Why there isn't any revolution in PC Speakers

There are design goals with any speakers worth owning.

Low distortion. (drivers and cabinets). Correct phasing of drivers. A flat frequency response. An extended frequency response. taming of artifacts from drivers in the crossover etc etc. Fundamentally it is an engineering problem. It is an art too though.

Speakers aren't a solved problem and likely never will be (Though we are getting closer). When there is a frequency response of + or - 0 Db, no phase issues. Distortion below audibility etc etc etc Then maybe we are there. Good luck finding that. And if you ever do, you will still need to solve the problems of room interaction.

The real challenge with speakers is to maintain that flat frequency response also in a real world scenario. All those other properties are completely worthless if the speaker allows the room to mess the sound up.

This is a Bowers&Wilkins graph where the same speaker is measured directly in front of the speaker (nearfield) and in the listening position. As you can see, the flat response is completely meaningless unless the speaker can reduce the room reflections:

Frequency-response1.jpg


The green trace is an in-room measurement of the speaker close to a rear wall. The grey trace is with it closer to a corner. Both these measurements are very ragged and it’s difficult to decipher what is going on. They show the effects of reflections and resonance modes in the room and, while they are similar at higher frequencies, they’re very different in the bass. If you were to move to a different room, the results would be different again. Fortunately, the ear is able to cope with some of these differences, but even then, the professional engineer cannot optimise for just one room and even one position in that room. He needs to know what the speaker itself is doing and assign measurement targets that will make it sound good in the majority of practical listening rooms.

The red trace is an anechoic measurement of the speaker. It’s much smoother and, more importantly, it is consistent and repeatable. In the bad old days, it was considered the right thing to make the speaker’s response extend as far as possible into the bass and measure ruler flat as it did so. If nothing else, it was good for published specifications. Actually, that’s about all it was good for. Nowadays, most designers have more sense and make some allowance for the bass boost that virtually all rooms will provide.
 
you are building a speaker in an uncontrolled environment faults occur more often and skills of the DIY person is limited. Design of DIY speakers are generally very basic as so anyone can build them. Do you actually think you can build KEF 207/2 to the same degree as KEF??? no way.

I'm sorry but no... If you can get KEF components in your hand and have the crossover and enclosure designs, anyone can build a similar speaker using hand tools. There's no magic, it's simple technology. The drivers are the only thing you really can't build yourself but luckily they're mass produced by a few OEM manufacturers and most speaker drivers are sold to anyone interested. KEF drivers are an exception because they use their own patented technique so the factory is not allowed to sell them to anyone buf KEF. You can, however, buy drivers that operate in a much similar fashion with kefs.

KEF doesn't have an alien technology plant building the stuff, they're assembled on a mass production line. I've visited a couple of loudspeaker factories (Fusion, Gradient) and I've visited their measuring chambers, production lines etc. so I know this for a fact :)

Any hobbyist can invest 200 bucks to a microphone and a RTA software for time window based frequency response measurements - and even this is required only if you start doing your own designs that you need to test.
 
Last edited:
Those Kefs are great. Their Uni-Q driver has solved a lot of problems. I got a set of the Q100s a few weeks back, they really do everything very well. They don't have the sense of scale as my floorstanders, but they're accurate and sound damn good.

Uni-q solves a couple of problems but introduces others. Since it uses the bass cone as a wave guide, it introduces interference to the higher frequencies which result from the moving bass cone. The effect is more pronounced the louder you play naturally. Also due to the small physical size and lack of use of delayed porting (resistance enclosure) etc. the uni-q is not especially directive on lower frequencies.

My brother in law bought a pair of KEF LS50:s recently and they're very nice little speakers. But no match to ESLs for example.
 
Realistically we all hear things differently. Speaker sensitivity is related to power needed to drive them. PC speakers for the most part meet the demands of most people. Those who truly enjoy listening to music will receive a greater reward by investing in proper speakers that meet your ears happiness. To drive those speakers you need proper amplication, which in turn require a good source.

There's many ways to get there , obviously some better than others. Some like these shitty computer speakers and others don't. When I see people wanting USB dac''s and what not, it tells me they care about their music... So why compromise it with speakers that aren't giving you the entire frequency range? That's how I look at it but others have their own views.

