Focusright VRM Box Virtual Reference Monitoring Page 2

METHOD
First, let's talk about how it works. Focusrite explains it in simple terms HERE, with a link to a more detailed explanation in the form of an AES paper. The process starts with room measurements: for each of the three room choices Focusrite uses a complex calculation to determine the behavior of the space in terms of reflections. Obviously a large open area will behave differently than a small well-treated studio, and the algorithm aims to capture those differences. The second ingredient is the measurement of the actual speakers--Focusrite attempts to capture the frequency response, phase response, and directivity of each model. This is accomplished by taking a precise series of polar plots at 15 degree intervals. The theory is that a speaker sounds the way it sounds largely because of the radiation pattern it produces, so capturing that and combining it with the room model allows the sound to be recreated over a standard pair of headphones.This is a very simplistic explanation--I highly recommend reading the AES paper for the specifics, as it is a fascinating topic.

The VRM algorithm itself was developed by Focusrite audio engineer Ben Supper. Ben has a PhD in spacial psychoacoustics, holds multiple patents related to environmental processing (links HEREand HERE), and is generally an all around smart guy. Although he was not involved in the design of the VRM Box hardware, he tells me he's pleased with the way it turned out, with a few exceptions--he would have liked a hardware on/off button for example. I was surprised when he mentioned the headphones used during development of the VRM technique--entry level Sennheiser and Grados among others. He in turn was interested in hearing my experience regarding which headphones I found least effective. Ben happens to be extremely busy at the moment so I didn't bother him with as many questions as I would have liked to. Perhaps another time.

Interestingly, this particular version of VRM runs purely in software. It does use some system resources, but seemed to function well enough even on a low powered netbook computer. Focusrite makes a far more expensive product called the Saffire PRO 24 DSP which runs the process via hardware. This allows more versatility such as choosing different listening positions. Theoretically, the company could release this software VRM that could work with any DAC. They advised me that no plans exist to do so in the foreseeable future.

LISTENING
The big question for the VRM box: is it really convincing? Does it sound like being in the selected room, listening to the chosen speakers? The short answer is...sort of...potentially...sometimes...at best, yes it can. It really depends on the headphones used, and how well they mate with the selected virtual speakers. Even the type of music being listened to plays a role. Another key factor is the ability of the user to mentally "let go" and accept the illusion rather than fighting it---apologies if that sounds too "touchy-feely," but it really is the case.

The best way I can describe the effect is that it has a lot in common with crossfeed, yet is also completely different than any crossfeed implementation I've ever heard. It is similar in that it does blend everything together and the result can be rather drastic. Using Giant Steps by John Coltrane as a classic example--piano and sax get hard panned to the left, drums and bass to the right. This can be awkward to listen to with headphones. Any sort of crossfeed will blend the sounds creating more of a center image and better approximating the sound you would hear through speakers. Some implementations will do better than others, but generally speaking it's the type of thing that you either love or hate. Personally I like crossfeed with extreme examples like Coltrane or Hendrix but not much else. I find that the drawbacks (soundstage drawn in too closely, muddy bass) usually outweigh the benefits.

VRM processing is different in that it doesn't just blend the channels but applies processing to the sound as well. Some of the virtual speakers do have muddy bass, but some are very well articulated. Some have grainy highs, or boxy mids, or various other issues, and some just sound pretty darn good all the way around. I tended to like the Adam and Genelec monitors best although a few of the others had their charm as well. In terms of not just liking the sound, but finding it convincing: If the mood was right and I was focused on enjoying the music rather than analytically breaking down the effectiveness of the VRM processing, I did in fact have moments where I could actually believe the sound was emanating from speakers in front of me rather than headphones around my ears. It helped to close my eyes and imagine real speakers in the room with me. It was never overwhelmingly realistic, and the illusion was easily broken by any number of distractions, but at times the VRM Box did exactly what it promised...but only when I would let it.

The issues that caused distraction were many. For one, comfortable headphones are a must. Something that practically disappears on your head stands a far greater chance of working than a tight clamping, ear tickling, sweat inducing headphone which continually calls attention to its presence. You also want headphones with a relatively flat and smooth frequency response. Brightness is not your friend in this case. For example, my Audio-Technica W1000x (which I really enjoy when paired with the right amp) had the comfort thing nailed but seemed a bit too hot up top, so it never really worked all that well in this application. It tended to make the virtual speakers have less differentiation from one another, imparting each with the same W1000x flavor. In that case I preferred to leave VRM unselected. I got similar results from the AKG K701 and a few different Grados I tried, so the theme seems to be that lit up highs don't fare so well with the VRM processing. An older Sennheiser HD650 with the black drivers and darker sound did a pretty good job, so a little darkness is acceptable, though I would think going much further than that would have drawbacks as well. I suspect the current version of the HD650 with the latest drivers would do a very good job but I didn't have one handy to confirm that.

