Between Ergo's, DIY_newbie's and Gary P's reply you have all the wisdom regarding DCX unit. My advice is - build your pasive crossover while you are modding DCX. It is a long standing project that is not cheap. Starting with volume control, and elimination of any original analog circuitry within DCX because it really sucks. If you have any decent system with good DA converter you will hear huge downgrade by suplementing it with DCX as a DA converter. Original circuitry is bad so I have to say modding is a must. But think in the terms of modding 12 channels. That is 6 ballanced channels of everything you do. I never used analog inputs simply because that will make double DA/AD conversion which is to be avoided at any cost.
Here is the link for my mod:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=76095
Just like Gary P I had a great success with line level transformers right after DA converters. I would recomennd it to anyone.
Regarding features on DCX - there is nothing like sitting comfortly in the sofa in front of your speakers with lap top computer and changing all kinds of settings live!
Delay, EQ, crossover points, mute single drivers, attenute separate drivers... All while you listening effects. Even if you are not going to use DCX later you could use it just to test design of your future pasive XOver. Not to mention that DCX and Sound Easy are talking to each other.
Here is the link for my mod:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=76095
Just like Gary P I had a great success with line level transformers right after DA converters. I would recomennd it to anyone.
Regarding features on DCX - there is nothing like sitting comfortly in the sofa in front of your speakers with lap top computer and changing all kinds of settings live!
Delay, EQ, crossover points, mute single drivers, attenute separate drivers... All while you listening effects. Even if you are not going to use DCX later you could use it just to test design of your future pasive XOver. Not to mention that DCX and Sound Easy are talking to each other.
Hi Raka,
long time since I spoke with you. How you've been?
Yes you are right, but that was just one feature out of many presented with DCX. Obviously with SB you will have your computer forever tied to your system if you would like to use it permanently as active crossover. As I mentioned in the text all the mods and volume control will cost multiples of price for DCX. DCX itself has rally bargan price. The only unit that could be compared to modded DCX would be DEQX that goes for $ 3,000.00. I had oportunity to listen DEQX with analog outputs and it doesn't sound as good as DCX with transformers in the output but it is certainly convenient for the people that do not want to DIY. I do not recall any other product that will compare to DCX with modifications. Correction - you could order DEQX with digital outs which is something that I would prefer but not for that price.
All in all you really have to be determined to make this to work and if you do you will be happy. Are there more expensive ways to do it - yes, less expensive - yes, but with moded DCX you will have a serious audiophile tool.
long time since I spoke with you. How you've been?
Yes you are right, but that was just one feature out of many presented with DCX. Obviously with SB you will have your computer forever tied to your system if you would like to use it permanently as active crossover. As I mentioned in the text all the mods and volume control will cost multiples of price for DCX. DCX itself has rally bargan price. The only unit that could be compared to modded DCX would be DEQX that goes for $ 3,000.00. I had oportunity to listen DEQX with analog outputs and it doesn't sound as good as DCX with transformers in the output but it is certainly convenient for the people that do not want to DIY. I do not recall any other product that will compare to DCX with modifications. Correction - you could order DEQX with digital outs which is something that I would prefer but not for that price.
All in all you really have to be determined to make this to work and if you do you will be happy. Are there more expensive ways to do it - yes, less expensive - yes, but with moded DCX you will have a serious audiophile tool.
Thanks for everyone's input on this.
I am a bit old school and all this modification and op amps after the DCX makes me happy tweaking my passive crossover.
The speakers afterall sound pretty damn good as they are and the added expense of extra amps... well I don't know if I can justify it. So thanks again everyone .
I am a bit old school and all this modification and op amps after the DCX makes me happy tweaking my passive crossover.
The speakers afterall sound pretty damn good as they are and the added expense of extra amps... well I don't know if I can justify it. So thanks again everyone .
You don't need to mod the DCX.
You don't need to use only the digital input.
You don't need output attenuators.
Paranoid people need pills
I'll bet my paycheck that if I conduct an ABX test using
a modded DCX vs. stock DCX, you will fail the test.
