Monolith by Monoprice HTP-1 16 Channel Dolby Atmos & DTS:X Home Theater Processor w/Dirac & Alexa Compatibility

[Erod]

Congrats. I saw over on that other site you bought. I'm not going to make a commitment until I see what the HDMI 2.1 solution will be. I know it's not going to effect anything now, but this will be a 10-yr purchase for me, if I choose it.

I have a projector, so the 2.1 thing means nothing to me.

And 8K is absolutely pointless. You can't see it. Might as well be 1000K. The human eye can't even really see 4K from 12 feet.

 

[ryanosaur]

I have a projector, so the 2.1 thing means nothing to me.

And 8K is absolutely pointless. You can't see it. Might as well be 1000K. The human eye can't even really see 4K from 12 feet.

I do game, but recognize 2.1 is only half-baked right now. 8k, could care less about, now, but in 4 years? Maybe I care, then. The 4K/120Hz is more pertinent in the near term, though.
I probably won't buy a console until late this year at the earliest.
By then, the next gen of AVRs and Processors will be coming out. We'll see then, I think, whether the 2.1 situation gets settled properly.
But my other upgrade desire is for Dirac and more expansive options for Atmos or Auro which is why I am most interested in the HTP-1.

 

[shadyJ]

I probably won't buy a console until late this year at the earliest.

Console? Pff. Get a gaming PC!

 

[bobof]

READ: Monoprice Monolith HTP-1 AV Processor Review

@shadyJ How confident are you in the measurements of the max level output measured before the performance colapses due to clipping? I have one of these processors and a few analysers and I've not measured much over 4V RMS balanced before the output clips and performance nosedives. I'm not familiar with the QA gear - but are you performing your measurements for level with the QA480 notch filter in circuit - and if so are you fully accounting for the notch filter gain?

 

[shadyJ]

@shadyJ How confident are you in the measurements of the max level output measured before the performance colapses due to clipping? I have one of these processors and a few analysers and I've not measured much over 4V RMS balanced before the output clips and performance nosedives. I'm not familiar with the QA gear - but are you performing your measurements for level with the QA480 notch filter in circuit - and if so are you fully accounting for the notch filter gain?

I only performed the acoustic measurements. Matthew Poes performed the electrical measurements, so you would have to ask him for those details.

 

[AcuDefTechGuy]

Not interested in anything in that price range but happy to see some competition from Monoprice.

All we need is Yamaha. All we need.

 

[John Daddabbo]

James WOW, such a thorough review as usual. Very much appreciate all the effort you put into this and the many detailed explanations and points made. Thanks!

 

[gene]

@shadyJ How confident are you in the measurements of the max level output measured before the performance colapses due to clipping? I have one of these processors and a few analysers and I've not measured much over 4V RMS balanced before the output clips and performance nosedives. I'm not familiar with the QA gear - but are you performing your measurements for level with the QA480 notch filter in circuit - and if so are you fully accounting for the notch filter gain?

Matt just rechecked and noted his Voltage measurements were all off by a factor of 3.16. I updated the measurement page with the correct voltage values and put an editorial note. Thanks for bringing this to our attention.

 

[bobof]

I only performed the acoustic measurements. Matthew Poes performed the electrical measurements, so you would have to ask him for those details.

Thanks & sorry, it wasn't obvious from the review.

Matt just rechecked and noted his Voltage measurements were all off by a factor of 3.16. I updated the measurement page with the correct voltage values and put an editorial note. Thanks for bringing this to our attention.

Glad it was helpful.

 

[Matthew J Poes]

Thanks & sorry, it wasn't obvious from the review.

Glad it was helpful.

I apologize. I was trying to get this measured the same week we were packing to move to Florida. I was hurrying and didn't realize the mistake I made. The QA401 is like an AP or any other analyzer in that it's accuracy is absolute. However, it has no idea if you stick something between it and the DUT. In this case, the QA480. The QA480 has a gain of 10dB in theory, but in practice you have to measure it to get the exact value (including the notch filter compensates to some extent for that). I did measure it and come up with a correction value, but in going back and forth to check the value, I forgot to put the correction back in.

