Violet DSP Evolution - an Open Baffle Project

[keyser]

Cuibono, you are right I wasn't very curious about how you think your speaker sounds. The concept is good and it seems well executed: it has to sound good! I guess we all have systems with very well controlled dispersion. I have the exact same experience as you guys: good recordings sound even better, bad recordings are listenable. Mission accomplished, case closed? Or could the concept still be improved upon?

"And now for something completely different" - Waveguides and high directivity anyone? I'd love to be able to compare it (a system like the Summa) to a good dipole. Most of the reviews on the Gedlee site are from (former) dipole aficionado's who either once heard or used to own Orions. Having heard the Orion myself a couple of times, I think it's a tough act to better a good dipole to an extent as said reviews are trying to make me believe the Summa does. Perhaps the Orion is no longer the non plus ultra on dipoles, but it sure sound very good. What do you think, cuibono?

[AJinFLA]

Originally Posted by Rudolf

Not long ago things started to change:
More and more I got interested in integration or - better phrased - 'balance'. A certain degree of differentiation had become a matter of course in my system, but it did not all fall together harmoniously and with ease. This has improved. I very much ascribe it to my better understanding of room related matters like power response, directivity and early reflections. As of now I have achieved a much better integration of the musical performance in front of me. And - lo and behold - many of those 'bad' recordings have become listenable again. Of course they have not changed in quality, but they don't annoy any longer.

Sounds like a classic case of "burn in" to me

p.s. remember, a normal person owns a stereo to listen to the music, whereas an audiophile owns music to listen to a stereo

[cuibono]

Thanks for your encouragement everyone.

Originally Posted by CLS

...Such conflict makes it a 'forbidden' pleasure ...

I censure my listening desires too, more often than not - that song "the sweetest taboo" comes to mind...

Originally Posted by sqlkev

...
i built myself a set of LR's Thor dipole woofers but they weren't loud enough for my taste. Looks like I'll be trying out the woofers you're using soon.

Do you mean SL's Thor monopole subs? The GWx4 is a different beast, and is no way capable of competing with the Thor. If you're interested in the BAMTMs, drop me a line. I'll probably arrange auditions of the Violet at sometime too - I've been wanting to start an LA area DIY club...

Originally Posted by Rudolf

cuibono,

I'm wondering if this is some end of the journey or what will come next.
Thank you for giving us such a significant insight into the 'sound' of your system.

Rudolf

It seems there may be an end in sight, possibly. With more people finding the same results with dipoles as you, stigerik and others here, I'm not surprised. I'm not too sure what I will tackle next. I'm interested in the ASP version of the DSP, but the cost and complexity are daunting, in the face of the possibility of not much gain, or none at all. I'm looking into linearizing the group delay with DSP currently, and I'll begin experimenting with in room EQ to see what there is to gain there. The back-end electronics could definitely use some attention - they're not very solid at this point. Subs would be nice sometime. I'll build nicer cabinets and take some nicer pics. Plenty to do, but nothing that is really going to change the sound tons, I expect.

Originally Posted by keyser

... it has to sound good! ...

...Waveguides and high directivity anyone? I'd love to be able to compare it (a system like the Summa) to a good dipole. ...

You know, I really consider this a win for basic physics and audio - I didn't have to find the most exotic XYZ, or get things down to the last 0.01%, no miles of expensive gear - just a basic principle - flat on and off axis response.

I'm really interested in comparing dipoles to a well done horn like the Summa - if I had the cash, I would just build them. There are so many differences between them, both claimed and physical, and it seems a lot could be learned - but there is also a lot of potential for confusion. I assume the Summas sound very smooth - their FR graphs say that pretty clearly. But they should also have a classic 'box' sound - they are omni in the low midrange and below, while the become monopole higher up. It seems a simple (and likely) conclusion that there is a sonic emphasis on the lower midrange (like any boxed speaker not against a wall). Similarly, the Summas will probably present a more detailed/direct/less ambient sound. But I'd love to find out in person!

The Orions are highly exceptional speakers - I don't feel like my speakers have 1/10 the quality of the Orions. Linkwitz also knows way more about them than I probably ever will. The Orions are engineered and built better, have nicer drivers, deeper bass, and greater output. They look better too. I'm sure more could be added to that list. In terms of commercially available dipoles, I can't think of any higher quality than the Orions. But I'm guessing my speakers do have a more regular dipole response above 700Hz or so. They cost a lot less too. Either way, I'm happy with what I've got, and building/designing/measuring/discovering has been tons of fun. I learned lots, especially from Linkwitz's very generous website, as well as from people like JohnK and MJK, to start.

