2023 Parts Express [Speaker Design Competition]

[DanielT]

I am posting the videos I found about it. I just started checking them out. This will be fun to see what the DIY participants have to show this year.

Feel free to comment on what you see.

 

[Trdat]

I would suggest that Parts Express invest in a Klippel machine not just to objectively measure these participants speakers but also help increase sales of DIY designs after they have achieved a pass in measurements.

 

[DanielT]

Damn, it was fun watching those videos. The creativity flows and they seem to have had fun, which is the most important thing. Well done I say to all the participants.

____
More about who won here. Plus who the judges were. By the way, one of the judges was Vance Dickason:

Speaker Design Competition

www.parts-express.com

Vance Dickason is probably best known for this book:

Loudspeaker Design Cookbook,Vance Dickason- 9781882580477 | eBay

Publisher: Audio Amateur Pubns ISBN 13: 9781882580477. Title: Loudspeaker Design Cookbook Item Condition: used item in a good condition. Will be clean, not soiled or stained. Books will be free of page markings.

www.ebay.co.uk

It can probably be fun to compete, but I suspect that the most important thing was to meet, share experiences and have a good time.
____

Perry Marshall is back this year and he won the category:Open Unlimited.

Here is his design. Pictures of the speakers plus various measurements:

Photo 3: For speakers, I chose the Eminence Kappa Pro 18LF woofer and the 8” Radian 5208 coaxial mid bass + Constant Directivity horn tweeter.

You Can DIY! The Live-Edge Dipoles

For the speaker builders, Perry Marshall details everything there is to know about his beautiful Live-Edge Dipoles project. A design that was inspired by two slabs of birch the author brought home, allowing to finally bring to fruition his idea to build these dipoles. Marshall reveals all the...

audioxpress.com

 

[Wolf]

It was another good year! I've been to every one since inception in 2002. Had my Bottleships and Monoculus present. I didn't do the video thing this year.

 

[DaveFred]

I would suggest that Parts Express invest in a Klippel machine not just to objectively measure these participants speakers but also help increase sales of DIY designs after they have achieved a pass in measurements.

<donning flame proof suit>

For the love of God, No.

I will take the impressions of three recognized long time industry experts who can appreciate the entire package over a graph thank you.

<running away now...>

 

[kemmler3D]

<donning flame proof suit>

For the love of God, No.

I will take the impressions of three recognized long time industry experts who can appreciate the entire package over a graph thank you.

<running away now...>

It's a reasonable thought. I think they should scan at least the top 3 in each category to satisfy the curiosity of the people. However, I think judging ought to be by ear and maybe use the NFS as a tiebreaker...

 

[DaveFred]

It's a reasonable thought. I think they should scan at least the top 3 in each category to satisfy the curiosity of the people. However, I think judging ought to be by ear and maybe use the NFS as a tiebreaker...

Imagine this scenario, someone writes a program to analyze every book ever written and comes up with a formula for what makes a good book, any book that doesn't score 100% is considered a "lesser" book. People who like "lesser" books being looked down on as being stupid for liking them. This is where I see pure numbers taking us. And, heaven forbid, what if the equation is/was wrong?

I would like to leave the variable of subjective human "taste" in the equation for as long as we can.

Just one dummyheads opinion.

Edit: Also, no one (that I have seen) has "weighted" the graphs. Ideally, best response and lowest distortion would be the better sounding speaker. But, what if between two speakers, one has better frequency response, but worse distortion and the other is the reverse, which is better?

 

[kemmler3D]

Imagine this scenario, someone writes a program to analyze every book ever written and comes up with a formula for what makes a good book, any book that doesn't score 100% is considered a "lesser" book. People who like "lesser" books being looked down on as being stupid for liking them. This is where I see pure numbers taking us. And, heaven forbid, what if the equation is/was wrong?

I would like to leave the variable of subjective human "taste" in the equation for as long as we can.

Just one dummyheads opinion.

Edit: Also, no one (that I have seen) has "weighted" the graphs. Ideally, best response and lowest distortion would be the better sounding speaker. But, what if between two speakers, one has better frequency response, but worse distortion and the other is the reverse, which is better?

I see what you mean but I think the analogy is just okay.

Writing is an art form. You can ARGUE that speaker building is also an art form, (certainly it is in terms of designing the housing) but the range of worthwhile expressions in books is much, much, much wider than it is in speakers.

You can have a reasonable argument as to whether Ender's Game or Finnegan's Wake is the better book, even though they're extremely different.

You can sort of have a reasonable argument as to whether a Wilson Tune Tot or Elac DBR2 is a better speaker.

But it would be very difficult to argue that a Tune Tot beats a KEF LS50.

My long-winded point is that speakers have less controversy around markers of quality than most art forms. Even if a competition like this only used ears-on testing, we could probably guess the rankings using Klippel-derived preference scores with reasonably good accuracy. The research shows there's a good correlation between what people's ears say is a good speaker, and what the machine says is a good speaker.

To put it another way, there is almost no such thing as an objectively good or bad book. However, if we start with the premise that a speaker's basic job is to reproduce input signals accurately, (I think we need to accept that premise to even talk about speakers) then there is such a thing as an objectively better or worse speaker.

This is not to say there isn't room for creativity or personal preference, but the creativity that goes into speaker building isn't all about the final gross/overall sound signature, either.