I could sit here and talk about sensitivity, phase ( more important when dealing with subwoofers), tweeters, mids, room correction and the other 100 things that make up home audio. The reality is this is a computer hardware forum rather than an AV forum which needs simple explanation rather than a debate based on ones opinion.

Keeping it simple is probably best as home audio is a giant animal all in itself.

Once again Venomous is "dead-on" He's right, this isn't a AV forum. I think I would have been happy with those Definitive Technology Inclines in the end. All-in-one set up doesn't mean it's a bad thing. I'm starting to get a little crazy on my end as I just received a nice Sewell T-AMP & I'm thinking about the NAD D3020 purchase. Also....As Venomous mentioned, sound is subjective. So there you have it.
 
The real challenge with speakers is to maintain that flat frequency response also in a real world scenario. All those other properties are completely worthless if the speaker allows the room to mess the sound up.

This is a Bowers&Wilkins graph where the same speaker is measured directly in front of the speaker (nearfield) and in the listening position. As you can see, the flat response is completely meaningless unless the speaker can reduce the room reflections:

Frequency-response1.jpg

Yep. But room is a different problem to solve. The goals for speaker design do not change, you can't design a speaker with a certain room interaction in mind as all rooms are different.
 
Uni-q solves a couple of problems but introduces others. Since it uses the bass cone as a wave guide, it introduces interference to the higher frequencies which result from the moving bass cone. The effect is more pronounced the louder you play naturally. Also due to the small physical size and lack of use of delayed porting (resistance enclosure) etc. the uni-q is not especially directive on lower frequencies.

My brother in law bought a pair of KEF LS50:s recently and they're very nice little speakers. But no match to ESLs for example.

Maybe on the Q100s, but the higher up you go on the KEF chain the UniQ driver becomes a dedicated midrange rather than trying to do the bass to. On the R900s I think it gets crossed over at 400hz to two 8" drivers. So this effect is much less pronounced
At the moment I cross mine over at 80hz to a sub. TBH, I understand what you are saying in theory, but I have yet to actually notice it. But I don't play these super loud due to neighbour constraints.
 
I'm sorry but no... If you can get KEF components in your hand and have the crossover and enclosure designs, anyone can build a similar speaker using hand tools. There's no magic, it's simple technology. The drivers are the only thing you really can't build yourself but luckily they're mass produced by a few OEM manufacturers and most speaker drivers are sold to anyone interested. KEF drivers are an exception because they use their own patented technique so the factory is not allowed to sell them to anyone buf KEF. You can, however, buy drivers that operate in a much similar fashion with kefs.

KEF doesn't have an alien technology plant building the stuff, they're assembled on a mass production line. I've visited a couple of loudspeaker factories (Fusion, Gradient) and I've visited their measuring chambers, production lines etc. so I know this for a fact :)

Any hobbyist can invest 200 bucks to a microphone and a RTA software for time window based frequency response measurements - and even this is required only if you start doing your own designs that you need to test.

sorry but I find it hard to DIY speakers to high end speakers build quality, curing, shaping, paint, finish, extensive research through decades of production, development, constant testing and refining. when you build a speaker from a kit that has been constructed in an uncontrolled environment and untested end product which means every DIY speaker can sound different even constructed with the exactly the same materials.
 
Last edited:
DIY is not simply building a nice looking box to house a couple of off shelf components.

I mean, everything is so much more complicated than that. You have to design crossovers, maybe build Zobel networks to flatten impedance and this is just the basics, this is before you get into quarter wave designs or open baffle.

There's just so many ways to build a loudspeaker, there's more to it than getting a correct sized box and plugging in Thiele/Small parameters.

What drivers are you going to use? Have you tamed cone breakup at high frequencies if you use certain metal coned drivers? Have you crossed over at a frequency low enough to prevent beaming but high enough to minimize distortion (such as on ribbon tweeters). Have you corrected phase issues that crossovers create.

That is just scratching the surface, I was interested in it in a while, but I never started because there is just so much to know and implement correctly if you want to end up with a speaker that can hang with ones designed by guys who know what they're doing.
 
that is why i would find it incredibly difficult to design, construct, my owner speaker and i would rather pay for a used high end, well engineered / designed set of speakers.
 
