Site Links
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
Categories
In this Discussion
Please review the site Rules, Terms of Service, and Privacy Policy at your convenience. Rules, TOS, Privacy
Get familiar with the reaction system: Introducing the Reaction System
Ping JR and everyone else, The "Sound" of DIY speakers...
Johnny,
When I was in the Lobby chatting with you and the gentleman who brought the Helios speakers (is he on here), the comment was made that a common characteristic of DIY speakers in that they all sound Neutral (Could that be interpreted as boring?). It seemed like there might have been more to be said, but the conversation ended.
Is there a case to be made for DIY not all making the same sounding speaker?
Or am I reading things into comments that were not there?
David, who just got home...
Comments
I'm out and about, when I get back to my real computer I'll expand a bit on my comment.
I think most of us would agree that a neutral speaker lends itself to accurate reproduction of recordings.
If you are looking for spice play with the EQ. (It'll be O.K., don't let the internet fool you)
Or design your own. BBC dip-get sum! In your face bright and analytical, tilt the response up a little above flat (overdo it and you will have regret). Want it to be a little warmer go with a little downward slope in the response.
Many of us have a preference.
At a certain point you start to realize that no speaker is perfect for every situation/room or recording.
A well engineered speaker helps you controll the room variables that you can not control.
You'll find a truly neutral speaker on display right next to a unicorn.
😀 I've heard a number of guy's say that they are bored with most of the diy speakers, and let's face it, most of the speakers are basically similar in mildly different flavors. At each event there are usually a couple of exceptional designs that I find more interesting than the rest. This will come as a great shock to some of you, but I love the unusual stuff like Meredith brings. Overall it's my belief that for the most part all the guy's are improving, but everyone builds what they interpret as correct and that's not wrong or boring for that individual.
So would you design a speaker to ‘sell’ (10 min demo) or one you’d enjoy day after day…
We get to ‘voice’ per our preferences.
A lot of speakers are not flat and should be. A lot of non flat speakers have to be to make them nonfatiguing or skirt the compromises. Overly neutral becomes sterile, and then they are boring.
InDIYana Event Website
Many of the speakers that sound good for 10 min don't sound good after an hour. This is a benefit for us in the DIY, we don't have to make a speaker that will sell.
We have seen where a little heat in the high end has helped the perception of a speaker. I picked that up on a few designs this weekend where the highs sounded good until a certain passage and the tweeters screamed a little under stress. All of a sudden what was initially definition and resolution, became distortion and distress. The same speaker on the next passage sounded good again. Most likely the
Stress was a peek in distortion as the tweeter was pushed to its limits.
There are many that think the 5k distortion peek in many AMT and plainer tweeters is the reason for their perception of resolution. Many don't find that peek offensive and my guess is because it is controlled better than most distortion peaks.
These are all theories that I have , please don't hesitate to give your input.
I would tend to agree. As most of the testing and event themes I've participated in, the slightly emphasized treble sounds good- at first. Then it grates on you, and it needs reduced. Most of the builder/designer members of the regular groups make sure the fatigue is minimal, flat or not. Distortion should also be minimized.
InDIYana Event Website
As much as I disagree with some of Steve Gibson's ideas, I regularly tune into his Youtube channel and laugh at how he shows high dollar commercial, successful speakers often have a horrendous frequency response. I tell my girlfriend, my speakers are flat and I can prove it. I love DIY!
https://youtu.be/Olo-dDouZ3c
I've got a couple of test tracks with serious sibilance (Bonnie Raitt usually) and have often used those to tune crossover until it mostly goes away. There are some big dynamic tracks that I also use to stress test a design, and Mozart's Requiem will show lack of resolution.
Flat on axis is not necessary correct. Overall power response is far more important. What's going on off axis is just as important as the on axis unless you listen in an anechoic chamber. In fact I here a great many diy designs that have off axis issues around the crossover frequency, but it seems like the guy's just get used to that sound profile and it becomes their preferred balance. This is not limited to diy however, just walk around Axpona and you will hear some great speakers and some complete garbage, but both will have their admirers. To each his own. Turn up the brightness or color control on your TV if it makes you happy, it's nobody's business unless you ask for an opinion.
As to the original question, I think there's "accurate", and then there's "pleasing". In my opinion, pleasing should win. Like Meredith was saying yesterday--stereo is an illusion anyway.
I backed into that very same conclusion. Traditionally, I've measured and designed to get as flat as possible on axis, and then tweak by ear from there in-room with the occasional measurement to make sure phase issues aren't sneaking in. In doing so, I suspect I've been adjusting for a preferred overall power response.