 

[DaveFred]

@kemmler3D

I'm a better speaker builder than analogy maker... It was more the gist of the analogy then the specificity of it.

David.

Edit: Events like drag racing only have one criterion, first across the line. I would like to think there is still some secret mojo to speaker building that is more than the sum of some cold graphs.

 

[D!sco]

It is very nice to consider speakers within very specific ideals and yet another thing entirely to wrestle with the compromises that constraints bring. Maybe they should leave preference scores to judges and measurements to machines?

 

[kemmler3D]

It is very nice to consider speakers within very specific ideals and yet another thing entirely to wrestle with the compromises that constraints bring. Maybe they should leave preference scores to judges and measurements to machines?

I actually fully agree that judges should judge these by ear. I wouldn't say there couldn't be objective criteria on top of that if they wanted to do it that way, but it would be weird to do a speaker contest that didn't involve listening. Would be like a drag race conducted on a dyno.

My thought on the Klippel scans was simply that I want to see them out of curiosity, so it would be cool if they scanned the winners. If you can't hear a speaker in person, at least you can get a general idea of what it sounds like from the graphs.

 

[D!sco]

Would be like a drag race conducted on a dyno.

Lmao that’s literally the state of muscle cars, but I guess if I were to expand, the serious limitation of having a sixth figure in the bank is a big one towards standardization. A cool ideal but maybe difficult for many to justify.

 

[Prana Ferox]

I would suggest that Parts Express invest in a Klippel machine not just to objectively measure these participants speakers but also help increase sales of DIY designs after they have achieved a pass in measurements.

People are fairly, uh, vocal about voicing DIY speakers to their preference. On the one hand I feel it would be good to Klippel winning designs to see how that voicing translates to specific frequency response and directivity behavior, it would be a good learning opportunity to see how different approaches result in different real-world measurable result. On the other hand I suspect it would be ruthlessly misinterpreted as deviation from the Harman curve would be seen as pure defect.

I'd like to see it a lot more on the kits that PE monetizes, but it's a bit self-defeating; most consumers aren't going to grok a spinorama and just think 'flat FR=good, deviation from flat=bad' and more precise measurement is just going to turn them away, especially if competitors and alternatives don't follow up with measurements of their own and instead deploy audiophool prattle in response.

 

[boXem]

Klippel winning designs

Uh?

 

[DanielT]

But, what if between two speakers, one has better frequency response, but worse distortion and the other is the reverse, which is better?

I am convinced that most people believes that a good frequency response beats a bit higher distortion any day of the week.

In descending scale starting with the most important:
1.Frequency response
2.Dispersion
3.Distortion
I think most people think that is the case.

Having said that, an amplifier driven really into clippig sounds too damn terrible. It is then impossible to listen to without getting an earache.
For people not interested in sound/HiFi what they would be annoyed by in addition to this with extreme amp clipping is clearly audible noise, something that rattles, persistent beeping sounds, distinct humming. But okay now I'm more or less talking about stuff being broken or on the verge of being considered broken.

 

[Trdat]

People are fairly, uh, vocal about voicing DIY speakers to their preference. On the one hand I feel it would be good to Klippel winning designs to see how that voicing translates to specific frequency response and directivity behavior, it would be a good learning opportunity to see how different approaches result in different real-world measurable result. On the other hand I suspect it would be ruthlessly misinterpreted as deviation from the Harman curve would be seen as pure defect.

I'd like to see it a lot more on the kits that PE monetizes, but it's a bit self-defeating; most consumers aren't going to grok a spinorama and just think 'flat FR=good, deviation from flat=bad' and more precise measurement is just going to turn them away, especially if competitors and alternatives don't follow up with measurements of their own and instead deploy audiophool prattle in response.

Its a strong point, and to be honest I was thinking that to. Why would they deter customers with bad Klippel measurements especially when most can't interpret the results, but on the flip side if the Amiga(DIY) or any other kit measured say as good as the PSB XB Bookshelf, then it would increase sales. Not all DIY enthusiasts are fathers with there kids, there is a percentage that enjoys making speakers and listening to hifi.

 

[D!sco]

Or decrease sales when people figure out that the magic of DIY is what you bring yourself, and that similar quality can come right off a shelf? (With resale value!)

Love of DIY is not often the love of graphs, but sawdust and obsession. Being a part of the process is what makes diy, literally. It isn’t for everyone, certainly not value seekers.

On the other hand, sending any design to a klippel owner is really cool, hard as that actually is to get the measurer to agree. I have a couple builds from known designers that I’ve never been able to send off due to lack of interest. Erin’s Audio Corner tends to be better about this, imo. There are designs on his site with all the VCAD data of each driver measured in box without crossovers. It ruined the Samba for me. Unfixable.

 

[jaakkopetteri]

I agree that the winner should be decided by listening, but even multiple trained listeners can't remove the room from the equation or do much of a job describing whether a speaker would suit you (and your room) personally, which is where the Klippel has a clear advantage

 

[Rick Sykora]

Was contemplating submitting the passive version of Directiva r1 this year, but the timing does not work out.

My elder son is getting married in mid-August.

 

[D!sco]

can't remove the room

And bigger designs with wider dispersion and high SPL probably win out in a large hall like that. I wouldn’t want to show off my desktop speakers in that format.