Emotiva XMC-2 Review (AV Processor)

[amirm]

Why doesn’t ASR take his offer?

First, the offer was not made to me and this is the first I hear about it. Even if it were, I have a bunch of other work to do than spending a ton of time measuring their products yet again especially when their measurement matches mine as far as SINAD.

 

[TimoJ]

See my post above https://www.audiosciencereview.com/...review-av-processor.26378/page-17#post-906491

But isn't Lonnie talking about input signal level, not volume control position? Or does the XMC-2 also adjust digital levels (headroom) with its volume control (like the HTP-1)?

 

[Lsc]

First, the offer was not made to me and this is the first I hear about it. Even if it were, I have a bunch of other work to do than spending a ton of time measuring their products yet again especially when their measurement matches mine as far as SINAD.

I posted on the Emotiva website asking Lonnie if he was willing to ship the processors to ASR and he said he would be willing. So correct, no official offer was made.

It’s all good. I think the feelings are mutual.

 

[GXAlan]

I posted on the Emotiva website asking Lonnie if he was willing to ship the processors to ASR and he said he would be willing. So correct, no official offer was made.

What Lonnie should really do is go to Best Buy, pick up an AVR X8500H, throw it into preamp mode and test. Then his engineers have something to compare against at -20 and 0. Automotive manufacturers buy competitor cars all the time to disasssemble and reverse engineer. As CTO, he has a team under him to do the actual work, so he may not be directly involved…

Because what I want is a AVP-8500 from a company like Emotiva. Without all 11 channels of amplification, it should be cheaper…

 

[David_M]

But isn't Lonnie talking about input signal level, not volume control position? Or does the XMC-2 also adjust digital levels (headroom) with its volume control (like the HTP-1)?

He was referencing customer levels, so that would be the volume control I believe. Either he sets the AP to produce a -20dBfs signal with the XMC-2 volume control set to 0dB or he sets the AP at 0dBfs with the volume control set to -20dB. Either way, same result but I think it might be using the former configuration and not the latter.

 

[Lsc]

What Lonnie should really do is go to Best Buy, pick up an AVR X8500H, throw it into preamp mode and test. Then his engineers have something to compare against at -20 and 0. Automotive manufacturers buy competitor cars all the time to disasssemble and reverse engineer. As CTO, he has a team under him to do the actual work, so he may not be directly involved…

Because what I want is a AVP-8500 from a company like Emotiva. Without all 11 channels of amplification, it should be cheaper…

Lonnie and every other processor manufacturer since the 8500 is tops in SINAD.

I mean who the hell would pay $15-20k for a processor that’s inferior to a receiver? All those processors should get a do not recommend! Buy a Denon!!! Essentially according to ASR, all processors suck because they are inferior to the Denon. Trinnov? Storm? Datasat who? Why can’t they engineer a processor for 5 figures and not outperform a receiver?

I don’t really believe any of this but anyone can play devils advocate is my point.

 

[amirm]

Why can’t they engineer a processor for 5 figures and not outperform a receiver?

I think until recently buyers assumed superior hardware performance just based on price, brand, etc. Now that our testing has shown otherwise, they actually need to deliver on this or see fading sales for reason you mention. There is so much standardized in these products that there isn't enough differentiation to market them otherwise.

 

[wseroyer]

This is a review and detailed measurements of the Emotiva XMC-2 Audio/Video Processor (AVP). It is on kind loan from a member and costs US $3,000.

The XMC-2 is surprisingly light for a high-end processor which my back appreciated as I carried it around:

View attachment 151604

Ironically, the remote is made out of metal and weighs a ton as remotes go.

Back panel is as you expect for this class product with balanced XLR outputs:

View attachment 151605

For testing, I performed a factory reset. I was hoping that this would get it to output video to my Samsung monitor but it would not. My HDMI interface would recognize it and I could pump audio through it but no picture.

Configuring the unit through the front panel was painful as some options would take as much as 5 seconds to take effect (e.g. User EQ). Worse yet, it would buffer all the clicks and then act on them in sequence making it maddening to operate.

The volume control as very soft detents that did not allow precise adjustments of the volume control. 0 dB for example sometimes would be at the peak of a detent rather than the notch. When in this situation, it would alternate between -1 and +1 dB and even +2 dB at times. Strangely it would change as I would rotate the volume.

There were other strange things like selecting the Direct mode and then back causing the output voltage to jump to 7+ volts. But the moment I would touch the volume control, it would go back to near 4 volts as I had it set.

Considering how little I use these products for just testing them, this level of problems is way too excessive and indicates a product that simply is not ready for distribution and needs to have a complete test run on it and many bugs fixed.