[Bratislav]

Originally Posted by cuibono

The Orions are highly exceptional speakers - I don't feel like my speakers have 1/10 the quality of the Orions.

I think you are being too harsh on yourself. I haven't heard your speaker but I'd be surprised if you haven't reached 90+% of Orion's quality (and even more in some areas). There is not much left to aspire to after you get flat on and off axis (and if anything your speaker will have in some areas even better polars than Orion).

[keyser]

Originally Posted by Bratislav

I think you are being too harsh on yourself. I haven't heard your speaker but I'd be surprised if you haven't reached 90+% of Orion's quality (and even more in some areas). There is not much left to aspire to after you get flat on and off axis (and if anything your speaker will have in some areas even better polars than Orion).

This makes me very curious. What could there be more than frequency response? Okay, some dynamic range is necessary, but in my book nice drivers and even a particularly solid construction don't matter half as much as getting the frequency response and dispersion right.
Also, in my experience the Pluto surpasses the Orion in terms of timbre. But of course with those you miss that low end you can chew on and they're not as loud as the Orion. The dipole built by a friend surpassed the Orion in a direct comparison.

Re-reading what I just typed, it looks as if I don't like the Orion all that much. I do! When I first heard it, it was the best speaker I had heard until then. It's just that I think it could be better still.

[twest820]

Congratulations on a nicely done project! I look forward to foaming my Neo3s once my power amp build is done.

Originally Posted by cuibono

I'm looking into linearizing the group delay with DSP currently, and I'll begin experimenting with in room EQ to see what there is to gain there.

Getting flat phase and time alignment in the analog domain is far harder than dealing with PC hassles. DAC to the power amp for the forseeable future for me; I can't see any way I'd ever go back to warped phase crossovers. But then I seem to be fussier about phase than most.

Personally I've settled on Allocator+PLParEQX3+Bidule. Cost effective and provides a good mix of crossovers, equalization, and phase control options. The DACs in Saffire Pro 40 I'm using sound better than my CD player, too (somewhat surprising as the Azur 640C uses a WM8740).

[Bratislav]

Originally Posted by keyser

This makes me very curious. What could there be more than frequency response?

Well, ideally you'd want to have flat phase, ultra low distortion, reach sub 20Hz and have huge headroom as well. But those are more like icing on the cake, if you wish ... Not too terribly important in my book.
What was the dipole that your friend built ? Can you tell us a little more about it ?

[keyser]

You can read about it here: MOB3W - my open baffle three way

[StigErik]

Originally Posted by twest820

Congratulations on a nicely done project! I look forward to foaming my Neo3s once my power amp build is done.


Getting flat phase and time alignment in the analog domain is far harder than dealing with PC hassles. DAC to the power amp for the forseeable future for me; I can't see any way I'd ever go back to warped phase crossovers. But then I seem to be fussier about phase than most.

Personally I've settled on Allocator+PLParEQX3+Bidule. Cost effective and provides a good mix of crossovers, equalization, and phase control options. The DACs in Saffire Pro 40 I'm using sound better than my CD player, too (somewhat surprising as the Azur 640C uses a WM8740).

Interesting. I use the PLParEQ myself (the full version). What do you think of its sound? But why not skip Allocator and do it all with PLParEQ?

[twest820]

Latency, CPU utilization, and cost. Allocator beats PLParEQ[X][3] on all three. PLParEQ is out of my budget. Testing showed my old laptop doesn't have enough CPU power to comfortably run a five way crossover with PLParEQ3 or PLParEQX3. Only problem is Allocator's equalization is not linear phase, but running Allocator+PLParEQX3+Bidule provides that option for less than it'd cost to use PLParEQ3 and uses something like half the CPU. I considered PLParEQX but wanted the ability to do warped phase equalization if I needed it. Plus I like the conveniences Allocator provides, such as showing up to a four way crossover all at once, importing measurement data, and easier implementation of high order crossovers.

PLParEQX3's a CPU hog even at its lowest quality setting, so that's all I've tried. I'm slowly pulling together the power amps and speaker parts for the dipole build but haven't run dipole yet. In the meantime my box speakers don't need much in the way of EQ so I've not made routine use of PLParEQX3, but in auditioning the demo version I found the sound to be transparent and clear. With Allocator running the crossover and port correction the boxes are linear phase within 20 degrees (probably better; I need to haul the system outside and remeasure on a day when there's not old snow). I found applying equalization in Allocator that caused 30 to 60 degree phase shifts noticeably degraded the sound relative to doing the same equalization in PLParEQX3.