I onlu use 2 channel fullrange to my headphone amp,and I use he digital in from the DEQ2496.All other stuff in DCX is bypassed.In DEQ i only use the EQ part all other parts are bypassed.I think its the analog out put in DCX that makes it sound not so god,at least I hope so.I have not given up hope on the DCX I am going to better the power supply change the output board (hope the new one from SELECTRONIC is a good one!And change the clock,hope that does it.
And if the DEQ can sound so good then I don´t want a weak link after it...
And if the DEQ can sound so good then I don´t want a weak link after it...
perhaps should have read more, but hearing is believing
Some of the bheringer stuff was interesting, and for 100 bucks you're right, you probably cant get another one like it. I have heard some high end units that create wonderful, sparkley highs, and produce beautiful imaging. As a crossover for low end, the behringer would be a good cost effective thing, and double your amps power by not using a passive device at a 3 or 6db cost. As far as it goes however, using a "wonderful" op amp at the output, correcting impedances either on the amp or crossover or both, are really necessary for audiophile reproduction. a 714 op amp just ain't cutting it (although I don't know if that's what they use.)
If you really want to get down to it a discrete drive set sounds even fatter, with the proper power supply. I has been my experience that the first thing that goes down in quality in a cheap design is the pots , then the capacitors , then the quality of the pc board. Power supplies usually are enough to do the job, but perhaps without much head room. I'm sure there's plenty of room for modding, and improvement as the a/d chips themselves are common, but as usual sampling rates, and then the surrounding components will have the largest effect on quality.
Some of the bheringer stuff was interesting, and for 100 bucks you're right, you probably cant get another one like it. I have heard some high end units that create wonderful, sparkley highs, and produce beautiful imaging. As a crossover for low end, the behringer would be a good cost effective thing, and double your amps power by not using a passive device at a 3 or 6db cost. As far as it goes however, using a "wonderful" op amp at the output, correcting impedances either on the amp or crossover or both, are really necessary for audiophile reproduction. a 714 op amp just ain't cutting it (although I don't know if that's what they use.)
If you really want to get down to it a discrete drive set sounds even fatter, with the proper power supply. I has been my experience that the first thing that goes down in quality in a cheap design is the pots , then the capacitors , then the quality of the pc board. Power supplies usually are enough to do the job, but perhaps without much head room. I'm sure there's plenty of room for modding, and improvement as the a/d chips themselves are common, but as usual sampling rates, and then the surrounding components will have the largest effect on quality.
read more, thought more
Well all said and done, what I'd really like to see is a way to program the thing. There's a lot of open source software out there for various filters and stuff. I'm not an electronics engineer, but have been involved for a pretty long time with pro audio. They are dealing with several orders of magnitude difference in price, the Lake crossover with basically the same setup is like 7000 US, so spending a little to improve the bheringer is no biggie. We were selling the DBX crossover of the same sort for 500 US, and it was a better sounding unit, although maybe not worth another 250 dollars (I stand corrected you can get the bheringer unit at Musicians friend for 250 US, the cheaper analog unit is 129.)
I can tell you that the lake unit is fantastic, with more parameters than you can shake a stick at, which will phase correct, eq with 100 db of cut and boost, and infinite crossover slope. this thing is no joke.
You can also get them to tailor it to your speaker by providing the data from tef testing etc.
what would be cool, would be a device somewhere in the middle. I believe that the lake runs 24 bit 96khz, but it possibly is running 192khz in the later models. It is risc based, and uses a very interesting power supply scheme (it uses ac and switching power supplies at the output transistors) yes, get one, open it. figgure it out. I only know what I had heard at the shop.
Well all said and done, what I'd really like to see is a way to program the thing. There's a lot of open source software out there for various filters and stuff. I'm not an electronics engineer, but have been involved for a pretty long time with pro audio. They are dealing with several orders of magnitude difference in price, the Lake crossover with basically the same setup is like 7000 US, so spending a little to improve the bheringer is no biggie. We were selling the DBX crossover of the same sort for 500 US, and it was a better sounding unit, although maybe not worth another 250 dollars (I stand corrected you can get the bheringer unit at Musicians friend for 250 US, the cheaper analog unit is 129.)
I can tell you that the lake unit is fantastic, with more parameters than you can shake a stick at, which will phase correct, eq with 100 db of cut and boost, and infinite crossover slope. this thing is no joke.