Yep. But room is a different problem to solve. The goals for speaker design do not change, you can't design a speaker with a certain room interaction in mind as all rooms are different.

LOL learn a little about the subject before you write stuff like that. Learn about controlled directivity a bit. It negates the effect of the room the more efficiently the higher the directivity is. When room is removed from the equation it doesn't matter which kind of room you have. For example if you measure the listening position room response of an electrostatic panel, it will closely match the response in the anechoic chamber. Outside listening position naturally the sound is destroyed just as with common dynamic speakers.

The back radiation of a dipole ESL can give you some interesting options though, you can for example reflect the back beam of your listening setup so that it hits your computer seat. It makes an amazing effect of having a wall surface 'play' sound to you while you're at the computer :)

The same back beam also needs controlling, you should never place any dipole panel so that the back reflection hits the listening position or radiates directly back to the panel itself.
 
Last edited:
sorry but I find it hard to DIY speakers to high end speakers build quality, curing, shaping, paint, finish, extensive research through decades of production, development, constant testing and refining. when you build a speaker from a kit that has been constructed in an uncontrolled environment and untested end product which means every DIY speaker can sound different even constructed with the exactly the same materials.

If you build speakers to look at them, yes. But if you build them for pure sonic properties it is really simple for anyone to build a speaker given the correct components.

If you think that serial produced hifi/high-end speakers are hand tuned and tested after assembly, you're wrong. The passive crossover would be insanely expensive to tune box by box as it requires not only disassembly of the speaker but also manual labour. The manufacturing tolerances in the quality components are so low that there's no point testing each individual. The room effect will be much bigger than a component tolerance anyway in the end.

There is no 'magic' in building boxed speakers folks. It's 1930's technology. The basic transducer has not evolved a bit after that. The first revolution was 1950s when electrostatic panels were discovered. Again, after that extremely little has happened in the tech, only refining the existing techniques. Don't get fooled by the bed time stories high-end product manufacturers try to feed you. The marketers know that in audio, a product that has a positive image will always sound better to the listener. It's enough to feed an unconcious idea of mysterious technology for the listener to get bias already. Super expensive cables are a fine example of sonic differences which very listener to listener, cannot be measured with any known measuring device and have disproportionate cost.

Boxed speakers with dynamic drivers are extremly simple. The only magic happens at the design table, once the construction time starts it's a simple assembly job that anyone can accomplish. The hardest part really is to build non rectangular boxes for the more advanced designs. The curved enclosure shape is very hard to do, which is why most speaker brands avoid it despite the sonic advantages it offers. It's just so expensive to mass produce.
 
Last edited:
DIY is not simply building a nice looking box to house a couple of off shelf components.

I mean, everything is so much more complicated than that. You have to design crossovers, maybe build Zobel networks to flatten impedance and this is just the basics, this is before you get into quarter wave designs or open baffle.

There's just so many ways to build a loudspeaker, there's more to it than getting a correct sized box and plugging in Thiele/Small parameters.

What drivers are you going to use? Have you tamed cone breakup at high frequencies if you use certain metal coned drivers? Have you crossed over at a frequency low enough to prevent beaming but high enough to minimize distortion (such as on ribbon tweeters). Have you corrected phase issues that crossovers create.

That is just scratching the surface, I was interested in it in a while, but I never started because there is just so much to know and implement correctly if you want to end up with a speaker that can hang with ones designed by guys who know what they're doing.

NO NO and NO! You do NOT do DIY hifi by starting to design your own speakers from the scratch. You take a design of a respected designer and assemble it at home. Some DIY designs are clones of factory speakers, it's really simple to disassemble a factory speaker, measure the enclosures, check the crossover components and the types of the drivers used. After you know all that you can already build an identical copy. Only the surface finish is outside the reach of most hobbyists.

You have to study speaker design for years before you reach the level of starting to design your own speakers. For most DIY builders this is never the goal in the first place. They just want a good sounding speaker for cheap - and have a building hobby as a side bonus. If you're interested in actually designing speakers, you're going to have to invest heavily on studying the theory and buying RTA measuring tools. That's only a short step away from starting your own commercial venture.
 