I stuck the measurements away, moved, and then James reminded me I needed to write up the measurements report and for whatever reason I didn't double check the voltages and note they were off. Since the voltages are not an actual gain issue but a correction factor, the measurements are right (In terms of S/N and Distortion). The voltage, on the other hand, was wrong. Since its a fixed and known value, I was able to back into the correct value. I did some quick double-checking last night to be sure using a fixed voltage oscillator.

To make matters worse on this, I was using beta software for this since the QA480 was not a regular product and was fairly new. The QA401 software and the QA480 software have had at least 2 updates since I used it and corrected a few of the errors that caused this problem. I've decided that to be sure we don't have more problems as I publish more results using this rig, Gene and I will do some back to back measurements on the same DUT and come up with a protocol to follow that will prevent further errors. This was mostly carelessness on my part, but I pretty routinely make mistakes since I'm new to measuring electronics using gear like this.

Gene had rightly asked why I didn't check the voltages with an RMS volt meter. I did actually do that, it was partly how I established the correction value. This was simply me forgetting to put the value back into the software after taking the QA480 out of the signal chain to confirm measurements.

I have measurements of the Marantz AV7706 and the Purifi amplifier coming. I want to do some double checking with Gene before we publish those, but its pretty informative.

Matt

 

[Matthew J Poes]

Gene also asked me how I measured Signal to Noise ratio. I used the same AES method that ATI used without the a-weighting filter. It's the ratio of noise with no signal to the maximum clean signal output of 4.2 volts. That gave me 112.9dB. Adding the A-weighting filter probably improved things by 3-6dB and the manufacturers spec is 116dBa so I am in the ball park.

Again, moving forward, I will be creating a custom notch filter compensation file that includes an A-weighting filter added to it. That will allow me to use the QA480 notch filter for very low distortion and noise measurements and give a proper A-weighted S/N.

One other thing I'll note. From what I have read in the AP tech papers, the AES17 protocol, and various other handbooks, Signal to noise ratio is not well defined. It is the ratio of some arbitrary signal against some undefined noise measurement. Why I say this is that there are dozens of ways to measure the noise floor that all yield different results with real world equipment. For example, the QA401 software takes the average of the noise spectrum when a signal is present (so removing the fundamental and all it's harmonics and just looking at the remaining noise). This gives a lower signal to noise ratio than the device may be capable of because most devices (including the QA401 itself) raise it's noise floor with increasing signal levels. I think we conceptualize signal to noise ratio as the better defined term dynamic range and in that case we really mean the lowest possible noise floor when no signal is present relative to highest possible clean signal output. The way to achieve that is to short the input of the device and measure it's noise output with no signal, then connect the signal generator and measure a peak output. With a DAC or HT processor that may not be exactly how you do it, but close enough. You present no signal at all and measure just the noise. Well, on the HTP-1 and most devices I've tested, that actually gives an inaccurate result too because the devices often mute their output. To get an accurate result, you have to force it to unmute itself with a signal big enough to unmute the output while small enough to not corrupt the noise floor measurement. Easier said than done to be honest.

I am not totally clear how the AP software does this (I have a copy of the version of the software that can work with non-AP hardware), but from what I can tell, it seems like it measure S/N by first measuring a no signal condition and then measuring a max clean signal condition you set. I have been getting nearly identical results between the AP software and the QA software so far, but the two methods aren't the same.