Overall the best results I got came from the HiFiMAN HE400, Smeggy Thunderpants, and Sennheiser HD600. Each of these was capable of sounding very convincing at times and they also sounded pretty good even with the VRM turned off. It would take a few minutes for my brain to adjust to the VRM sound, but once it did I really enjoyed it. Even at the moments where I was fully aware of the illusion it could still be fun to listen to. I noticed that I tended to favor music which already contained realistic imaging. Classical, jazz, and live recordings were a better fit for VRM processing than typical rock or pop music, especially newer stuff which doesn't contain much real spacial information anyway.

GOOD AND BAD
Which brings me to the biggest weakness of the device: output impedance. At 10 ohms, the chances are good that some serious interactions are going to happen with a variety of headphones. This not only hinders its use as a standard DAC/amp but also impedes the chances of VRM sounding convincing. Think about it: if the key to believable VRM is smooth, neutral frequency response, and a higher output impedance creates unpredictable interactions in the frequency response...it just makes it that much more difficult to find a good match. Most IEMs are out, for example. My Denon D7000 (25 ohms) had loose, flabby bass due to a low damping factor. Audio-Technica, Grado, and Ultrasone all primarily make low impedance headphones as well, so those are out. Perhaps my W1000x was never the ideal candidate, and I agree with Tyll's review where he said imaging was hindered by non-articulate highest octaves. But it may have been exacerbated by impedance interactions. The reason HD600 and HD650 worked so well is likely because of their higher impedance. The planar HiFiMAN and Thunderpants models, while not quite achieving the ideal damping factor, have reactive loads and thus aren't too bothered by the higher output impedance.

Aside from the unfortunate output impedance (which certainly could have been worse), there are some other quirks to the unit but nothing too serious. For example, the VRM presented the KRK RP6 monitors with significantly more low frequency impact than the larger KRK VXT8. I find that somewhat hard to believe but I haven't heard those particular speakers so I can't say for sure. Likewise, the two different brands of LS3/5a monitors sounded like completely different speakers rather than subtle variations of the classic BBC design. But again I have no personal experience with those two so maybe that's just the way they really sound.

SUMMARY
All things considered (especially the $99 price tag), I still think Focusrite has a great little product on their hands. Would I like to see a lower output impedance and slightly higher gain? Sure. Yet the VRM Box remains a compelling device--for the unique VRM feature as well as simply for the compact DAC/amp functionality. It's no Smyth Realizer, but I think it's worth giving a shot if this sort of thing even remotely interests you.

RESOURCES
Focusright VRM Box product page.
List of all virtual speakers implemented.
List of rooms and which speakers are available in that room.
Personal website of VRM creator Ben Supper.

COMPANY INFO
Focusrite Novation Inc
840 Apollo Street
Suite 312
El Segundo, CA 90245
us.sales@focusrite.com
(310) 322-5500
ARTICLE CONTENTS

COMMENTS
mir's picture

Thanks for the review.

However I have to totally disagree that AKG K70x sound similar with VRM Box turned on with all modelled speakers. I've been using VRM Box for half year now with K702 in studio and I can confidently tell you that all of the modelled speakers sound very, very differently on K702. If you can't hear strong difference on K70X between car-speaker powered mid-fi Auratone and bass heavy Quested (British Studio) or even bass heavier KRK Rokit (which, for your curiosity, is widely criticized in pro audio circles for their mud - yes, they have too much bass in low end, that's fact, some hate them, some love them), then there must be something wrong, sorry. And for too much high end - yes, listening with K70X in overall isn't fun (even with VRM Box used only as DAC with processing turned off), unless the recording isn't bright/flawed itself. 

No offense, but I just don't get it how you can write such thing. I've used VRM Box with K702s to successfully help me balance low end in some mixes and it definitely worked/translated to the outside world. It just doesn't make sense what you write - if that would be true, VRM Box with K702 would have never worked and would be useless - I wouldn't have known then how crappy some mixes were!

John Grandberg's picture

Seriously, read what I posted again, specifically about the K701 (and W1000X, and Grados).

I didn't say they make all virtual speakers sound the same, as if the VRM processing is rendered completely ineffective. I simply said the differentiation is less than it is with certain other headphones. And I stand by that. Obviously it works well enough for you to mix with, and for that I am glad. 

Have you tried any other headphones with the VRM Box? If so, which ones? I'm curious to hear what sort of results different people are getting with different models. I'm sure VRM creator Ben Supper will like to hear that info as well. 

mir's picture

I don't have another studio headphone to compare with, because I'm pretty happy with them actually. They're not easy headphones, they require some hard time to get used to their sound, but I think it's worth it. I've learned to not blame them if something's wrong in the mix - it turns out it's the mix that is wrong :)

As for the less/more differentiation - I can't compare whether the results would be better with another headphones, but my experience shows this combination just works.

donunus's picture

Not all k70X cans sound the same. The one I got for instance was not an accurate headphone at all due to the upper midrange and lower treble peak that it had. 

