Kind, I think we need to know what kind of driver/amp combo we're looking at for a starting point to create a box for the dsp parameters.
Binely,
Could you be more specific in what you are asking? I am working with class AB amplifiers myself, hut others may be looking at class D type amps or even a truly digital amplifier solution. I don't know why that would really matter that much, a dsp shouldn't have a clue what type of amp it is working with.
Could you be more specific in what you are asking? I am working with class AB amplifiers myself, hut others may be looking at class D type amps or even a truly digital amplifier solution. I don't know why that would really matter that much, a dsp shouldn't have a clue what type of amp it is working with.
Why not have balanced XLR i/ps?
There are very simple & inexpensive ways of connecting unbalanced preamp o/ps to a properly balanced input with VERY high quality.
Richard,
Besides being able to hook up a balanced input what advantage would there be to attaching a single ended system to a powered speaker? Is this just so you have both options? Unless we are talking pro audio types of outputs or some rare audiophile equipment where and why would you do that?
Besides being able to hook up a balanced input what advantage would there be to attaching a single ended system to a powered speaker? Is this just so you have both options? Unless we are talking pro audio types of outputs or some rare audiophile equipment where and why would you do that?
Well, we've tested this.
AES E-Library Simulation and Investigation of Doppler Distortion
As one of the guinea pigs, I can say you need HUGE levels (akin to your 3" driver playing loud organ music with orchestra & other stuff happening too) before its audible ... and when its audible, it's not objectionable.
So if you wanna work on solving AUDIBLE & OBJECTIONABLE speaker faults, Doppler is well down the list.
Well it is nice to know you have empirical data to confirm what I have thought from listening to my larger speaker with high excursion and I couldn't hear it myself. So many others will always go to the theoretical and say it has to be audible, I say you can't hear that when it is moving on axis like a speaker does, now if the speaker was moving like a train across our ears okay, but not the limited displacement that we are talking about. So many use this theory of Doppler shift to justify the use of large 15" speakers, my feeling is it is just easy to make bass that way with a high efficiency speaker and low power requirement. But then they try and push the speaker so high above the pistonic range that the cone break up is much more of a problem than some Doppler shift from a smaller cone working higher into the frequency band. I know I am an outlier in my thinking, I just don't always believe what is generally gospel in audiophile circles.
No.
But you definitely want a balanced line if your preamp is at one end of the room and the amp/speaker is at the other. You can do it with 2 resistors per channel and high quality 2 core mike cable. The 2 resistors convert your consumer unbalanced gear output to 'pro' balanced line.
Thank you, I wasn't thinking about the possible long length of the interconnect cables. I would need to use the mini XLR connectors, a full scale connector is just so big in diameter.
I'm sorry I didn't see this thread but these days, I only look at stuff which I've been previously involved in. Too much "mine is bigger than yours" as is usual ... but I do see some pearls of wisdom ... 🙂
I played with digital speaker EQ in da previous Millenium firstly with a Loughborough Sound Images card (which dates me) and later was involved with Malc. Hawkesford & Richard Greenfield at Essex.
The first effort never led to good sound cos lack of good DACs. The Essex stuff was reasonably successful cos by that time CD was rampant and you could modify the S/PDIF stream.
If I was making a 2-way 'digital' speaker with xover at the usual 3-5kHz, it would have a single big amp and passive xover. This would NOT be flat but analogue EQ for this gives excellent results. Today, I would use DSP to EQ but the rest would be the same as it brings out the best of passive, digital and analogue.
If it was a 3-way system, I would bi-amp with an active xover about 1kHz to maximises dynamic range. Rod Elliot has a page explaining this which is quite accurate .. and more importantly, its easy to demo and hear the advantage.
Even a small speaker would be EQ'd flat to either 40 or 50Hz and they would be ported. My patented Powered Integrated Super Sub tech. would be used to protect and get the best out of both big & small units. There are digital versions of this but it is probably still easier today to do this analogue.