With Amplifiers, I realized just how flawed dynamic range, as a measurement of S/N, is. Comparing a 100 wat amplifier to a 10,000 watt amplifier can give a very false idea that the crazy high output amplifier is somehow superior, when in fact, the 100 watt amp may be the lower noise device (and of course, in common use, will be sitting in that low output and lower noise scenario far more. While 10,000 watts is an extreme case, there exist plenty of pro amps that people use in their home audio systems with crazy amounts of power. Gene measures S/N at 1 watt and then calculates it for full power from that. I plan to do the same. I remain convinced that one of the biggest audible detriments in common amps is noise floor differences, with lots of amps being pretty noisy.

 

[ryanosaur]

I just want to thank Team AH for going back in and addressing this, making the edit! I appreciate the efforts and am super glad to be a part of this forum where you guys are willing to discuss this openly.
@gene , @shadyJ , @Matthew J Poes : seriously, you guys rock!

 

[Matthew J Poes]

I just want to thank Team AH for going back in and addressing this, making the edit! I appreciate the efforts and am super glad to be a part of this forum where you guys are willing to discuss this openly.
@gene , @shadyJ , @Matthew J Poes : seriously, you guys rock!

As long as you all can forgive me for the occasional bone-headed move like that, I'm happy to admit my mistakes and correct them.

 

[bobof]

I apologize....

Thanks for the explanation, I don't think it needed an apology (not to me anyway!). I was secretly hoping (unlikely though I knew it to be) that you'd found the go faster button...!

The "correction factor" and the numbers being out "by a factor of" 3.16 sounded a bit odd (is that loose use of language for factor?) but I guess it may make sense if you have to derive it using the method you describe. I thought the QA480 was touted as having 12dB of gain not 10, but I know all I know about that just from their webpage for the product, so I defer to you on that.

Using whole number test stimulus (ie 0dbFS played through at 0 on the master volume) I think I only saw about 4.08V RMS, before clipping, but admittedly I didn't try increasing the MV to +1dB and using -1dBFS tones + 0.1dB increments to find exactly the point that yields clipping of the waveform (I just know that 0dBFS tone with MV at +1dB is clipped).

Product like the QA401/480 I guess is a double edged sword; it's great in that you get access to functionality that is typically only available in more expensive equipment, but some of that and the way features are implemented comes at the cost of extra complication that gives you enough rope to hang with! Though once you get it right you are perhaps likely understand it better for not having had it done for you...

Anyway, thanks for including measurements in your review, it's nice to see this side of things getting a bit more coverage than has typically been the case.

 

[Erod]

As long as you all can forgive me for the occasional bone-headed move like that, I'm happy to admit my mistakes and correct them.

No problem, Matt.

Does any of this change your or James opinion of the HTP-1?

 

[Matthew J Poes]

No problem, Matt.

Does any of this change your or James opinion of the HTP-1?

No, why would it? It was just a voltage out issue. No consumer amplifiers can make use of more than 4 volts anyway. It's distortion and noise is still as good as it gets in this sector. In fact, at this point, the distortion is so low I would be shocked if anyone could hear it. Even if we consider that some highly trained listeners might be sensitive to distortion at around this level (Something I would highly doubt), the likelihood they could discern that in a room playing multichannel information is nill.

The Signal to noise ratio is excellent. When I hooked it up in my system using the Purifi amplifier, I heard no hiss with my hear to the speaker. My speakers are the Gedlee Abbey, so 95dB sensitivity at 1 watt or so. That sensitivity makes me more likely to hear noise.

If it could output more voltage, I'm sure that would make some happy, but at the end of the day, what would they do with that? Amir at ASR argues that it would allow us to lower the gain of the amplifier and get a lower signal to noise ratio. Ok, but its a shell game then, isn't it? First, consumer gear isn't built that way, so we would need to fundamentally change how we build gear. For a company to do that, they would need to build an entire system around it. Second, it just means the high voltage gain stage now is moved from the input of the amplifier to the output of the preamplifier. Is that better? Probably, but in practical terms I don't think it would make much difference.

 

[Matthew J Poes]

Thanks for the explanation, I don't think it needed an apology (not to me anyway!). I was secretly hoping (unlikely though I knew it to be) that you'd found the go faster button...!

