Review and Measurements of Emotiva XMC-1 Gen 2 Pre/Pro

[hutt132]

It really is disappointing how all AVRs measure so poorly, especially the analog inputs to use the AVR as a preamp. My whole plan for my new theater setup was to get a high-end 11.2 pre/pro (possibly marrantz) with a nice external amp for movies, and then feed the pre/pro with a desktop dac to use it as a preamp for music in my living room.

But since this measures so poorly even with the analog inputs, it’s really making me think it is not possible to have a nice movie/music listening setup.

 

[RichB]

The muting appears to occur when ONLY below 90 db audio is present so during that you won't be able to calculate linearity below that. That doesn't mean that it doesn't resolve audio below 90 db. It means that if ALL the audio is below 90 db that it will mute.

If that is the case, then should we be gluing on the panther's head?

- Rich

 

[Dj7675]

Can someone explain a bit more on what the broken linearity tests mean in real world use? I'm just not following.

 

[DonH56]

Anything below -90 dB FS or so is (potentially*) going to be muted. How much real-world impact that has I could not say. But it means your resolution is limited to about 15 bits.

(*) I do not know if there is hysteresis or other programming that does not mute quick dips below -90 dB but only when it dwells there for some time. The aforementioned "pop" issue was mainly when changing sources or when the bitstream died, e.g. between tracks of a CD or when changing formats on a movie or live cable link, and thus the data drops to 0 for a period of time. I do not recall it being a problem say during a movie or in the middle of a song, but it could well be they simply never dropped to complete 0.

 

[audimus]

The muting appears to occur when ONLY below 90 db audio is present so during that you won't be able to calculate linearity below that. That doesn't mean that it doesn't resolve audio below 90 db. It means that if ALL the audio is below 90 db that it will mute.

In English please. Does that make the unit sound bad or good relative to the NAD?

 

[audimus]

If that is the case, then should we be gluing on the panthers head?

That depends on whether the head was blown off by what the panther heard or by what it saw. A blown off head is ambiguous as to the reason.

 

[DonH56]

If it is only muted when a 0's are detected "for a while" then probably no impact. I believe that to be the case since muting takes a little time to engage and disengage so would be clearly audible in the middle of a song or movie. But it may be challenging to figure out exactly how long 0's must stream by before it mutes. I am hoping the AP dwells there for a while in their test and thus the mute circuit is activated. I suspect it would be difficult to restructure the test to avoid that.

Don't have a NAD; the other measurements are better, and I own an XMC-1, so I'm gonna' have to go with "good".

 

[audimus]

If it is only muted when a 0's are detected "for a while" then probably no impact. I believe that to be the case since muting takes a little time to engage and disengage so would be clearly audible in the middle of a song or movie. But it may be challenging to figure out exactly how long 0's must stream by before it mutes. I am hoping the AP dwells there for a while in their test and thus the mute circuit is activated. I suspect it would be difficult to restructure the test to avoid that.

Do we know if it is muting (of output) or some kind of DSP filtering of input below certain level to be converted to silence output? Looking at any digital out should differentiate between packets coming out with silence (zeros) if filtering and no packets coming out if muting when a low continuous input is provided. Audio out should show floor noise or no noise depending on what it is. May be that information is already there but my eyes glazed over amongst all those graphs.

Muting has larger implications depending on the latency to unmute. So, even if it goes to mute on a lot of zeros, unmuting may result in loss of initial content when the level goes higher. Some players have an option to output low noise or silence packets to prevent muting if that latency is high downstream. That would become ineffective also.

I think, more exploration is necessary before prematurely lopping off panther’s heads.

 

[Dj7675]

If it is only muted when a 0's are detected "for a while" then probably no impact. I believe that to be the case since muting takes a little time to engage and disengage so would be clearly audible in the middle of a song or movie. But it may be challenging to figure out exactly how long 0's must stream by before it mutes. I am hoping the AP dwells there for a while in their test and thus the mute circuit is activated. I suspect it would be difficult to restructure the test to avoid that.

Don't have a NAD; the other measurements are better, and I own an XMC-1, so I'm gonna' have to go with "good".

Thanks Don, that helps.

 

[folzag]

Agreed. This muting (feature or a bug ) needs more investigation.

I thought something was broken when the test started from -120 dB with no output from the unit. Then all of a sudden it output the -90 dB signal.

For starters, can you run the linearity test in reverse?

 

[folzag]

Not that anyone asked, but I think giving the (to date) Best-in-Class AVR the decap panther treatment over the linearity test is a bit harsh.

Unfair even, if it performs acceptably under a more realistic test-case... ie. prime the pump with some realistic signal level.