If you are still confused then go and buy a DBX product
over Behringer as the price of a modded DCX can probably
exceed the cost of a different brand.
You don't need to use only the digital input.
You don't need output attenuators.
Paranoid people need pills
I'll bet my paycheck that if I conduct an ABX test using
a modded DCX vs. stock DCX, you will fail the test.
If you are still confused then go and buy a DBX product
over Behringer as the price of a modded DCX can probably
exceed the cost of a different brand.
LOL,
I was going to stay out of this end part of the discussion but now you've got me started Thy... Why do you have to drag me into this???
On the frist one : Agreed, unless you just like to mod things
One the second and third: Agreed, unless you just want the best sound possible.
Wether or not people can hear the actual difference between using the digital inputs or the A->D converter is debatebale. Regardless of onee's ability to hear this, it is another source of adding error/noise to the sytem. Why do it unless you have to??
The same goes for the attenuators. If you want to keep the maximum dynamic range and the highest signal to noise ratio you want to run the D->A converter at the highest level. I have no idea where the human threshold for noise is -90db, -75db??? I have no clue, same reasoning as above holds true: It is another source of adding error/noise to the sytem. Why do it unless you have to??
Pills area always good
I would actually like to see this. I would agree with you on the inability to hear the difference between a stock and modded DCX.
One a side note, a buddy of mine and I ran blind ABX testing on the DEQ (no adjustment, just A->D, D->A) with the levels matched and we could definetly tell the difference with the DEQ in the loop. Funny thing is neither of us preferred one or the other but we could definetly tell when it was there..
[/SoapBox]
--Chris
I was going to stay out of this end part of the discussion but now you've got me started Thy... Why do you have to drag me into this???
thylantyr said:
On the frist one : Agreed, unless you just like to mod things
One the second and third: Agreed, unless you just want the best sound possible.
Wether or not people can hear the actual difference between using the digital inputs or the A->D converter is debatebale. Regardless of onee's ability to hear this, it is another source of adding error/noise to the sytem. Why do it unless you have to??
The same goes for the attenuators. If you want to keep the maximum dynamic range and the highest signal to noise ratio you want to run the D->A converter at the highest level. I have no idea where the human threshold for noise is -90db, -75db??? I have no clue, same reasoning as above holds true: It is another source of adding error/noise to the sytem. Why do it unless you have to??
Originally posted by thylantyr
Paranoid people need pills
Pills area always good
I would actually like to see this. I would agree with you on the inability to hear the difference between a stock and modded DCX.
One a side note, a buddy of mine and I ran blind ABX testing on the DEQ (no adjustment, just A->D, D->A) with the levels matched and we could definetly tell the difference with the DEQ in the loop. Funny thing is neither of us preferred one or the other but we could definetly tell when it was there..
Originally posted by thylantyr
If you are still confused then go and buy a DBX product
over Behringer as the price of a modded DCX can probably
exceed the cost of a different brand.
[/SoapBox]
--Chris
DIY_newbie said:
I'd have to say bollocks to that
I had an unmodded DEQ2496 as DAC. I found it slightly harsh and distorted. I bypassed the output stage with transformers and it was much better. I could really hear the higher IMD before, now I could play it much louder (as sown by the SPL meter on the DEQ!) without it sounding too loud. There were many other improvements but in a blind test this IMD is what I think I would hear as the give away.
Tenson said:
I'd have to say bollocks to that
I had an unmodded DEQ2496 as DAC. I found it slightly harsh and distorted. I bypassed the output stage with transformers and it was much better. I could really hear the higher IMD before, now I could play it much louder (as sown by the SPL meter on the DEQ!) without it sounding too loud. There were many other improvements but in a blind test this IMD is what I think I would hear as the give away.
You where using the DEQ purly for DAC? or AD->DA conversion?
Since you say "I could play it much louder (as sown by the SPL meter on the DEQ!)", I'm assuming your driving it analog?