XMC-2 Measurements
As usual, I start with testing the HDMI input. Here is our dashboard there:

View attachment 151606

We have so many issues here. Starting with top left we see that the two channels are out of sync. This issue was frequency sensitive which likely indicates one channel being N samples behind the other. Remember that I performed a factory reset so there should be no programmed delay.

Rotating to top right we see heavily rising noise floor as frequencies get lower. Why??? Distortion by itself is bad news enough at -88 dB or so. It doesn't need extra noise to make the situation worse. Speaking of noise we can measure it and see the frequency dependence of it:
View attachment 151607

We lose nearly 25 dB of dynamic range as frequencies go from 1 kHz to 20 Hz! [right side should say 15 bits, not 15 dB]

I figured I try Coax/Toslink inputs to see if there is a problem specific to HDMI but got more problems:

View attachment 151608



Using the better of the two SINAD scores, the XMC-2 lands in the poor category of all AV products tested:

View attachment 151609

It is also much worse than XMC-1!

I ran a jitter test and sure enough it also produces a ton of jitter and low frequency noise:

View attachment 151610

I didn't see a reason to keep testing the XMC-2.

Conclusions
The XMC-2 reminds me of getting a new job at a company and taking over a messy project with tons of problems. Lesson I learned there was to just ditch then thing and start over with a clean design. I think such is the situation with XMC-2. It clearly has a list of hardware and software design problems.

Needless to say, I can't recommend the Emotiva XMC-2. Get yourself a Denon AVR and use its pre-amp outs. Yes, you don't get balanced output but if you are not facing ground loops and connections are short, you should be OK. At least you get a working product with very respectable measurements.

----------
As always, questions, comments, recommendations, etc. are welcome.

You may have noticed that the head of the headless panther has made an entrance here. It pleaded with me to get it glued back to its former body. I told it that I can't afford the money but if he starts "working" again as a model, maybe the membership starts to feel sorry for him and donates money toward this cause using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

What's the best measuring 16ch processor, i'm building a custom house for a client of mine who wants a balls to to wall home theater, and i'm struggling to find him something that's not going to be a buggy mess with 16ch, I really want to tell him to just get a AVR-X8500H but we need more then 13 channels.

 

[amirm]

What's the best measuring 16ch processor, i'm building a custom house for a client of mine who wants a balls to to wall home theater, and i'm struggling to find him something that's not going to be a buggy mess with 16ch, I really want to tell him to just get a AVR-X8500H but we need more then 13 channels.

I can't speak to how buggy they are as that is not something I check. Membership can advise about that. As for performance, you have just a few choices in the green range:

 

[wseroyer]

I can't speak to how buggy they are as that is not something I check. Membership can advise about that. As for performance, you have just a few choices in the green range:

ironically it Looks like it will be the RMC-1 since the others will blow the budget. I really wish Denon/Marantz would come out with a 16ch processor so my choice would be easy.

 

[TimoJ]

He was referencing customer levels, so that would be the volume control I believe. Either he sets the AP to produce a -20dBfs signal with the XMC-2 volume control set to 0dB or he sets the AP at 0dBfs with the volume control set to -20dB. Either way, same result but I think it might be using the former configuration and not the latter.

But he is then not setting XMC-2's output for 4V voltage like Amir does for his tests?

 

[TimoJ]

I can't speak to how buggy they are as that is not something I check. Membership can advise about that. As for performance, you have just a few choices in the green range:

Right choice is the Monoprice HTP-1, using 2.7V output SINAD is around 102dB.

 

[poopy]

I acknowledge that I should link to the measurement article to help with that. I just hate stuffing boilerplate stuff in reviews but being forced to do so (been doing it for speakers and headphones). Beyond that, it is hard to create comprehensive guides in text. It takes me way too long to do so. I have done some videos and will do more of them as that is easier to do. So your larger point is taken.

@Amir - I know it could be very time consuming, but why not considering making a video including all measurements for an AVP or an AVR of your choice ? So basically you start off showing and describing all your equipment you'll use for the review. Then for each test, you explain the goal, describe the measurement protocol, how you connect your equipment, the setting you use for your measurements etc... and then you give an application to an AVP or AVR by showing and commenting the results. A comparison with what ideally we should get could be also included. You would be doing that for every single test you usually perform for an AVP. Each test could be seen independently from each other.

 

[SimpleTheater]

What's the best measuring 16ch processor, i'm building a custom house for a client of mine who wants a balls to to wall home theater, and i'm struggling to find him something that's not going to be a buggy mess with 16ch, I really want to tell him to just get a AVR-X8500H but we need more then 13 channels.

I've been waiting for years for an acceptable 16ch processor (SINAD > 100 db) to upgrade my current 13 channel system. The longer I wait, the less interested I get in upgrading because as of right now everything just works. The last couple of years has taught me that having a system that I can enjoy, instead of dealing with buggy software, is definitely worth the loss of a couple of channels.

PS: Sorry to the moderator if this post is considered off topic.