It's also kinda risky, when the speaker is transported to another room and things fall apart. Perhaps that's what got some of them Axponants?
Yeah, going to Vituix and a turnable to try to tune for "correct" power response for this build.
I was referring to those commercial units like Wilson that have spiking treble when I said "need to be flat." I agree that flat is not always correct, maybe 'flatter' is the better word, as non-flat sometimes works better everywhere but on-axis. Power response is very important. My Zingers and Stance are non-flat, but sound right.
For instance, the Opaz I took this weekend for the Tweeter Yoga theme; I'm betting it was not flat, but the power response looked very good in sim. I have not measured the final responses yet. I will update that thread with the measurements. This is where the ears come into play and let me know that they were very fatiguing with a flat simulation on axis.
InDIYana Event Website
OK, to clarify my comment... DIY'ers tend to voice to a neutral sound. Things such as port resonance, panel resonance, and driver linear and non-linear distortion profiles are all factors that get in the way of that, however. True neutrality is not possible - some drivers just have inherent limitations.
In my opinion, it is a good thing that we gravitate towards neutral sounding speakers. Sure, speakers with some fairly obvious issues can often win kudos (both DIY and commercial) but at the end of the day those are speakers we just can't live with.
I tend to voice on the warm side of neutral - a slight downward spectral tilt - but that often leaves me behind on the typical short demos we use. I am not a competitive person, so that in itself is alright.
First arrival sound should be heavily weighted compared to the off-axis. Both should be smooth and controlled, but since most of use 4th order LR topology it is impossible to get that perfect power response - off-axis bloom is inherent to the network. Waveguides won't fix that, diffraction control won't fix that. Still, some of the best speakers I have heard have been nearly ruler flat on-axis and use 4th LR so I am not sure what is happening there if off-axis is so damn critical. That being said, I actually do concern myself with off-axis response, I just do not abandon other proven design principles in favor of the current spinorama trend.
If we had used the flute track I originally intended to do at Indy this year, I guarantee we all would agree there are things that color a speakers response more than off-axis performance. That track will excite panel resonance and stress tweeters like no other track I have found. Both are issues quite a few of us just do not address adequately. Usually those issues are moot since quite a bit of demo music, and I suspect the stuff we use for daily listening, tends not to be very challenging so the issues rarely raise their heads.
Anyways, when possible design towards neutral unless you have specific in-room needs. If that is the case, don't be surprised when the speaker performs poorly at a DIY event, for example.
As far as commercial speakers go, most of those guys are going to be challenged with the same issues we are. Most of them are not paying much attention to cabinet stiffness or port resonance, and definitely are not concerning themselves with off-axis response. Maybe they are just not as innovative as Stereophile wants them to be but my experiences at Axpona do not give me confidence that most commercial guys possess design magic. I'll take most DIY speakers in a given size class over commercial, price notwithstanding.
As a mere builder, not designer, I'm chuffed at the projects which I've lucky enough to be able to make courtesy of the kind and generous people on this Forum and elsewhere. I can honestly say that I'm happy with all of them and don't at all find that they sound similar.
The only build that was not quite right was when I tried to design the crossover myself, or at least, changed an existing commercial design which appeared that it would be too bright, according to my sim. It was still a good learning experience and I'll re-use most of the bits eventually.
Geoff
JR said: "That being said, I actually do concern myself with off-axis response, I just do not abandon other proven design principles in favor of the current spinorama trend."
This!!!^^^
InDIYana Event Website
IMHO it's true - commonly DIY speakers DO likely sound very similar. This might be because there are many builds that follow the same theme: a 2-way vented box of modest size using a 6-7" woofer and dome tweeter. Once you start to listen to a wider variety of build styles (3-way, horns, open baffle or dipole, etc) things will start to sound more varied. Part of the basis for the sameness in similar speakers (e.g. the 2-way speaker typical of many DIY builds) is because the drivers will have similar crossover points (out of necessity) and this imparts a sameness to the power response (the off axis responses of the drivers). Pushing this envelope around, such as with the theme from InDIYana, or with a very different type of speaker like Nick's "Portals" will get you a different sound.
Also, maybe because I primarily listen to my own DIY build (no baffle dipole speakers) one other characteristic of many of the speakers I heard last weekend was that "box" sound. It struck me as a difference in the lower midrange, where (to me it seemed that) I could perceive the hollow sound of the internals of the box. With insufficient wall and internal damping to the enclosure, the sound can be re-radiated back out through the cone and through the sidewalls of the enclosure. I found that I could identify this characteristic "sound" of the boxed speaker for about half of the speakers I heard. It lent a roughness and heaviness to the lower mids and upper bass region.
