass management that mains play together with sub. I think you need to mute the channels in bass management to not do that before running dirac and unmuting after). Still I am not sure which is more correct (measure mains in dirac with/without sub, the lfe runs independently of course).

I think I have set my Flex HT up incorrectly with my 7.1 system.
It sounds great; much clearer than my previous nanoAVR HD & DL combination, but I don't think my Bass Management is right.
I used a preset with all processing disabled (no BM etc.) and ran the Dirac measurements, first equalising all of the speakers including the sub.
I then setup Bass Management with -10dB on each channel except LFE at 0dB, I then added the bass management in on the channel routing at 0dB.
I have set the BM LPF/HPF quite low on the front left/right/centre channels (30Hz) as they measure quite well at lower frequencies, the other surrounds are at 60-75Hz.
The curve I have in Dirac is a +5dB bass shelf, my PMC front left/right/centre speakers are corrected up to 600Hz, the PMC surrounds up to 1000Hz and the KEFs full range.
The Dirac tab is saying the gain of the sub is -15.4dB, so perhaps I had +10dB on the sub from my previous set up (I should have kept notes).

Do I need to turn up the gain on my subwoofer +10dB, or do I need to make an adjustment in BM, Routing, Output channels ...or in Dirac and rerun the measurements?
I should set up the mic in REW and test what the corrected result is, but I don't have as much tinkering time as hoped for at the moment.

 

[Fredygump]

I think I have set my Flex HT up incorrectly with my 7.1 system.
It sounds great; much clearer than my previous nanoAVR HD & DL combination, but I don't think my Bass Management is right.
I used a preset with all processing disabled (no BM etc.) and ran the Dirac measurements, first equalising all of the speakers including the sub.
I then setup Bass Management with -10dB on each channel except LFE at 0dB, I then added the bass management in on the channel routing at 0dB.
I have set the BM LPF/HPF quite low on the front left/right/centre channels (30Hz) as they measure quite well at lower frequencies, the other surrounds are at 60-75Hz.
The curve I have in Dirac is a +5dB bass shelf, my PMC front left/right/centre speakers are corrected up to 600Hz, the PMC surrounds up to 1000Hz and the KEFs full range.
The Dirac tab is saying the gain of the sub is -15.4dB, so perhaps I had +10dB on the sub from my previous set up (I should have kept notes).

Do I need to turn up the gain on my subwoofer +10dB, or do I need to make an adjustment in BM, Routing, Output channels ...or in Dirac and rerun the measurements?
I should set up the mic in REW and test what the corrected result is, but I don't have as much tinkering time as hoped for at the moment.

I would run sweeps in REW to check the balance in stereo + sub, and work from there. You might disagree with the calibration.

With such a low crossover on your main speakers, your sub won't be doing a whole lot. So keep that in mind also.

If you want to be a little creative, you can give yourself some of the benefits of multiple subwoofer optimization by changing a few settings.

This is the kind of thing that I'm interested in, so disregard if you don't want to experiment....

1) I would go into Bass Management and set a high pass on every speaker to 100hz. (For the multiple sub concept to work right, you need low frequencies summed, which is what you're creating with bass management.)
2) In Channel Routing, I would assign "bass management" and "LFE" to the LFE channel
3) Then in Matrix Mixer, I would assign LFE to front right, front left, and LFE (Subwoofer)
4) In Output Channels, I would add high pass filters to left and right channels at 30hz, or whatever their practical limit is. You don't need a high pass on the other channels, because you already set one in Bass Management.
5) For extra credit, I would spend some time in REW to optimize the subwoofer in relation to the left and right. You can effectively adjust the phase with delays or inverting the signal, which will change how even bass response is throughout the room.

Voila! Now you have most of the functionality of multiple subs, and you only bought one sub!

 

[bungle]

The Dirac tab is saying the gain of the sub is -15.4dB, so perhaps I had +10dB on the sub from my previous set up (I should have kept notes).

Before running Dirac you need to set LFE -10db in output settings. After Dirac you should return it to 0.

Can you show your BM screen?

 

[ARNiTECT]

Before running Dirac you need to set LFE -10db in output settings. After Dirac you should return it to 0.