Armaegis's picture

I've been using the VRMbox as a secondary rig for a while now and most frequently pair it with the Senn PX100-ii. Perhaps the lightness of the Senns just makes them disappear, but I really do forget that I'm wearing headphones sometimes and have on more than one occasion walked away from the computer and had the headphones rather harshly yanked from my head. 

In general I find the VRMbox works best with open headphones. Closed ones wind up sounding a bit boxy to me. 

I do wish it had more power though. Actually, it never occured to me to try feeding the VRM output into another amp. Hmm...

John Grandberg's picture

Because we keep ending up with the same gear. Creepy. 

Joking aside, I do see your point about open headphones generally working better. My particular Thunderpants (because they aren't all the same) has an unusually open sound for a closed can, which is probably why it is an exception to that rule. 

It's a good thing you weren't using IEMs when your "incident" happened. That could hurt. 

Armaegis's picture

Well considering just how much gear you've got... heh. I wish I could be able to hear some of the big toys that you've gone through though. How about you send me all your extra stuff and I'll review them too? =P

I preferred the sound with open headphones since in general they have fewer reflections and resonances. The whole VRM thing basically adds that artificial room resonance (sort of not really; you know what I mean), which when re-reflected in a closed headphone makes it sound a bit odd. 

I'm in the middle of reviewing the NuForce Icon2 and S-X & W-1 speaker set. I'm going to see if *maybe* I can get the DAC-100 from them. Maybe if I tell them I've got an HE-6 coming in (I traded 2/3 of my inventory for it!), they'll send me one?

One issue I've had with reviewing is getting stuff across the border into Canada. I've actually once had to send an item back because customs was being a pain and after a week of paperwork and wasting so much time driving back and forth to the customs office and the depot, I just gave up. 

khaos's picture

The concept of simulating speakers via headphones is very nice, the aim, as I understand it is to allow sound engineers to hear how their music translates on various speakers.

Unfortunately, the Focusright Box misses a crucial feature, measuring the headphones themselves, the signature of the headphones is still too embedded in the final result. Conceptually wise, it's not too different from the ambiance settings (Hall, Bathroom, Jazz, Chruch) on an AV Receiver, except tailored to speaker models instead of ambiances.

It's a first step, but still incomplete (though it's understandable given the price).

mir's picture

It's not 100%, but I find it valuable and somehow "accurate" (if I can use such word, you know what I mean) - if several perfectly polished/balanced commercial mixes/masters sound pretty much the same (tonally) on all modelled speakers, and bad mixes sound very differently and unbalanced, that speaks for something great. That's my experience with K702s. I think Focusrite has done a really nice job with VRM Box. It can really help balance mixes and discover flaws/details (to some extent, of course). Well done, Focusrite. I would definitely recommend it.

ultrabike's picture

I was salivating over this product in the past for a while. The 10 ohm impendance is sort of a turn off though, but the price and coolness factor is quite good. It's also nice to see a competent DAC in there (same as the one inside the well regarded Leckerton UHA-6S MKII - Cirrus Logic's flagship!).

Hopefully Focusrite will rev it up and lower the impedance in order to better match lower impedance HPs (maybe improve the amp specs.) Or maybe add a DAC output option given their DAC IC choice.

BTW, they did mention they used the HD650, Beyer DT770, and the Proline 650, which may sort of work OK with 10 ohms like you metioned: http://www.focusrite.com/answerbase/en/article.php?id=1134.

John Grandberg's picture

I agree and definitely see room for a "rev 2" or similar. The ability to feed an external amp would take this to a whole different level imo, and I could see it becoming rather popular. Hopefully Focusrite is listening. 

Lunatique's picture

There are a few software alternatives that you really should try:

TB Isone - http://www.toneboosters.com/tb-isone/

Redline Monitor - http://112db.com/redline/monitor/

Monitor MSX5 - http://www.g-sonique.com/msx5headphonemonitoring.html

They are all pro audio plugins that were designed for headphones, and TB Isone in particular, uses sophisticated HRTF algorithm, and its creator (Jeroen Breebaart) is one of the most respected audio engineers in the world (just read his bio and see). The other two are more conventional crossfeeds, and aren't as sophisticated, but they alter the frequency reponse less than HRTF.

TB Isone is the most realistic of the three, and I never listen to headphones without it anymore. I personally prefer the previous version called Isone Pro, but it's possible to tweak TB Isone to sound like Isone Pro (althought you'd need a copy of Isone Pro as reference). I'm sure Joroen will send you Isone Pro if you ask him.

dunky's picture

This is an impressive blog. Find it very interesting. You've provided a great source of information. A big thumbs up for having created this blog. Thanks for sharing..

Buy Reverbnation Plays

rubyproduction's picture

Thanks for useful post , btw here is some useful websites for music production
https://www.lucidsamples.com
https://www.looperman.com/forum

X