You can also get them to tailor it to your speaker by providing the data from tef testing etc.
what would be cool, would be a device somewhere in the middle. I believe that the lake runs 24 bit 96khz, but it possibly is running 192khz in the later models. It is risc based, and uses a very interesting power supply scheme (it uses ac and switching power supplies at the output transistors) yes, get one, open it. figgure it out. I only know what I had heard at the shop.
coupla notes
musiciansfriend.com ships anywhere, not a salesman, but really cheap prices.
Lake crossover uses ac power and transformers and caps (and I guess a regulator) at outputs and other analog nodes
toshiba cell processor has an audio coprocessor available in the chipset, and I have a feeling that it will make an amazing crossover, as it has 8 processsors (1 per channel of i/o) al little bit of linux, some free audio programs, already available and some decent output stages, (I understand that there are premium chips available for cheap in the 15 dollar range) and viola the superwonderful, fully configurable (I mean anything) crossover. If you're interested in building REAL audio, well let's give it a shot. I can't do it by myself, as I said I'm not an engineering student, but have a great repoiture of many of the disciplines necessary, and the folks here seem to know what they're talking about. (cell processor think xbox 360)
digital crossover and processing is the way to make your boxes work, I have seen it, and the cost is dropping. Your bheringer is great for prototyping, and ballparking, but start with something better if you're going to spend all that time and effort
education is expensive, and work is often thrown away
just one person's opinion.
cheers
musiciansfriend.com ships anywhere, not a salesman, but really cheap prices.
Lake crossover uses ac power and transformers and caps (and I guess a regulator) at outputs and other analog nodes
toshiba cell processor has an audio coprocessor available in the chipset, and I have a feeling that it will make an amazing crossover, as it has 8 processsors (1 per channel of i/o) al little bit of linux, some free audio programs, already available and some decent output stages, (I understand that there are premium chips available for cheap in the 15 dollar range) and viola the superwonderful, fully configurable (I mean anything) crossover. If you're interested in building REAL audio, well let's give it a shot. I can't do it by myself, as I said I'm not an engineering student, but have a great repoiture of many of the disciplines necessary, and the folks here seem to know what they're talking about. (cell processor think xbox 360)
digital crossover and processing is the way to make your boxes work, I have seen it, and the cost is dropping. Your bheringer is great for prototyping, and ballparking, but start with something better if you're going to spend all that time and effort
education is expensive, and work is often thrown away
just one person's opinion.
cheers
Pesky are you just being a troll? Talking to yourself, bashing the DCX w/out knowing what's inside it, etc.
Most of the people here have done their homework or have got a meaningful problem to solve. What is your contribution?
There is no need to programme a digital xover for pc, do a search for Thuneau Allocator or Brute FIR (given that you are mentioning Linux).
http://www.duffroomcorrection.com/wiki/BruteFIR_on_a_USB_memory_stick
But not many soundcards have got DACs as good as the Behringer and those which have are more expensive than the DCX. Furthermore, I am not sure how easy it is to mod the output stage of a soundcard given the limited space inside.
Best
Giulio
Most of the people here have done their homework or have got a meaningful problem to solve. What is your contribution?
There is no need to programme a digital xover for pc, do a search for Thuneau Allocator or Brute FIR (given that you are mentioning Linux).
http://www.duffroomcorrection.com/wiki/BruteFIR_on_a_USB_memory_stick
But not many soundcards have got DACs as good as the Behringer and those which have are more expensive than the DCX. Furthermore, I am not sure how easy it is to mod the output stage of a soundcard given the limited space inside.
Best
Giulio
Hi pesky,
The Lake contour 26D supports classical LR48 crossovers (with phase shift I assume) and impressive 180dB linear 'brick wall' crossovers I learned.
I do not expect too much from the higher slope with my standard 3-way speaker (might be different with several speakers spread in a room), since the difference between a 24dB slope and a 48dB slope isn't such big.
But I've never heard a linear crossover. Has anybody experience with this Lake crossover and what is the audible difference between a phase shifting and a linear crossover? Possibly BruteFIR supports both kinds of crossovers too.
The only disadvantage of this Lake is the 20 times higher price compared with the DCX.
Regards,
Frank
The Lake contour 26D supports classical LR48 crossovers (with phase shift I assume) and impressive 180dB linear 'brick wall' crossovers I learned.
I do not expect too much from the higher slope with my standard 3-way speaker (might be different with several speakers spread in a room), since the difference between a 24dB slope and a 48dB slope isn't such big.
But I've never heard a linear crossover. Has anybody experience with this Lake crossover and what is the audible difference between a phase shifting and a linear crossover? Possibly BruteFIR supports both kinds of crossovers too.