This makes sense, rather than messing about with the gain on the back of the sub.
Does mean rerunning Dirac measurements, though.

Can you show your BM screen?

Its on a different PC next to Hifi - will try this evening.

Voila! Now you have most of the functionality of multiple subs, and you only bought one sub!

This is interesting ...the sub is right next to the right speaker, so maybe only benefit from the left.

 

[bungle]

This makes sense, rather than messing about with the gain on the back of the sub.
Does mean rerunning Dirac measurements, though.

I think it does, as otherwise you cannot remove Dirac's +10db correction, not sure if it is possible somehow, but at least I don't know how, or have dived deeper. You can perhaps load the setting and remove the "adjust volume and delay", but it is probably easiest to just run it again, i know, not something you want to do many times.

 

[bungle]

2) In Channel Routing, I would assign "bass management" and "LFE" to the LFE channel

Would't it be better to low-pass the the LFE (say 120 Hz) in BM and only route BM to LFE in channel routing. As then you can do for other channels the -10db (to get LFE to match the other channels volume - as it usually comes with -10db on input, of course if it doesn't then ignore this comment).

 

[bungle]

Voila! Now you have most of the functionality of multiple subs, and you only bought one sub!

Except perhaps more distortion, depending of course your mains capabilities (while they can do 30Hz, how good are they at it?)

 

[ARNiTECT]

I think it does, as otherwise you cannot remove Dirac's +10db correction, not sure if it is possible somehow, but at least I don't know how, or have dived deeper. You can perhaps load the setting and remove the "adjust volume and delay", but it is probably easiest to just run it again, i know, not something you want to do many times.

If I instead +10dB in one of the following, would this have the same effect?
Bass Management (speakers at 0dB, LFE at +10dB)
Channel Routing (Bass mgnt +10dB)
Output Channels (Sub gain +10dB)

 

[Fredygump]

Except perhaps more distortion, depending of course your mains capabilities (while they can do 30Hz, how good are they at it?)

This is true.

 

[Fredygump]

This makes sense, rather than messing about with the gain on the back of the sub.
Does mean rerunning Dirac measurements, though.

Its on a different PC next to Hifi - will try this evening.


This is interesting ...the sub is right next to the right speaker, so maybe only benefit from the left.

Based on my experiments, I believe it is possible to co-locate subwoofers while still getting the benefits of multiple subwoofers. It does not matter that the sub is next to the main speaker.

The maxim that 2 drivers located within 1/4 wave length will sum as a single source assumes that the two drivers are co-planar and in phase. If the two drivers are out of phase, they interact like 2 different sources. We don't normally do this, because the interaction of these two out of phase sources is an acoustic mess when measured in free space.

But sound in a small room is a mess at the best of times.

I have a pair of custom 4 way stero speakers, and each speaker has 2 subwoofers. The physical design was inspired by Genelec's W371A. I have found that by carefully adjusting the phase (i.e. setting delays) of the co-located low frequency drivers, I can effectively fix low frequency peaks and nulls.

Based on my experiments with my speakers, I believe that co-located subwoofers can achieve the same results as multiple subwoofers spread out around a room. The subs spread out around the room are automatically out of phase due to the distance between each other, and co-located subs can produce the same effect with delays. Both approaches do the same thing, which is break up large standing waves, by creating many small peaks and nulls that are evenly distributed throughot the room. It doesn't matter where the subs are located; what matters is the relative phase of the sound waves as they interact with each other in the room.

 

[er|κzvio1in]

DLBC is way superior to anything you can do on your own as it phase corrects the crossover region for multiple seats.

You can craft your own.manual all pass filters and put them in the miniDSP as biqiads

 

[Carousel]

Hello, a few questions:

Anyone use the Flex HTx with Roon? I do not see anything about Roon Ready. If it works with Roon at all, does that mean over a USB cable from a PC / Mac running Roon server?

Anyone using the Flex HTx for active crossover? How is your experience running Dirac room correction with such a set-up? I assume a USB mic has to attached to the PC.

Does the miniDSP remote control work for volume control in such a set-up? I prefer to control volume later in the digital flow, not up-front in Roon.

Thanks