Smyth SVS Realiser Virtual Surround Technology for Headphones

[wavoman]

At Can Jam '08, no one could tell the difference between the speakers Smyth had set up, and the Stax. The DSP had the Stax reproduce exactly the sound waves that arrived at your ears when you heard the speakers (since they measured it!).

But the Stax can do this, and cheap-o HP's can't. Moreover, there are most likely uber-expensive speakers that produce sound waves at your ears that the Stax can't reproduce, even though the DSP wants them to.

There is nothing artificial or over-processed in the Smyth sound. The sound through the Stax is exactly like the speakers. The Smyth tilt-switch cleverly lets you prove this for yourself in an air-tight A/B test.

As you can tell from various threads, I am as far from a Smyth fanboy as you can get. But what I said above is true. This unit does exactly what they claim.

When it is marketed, and you listen to it, you can, and will (I predict) use it for 2-channel audio. After all, that's what you use your speakers for, no? I plan to get measured in a room at a dealer with $20,000 speakers, and take the memory card home with me. I hope they offer the unit with top-of-the-line Stax, not entry level.

Think of it as customizing Stax HP's so that the actual freq response you hear is perfect, and matching the soundstage you get from speakers. If it was sold as an option built in to the Stax amp, we all would be buying it now, not arguing over it. Yea, the SVS can also be used for 5.1 Dolby or DTS movies, but that's not the breakthru in any way.

Give in to the magic now ... it will save you time in the long run. Any purist who listens to speakers will be convinced in a heartbeat.

[earwicker7]

Has anyone tried these without the speakers that the manufacturer provided? (Sorry if someone has and I missed it in the thread!) I'm wondering if it's possible that the "It sounds exactly the same" result is coming from using speakers and headphones with similar coloration. I would be far more convinced by a demo with, say, three speaker setups using radically different speakers.

This being said, I still think it sounds awesome for movies/games, and I'd certainly consider buying one. I'm just not convinced that it's physically possible for the setup to do what it claims.

Then again, I have a "magic power cord" (the Shunyata Python VX) that solved every interference problem that skeptics said it couldn't possibly solve, so maybe I'm way off base.

[wavoman]

We are not talking similar, we are talking identical, as in "you think the speakers are playing".

We witnessed Smyth fool a Head-Fi'er who shouted "turn off the speakers and turn on the headphones" when in fact the headphones were playing ... I won't go in to all the details of the ruse but the set-up is perfect. Legend says Smyth won a $1000 bet (but the bloke didn't pay) during CES in Vegas when someone wagered they could tell the speakers from the headphones.

You can't. The soundwaves the speakers produce at your ears are being reproduced by the headphones.

I did listen to Smyth's raw speakers and the Stax raw too -- they sounded nothing alike, that's not the explanation.

The explanation is simple -- the mikes measure the effect the speakers have on your ears, and the DSP imitates it, compensating for the chamber of the headphones and the turning of your head.

Twenty of us heard, and witnessed.

[earwicker7]

Originally Posted by wavoman

We are not talking similar, we are talking identical, as in "you think the speakers are playing".

We witnessed Smyth fool a Head-Fi'er who shouted "turn off the speakers and turn on the headphones" when in fact the headphones were playing ... I won't go in to all the details of the ruse but the set-up is perfect. Legend says Smyth won a $1000 bet (but the bloke didn't pay) during CES in Vegas when someone wagered they could tell the speakers from the headphones.

You can't. The soundwaves the speakers produce at your ears are being reproduced by the headphones.

I did listen to Smyth's raw speakers and the Stax raw too -- they sounded nothing alike, that's not the explanation.

The explanation is simple -- the mikes measure the effect the speakers have on your ears, and the DSP imitates it, compensating for the chamber of the headphones and the turning of your head.

Twenty of us heard, and witnessed.

I'm not doubting for a second that what you're saying is true, but from a statistical standpoint, if you were doing a scientific study on this thing (god, I'm starting to sound like a chart nazi, sorry) I don't think that one set of headphones and one set of speakers is enough to validly say that this thing can make a set of headphones sound like any speaker setup in the world.

[Mazz]

Originally Posted by wavoman

[...]when someone wagered they could tell the speakers from the headphones.

You can't. The soundwaves the speakers produce at your ears are being reproduced by the headphones.

I beg to differ - but only a tiny bit.

I could hear a slight difference, but it took me several minutes of careful listening to hear it. And even so Smyth said the difference would probably get smaller with a full calibration instead of the abbreviated version he was running due to time constraints - there was quite a queue building up.

Originally Posted by wavoman

We are not talking similar, we are talking identical, as in "you think the speakers are playing".

Most of the time that I was listening to the headphones my experience was such that I would have been able to stand up in a court of law and swear truthfully that I heard the speakers playing.

Just after the basic calibration a test sound was played through the center channel via the headphones. I almost jumped because the sound was so obviously emanating from the center channel speaker.

Yes, the transducers in the headphones place limitations on what waveforms can be reproduced (no matter how much processing and amplifying you do), but with that speaker setup (which seemed decent enough), and under those listening conditions it was an impressively effective simulation of the speaker rig.

[SunDevil]

It's very interesting to read that some people (most of them actually tested the SVS and/or Headzone) "defend" these systems and some (most of them haven't heard any of the systems) think it's impossible

Does everybody actually know how we (humans) are able to "detect" from which direction a sound is coming or how we are able to "detect" the distance. Maybe most of you would answer this question with: "of course I know!", but do you really know??