 

[DonH56]

Have not read all the posts, sorry, busy this morning...

Lonnie responded to my post in the Emotiva Lounge and clarified that -20 dBFS really is the digital reference, not any sort of analog signal, so it is -20 dB relative to a digital full-scale input. Note dBFS is dB with respect to full scale, typically of a a data converter (ADC or DAC), so generally applies to the digital input (bits, numbers) fed to a DAC. In my world it is also used to indicate the analog input to an ADC relative to full-scale digital output, and to the analog output from a DAC relative to a full-scale digital input. Maybe wrongly, but it's how I was weaned, and how it was used at the companies I have worked through the years.

This is not how an ADC or DAC is typically tested, at least IME; Lonnie says -20 dBFS is more typical of real-world conditions to allow for headroom, perhaps more for HT processors? Not really my field, but I would have expected them to be tested like anything else since you control the input to prevent overruns (data converters are typically tested at -1 dBFS to provide a small buffer for noise and such). I defer to those who know.

Running at -20 dBFS should significantly improve distortion, mainly from the analog stages at and after the DAC , at the cost of reduced SNR since the noise will be higher relative to the signal.

FWIWFM - Don

 

[muslhead]

Have not read all the posts, sorry, busy this morning...

Lonnie responded to my post in the Emotiva Lounge and clarified that -20 dBFS really is the digital reference, not any sort of analog signal, so it is -20 dB relative to a digital full-scale input. Note dBFS is dB with respect to full scale, typically of a a data converter (ADC or DAC), so generally applies to the digital input (bits, numbers) fed to a DAC. In my world it is also used to indicate the analog input to an ADC relative to full-scale digital output, and to the analog output from a DAC relative to a full-scale digital input. Maybe wrongly, but it's how I was weaned, and how it was used and the companies I have worked through the years.

This is not how an ADC or DAC is typically tested, at least IME; Lonnie says -20 dBFS is more typical of real-world conditions to allow for headroom, perhaps more for HT processors? Not really my field, but I would have expected them to be tested like anything else since you control the input to prevent overruns (data converters are typically tested at -1 dBFS to provide a small buffer for noise and such). I defer to those who know.

Running at -20 dBFS should significantly improve distortion, mainly from the analog stages at and after the DAC , at the cost of reduced SNR since the noise will be higher relative to the signal.

FWIWFM - Don

Its people like Don who make this website so valuable.
Thanks for your follow through and explanation.

 

[David_M]

But he is then not setting XMC-2's output for 4V voltage like Amir does for his tests?

No, he's not.

 

[bobfriedman]

Because what I want is a AVP-8500 from a company like Emotiva. Without all 11 channels of amplification, it should be cheaper…

Or Denon AVR-8500h with XLR.... I would buy that!

 

[abdo123]

Have not read all the posts, sorry, busy this morning...

Lonnie responded to my post in the Emotiva Lounge and clarified that -20 dBFS really is the digital reference, not any sort of analog signal, so it is -20 dB relative to a digital full-scale input. Note dBFS is dB with respect to full scale, typically of a a data converter (ADC or DAC), so generally applies to the digital input (bits, numbers) fed to a DAC. In my world it is also used to indicate the analog input to an ADC relative to full-scale digital output, and to the analog output from a DAC relative to a full-scale digital input. Maybe wrongly, but it's how I was weaned, and how it was used and the companies I have worked through the years.

This is not how an ADC or DAC is typically tested, at least IME; Lonnie says -20 dBFS is more typical of real-world conditions to allow for headroom, perhaps more for HT processors? Not really my field, but I would have expected them to be tested like anything else since you control the input to prevent overruns (data converters are typically tested at -1 dBFS to provide a small buffer for noise and such). I defer to those who know.

Running at -20 dBFS should significantly improve distortion, mainly from the analog stages at and after the DAC , at the cost of reduced SNR since the noise will be higher relative to the signal.

FWIWFM - Don

The RMS value of audio tracks for most movies, series .etc is actually arround -28 to -30 dB. that's why for Home Cinema I think S/N is absolutely essential because the moments you get even close to 4Vrms or 2Vrms of output are extremely rare.

But at the same time i noticed a pattern with @amirm 's AVR reviews, I think since almost all AVRs and AVPs are +/- 10 dB of SINAD between each other i feel like his overall experience with the device has much more impact on his final recommendation. Considering how the majority of users here are also complaining about a plethora of problems I kind of trust Amir's verdict on this one. this is not something I would pay 3000$ for.

 

[TimoJ]

Emotiva's Lonnie writes: "I would have been happy to explain that it was the Dolby up mixer sitting in the background and if he lowered his test signal to -20dbfs it would all go away."
Now, I wonder what this means.. Didn't Amir use Direct Mode, so why is Dolby upmixer active in the background?