The other impression I got from some speakers was of vented box overhang. This tended to be in some of the larger speakers. A prime example of that was Paul Carmody's speaker that was played near the end of Saturday IIRC. It tends to make the bass sound a bit more muddy, but that is somewhat of an inevitable tradeoff with a vented party speaker type design of a certain size.
These are just my impressions of some builds and none of this makes one sound or loudspeaker type superior or inferior. A loudspeaker is an adventure in tradeoffs. There are always stellar examples of any type of loudspeaker that will tick all the boxes. It was nice that there were some of these as well.
I agree with most everything said above regarding neutrality, on-axis vs off-axis, power response smoothness & slope, blooming, cabinet stiffness, port resonance, etc., etc. I am always amazed at how much of a difference the room makes to the sound. Change the room and the sound changes not a little bit, but significantly. Speaker building is like chasing a moving target. When you are designing for a competition, like MWAF (SDC) or InDIYana, you need to design with full BSC and for the best sound in a somewhat wide listening window; not a narrow sweet spot. To improve my design guesswork, I'm trying to come up with better ways to measure my speakers. Maybe use some type of RTA averaging of multiple seating positions to determine the response below 500Hz instead of setting up a quasi-anechoic model of the low frequencies. A quasi-anehoic model tells me nothing about how room placement, or woofer height above the floor, is going to alter the sound.
One of things I've always found striking at DIY events is that even when a majority of the speakers sound good the noise track that is used to set levels sounds considerably different. I've often wondered why that is. It might make a good project for me to investigate.
Ron
One aspect I don’t believe has been mentioned (unless it is being assumed in the statement of speaker placement) is listener position. I have not been to one of these speaker events but from car audio and my own A/Bing of speaker builds i have found some speakers do a greater job of minimizing the listening position variable and some do a lesser job. In my home theater room i have been able to lean forward in my seat and notice a drastic difference some times as the loudspeakers are playing and some units i cannot tell if there is a difference. I understand this is part of the room effect but if people are evaluating from different positions there is going to be different results.
In my humblest of opinions…..and believed experience.
I try to use the same noise cut every year. Somehow it got normalized this year on the house tracks. I thought sure I did right to avoid it happening.
The Green Noise sounded "not 500Hz centered" like it should be, so I will investigate that too. FWIW, it is difficult to get a recording of Green Noise. I could not find a download or disc of it, and ended up ripping an MP3 from Youtube and converted to a .wav file. This may have hampered the sound of it. I'll try again to get a better version.
The Brown Noise was immensely fun! Watching woofers excurse is never boring.
InDIYana Event Website
Not too long ago Ed Perkins did a great job of fleshing out why big speakers vs little speakers are so different in regards to the noise cuts and he proposed a better method to level (pun intended) the playing field. My search foo isn't all that great. Hopefully somebody can find his post and link it here.
Found it. Toggle down to the 5th post in this thread (IowaDIY 2022):
https://diy.midwestaudio.club/discussion/1932/iowa-diy-2022/p1
Absolutely, I have been to dozens of these events and I walk around the room as speakers play, what the guys in the back are hearing is different from the guys in the middle to the guys near the speakers left, center and right, etc. Because if this it’s hard to get a good consensus of what everyone is hearing, sometimes I’ll be listening and someone will solit in front of me and I can get a very different sound if I move my head from behind them to the left or right. The room even sounds different as people come and go, a mass of human bodies make pretty good sound absorbers and diffusers.
Listening in a big room filled with people like what we do at these events is fun, informative and very educational, but I’ve always been careful to use what I hear to form too strong of opinions, it’s good to keep an open mind until you can listen to a speaker in a good listening environment where you can be in the right position to hear what the designer intended.
Javad
Drawn from the referenced thread:
Lol I was just thinking of something like this last night. I found my build sounded rather dark, but also were pretty quiet. I suspected the energy below 1k was boosting the spl measurement and resulting overall volume was reduced. Just based on how I know they measure in that range.
Since the only other experience I had to compare to was last year's PE speaker builder event.. I was designing for elevated volume levels. Basically trying to ensure they don't "hurt" when they are making more spl. Now that I have them setup in my listening space they do seem dark. The tweeters don't wake up until I start to crank it. But I guess cranked is kinda how I like it.