Bonnie,

i appreciate your comments on DIY speakers but im one who will never attempt this as i dont have the skill or knowledge.

me i would rather purchase a used high end speaker like the KEF 207/2, Dynaudio C4, Revel Salon 2 - looks great and sounds great as well.
 
Bonnie,

i appreciate your comments on DIY speakers but im one who will never attempt this as i dont have the skill or knowledge.

me i would rather purchase a used high end speaker like the KEF 207/2, Dynaudio C4, Revel Salon 2 - looks great and sounds great as well.

And this is fine, DIY is not for everyone. I just wanted to point out that you don't need more than basic soldering and wood work skill to build DIY speakers. There are ready designs, tried, measured and tested for that purpose.
 
LOL learn a little about the subject before you write stuff like that. Learn about controlled directivity a bit. It negates the effect of the room the more efficiently the higher the directivity is. When room is removed from the equation it doesn't matter which kind of room you have. For example if you measure the listening position room response of an electrostatic panel, it will closely match the response in the anechoic chamber. Outside listening position naturally the sound is destroyed just as with common dynamic speakers.

The back radiation of a dipole ESL can give you some interesting options though, you can for example reflect the back beam of your listening setup so that it hits your computer seat. It makes an amazing effect of having a wall surface 'play' sound to you while you're at the computer :)

The same back beam also needs controlling, you should never place any dipole panel so that the back reflection hits the listening position or radiates directly back to the panel itself.

Don't patronize me sunshine. No one elected you the source of all knowledge and in this case you are wrong.

It's mostly idiocy to design a speaker with room interaction in mind. Because rooms vary. You could have a 100 sq foot room in the shape of a cylinder or an 1800sq ft room. How can you build a speaker that will play the same in both? You can't

Room treatments are an entirely different subject. I can think maybe Bose designed some speakers where bouncing off walls was important. The 901's and maybe dipoles bounce sound back, but this is entirely not the same as what I'm saying and you well know it. You are being deliberately obtuse.
 
NO NO and NO! You do NOT do DIY hifi by starting to design your own speakers from the scratch. You take a design of a respected designer and assemble it at home. Some DIY designs are clones of factory speakers, it's really simple to disassemble a factory speaker, measure the enclosures, check the crossover components and the types of the drivers used. After you know all that you can already build an identical copy. Only the surface finish is outside the reach of most hobbyists.

You have to study speaker design for years before you reach the level of starting to design your own speakers. For most DIY builders this is never the goal in the first place. They just want a good sounding speaker for cheap - and have a building hobby as a side bonus. If you're interested in actually designing speakers, you're going to have to invest heavily on studying the theory and buying RTA measuring tools. That's only a short step away from starting your own commercial venture.

You tell that to all the DIYers who design and build their own speakers.

Copying someone elses design is little different to assembling a kit. Yes it's DIY. They are both DIY, but one is done in order to save a small amount of cash, the other is done because of love for audio, engineering and art. In this day and age you can buy speakers built in China by people being paid 12 cents an hour and that is designed by someone who knows their stuff and you are going to spend pennies more realistically buying over DIYing.
 
If you build speakers to look at them, yes. But if you build them for pure sonic properties it is really simple for anyone to build a speaker given the correct components.

If you think that serial produced hifi/high-end speakers are hand tuned and tested after assembly, you're wrong. The passive crossover would be insanely expensive to tune box by box as it requires not only disassembly of the speaker but also manual labour. The manufacturing tolerances in the quality components are so low that there's no point testing each individual. The room effect will be much bigger than a component tolerance anyway in the end.

There is no 'magic' in building boxed speakers folks. It's 1930's technology. The basic transducer has not evolved a bit after that. The first revolution was 1950s when electrostatic panels were discovered. Again, after that extremely little has happened in the tech, only refining the existing techniques. Don't get fooled by the bed time stories high-end product manufacturers try to feed you. The marketers know that in audio, a product that has a positive image will always sound better to the listener. It's enough to feed an unconcious idea of mysterious technology for the listener to get bias already. Super expensive cables are a fine example of sonic differences which very listener to listener, cannot be measured with any known measuring device and have disproportionate cost.

Boxed speakers with dynamic drivers are extremly simple. The only magic happens at the design table, once the construction time starts it's a simple assembly job that anyone can accomplish. The hardest part really is to build non rectangular boxes for the more advanced designs. The curved enclosure shape is very hard to do, which is why most speaker brands avoid it despite the sonic advantages it offers. It's just so expensive to mass produce.