The only disadvantage of this Lake is the 20 times higher price compared with the DCX.
Regards,
Frank
Ryssen,
Yes, using a good headphone amp and high quality headphone makes the difference in analog out of the DCX2496 and other equipment standout. It was my experience with headphones on the DCX2496 versus an MAudio SuperDAC that made me look for other alternatives. (The stock SuperDAC was waaaaay better and my DCX2496 is already semi-modified) The solution I am using now is Thuneau Allocator Lite PC software with better DAC's.
It was also this experience that made me consider whether I shouldn't just be doing either
a) active crossover Marchand style
b) passive crossover
My thinking is... if I am ultimately going to have to mod something to get better sound.... why not build a complete solution from scratch. The active crossover GB on DIYAudio for example?
Oettle,
Phase shifting versus linear. I briefly used a Voxengo VST FIR crossover in a two-way system. I then compared to Voxengo's IIR style filter and then to Thuneau's. I found the phase-shifting IIR's to sound like perfect passive crossovers. The linear Phase FIR filters sounded "processed" and "electronic". I should have given them a chance to grow on me. Maybe years of listening to IIR filters has programmed me to what the music _should_ sound like. I am going to give FIR filters one more shot with the Accourate software. Haven't had time yet to play with that.
Yes, using a good headphone amp and high quality headphone makes the difference in analog out of the DCX2496 and other equipment standout. It was my experience with headphones on the DCX2496 versus an MAudio SuperDAC that made me look for other alternatives. (The stock SuperDAC was waaaaay better and my DCX2496 is already semi-modified) The solution I am using now is Thuneau Allocator Lite PC software with better DAC's.
It was also this experience that made me consider whether I shouldn't just be doing either
a) active crossover Marchand style
b) passive crossover
My thinking is... if I am ultimately going to have to mod something to get better sound.... why not build a complete solution from scratch. The active crossover GB on DIYAudio for example?
Oettle,
Phase shifting versus linear. I briefly used a Voxengo VST FIR crossover in a two-way system. I then compared to Voxengo's IIR style filter and then to Thuneau's. I found the phase-shifting IIR's to sound like perfect passive crossovers. The linear Phase FIR filters sounded "processed" and "electronic". I should have given them a chance to grow on me. Maybe years of listening to IIR filters has programmed me to what the music _should_ sound like. I am going to give FIR filters one more shot with the Accourate software. Haven't had time yet to play with that.
trolling
no, I'm not putting down anyone's project, I have had however experience with a variety of crossovers, including the lake, and have experience listening in a fairly controlled environment to some of bheringer's stuff. Usable, but not really that good sounding. As I stated, I listened to the analog version, and I believe this version as well since we had a speaker/amp/crossover wall that I built and listened to and demo'd on a regular basis. I worked at Clair Bros audio, and heard the lake, in fact many of them, as they were the standard item going on tour. As far as modding the bheringer, perhaps a larger enclosure would make adding parts easier.
soundwise the dbx was a better unit, I have no idea what dac's or whatever is inside, just going to show that the input and output stages, as well as all the other stuff inside is just as important as the other. it's a system after all.
no, I'm not putting down anyone's project, I have had however experience with a variety of crossovers, including the lake, and have experience listening in a fairly controlled environment to some of bheringer's stuff. Usable, but not really that good sounding. As I stated, I listened to the analog version, and I believe this version as well since we had a speaker/amp/crossover wall that I built and listened to and demo'd on a regular basis. I worked at Clair Bros audio, and heard the lake, in fact many of them, as they were the standard item going on tour. As far as modding the bheringer, perhaps a larger enclosure would make adding parts easier.
soundwise the dbx was a better unit, I have no idea what dac's or whatever is inside, just going to show that the input and output stages, as well as all the other stuff inside is just as important as the other. it's a system after all.
Then I´d have to have mu PC beside the stereo and goood soundcard.
That´s not the solution to me right,now.
I think it´s plenty of space inside the case to do the mods,new powersupply a clock and more,and I´m waiting for the NEW output bard at SELECTRONIC to se the light,and someone to make a rewivew of it...Idon´t think it´s a good idea to use a couple of not so good sounding IC´s in the outputstage.Why do you think people change the outputput IC´S in their CD players?Just hope the DAC is good enough to use...
Ryssen said:
Then I´d have to have mu PC beside the stereo and goood soundcard.
That´s not the solution to me right,now.