A separate sub(s) would be required to go down to 20Hz or 16Hz for da organ fans. There are too many compromises in making a 'main' bass unit also do this last 8ve. We designed & built the speakers for the Bradford Computer Organ with installations from 2 subs + 2 mains for small churches to 20 subs + 20 mains in the nave of Worcester Cathedral.
What I have no experience of and would love to try is Room EQ though I have ideas. I don't think any of the 'free' solutions is optimum and I've had long & interesting discussions with Dennis on Accourate. My personal prejudice is that Benjamin's method (as used by Meridian) is probably the way to go
The Effect of Room Acoustics on Subwoofer Performance and Level Setting
There's also B&O's method but there's much less info on this in public.
For Digital EQ, Simple Arbitary IIRs is as good as any and better than most .. but I'm biased. 😀 It has very low latency.
my $0.02
I played with digital speaker EQ in da previous Millenium firstly with a Loughborough Sound Images card (which dates me) and later was involved with Malc. Hawkesford & Richard Greenfield at Essex.
The first effort never led to good sound cos lack of good DACs. The Essex stuff was reasonably successful cos by that time CD was rampant and you could modify the S/PDIF stream.
If I was making a 2-way 'digital' speaker with xover at the usual 3-5kHz, it would have a single big amp and passive xover. This would NOT be flat but analogue EQ for this gives excellent results. Today, I would use DSP to EQ but the rest would be the same as it brings out the best of passive, digital and analogue.
If it was a 3-way system, I would bi-amp with an active xover about 1kHz to maximises dynamic range. Rod Elliot has a page explaining this which is quite accurate .. and more importantly, its easy to demo and hear the advantage.
Even a small speaker would be EQ'd flat to either 40 or 50Hz and they would be ported. My patented Powered Integrated Super Sub tech. would be used to protect and get the best out of both big & small units. There are digital versions of this but it is probably still easier today to do this analogue.
A separate sub(s) would be required to go down to 20Hz or 16Hz for da organ fans. There are too many compromises in making a 'main' bass unit also do this last 8ve. We designed & built the speakers for the Bradford Computer Organ with installations from 2 subs + 2 mains for small churches to 20 subs + 20 mains in the nave of Worcester Cathedral.
What I have no experience of and would love to try is Room EQ though I have ideas. I don't think any of the 'free' solutions is optimum and I've had long & interesting discussions with Dennis on Accourate. My personal prejudice is that Benjamin's method (as used by Meridian) is probably the way to go
The Effect of Room Acoustics on Subwoofer Performance and Level Setting
There's also B&O's method but there's much less info on this in public.
For Digital EQ, Simple Arbitary IIRs is as good as any and better than most .. but I'm biased. 😀 It has very low latency.
my $0.02
Last edited:
This is where it has been hard to not only come up to speed quickly but also to try and separate the opinions from the reality. Trying to read different threads there is so much disagreement between using iir and fir filters. Some say the iir filter is not enough, probably more of the mine is bigger than your thinking, but perhaps in specific cases where phase shift is large at the crossover point an fir filter would be called for? One thing that has made me think the fir filter could be a problem is the pre-ringing of the filter and also the latency that can be a real problem if you have to sync with a video stream.
George, gpapag, over on the Blowtorch thread has suggested I obtain a mini-dsp to play with and see what can be done. Learning all this and new software applications is a lot to learn and understand but what else can you do.
Now I need to spend $40.00 for those two papers!
George, gpapag, over on the Blowtorch thread has suggested I obtain a mini-dsp to play with and see what can be done. Learning all this and new software applications is a lot to learn and understand but what else can you do.
Now I need to spend $40.00 for those two papers!
First congratulation what sounds as very fine performing drivers developed.