Curved enclosures are very simple to do, they are often made by bending the MDF under pressure or using a series of cuts. The Pioneer BS22 used curved cabinets and they are about $120 a pair.
 
Boxed speakers with dynamic drivers are extremly simple. The only magic happens at the design table, once the construction time starts it's a simple assembly job that anyone can accomplish. The hardest part really is to build non rectangular boxes for the more advanced designs. The curved enclosure shape is very hard to do, which is why most speaker brands avoid it despite the sonic advantages it offers. It's just so expensive to mass produce.

This pretty much sums it up... http://www.paradigm.com/support/articles.php?topic=quiet-enclosures
 
Curved enclosures are very simple to do, they are often made by bending the MDF under pressure or using a series of cuts. The Pioneer BS22 used curved cabinets and they are about $120 a pair.

the use of MDF is on entry level speakers mainly - high end generally use plywood i believe.
 
No, MDF is the primary materiel of almost all commercial speakers, even on the high end. It has good dampening characteristics, is consistent, has no grain, and is inexpensive. It's heavy and doesn't have great strength, but is generally "good enough". Plywood has voids, tears out, bends poorly, and can de-laminate. You can find voidless Baltic Birch ply but that only solves one problem. That being said, it's better to use where low weight and high strength are needed like in PA or music cabs, but these are usually finished with paint or carpet.
 
Don't patronize me sunshine. No one elected you the source of all knowledge and in this case you are wrong.

It's mostly idiocy to design a speaker with room interaction in mind. Because rooms vary. You could have a 100 sq foot room in the shape of a cylinder or an 1800sq ft room. How can you build a speaker that will play the same in both? You can't

Room treatments are an entirely different subject. I can think maybe Bose designed some speakers where bouncing off walls was important. The 901's and maybe dipoles bounce sound back, but this is entirely not the same as what I'm saying and you well know it. You are being deliberately obtuse.

It is extremely simple to build a speaker that works well in both rooms: DIRECTIVITY. If you haven't learned about this tehcnique yet it's time to study, sunshine lol.

When room is removed from the equation, it no longer affects the outcome. 1000x0 = 0.
 
the use of MDF is on entry level speakers mainly - high end generally use plywood i believe.

High end speakers use all sorts of exotic materials because they don't have to worry about mass production costs.
 
You tell that to all the DIYers who design and build their own speakers.

Copying someone elses design is little different to assembling a kit. Yes it's DIY. They are both DIY, but one is done in order to save a small amount of cash, the other is done because of love for audio, engineering and art. In this day and age you can buy speakers built in China by people being paid 12 cents an hour and that is designed by someone who knows their stuff and you are going to spend pennies more realistically buying over DIYing.

Yes I'm telling that to them, as I'm telling that to you now. Designing your own speakers from scratch is a fools way to enter DIY audio. If you're naive enough to think that by studying the subject for a week or two, you're ready to be a speaker designer what can I say lol.

The more sensible DIY builders start with predesigned plans, do their first projects and then advance from that. A little by little you can progress in the hobby if interest remains and you can invest to RTA tools, tooling etc. to do more demanding projects. But 99% of DIY builders never enter that stage. Very rarely if ever people are ready to invest into high tech tooling and analyzers right from the start. Most likely they do not even understand that's what it takes to get anything serious done.
 
Last edited:

Was there something special about that enclosure? Internal bracing is extremely simple to make in DIY products. The aluminium material in the enclosure is most likely a cost saving method in order to be able to mass produce the preferred design. All those grooves and voids are required to get rid of the unwanted ringing characteristics of the metal used. No such problems with more inert materials.

Same thing with matteos' image, nothing special or hard to make in the enclosure except the curved front panel.

As for matteos' comments about curved enclosures being easy to make.. ROFL! Yeah try creating that pressure environment at home first, then let's discuss. Using custom made tooling its naturally a very simple process, insert mdf, press button and a curved piece pops out after some steaming. At home there's absolutely nothing thats simple about the process.

There is a reason why most speakers are boxed and why also matteos' KEF only has a curved front panel instead of having the whole enclosure at optimal shape: cost.