That's the Behringers main advantage as I see it. It's dead quiet and there's no messy cabling. Also, if you want to play a game on the PC it won't effect someone else playing music in another room. I am able to do light stuff like read email, surf the web, etc. with music or DVD playing. But I wouldnt dare try to play a game and listen to music in the background, even with dual core Intel.
In my case, I had to either
a) build a quiet PC
or b) remote the PC from the stereo.
I decided to remote the PC from the the listening room with 60 feet of ADAT toslink.
I'd still like to see someone offer to put 3 SPDIF outputs onto the Behringer units. I'd probably drop the PC at that point.
since you asked..(for some ideas)
the reason that the lake crossover uses a switching power supply on the output is to raise the 5volt computer supply to +/- 24 volts (I believe) for the output stages. If you check the voltage to your output chips, they probably are 5 volt, which is good for +6db and a couple of db of headroom. Traditionally good crossovers were capable of +15 or 18 db, which requires from +/- 15 to 24v. the lake had two boards inside the very tightly packed single rack space case, one on the top and one on the bottom. The processor(s) were on one and the remainder of the stuff was on the other.
if you change your power supply, you'll probably want to increase the voltage to the output stage, since most "pro" amps use +6db, so this leaves really not enough head room for real audio quality. I was reading another audio thread somewhere about neive op amps being available for like 15 dollars each, but I don't know where that was. Apparently there was a glut of some sort of module that had these on them, and they were selling them dirt cheap on e-bay.
hope this helps, rather than provokes
pesky
the reason that the lake crossover uses a switching power supply on the output is to raise the 5volt computer supply to +/- 24 volts (I believe) for the output stages. If you check the voltage to your output chips, they probably are 5 volt, which is good for +6db and a couple of db of headroom. Traditionally good crossovers were capable of +15 or 18 db, which requires from +/- 15 to 24v. the lake had two boards inside the very tightly packed single rack space case, one on the top and one on the bottom. The processor(s) were on one and the remainder of the stuff was on the other.
if you change your power supply, you'll probably want to increase the voltage to the output stage, since most "pro" amps use +6db, so this leaves really not enough head room for real audio quality. I was reading another audio thread somewhere about neive op amps being available for like 15 dollars each, but I don't know where that was. Apparently there was a glut of some sort of module that had these on them, and they were selling them dirt cheap on e-bay.
hope this helps, rather than provokes
pesky
Daveis said:
There has been a PC in my stereo rack for almost 4 years now. Reason for me to take the DEQ > DCX path now is that there is almost always some glitches somewhere in the chain when you play with htpc. Nice thing though - there is always a software update soon to be released. Problem is that when it finally is released and you have updated your system there is all to often something else that now has stoped working.
I can't see myself to go back to multi functional htpc again. Multiple computers maybe, but....
Daveis said:
Something like this you are looking for?
Sorry, I can't for the life of me remember where I copyed it from. I do believe it was a French site. Someone that can help here?
EDIT: Where to look other than diyaudio ?
Posted by Tanto
http://www.diyaudio.com/forums/showthread.php?s=&threadid=36252
Attachments
one last comment
read more of the thread, maybe should have first, since it seems like there was a fair amount of activity during the first weeks of the post. A real edumakation there. Seems like you have +/- 15 volt rails in the thing, so little to say there. Really seems like it's all been said at this point, so will check back and see the result, although would be nice to hear it. Difficult part about audio forums, graphs and charts just don't do it for me.
good luck.
pesky
read more of the thread, maybe should have first, since it seems like there was a fair amount of activity during the first weeks of the post. A real edumakation there. Seems like you have +/- 15 volt rails in the thing, so little to say there. Really seems like it's all been said at this point, so will check back and see the result, although would be nice to hear it. Difficult part about audio forums, graphs and charts just don't do it for me.
good luck.
pesky
DCX digital out
Reading trough Tanto's thread I'm thinking there are two different path one can go with a digital upgrade for the DCX.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=36252
The CS8404 Tanto use is a rather expensive chip. If our implementation is not for digital amp's it would make sense just to build in better DAC chips rather then use three external DAC's . If the DCX main board is capable of driving 3 x CS8404 in parallel with the internal DAC's it might also be capable driving 3 parallel DAC's inside the DCX?