Think there is no shortcut other than put in the time it takes to understand and navigate some software/hardware package. From own experience seems for me you own yourself to listen your new speakers with at least the XO phase turn neutralized by a FIR filter and drivers time aligned. Suggest as Barleywater a fast cheap solution to do that by using JRiver as temporary exercise and a mono speaker. That experience could point you to what filters is needed for industrial product and then investigate what hardware/software package can deliver that to your final product.
If you need help starting up JRiver/Rephase/REW software just post a request or PM.
Byrtt,
Thank you very much for the comments and the offer of help. I have tried to read through a bunch if the RePhase thread, it is like many others long and convoluted. It is time to update my desktop computer and also get a much higher quality sound card than what is now in that computer, much to old now to use a current card, no pcie slots so a no go on that. I guess I could use an external sound card. I have actually been wondering if I can use some of this software with my Clio lite analysis setup, they want another $400.00 just to add the waterfall plotting functions to the software. I've had to have a friend do waterfall plots for me but he is in another state and takes forever to get something done.
Thank you very much for the comments and the offer of help. I have tried to read through a bunch if the RePhase thread, it is like many others long and convoluted. It is time to update my desktop computer and also get a much higher quality sound card than what is now in that computer, much to old now to use a current card, no pcie slots so a no go on that. I guess I could use an external sound card. I have actually been wondering if I can use some of this software with my Clio lite analysis setup, they want another $400.00 just to add the waterfall plotting functions to the software. I've had to have a friend do waterfall plots for me but he is in another state and takes forever to get something done.
Kindhornman,
Know Rephase and REW are free and not pro expensive software but think they can bring one very long down the road (see the plots that Barleywater show us from time to time they impressive). If Clio lite have import/export functions as REW and Rephase they should be able to support each other.
Regarding soundcards think external ones are great nowadays and relative low cost as example a ASUS U7 which can bring you four channels for real stereo setup to feed four slewmasters bonded to two tweeters and two basses. A simple stereo soundcard is enough to develop filters for a mono speaker in JRiver its possible to mix L and R to mono then route to HP/LP sections filters that go to soundcard L/R feeding two slewmasters which bonded a tweeter and bass driver. Regarding FIR filters your concern for pre ringing and latency is from my experience not a problem if only used to either linearize a IIR XO or using a real FIR XO, but immediately pay back with a impressive step response that over time makes one smile when listening. Think the worse pre ringing and latency is when one is manipulating the speaker system bandpass ends especially the lowend.
A step further than us diy can afford could be to go for LspCAD or SoundEasy, not shure but think they have possibility set target slopes and calculate the filters automatic.
Know Rephase and REW are free and not pro expensive software but think they can bring one very long down the road (see the plots that Barleywater show us from time to time they impressive). If Clio lite have import/export functions as REW and Rephase they should be able to support each other.
Regarding soundcards think external ones are great nowadays and relative low cost as example a ASUS U7 which can bring you four channels for real stereo setup to feed four slewmasters bonded to two tweeters and two basses. A simple stereo soundcard is enough to develop filters for a mono speaker in JRiver its possible to mix L and R to mono then route to HP/LP sections filters that go to soundcard L/R feeding two slewmasters which bonded a tweeter and bass driver. Regarding FIR filters your concern for pre ringing and latency is from my experience not a problem if only used to either linearize a IIR XO or using a real FIR XO, but immediately pay back with a impressive step response that over time makes one smile when listening. Think the worse pre ringing and latency is when one is manipulating the speaker system bandpass ends especially the lowend.
A step further than us diy can afford could be to go for LspCAD or SoundEasy, not shure but think they have possibility set target slopes and calculate the filters automatic.
REW does waterfalls too.
re-Phase is much easier to learn by downloading and exporting a few filters and looking at them in REW or other software rather than reading thread. I'm certain that with your experience the light bulb will not take long to shine brightly.
Arbitrary IIR filters may be used to flatten frequency response. Each peak in response has matching filter for Q and a negated gain value. I'm sure automation is possible. Much easier to make inverse transfer function of measured response and then apply pass band filtering of choice.
re-Phase is much easier to learn by downloading and exporting a few filters and looking at them in REW or other software rather than reading thread. I'm certain that with your experience the light bulb will not take long to shine brightly.