The "correction factor" and the numbers being out "by a factor of" 3.16 sounded a bit odd (is that loose use of language for factor?) but I guess it may make sense if you have to derive it using the method you describe. I thought the QA480 was touted as having 12dB of gain not 10, but I know all I know about that just from their webpage for the product, so I defer to you on that.

Using whole number test stimulus (ie 0dbFS played through at 0 on the master volume) I think I only saw about 4.08V RMS, before clipping, but admittedly I didn't try increasing the MV to +1dB and using -1dBFS tones + 0.1dB increments to find exactly the point that yields clipping of the waveform (I just know that 0dBFS tone with MV at +1dB is clipped).

Product like the QA401/480 I guess is a double edged sword; it's great in that you get access to functionality that is typically only available in more expensive equipment, but some of that and the way features are implemented comes at the cost of extra complication that gives you enough rope to hang with! Though once you get it right you are perhaps likely understand it better for not having had it done for you...

Anyway, thanks for including measurements in your review, it's nice to see this side of things getting a bit more coverage than has typically been the case.

The weird number comes from the fact that the gain it shows after the filter isn't exactly what Matt Taylor claims. I guess I should talk to Matt more about that, but adding 12dB doesn't fix the problem. It ends up reading voltages that are too low. It was my understanding from him that this is or could be related to what happens after the dewarping is added in. Maybe I've misunderstood. At the end of the day, all I can say is I backed into that number by taking a fixed voltage tone generator, measuring its voltage, running it directly into the QA401 and confirming same voltage. Inserting the QA480, adding the filter, and continually compensating for the voltage until I got the same number.

As for the value, when did you measure? As of last night I was told by the ATI guys that some of the updates from the firmware helped slightly improve distortion at the limits. I'm just a tiny bit over 4 volts and I was already showing signs of clipping. I probably should have pulled back to 4 volts, but if I recall, when I was much below 4 volts, I was getting the same result as I got at 1.4 volts (And they measured at 1.8 volts). So I pushed it to the point where clipping was starting to come on but it was still usably clean. That is the value I reported.

The QA480 has been a learning experience. There are also weird scenarios with an external load, the QA451, QA480, and QA401 together that will yield totally bonkers numbers. At one point I somehow got an amplifier measurement that suggested 1000 watts with .0002% distortion from something not capable of 1000 watts. It was caused by hitting the limits of the QA451 and the entire circuit path masking what was going on. I still don't know why it did this, but it was repeatable and fixed by increasing the voltage divider of the external load. I'm just learning as you go that with something this complex, you have to really be extra careful and have some double checks in there.

Another oddity, I can't leave a volt meter attached to amps like the Purifi because it throws off the measurements. The leads I have contain steel and it causes distortion to rise. Talking to Bruno about this stuff basically goes to, "Buy an AP its easier" but of course, I can't afford that.

 

[bobof]

If it could output more voltage, I'm sure that would make some happy, but at the end of the day, what would they do with that? Amir at ASR argues that it would allow us to lower the gain of the amplifier and get a lower signal to noise ratio. Ok, but its a shell game then, isn't it? First, consumer gear isn't built that way, so we would need to fundamentally change how we build gear. For a company to do that, they would need to build an entire system around it. Second, it just means the high voltage gain stage now is moved from the input of the amplifier to the output of the preamplifier. Is that better? Probably, but in practical terms I don't think it would make much difference.

The one place where in the hobbyist world higher levels would seem "almost" useful in recent times has been using some of the Class D modules without external buffers - an option for instance when using the Purifi module on the Eval1 board, which results in a bit better measurements if I recall than with the buffer. But it's a pretty marginal application...