As I understand Tanto configuration's he set the CS8404 up for I2S mode. There are several threads dealing with the task of overcome S/PDIF's limitation. One approach is to interface external DAC direcly to the I2S 'bus'. I should think that one need to think trough how to implement synchronized clock if one like to use external DAC's? The I2S approach would take care of that.
enough babbling from me now
regards
Reading trough Tanto's thread I'm thinking there are two different path one can go with a digital upgrade for the DCX.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=36252
The CS8404 Tanto use is a rather expensive chip. If our implementation is not for digital amp's it would make sense just to build in better DAC chips rather then use three external DAC's . If the DCX main board is capable of driving 3 x CS8404 in parallel with the internal DAC's it might also be capable driving 3 parallel DAC's inside the DCX?
As I understand Tanto configuration's he set the CS8404 up for I2S mode. There are several threads dealing with the task of overcome S/PDIF's limitation. One approach is to interface external DAC direcly to the I2S 'bus'. I should think that one need to think trough how to implement synchronized clock if one like to use external DAC's? The I2S approach would take care of that.
enough babbling from me now
regards
off topic
hi
did some measurement to compare the digital and analog input:
Complete description here:
http://www.kinotechnik.edis.at/pages/dcx2496/DCX2496/DCX2496_analog_digital_input.html
"http://www.kinotechnik.edis.at/pages/dcx2496/DCX2496/DCX2496_analog_digital_input.html"
It's hard to tell which one is better. Digital has a lower noise floor but some +5dB higher peaks above and also the clocking seems to be have more low frequency jitter on digital input .
Soundwise I still go for the analog.
I may come back when I finished the clock or SPDIF mod's.
Anybody here how has upgraded the clock or the digital input and made some measurements?
Greetings
Michael
hi
did some measurement to compare the digital and analog input:
Complete description here:
http://www.kinotechnik.edis.at/pages/dcx2496/DCX2496/DCX2496_analog_digital_input.html
"http://www.kinotechnik.edis.at/pages/dcx2496/DCX2496/DCX2496_analog_digital_input.html"
It's hard to tell which one is better. Digital has a lower noise floor but some +5dB higher peaks above and also the clocking seems to be have more low frequency jitter on digital input .
Soundwise I still go for the analog.
I may come back when I finished the clock or SPDIF mod's.
Anybody here how has upgraded the clock or the digital input and made some measurements?
Greetings
Michael
Hi Michael,
Very interesting measurements. Comparing the two diagrams I would say above -115db they are very similar and below the digital input is still better.
I double-checked my two inputs (both modded) sound wise and couldn't hear real differences. Might be the digital one was a little bit more precise.
My experience with the SPDIF input was that the XLR connector causes sometimes trouble. Also standard SPDIF levels are too low for the Behringer and typical PCs produce a lot of noise which you have on the SPDIF line. Everything is analog even this SPDIF line.
My decision was to improve signal quality first and therfore used an optical cable with a converter which I powered from the +5V rail of the DCX. I modded the output of the converter due to better signal quality. On the DSP PCB there is a small transformer and a 110 ohm resistor which could be omitted (haven't done so far). Do you have a scope to measure what you get between the RxP and RxN line at the CS8420.
This converter mod is pretty easy. Next step could be the Dr. Oehlrich mod. Would be interested in your results.
regards,
Frank
Very interesting measurements. Comparing the two diagrams I would say above -115db they are very similar and below the digital input is still better.
I double-checked my two inputs (both modded) sound wise and couldn't hear real differences. Might be the digital one was a little bit more precise.
My experience with the SPDIF input was that the XLR connector causes sometimes trouble. Also standard SPDIF levels are too low for the Behringer and typical PCs produce a lot of noise which you have on the SPDIF line. Everything is analog even this SPDIF line.
My decision was to improve signal quality first and therfore used an optical cable with a converter which I powered from the +5V rail of the DCX. I modded the output of the converter due to better signal quality. On the DSP PCB there is a small transformer and a 110 ohm resistor which could be omitted (haven't done so far). Do you have a scope to measure what you get between the RxP and RxN line at the CS8420.
This converter mod is pretty easy. Next step could be the Dr. Oehlrich mod. Would be interested in your results.
regards,
Frank
oettle said:
Hi Frank,
The 110 ohms you mentioned is there for proper cable termination and if you take it out you may have higher level but worse signal (bad square waves, ringing, jitter etc.).
My own solution was to built a small S/PDIF to AES/EBU converter because the Behringer's CS8420 expects AES/EBU format, it works often with S/PDIF but that's just luck.
See the yahoo DCX2496 group for schematic details and a small PCB.
Jan Didden
oettle said:
Hello guys,
Here's a modification for converting the digital input to SP/DIF SPDIF conversion
It's from my DCX296 website which by the way is being translated into english.
Lot of work indeed : http://www.dcx2496.fr
TM