Arbitrary IIR filters may be used to flatten frequency response. Each peak in response has matching filter for Q and a negated gain value. I'm sure automation is possible. Much easier to make inverse transfer function of measured response and then apply pass band filtering of choice.
Vacuphile,
Thanks I just downloaded the latest version of ARTA last night but didn't have time to look at it yet. Where is the $100.00 charge coming in? I have an Earthworks mic and Earthworks single channel mic preamp that I use with my Clio setup. I have part of what I need but not yet a complete system so I can move up to a 1/4" B&K measurement mic but that is still going to cost me, I also need the power supply to go with the mic. That is why I use my friend, he has a complete B&K lab, very impressive, he even had the B&K speaker turntable which will take something like 450 lbs.
Brytt,
I'll have to check and see if I can interface with the Clio system with secondary software. I think that may be possible as I think you can add outside wave functions, but again it may be that they crippled the software so you have to upgrade it to do that.
If I was a programmer I guess you could look at the source code and modify it, but that isn't something I can do.
Thanks I just downloaded the latest version of ARTA last night but didn't have time to look at it yet. Where is the $100.00 charge coming in? I have an Earthworks mic and Earthworks single channel mic preamp that I use with my Clio setup. I have part of what I need but not yet a complete system so I can move up to a 1/4" B&K measurement mic but that is still going to cost me, I also need the power supply to go with the mic. That is why I use my friend, he has a complete B&K lab, very impressive, he even had the B&K speaker turntable which will take something like 450 lbs.
Brytt,
I'll have to check and see if I can interface with the Clio system with secondary software. I think that may be possible as I think you can add outside wave functions, but again it may be that they crippled the software so you have to upgrade it to do that.
If I was a programmer I guess you could look at the source code and modify it, but that isn't something I can do.
Barleywater,
Thanks, you must have been writing at the same time as I was, I didn't see your response until I was done writing, I hate my internet connection, oh so many interrupts. Drops the connection constantly and AT&T doesn't care in the least. Damned DSL! I'm sure I have REW on this laptop also, so many downloads that I have to take time to go back and learn. I will start to play and see what I can learn. Unlike someone else on this site I don't expect any sound quality out of the speakers in my laptop, perhaps with headphones but not the built in Beats speakers, what a joke that is.
Thanks, you must have been writing at the same time as I was, I didn't see your response until I was done writing, I hate my internet connection, oh so many interrupts. Drops the connection constantly and AT&T doesn't care in the least. Damned DSL! I'm sure I have REW on this laptop also, so many downloads that I have to take time to go back and learn. I will start to play and see what I can learn. Unlike someone else on this site I don't expect any sound quality out of the speakers in my laptop, perhaps with headphones but not the built in Beats speakers, what a joke that is.
You can't save with the "demo" version, which locks you out of some of the features that can only be used with ARTA's .pir file format. I use REW exclusively these days, but I'll export the data to use ARTA's directivity plots.
Nate,
Thanks for the heads up on that. Sounds like my Clio system, even if you bought it they disable some features unless you pay more.
Thanks for the heads up on that. Sounds like my Clio system, even if you bought it they disable some features unless you pay more.
Derek, I know you are involved with
Signal Wizard Systems
but could you advise us itinerant beach bums on a suitable inexpensive hardware & development DSP platform, ?
After being in the bush for nearly 2 decades, I emerge to find the even the cheapest PC has more than enough computing power to do what I could only dream about in da last Millenium.
My thoughts for a 'digital speaker' tend towards using a populated ATX mini PC motherboard.
You can get them for US$50 with 8 channels of textbook 16b 48kHz sound outputs on board. Running Linux off a USB stick would be an easy development platform and the boards often have ethernet & USB to connect fancier soundcards too.
- Status
- Not open for further replies.