You cannot claim your output stage created improvement without runnin it exactly the same conditions as before. (same input level, same output volume). If your output stage changed the output level for the same input (which it sounds like it did) you may be hearing the benefits of an increased analog level at the A->D stage of the DEQ.
If you're feeding the DEQ or DCX with an analog signal you would probably "hear" the most benefit from installing attenuators on the outputs allowing you to drive the DEQ or DCX with the highest possible amplitude on its inputs..
If you're feeding it digitally, what are you using for volume control?
Regards,
--Chris
thylantyr said:You don't need to mod the DCX.
You don't need to use only the digital input.
You don't need output attenuators.
Paranoid people need pills
I'll bet my paycheck that if I conduct an ABX test using
a modded DCX vs. stock DCX, you will fail the test.
If you are still confused then go and buy a DBX product
over Behringer as the price of a modded DCX can probably
exceed the cost of a different brand.
I will take your paycheck.
I think you and I already debated this in some other tread. Regardless, I do not want to change your opinion but just to share what I have experienced. Everyone's system is different and sometimes what makes big difference for one it is not equal for someone else.
In my case I kept for a while two DCX channels unmoded and four channels with mod. I also have very good full range speakers with passive xover as my second set. That allows me to easily test anything that I do. That is how I made easy A/B test where i used DCX just as DA converter and not like crossover. I also have CalLab stand alone vacuum tube DA converter heavily improved that I consider as my reference for DA conversion. Certainly much better than DCX as a DA converter.
With all my listening tests unmoded DCX dolls out the sound and losses clarity. If you are not comparing that sound to something else than you would say, yes it's OK. But in the moment I switch to Cal Lab or to moded channel I have feeling that everything gets more open and clear.
When I compare just moded channel with transformer behind DA converter to Cal Lab DA it is better or let me say different. Very clear and open.
This is also question of taste. I prefer minimal equipment in the chain. Absolutely anything that is in the chain will affect sound. With DA converter and transformer behind and one resistor in relay controlled volume control before amp - it just cannot be shorter than that. And that is what I hear - minimal electronic and clear and natural wide sound. I will say again - there is a difference and it is big in my system.
If we speak about analog vs. digital input than it is not matter of taste anymore it is obvious difference. Why would you go for double DA and AD conversion? That affect sound more than analog circuitry.
Construction wise signal goes through electrolytic caps in the analog input and in analog output. As well in the output, signal is converted from balanced to unbalanced and than again back to balanced all through several opamps. Just that fact explains how much is done to the signal and we have to agree that is not best way how to preserve natural and lifelike sound.
I would love to take your check so give me a call if you are in SF area.
There is a high probability that you are hearing frequency
response anomalies or gain differences {power}. These two
variables alone contribute to ABX test failure. That said, if the
better setup is perceived because of those two issues then
a stock DCX could be programmed to mimick the mod. DCX
has gain control and parametric EQ. I bet I could program
the DCX to sound just like any modded DCX assuming
the mod changed those variables. There is no magic voodoo
inside electronics that gives it 'sound quality'. It's basic
science
response anomalies or gain differences {power}. These two
variables alone contribute to ABX test failure. That said, if the
better setup is perceived because of those two issues then
a stock DCX could be programmed to mimick the mod. DCX
has gain control and parametric EQ. I bet I could program
the DCX to sound just like any modded DCX assuming
the mod changed those variables. There is no magic voodoo
inside electronics that gives it 'sound quality'. It's basic
science
I know you are talking about the DCX but the stuff is very much the same in the DEQ I modded.
I was using it purely as a DAC with a digital input. Without the mod I could not play it as loud - 84dB A weighted. With the mod I was playing the same CD at more like 90dB without it feeling like it was 'too loud'. Yes the output is lower with the mod but I was comparing in-room SPL level.
Before the mod I had tried all sorts of fiddling with the EQ to bring down the 'offending' frequency but this was just not the issue. It was distortion.
Thorsten tells me that the normal analogue output stage uses three op-amps in series which load the DAC output too high an cause low levels of IMD. This is exactly what I could hear. The same thing I hear with valve amps when I drive them loud.
