"House sound" Do the designers here have a "house sound"?

[carlspeak]

Re: "House sound" Do the designers here have a "house sound"?

............... I wrote that OFTEN "voicing" is the manipulation done to hide basic errors such as selecting drive units that cannot match directivity with a reasonable crossover. If you disagree with my actual assertion, evidence to the contrary would be interesting to see.

How have you determined "OFTEN"? Did you take a survey of speaker builders? Or, is this you surmizing based on little or no factual evidence?

[philthien]

Re: "House sound" Do the designers here have a "house sound"?

Care to formulate a reply that reasonably relates to the words you quoted? You, Craig, and probably some others need to work on that.

I never wrote there's no role for voicing, which seems to be how you misinterpreted my post to read. I agree that some voicing is generally inevitable and useful. But I wrote that OFTEN "voicing" is the manipulation done (consciously or otherwise) to paper over basic errors such as selecting mid/tweeter that cannot match directivity with a reasonable crossover.

If you disagree with my actual assertion, evidence to the contrary would be interesting to see. For example, where in the frequency spectrum do you think most voicing happens?

[edit, which also answers Carl's post below] From what I've seen, in speakers with directivity errors most voicing tends to be within an octave of the nominal crossover. In better configurations the two octaves around the crossover tend to be as flat as the drivers will allow, minus a few diffraction related bumps that cover narrow angular regions of the polar response.

That doesn't really make sense though, does it? Because in the context of this thread, we're talking voicing in terms of consistent qualities found in a designer's work. So in order for those qualities to be consistent, the designer would seemingly have to make the same sorts of unworkable parings over and over (and over).

That is, if people can identify Jeff B's work from Wolf's, and their designs are so far ranging (from matchbox size to substantial), it can't be due to picking a midrange and tweeter that don't work together.

[isaeagle4031]

Re: "House sound" Do the designers here have a "house sound"?

How have you determined "OFTEN"? Did you take a survey of speaker builders? Or, is this you surmizing based on little or no factual evidence?

Exactly Carl. Pallas your initial statement leads most to assume that voicing is covering up errors. That is absolutely incorrect. But it is also a statement I would expect from someone that does not have the knowledge or experience of actual design. If anything, you should be qualifying your own words. The onus of proof is upon you sir.

[fastbike1]

Re: "House sound" Do the designers here have a "house sound"?

Bear in mind that Pallas is "often" critical about everyone's work, but has yet to show any product of his own. Just sayin'

That doesn't really make sense though, does it? Because in the context of this thread, we're talking voicing in terms of consistent qualities found in a designer's work. So in order for those qualities to be consistent, the designer would seemingly have to make the same sorts of unworkable parings over and over (and over).

That is, if people can identify Jeff B's work from Wolf's, and their designs are so far ranging (from matchbox size to substantial), it can't be due to picking a midrange and tweeter that don't work together.

[craigk]

Re: "House sound" Do the designers here have a "house sound"?

Care to formulate a reply that reasonably relates to the words you quoted? You, Craig, and probably some others need to work on that.

I never wrote there's no role for voicing, which seems to be how you misinterpreted my post to read. I agree that some voicing is generally inevitable and useful. But I wrote that OFTEN "voicing" is the manipulation done (consciously or otherwise) to paper over basic errors such as selecting mid/tweeter that cannot match directivity with a reasonable crossover.

If you disagree with my actual assertion, evidence to the contrary would be interesting to see. For example, where in the frequency spectrum do you think most voicing happens?

[edit, which also answers Carl's post below] From what I've seen, in speakers with directivity errors most voicing tends to be within an octave of the nominal crossover. In better configurations the two octaves around the crossover tend to be as flat as the drivers will allow at the design axis and over the front hemisphere, minus a few diffraction related bumps that cover narrow angular regions of the polar response.

maybe if you could clearly state a point and not leave open ended statements that always give you and escape we would understand what you are actually trying to say. no lets see some proof for your statement that voicing is covering up mistakes in the crossover design. you are the one making the claim that voicing is masking errors in design.

[Pete Schumacher]

Re: "House sound" Do the designers here have a "house sound"?

maybe if you could clearly state a point and not leave open ended statements that always give you and escape we would understand what you are actually trying to say. no lets see some proof for your statement that voicing is covering up mistakes in the crossover design. you are the one making the claim that voicing is masking errors in design.

I've got a suggestion for Pallas' new avatar.



As an example of voicing using directivity matched woofer and tweeter I will reference my 8" waveguide 2-way build. Every iteration I worked on had a focus on flat response, above, below, and through the crossover region. However, changes to slope, crossover point, baffle step, all played a role in determining where the final crossover configuration ended up. I could have taken response measurements, simulated a crossover, built it, measured to make sure it was flat, and called it a day. And I am dead certain that plenty of manufacturers do exactly that and don't spend weeks of extended listening to a wide range of recordings to "voice" the design (and you can hear it in the final product as cited by more than a couple people in this community who attend shows.)

I'd love to hear a speaker designed from the ground up by Pallas at one of the DIY events. Why do I have the feeling I'll be a great grandfather before that happens?

[craigk]

Re: "House sound" Do the designers here have a "house sound"?

Nice

[---k---]

Re: "House sound" Do the designers here have a "house sound"?

Voicing can and is also done at the simulation / design phase. It doesn't necessarily have to be tweaks done after the first speaker is built. ...

[DDF]

Re: "House sound" Do the designers here have a "house sound"?