If cost was no factor most boxed speakers would look something like this:
Homage5.jpg


Even this shape is a compromise resulting from the use of regular dynamic drivers which require a wide flat front plate.
 
pretty sure MDF cant be bent- they just snap

Wood, MDF and plywood can be bent using a simple process where the material is closed into pressured steam and then bench pressed into form.

Simple process, extremely hard for a DIY builder to achieve at reasonable cost. Or safely.

Lawrence Audio Mandolin/Violin are one of the best affordable boxed speakers I've heard in a while. They (like many other good speakers) have a very narrow sweet spot though.

Lawrence-Audio-Cello-Speaker-Lawrence-Liao.jpg
 
Last edited:
Was there something special about that enclosure? Internal bracing is extremely simple to make in DIY products. The aluminium material in the enclosure is most likely a cost saving method in order to be able to mass produce the preferred design. All those grooves and voids are required to get rid of the unwanted ringing characteristics of the metal used. No such problems with more inert materials.

Same thing with matteos' image, nothing special or hard to make in the enclosure except the curved front panel.

As for matteos' comments about curved enclosures being easy to make.. ROFL! Yeah try creating that pressure environment at home first, then let's discuss. Using custom made tooling its naturally a very simple process, insert mdf, press button and a curved piece pops out after some steaming. At home there's absolutely nothing thats simple about the process.

There is a reason why most speakers are boxed and why also matteos' KEF only has a curved front panel instead of having the whole enclosure at optimal shape: cost.

If cost was no factor most boxed speakers would look something like this:
Homage5.jpg


Even this shape is a compromise resulting from the use of regular dynamic drivers which require a wide flat front plate.

No I was simply pointing out curved enclosures aren't just run of the mill shit. High end speakers all have entry level to full reference. Within those designs type of materials and how they are assembled change at each tier...
 
Yes I'm telling that to them, as I'm telling that to you now. Designing your own speakers from scratch is a fools way to enter DIY audio. If you're naive enough to think that by studying the subject for a week or two, you're ready to be a speaker designer what can I say lol.

The more sensible DIY builders start with predesigned plans, do their first projects and then advance from that. A little by little you can progress in the hobby if interest remains and you can invest to RTA tools, tooling etc. to do more demanding projects. But 99% of DIY builders never enter that stage. Very rarely if ever people are ready to invest into high tech tooling and analyzers right from the start. Most likely they do not even understand that's what it takes to get anything serious done.

I don't believe you read anybodies comments, you just disagree with them.
 
Dipoles are nice, but I've got 2, 4, and 5 year old kids at home and those just wouldn't last long. As it is, every exposed dome tweeter and dust cap gets pushed in, toys go into bass ports, etc. It's not difficult or expensive to DIY your own speakers to get something better/cheaper/more personalized that what's otherwise available. My current computer speakers are DIY cornu spiral horns:
http://www.diyaudio.com/forums/full...k-building-cornu-spiral-horn-now-you-can.html
Mine can be seen in post #225. The cost is minimal and the sound is better than cheap normal speakers, but it's definitely not "audiophile".
 
Dipoles are nice, but I've got 2, 4, and 5 year old kids at home and those just wouldn't last long. As it is, every exposed dome tweeter and dust cap gets pushed in, toys go into bass ports, etc. It's not difficult or expensive to DIY your own speakers to get something better/cheaper/more personalized that what's otherwise available. My current computer speakers are DIY cornu spiral horns:
http://www.diyaudio.com/forums/full...k-building-cornu-spiral-horn-now-you-can.html
Mine can be seen in post #225. The cost is minimal and the sound is better than cheap normal speakers, but it's definitely not "audiophile".

Speaker porn is no good if i need to create an account to see it:D
 
start sending tweets or emails to pc speaker makers and say you want the older 5,1 systems back'

hit corsairs fourms and say you want to buy 5.1 systems and maybe the will make a limted number of them or they will make the old 5.1 ones again since they are just fine anyway
 
Proof? They sound pretty knowledgeable to me...

They have no concept of directivity and what it does to room interaction. He (matteos) gets stuck to some sort of brain lock where he thinks that it's not possible to design a speaker that would work better in any given room. That's simply false.
 
Back
Top