With the output stage bypassed and using attenuating transformers (amorphous core i.e. Django) directly connected to the DAC output via a single wire-wound resistor per phase and a parallel cap the distortion was significantly lowered.
There were other improvements like detail, soundstraging etc.. but the obvious thing was the lowered IMD.
There is no magic to audio it is science.
Of course.... it could have been the heat treated and cryogenically frozen pure silver cable I used for my custom cable!
By the way there are not any obvious frequency response changes with the mod. I have taken measurements and if there is a difference it is less than 2dB over any significant range.
I was using it purely as a DAC with a digital input. Without the mod I could not play it as loud - 84dB A weighted. With the mod I was playing the same CD at more like 90dB without it feeling like it was 'too loud'. Yes the output is lower with the mod but I was comparing in-room SPL level.
Before the mod I had tried all sorts of fiddling with the EQ to bring down the 'offending' frequency but this was just not the issue. It was distortion.
Thorsten tells me that the normal analogue output stage uses three op-amps in series which load the DAC output too high an cause low levels of IMD. This is exactly what I could hear. The same thing I hear with valve amps when I drive them loud.
With the output stage bypassed and using attenuating transformers (amorphous core i.e. Django) directly connected to the DAC output via a single wire-wound resistor per phase and a parallel cap the distortion was significantly lowered.
There were other improvements like detail, soundstraging etc.. but the obvious thing was the lowered IMD.
There is no magic to audio it is science.
Of course.... it could have been the heat treated and cryogenically frozen pure silver cable I used for my custom cable!
By the way there are not any obvious frequency response changes with the mod. I have taken measurements and if there is a difference it is less than 2dB over any significant range.
Thylantir,
I agree with you that if levels are not matched A/B test will fail. Lets turn this into amplifier comparison. A lots of people here made and use Aleph amps. What is it that makes Alephs so good? In my mind is simplicity and minimal amount of gain stages. Same stands for so popular single ended vacuum tube amps - minimum amount of electronics in the way. Where this is most obvious and where does it make difference is in high frequencys and to certain extent in mids. When I replaced other amps that I had with Aleph it was such a difference particularly in that region. That is exactly same difference that I experienced with modding DCX vs. original set up.
Just like you I do not beleive in vodoo and cable myths that are making day and night difference. I strongly believe that properly designed circuit should work well and it does. Opamps output circuits in DCX are working well and there are no problems with them. But there are lots of different designs and they will sound different no doubt. In my (listening) experience
I prefer minimalistyc designs with smallest amount of parts in the way. I am also finding this statement somewhat funny knowing that I am using three way active set up - instead of single speaker driver with no xover and with small single ended amp. I would certainly prefer that over my set up but speakers are different story all together.
Anyway if my set up is not so big I would be tempted to pack my moded DCX and send it to you. There is nothing like hearing and comparing the difference.
Best regards and stay well
AR2
I agree with you that if levels are not matched A/B test will fail. Lets turn this into amplifier comparison. A lots of people here made and use Aleph amps. What is it that makes Alephs so good? In my mind is simplicity and minimal amount of gain stages. Same stands for so popular single ended vacuum tube amps - minimum amount of electronics in the way. Where this is most obvious and where does it make difference is in high frequencys and to certain extent in mids. When I replaced other amps that I had with Aleph it was such a difference particularly in that region. That is exactly same difference that I experienced with modding DCX vs. original set up.
Just like you I do not beleive in vodoo and cable myths that are making day and night difference. I strongly believe that properly designed circuit should work well and it does. Opamps output circuits in DCX are working well and there are no problems with them. But there are lots of different designs and they will sound different no doubt. In my (listening) experience
I prefer minimalistyc designs with smallest amount of parts in the way. I am also finding this statement somewhat funny knowing that I am using three way active set up - instead of single speaker driver with no xover and with small single ended amp. I would certainly prefer that over my set up but speakers are different story all together.