[earwicker7]

Originally Posted by SunDevil

It's very interesting to read that some people (most of them actually tested the SVS and/or Headzone) "defend" these systems and some (most of them haven't heard any of the systems) think it's impossible

Does everybody actually know how we (humans) are able to "detect" from which direction a sound is coming or how we are able to "detect" the distance. Maybe most of you would answer this question with: "of course I know!", but do you really know??

We detect them based on the temporal difference between what the two ears hear. This isn't that hard to manipulate with delays, so I'm 100% behind the surround sound aspect of the system; it looks like it could easily become the standard and I wouldn't mind having one for movies/gaming.

My problem is with the "I can make this headphone and this amp do things it isn't physically designed to do" claim. I'm assuming that, if you can use this to imitate a speaker, there's no reason why you can't make it imitate a headphone. Can someone please explain to me how you make a lower end Stax with a stock amp sound like my Omega 2 with a $7,500 aftermarket amp? Stax had to do a ton of engineering to get these cans to have a bass response that no other electrostatic headphones can touch, and you're telling me all it takes is a little jimmying with the signal? A signal processor can undo the limitations of a (relatively) cheap headpone? I'm sorry, but this just sounds like a giant marketing gimmick.

I'd really appreciate it if someone could give me an explanation based on science, not just "We heard it so that makes it true."

[rimrocks]

A good dose of scepticism is perfectly reasonable and healthy in this debate.

The concept of SVS is to re-create the sound of loudspeakers through headphones - for the exact purpose of reproducing the intent of the artist. Music and movie soundtracks are mixed by monitoring through speakers in a semi-reverberant environment, so listening to these tracks over headphones is artistically 'incorrect'. And the acoustic processing (in the analogue domain) of the audio caused by the speakers within this mixing environment is pretty severe - but it's what the mixer/artist ordered. SVS is just trying to re-create this, and add it back into the source signal for presentation over headphones.
Mike

[earwicker7]

Originally Posted by rimrocks

The concept of SVS is to re-create the sound of loudspeakers through headphones - for the exact purpose of reproducing the intent of the artist.

That sounds totally reasonable; I get the idea of trying to reproduce a three dimensional space with headphones. That part makes complete sense and it really sounds like you guys have nailed it down pretty well.

The part where I get lost is being able to duplicate the sound of any speaker (as opposed to just speakers in general). I'm wondering... is this an officially stated feature or just something that just got twisted (not in a malicious way) by word of mouth?

Thanks for the response! It's good to hear it from the horse's mouth.

[Edwood]

We can debate the philosophy and physics behind all of this, but in the end you should really hear for yourself.

Hearing is believing, and I am a believer, I simply must have a Smyth SVS System in my Home Theater.

Almost ready to have us over for a visit to Smyth Research, Mike?

-Ed

[bozebuttons]

Originally Posted by Edwood

We can debate the philosophy and physics behind all of this, but in the end you should really hear for yourself.

Hearing is believing, and I am a believer, I simply must have a Smyth SVS System in my Home Theater.



-Ed

X2
I am a believer as well after hearing it at Can Jam, I want one as well.

[Mazz]

Originally Posted by earwicker7

We detect them based on the temporal difference between what the two ears hear.

Not just that, but differing locations sound different - i.e. as if the frequency response is different. That's not a very accurate sentence, so perhaps you could say instead that the HRTF is parameterized by location rather than being purely omnidirectional.

Originally Posted by earwicker7

My problem is with the "I can make this headphone and this amp do things it isn't physically designed to do" claim.

I hear what you're saying, and I'd say you are totally correct - and if you read my earlier comments I've been suggesting there are limits on how much it can do. There's only so much you can reproduce with a given headphone, no matter how much signal processing you do. For one thing you can't reproduce a mega-bass speaker system on a headphone that clips with low levels of bass. Other limitations exist too (frequency range, transient response...)

But I must also report that in the SVS demo I heard a lot more bass coming out of those little Stax headphones than I heard on (what I think was) the same model of headphones in the TTVJ room - so there appears to be room for significant change via signal processing algorithms, for at least some headphone models.

[HeadphoneAddict]

I would love to hear this system with my HEV70/HE60.

Is there a price set yet (without the Stax)? If the price is anything like the Beyer Headzone forget it...

[rimrocks]

Originally Posted by earwicker7

The part where I get lost is being able to duplicate the sound of any speaker (as opposed to just speakers in general). I'm wondering... is this an officially stated feature or just something that just got twisted (not in a malicious way) by word of mouth?

We haven't actually claimed this for SVS but, based on our own experiences, we still haven't found a loudspeaker+listening environment that 'exposes' flaws in the SVS+Stax combination - and this includes at least half a dozen professional studios. Will this continue to be true? Of course not, but it'll make for an interesting on-going challenge. Mazz is quite correct though in pointing out the potential low-frequency limitations - particularly for the entry-level Stax headphone+amp that we're typically demonstrating with.

The price for the initial unit is still going to be around the US$3000 mark. Edwood - we're still aiming for an October meet in Camarillo, but no actual date yet - I'll keep you notified.
Mike

[Edwood]

One feature that will be important for Head-Fiers, is the ability to turn off all EQ. So that all the spacial and surround effects (including head tracking) are in use, but preserving the native frequency response of the headphones being used.

I think many people here will be using headphones exclusively with the SVS, and not as an accessory to match an existing speaker system.

Of course, being able to tweak the EQ a bit would be nice, namely adding bass, as bass is more important for movies.

-Ed