The weird number comes from the fact that the gain it shows after the filter isn't exactly what Matt Taylor claims. I guess I should talk to Matt more about that, but adding 12dB doesn't fix the problem. It ends up reading voltages that are too low. It was my understanding from him that this is or could be related to what happens after the dewarping is added in. Maybe I've misunderstood. At the end of the day, all I can say is I backed into that number by taking a fixed voltage tone generator, measuring its voltage, running it directly into the QA401 and confirming same voltage. Inserting the QA480, adding the filter, and continually compensating for the voltage until I got the same number.

I thought the QA writeup seemed to indicate 12dB gain, but I must say I wasn't fully following it, no substitute for having the gear in front of you.

As for the value, when did you measure? As of last night I was told by the ATI guys that some of the updates from the firmware helped slightly improve distortion at the limits. I'm just a tiny bit over 4 volts and I was already showing signs of clipping. I probably should have pulled back to 4 volts, but if I recall, when I was much below 4 volts, I was getting the same result as I got at 1.4 volts (And they measured at 1.8 volts). So I pushed it to the point where clipping was starting to come on but it was still usably clean. That is the value I reported.

It was 4.08V RMS @1kHz for 0dBFS stimulus with 0 on the master volume control. I measured it again last night on almost the latest beta as you had me doubting myself... From what I understand the optimum lineup is for the output to be set to around 2.4V in the UI.

The QA480 has been a learning experience. There are also weird scenarios with an external load, the QA451, QA480, and QA401 together that will yield totally bonkers numbers. At one point I somehow got an amplifier measurement that suggested 1000 watts with .0002% distortion from something not capable of 1000 watts. It was caused by hitting the limits of the QA451 and the entire circuit path masking what was going on. I still don't know why it did this, but it was repeatable and fixed by increasing the voltage divider of the external load. I'm just learning as you go that with something this complex, you have to really be extra careful and have some double checks in there.

Another oddity, I can't leave a volt meter attached to amps like the Purifi because it throws off the measurements. The leads I have contain steel and it causes distortion to rise. Talking to Bruno about this stuff basically goes to, "Buy an AP its easier" but of course, I can't afford that.

It sounds like "fun"...

 

[Matthew J Poes]

The one place where in the hobbyist world higher levels would seem "almost" useful in recent times has been using some of the Class D modules without external buffers - an option for instance when using the Purifi module on the Eval1 board, which results in a bit better measurements if I recall than with the buffer. But it's a pretty marginal application...

I thought the QA writeup seemed to indicate 12dB gain, but I must say I wasn't fully following it, no substitute for having the gear in front of you.

It was 4.08V RMS @1kHz for 0dBFS stimulus with 0 on the master volume control. I measured it again last night on almost the latest beta as you had me doubting myself... From what I understand the optimum lineup is for the output to be set to around 2.4V in the UI.

It sounds like "fun"...

Well 4 volts is the official max. I thought I recalled Amir indicating he didn't see extreme clipping until 4.5 volts, so I didn't think to question it. Since I don't have the device anymore I can't recheck and have to go by what I think I did 3 months ago. I corrected the voltage values from 3 months ago based on testing a different device last night to figure out the difference if I didn't compensate for the gain of the QA480.

The QA480 does have 12dB of gain, but if you type in -12dB into the external gain of the analyzer software it won't give accurate voltages. I don't know why, Matt is looking into it. While I told him about it months ago, he apparently didn't understand what I was saying and now thinks its a glitch. I don't see how it could be a glitch given that this particular thing is pretty straight forward. I've measured a lot of devices that Matt or the manufacturer also has to measure and we have had good agreement once everything is calibrated right, so I can't figure out what would cause this gain problem. I think it's related to the dewarping process?

What are you using to measure?

 

[shadyJ]

I just want to thank Team AH for going back in and addressing this, making the edit! I appreciate the efforts and am super glad to be a part of this forum where you guys are willing to discuss this openly.
@gene , @shadyJ , @Matthew J Poes : seriously, you guys rock!

No! We must never let Matthew forget his mistakes! Remember this always, and hold it against him!