I'm not sure I understand the voicing process. After finishing a crossover, I use lossless FLAC files with female vocalists and instrumental music to test the system. My Chorales were tested using Sarah McLachlan's "Angel" from the CD "Absolute Voices II", one of the best recordings I've ever heard. To me, the sound is spot on. So when we talk about voicing, we're talking about a subjective process, and as such, what reference point is used to steer the crossover to produce the sound you desire?

Hi hong,
I find it's important to voice using many types of music. Great vocal recordings are notoriously flattering of a speaker and are much more tolerant to a non-ideal crossover than other types of music, in my experience. I even use slightly compressed but very spectraly wide stoner rock tracks and they often tell me more about what to tweak than wonderfully recorded voice.

The other thing that flatters a speaker crossover design is listening in a really big room where the side walls are far away, like most of the DIY meets do. They would by far reward a flat on axis through the xover range and be much more tolerant of iffy off axis through xover than listening in a small room. At least that's been my experience.

At an Ottawa DIY get together 2yrs back, I brought a very small 2-way speaker. I had just designed the xover and built it before the meet, and hadn't voiced it yet. Like many others here, I first target a family of on and off axis responses that worked for me in the past (my personal "house sound"). They sounded amazing at the get together in a large room with the side walls yards away. A bunch of people right there and then decided to order the parts to build it but I warned it hadn't been voiced yet. Once I got it home and listened in a very small room where it was intended for final use, I changed the crossover twice (I posted this on DIYAudio as the "Napoleons", and my sig there is DDF as well). Personally, it takes me months to finish voicing as things I don't notice at first become little nits weeks later and sometimes intolerable after a month of listening. And I can pick things off pretty readily. Twice other Ott folks brought well regarded DIY designs from the web and I immediately picked off peaks that I thought were big voicing flaws, and predicted what the frequencies were and how big (dB). They thought I was full of it, till we measured and there those peaks were. So, some of these effects are pretty subtle and take time to show themselves unless you've done allot of this. I think speaker "voicing" is allot like recording engineering and mastering. There are excellent courses, books and CDs available teaching these recording listening skills. I have one at my bed side that I listen to to help train myself for speaker design.

Sometimes this whole voicing process drives me away from flat on axis and away from things that would look correct on paper.

It's not that voicing corrects for a "mistake", just that any design chooses tradeoffs that required some coaxing in the xover to get the best out of it, and one size fits all doesn't exist. I posted another example on DIYAudio (the JR149) where I purposely had to veer away from flat for flat perceived response

I also once did a design where I started with on/off axis, BDC and woofer Q that were pretty close to my personal reference, but it sounded quite different. This exemplifies why, to me, voicing is so important, even if I have a family of curves that I try and hit each time.

Only once was I successful in voicing purely in the design phase, with no tweaks needed later.

[craigk]

Re: "House sound" Do the designers here have a "house sound"?

very good explanation.

[hongrn]

Re: "House sound" Do the designers here have a "house sound"?

DDF,

I second what Craig said. I can definitely hear what a room does to a pair of speakers, but I don't think I'm there yet in terms being able to identify the subtleties you discussed above. One thing we sorely lack in SoCal is a DIY club of the quality you guys have in the MidWest, so being able to listen to other designs and compare notes is virtually impossible. Thanks.

[peterpan]

Re: "House sound" Do the designers here have a "house sound"?

You get a gold star. When people talk about flat frequency response they are thinking one dimensionally. A speaker may be flat on one axis and have real problems off axis. I try to take that into consideration. As a result, over the years, I have developed a particular "recipe " in crossover design that I have found to sound the most natural and coherent to me. I tend to use this recipe in most of my designs. From the older Continuum to the new Spirit Wind the crossover approach is very similar, and this is case for most speakers in between.

Two speakers may have the identical on-axis response, but differences in crossover point, slopes, phase tracking, and the tonal balance of the power response will make them sound very different. In my case, most of my speakers share these design characteristics so they tend to sound very similar.

Wolf and I cook with very different recipes, as an example, so even if both of us target a flat axial response our speakers will likely still sound different , and each will reflect our individual design methods.

In addition to all of this, some people prefer a speaker that has a warmer balance while someone else may like one that is brighter and airier. How baffle step is handled and preferred bass tuning all factor into it as well.

Hello Jeff.

What are the SPIRIT WINDS? sounds nice, care to share more about this design? 2 way, 3 way, what drivers?

[DDF]

Re: "House sound" Do the designers here have a "house sound"?

DDF,

I second what Craig said. I can definitely hear what a room does to a pair of speakers, but I don't think I'm there yet in terms being able to identify the subtleties you discussed above. One thing we sorely lack in SoCal is a DIY club of the quality you guys have in the MidWest, so being able to listen to other designs and compare notes is virtually impossible. Thanks.

Hi hong, If you get a chance, visit a hifi shop with Revels. If decently set up, I find the higher end ones have a pretty neutral balance and can give a good point of comparison.

[Jeff B.]

Re: "House sound" Do the designers here have a "house sound"?

Hello Jeff.

What are the SPIRIT WINDS? sounds nice, care to share more about this design? 2 way, 3 way, what drivers?

Here's a pic -

It uses an Audio Technology 18H52S midwoofer and a Scanspeak Revelator 9900 tweeter. Seth Horton did the incredible cabinets. Clearly one of the best speakers I have ever had the pleasure of living with.

[bigfishe]

Re: "House sound" Do the designers here have a "house sound"?

Yeah those are as nice as they sound!!!