Anyway if my set up is not so big I would be tempted to pack my moded DCX and send it to you. There is nothing like hearing and comparing the difference.
Best regards and stay well
AR2
thylantyr said:
/off topic on
Has been done before. In the 90-ies Stanley Lipshitz conducted a double blind test by either inserting or bypassing a vintage SONY F1 (A/D -> PCM processor D->A) in the equipment. Impossible to reliably identify which was which. This was some kind of a bet with Ivor Tiefenbrun, the head honcho from Linn who always said that digital sucks. He failed the test.
(Later, Lipshitz realised that the noise level was slightly higher with the F1 inserted and that allowed him to score 100% by listening to that).
You'll find the original article on the Boston Audio Society website
/off topic off
Jan Didden
At the risk of sending this thread off-topic, what was the rest of the equipment used in the Tiefenbrun/Lipshitz test? We all know the engineering shortcomings of the F1, so they really ought to have been audible.
Edit: I've attended "shoot-outs" where the sound was such that I couldn't tell the difference between one device and another.
Edit: I've attended "shoot-outs" where the sound was such that I couldn't tell the difference between one device and another.
Indeed, it is nearly always possible to impose 'controls' which give the outcome you wanted.
My main argument against blind tests are that it is very often the case that only after prolonged listening can one start to hear the true character of equipment.
When I first changed from a Philips bit-stream player (mid-end thing) to an Arcam Alpha 9 I couldn't tell any difference! Only after a few days did I start to hear how it was different. When I then went back to the Phillips just to check, I could tell quite easily. I believe that a lot of these things which are supposedly not audible, when added together, and exposed to them over a long period are things which one can hear. But because they are so subtle, are easier described as 'it felt like this or that' than exact descriptions of what was different. i.e. I think it is the short exposure period where blind tests fall down.
Just my 2p worth.
My main argument against blind tests are that it is very often the case that only after prolonged listening can one start to hear the true character of equipment.
When I first changed from a Philips bit-stream player (mid-end thing) to an Arcam Alpha 9 I couldn't tell any difference! Only after a few days did I start to hear how it was different. When I then went back to the Phillips just to check, I could tell quite easily. I believe that a lot of these things which are supposedly not audible, when added together, and exposed to them over a long period are things which one can hear. But because they are so subtle, are easier described as 'it felt like this or that' than exact descriptions of what was different. i.e. I think it is the short exposure period where blind tests fall down.
Just my 2p worth.
EC8010 said:
Linn LP-12 turntable, arm and cartridge, Naim NAC 32 pre- and NAP 250 power amplifiers and Linn Isobarik loudspeakers) and his choice of program material (all LP records).
If you are interested in these things you should read the article at http://www.bostonaudiosociety.org/bas_speaker/abx_testing2.htm which is interesting if nothing else.
Jan Didden
Tenson said:
I don't doubt your perceptive experiences. The fact remains however that with reasonably controlled tests these differences seem to be absent. As you said, you can hear the differences when switching between old and new, and that is genarally what they do with blind tests, although many are set up to let the listener takes as long or as short, with switching between components as he feels like. That gap in perception should bother all of us. It does bother me.
Jan Didden
Why does it bother you that it takes one a long time to perceive the full reality of what is there? You can not analyse every aspect of sound in a split second.
By a long time, I mean a few days, not a few min or so in a blind test. If the test was conducted in that persons own house over a period of 1 week, I think there would be a quite different result.
..maybe not for equipment that measure the same but what I am talking about is different equipment.
I see no reason why I should not be able to tell the difference between three op-amps in series loading the DAC into IMD and a high quality transformer directly driven by and impedance matching the DAC directly to the power amps.
By a long time, I mean a few days, not a few min or so in a blind test. If the test was conducted in that persons own house over a period of 1 week, I think there would be a quite different result.
..maybe not for equipment that measure the same but what I am talking about is different equipment.
I see no reason why I should not be able to tell the difference between three op-amps in series loading the DAC into IMD and a high quality transformer directly driven by and impedance matching the DAC directly to the power amps.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.