 

[Dj7675]

Maybe it needs 2 panthers

 

[audimus]

Not that anyone asked, but I think giving the (to date) Best-in-Class AVR the decap panther treatment over the linearity test is a bit harsh.

Unfair even, if it performs acceptably under a more realistic test-case... ie. prime the pump with some realistic signal level.

Whether to decap a panther or not is a subjective evaluation, so there...

 

[Blumlein 88]

In English please. Does that make the unit sound bad or good relative to the NAD?

It has Dirac, so you should be happy. It seems likely the muting is a low signal condition only. It measures better than the NAD. So in real use it has better fidelity than the NAD. Does it sound better? I'll say it is a much better deal.

 

[amirm]

Agreed. This muting (feature or a bug ) needs more investigation.



For starters, can you run the linearity test in reverse?

This is discontinued product. I have lots to review. Does it make sense to spend more time on it?

 

[garbulky]

If that is the case, then should we be gluing on the panthers head?

- Rich

I've used the XMC-1. It sounded pretty great imo. If I was on the market for a surround sound processor the XMC-2 or the RMC-1 would be the only one on my list. Having said that, since I am in no hurry at all, I plan to ask If Amir thought the 3 second delay was much he is going to hate the delay on their Emotiva MC700 processor for HDMI switching was horrendous.

In English please. Does that make the unit sound bad or good relative to the NAD?

It does neither. Most modern DACs have measurements that - in regular music listening - don't play out as showing audible differences. So the difference between a dac with distortion at -80db and a dac with distortion at -120 db doesn't mean that the two dacs will sound different because of that measurement. I doubt I could hear a difference in a DAC with distortion at -65db (or really hear any sound at that level). Though arguably some people could.

 

[Dj7675]

This was a unit I sent in. I just purchased it used and sent it to Amir for testing. My goal was to see how it tested in comparison to the published specs and to see how it compared to other pre/pros such as Marantz and others. I also wanted to compare it to receivers that I could use for an 11.2 channel setup (using external amplification for 3-7 of the channels). The plan was to then trade in this unit for the XMC2 for the $1000 trade in price. Sounds like most of the issues are fixed on the XMC2, but Dirac still has not been released. With the relatively poor showing of receivers and pre/pros, not sure what direction to go. I may still move forward upgrading to the XMC2. It does have PEQ filters to use to tame room modes until Dirac is released.

 

[LLL]

This is discontinued product. I have lots to review. Does it make sense to spend more time on it?

Not at all. It's pretty sad that these components measure so poorly given that we are in 2019. We KNOW how to build these things right. It's not that hard, it just means someone needs to pay attention in designing.

 

[DonH56]

Do we know if it is muting (of output) or some kind of DSP filtering of input below certain level to be converted to silence output? Looking at any digital out should differentiate between packets coming out with silence (zeros) if filtering and no packets coming out if muting when a low continuous input is provided. Audio out should show floor noise or no noise depending on what it is. May be that information is already there but my eyes glazed over amongst all those graphs.

Muting has larger implications depending on the latency to unmute. So, even if it goes to mute on a lot of zeros, unmuting may result in loss of initial content when the level goes higher. Some players have an option to output low noise or silence packets to prevent muting if that latency is high downstream. That would become ineffective also.

I think, more exploration is necessary before prematurely lopping off panther’s heads.

It is muting and it takes time to recover. They said it was a digital 0 problem. You'd have to ask Emotiva for details. I'm an analog guy...

 

[DonH56]

Aren't those all capped at pre-Atmos surround though? If you want Dolby Atmos, DTS:X, or Auro-3D I think you need to go with a home theater receiver or a pre-pro like the Trinnov Altitude 32:

https://www.trinnov.com/altitude_32/

BTW anyone else love the fact that in the back of the Altitude 32 there's a standard PC motherboard port cutout, indicating the whole thing is based on a PC architecture of some kind:

It is PC-based. They upgrade the motherboard/processor/etc. essentially every year but one of the (Trinnov or Harman calibrator, forget which, think Harman) main guys has an original (S/N 5, I think he said) and it is still doing fine processing 32 channels and all that jazz. Kal recently reviewed one and gave it high praise. Specs, what they provide, are not all that great but probably in line with the rest of the AVR crowd. From my read it consistently gets high marks for performance and flexibility but comes with an exceedingly high price tag. Dataset is another, perhaps "the" other player in this field, but they stumbled badly from what I've read and are working to recover. Dataset uses Dirac Live while Trinnov does their own fancy (proprietary) room correction scheme. Trinnov and Harman have partnered so a lot of Harman (e.g. Dr. Toole's) research is flowing into the Trinnov/Harman processors, something I find appealing.