Digital active crossover listening tests

[sendler]

Blind

The fact that it wasn't a proper double blind test with statistical inferences to draw a proper valid conclusion. As well, the tester which is yourself shouldn't be part of the test so your memory issue gets thrown out of the window

Why do people always make excuses for not conducting these tests properly ? Don't you think it would add much more credibility to your aim of comparing equipment ?

regards
Trevor

You double blind guys are sooo funny. Even when the results say there was "no difference" you still want to jump in out of nowhere and say that the results of sighted listening tests have no relevance. In this case, there were no results. The testers are reporting that all of the units sounded "the same". You should be quite content this time.

 

[SY]

The testers are reporting that all of the units sounded "the same". You should be quite content this time.

No, the midrange dip ought to have been audible.

 

[Trevor White]

Trevor,

Have you set up and conducted a double blind ABX test yourself? One with instant switching and rigorously set up with a statistically significant sample and careful statistical analysis?

It wasn't our intention to set up such a test. This was an informal comparison where we made some attempt to reduce down uncontrolled variables. However, to do blind ABX and to do it right is a serious undertaking. I'm not making excuses - you are simply insisting on a level of "credibility" with others that was not part of our goal. To put it bluntly, we did it for our own interest, not yours. Later Keith decided to post it elsewhere and I decided to share it abroad as I knew many would be interested.

The problem with what you have done is people reading this and then drawing the wrong conclusions about a piece of equipment.

regards
Trevor

 

[paulspencer]

We had mixed results, ranging from a small difference to none noticed at all. There was no point where there was a large difference, except that it could be explained from the measurements or obvious clipping.

For most of us here, it will serve as confirmation that the affordable options perform quite well and are transparent enough to set up a high end system. The differences we are talking about here are minor and they will not detract in any significant way from sound quality or imaging.

It is unfortunate that we didn't really get to test DEQX vs the others effectively. It wouldn't surprise me if we didn't pick a sound quality difference had we managed to set it up perfectly matched. The real reason to get DEQX is the ability to better correct phase. The other units can only time align at crossover points, but not correct phase shift between crossover points.

 

[paulspencer]

The problem with what you have done is people reading this and then drawing the wrong conclusions about a piece of equipment.

No doubt as a general statement, that's possible.

The conclusion, if I can sum it all up is that differences were very small at most if present at all. I would not say that any of them were sufficient to direct anyone strongly towards one unit over the others. I was talking with Roger afterwards who owns a stock DCX. He was planning to get a MiniDSP as well as DCX, but had no plans to replace DCX. We both felt it was the winner mainly because it is cheaper and more flexible. Murphy seems still happy with his modified DCX even though we didn't find an improvement. Andi went away with his MiniDSP having his suspicion confirmed that it was transparent enough.

If we said MiniDSP blew them all out of the water, and yet it was in reality just a matching issue, then that would be drawing the wrong conclusion. The conclusion that I'm drawing is a pretty safe one. If someone were to do a more rigorous test, the only way I could see a radically different outcome would be if someone with very good listening skills easily picked a difference and called it massive. Otherwise, I suspect any subsequent test is likely to find differences that range from minor to none at all.

So I would have to ask how we could be misleading anyone.

 

[DualTriode]

Hello ,
I own an out of the box DCX. I have not explored the set of equalization tools yet. Are the equalization tools up to equalizing open baffle speakers or is a RTA unit required?

DT
All just for fun!

 

[paulspencer]

DT,

There is enough EQ within DCX that you don't need to get DEQX. I have used both together with a digital link, but I plan to sell my DEQ because I'm finding now that I don't need it.

Trevor,

I'll take it from your response that you have not run a blind test yourself. Perhaps you should consider it.

I have one planned next year. We have a goal which by it's very nature dictates that it must be done blind and to a much more rigorous standard.

 

[Pano]

The real reason to get DEQX is the ability to better correct phase. The other units can only time align at crossover points, but not correct phase shift between crossover points.

It's a little OT, but can you expand on that? The DCX has both delay and phase controls. What does the DEQX do that the DCX does not? I have not used the DEQX, so am curious.

 

[DualTriode]

DT,

There is enough EQ within DCX that you don't need to get DEQX. I have used both together with a digital link, but I plan to sell my DEQ because I'm finding now that I don't need it.

Trevor,

I'll take it from your response that you have not run a blind test yourself. Perhaps you should consider it.

I have one planned next year. We have a goal which by it's very nature dictates that it must be done blind and to a much more rigorous standard.

Hello ,
Thank you for your post.
You invited me to comment.
I view the complex experiential design for meaningful ABX testing of sound systems as getting the tire wrapped around the axel and missing the entire point. Picking a preference of sound systems is closer to wine tasting than Split-Plot ANOVA. Please take the labels off the bottles. If I buy at the end of the event it will be the cheap one that tastes good, not the one that 502 people out of a 1000 preferred.
The rest is marketing.


DT
All just for fun!

 

[DualTriode]

DT,

There is enough EQ within DCX that you don't need to get DEQX. I have used both together with a digital link, but I plan to sell my DEQ because I'm finding now that I don't need it.

Trevor,
<snip>

I am gulity of having too many toys, i have both, DCX and DEQX. Were there ill effects from running both in line? Or just redundant?
A little history, i used the DEQX with a Rane analog active crossover. The DCX looked like it could do both functions. You have done it , i am encouraged to try it.
thanks
DT

 

[MurphyUHF10]

I'm not sure why all the fuss??

I'ts fairly straight forward I believe..

Stock DCX, MiniDSP, modded DCX,, all sound pretty much the same..

My DCX is modded from one end to the other and cost a fair bit to do,, so trust me, if anyone was looking to "split hairs" it was me!! But I had to submit that is sounds just like the other 2 DSP's.. This doesn't worry me as I love the remote function, and I enjoy (electronics) DIY..

Hope this clears things up

Cheers..

 

[sendler]

Direct out mod?

I'm not sure why all the fuss??

I'ts fairly straight forward I believe..

Stock DCX, MiniDSP, modded DCX,, all sound pretty much the same..

My DCX is modded from one end to the other and cost a fair bit to do,, so trust me, if anyone was looking to "split hairs" it was me!! But I had to submit that is sounds just like the other 2 DSP's.. This doesn't worry me as I love the remote function, and I enjoy (electronics) DIY..

Hope this clears things up

Cheers..

Have you ever heard a DCX running it's dac chips direct out? I have found that any of the the V out dac chips I have tried, AK4393,4395,4396, CS43122,4397,4398 sound best by cap coupling one leg or doing a transformer balanced to single ended conversion directly to the output connector. It makes a nice improvement in sonics versus the over complicated stock DCX output scheme (and also in many other more purist opamp schemes such as with the Big Dac Board) and has the added benefit of lowering the output level out of the DCX by about 12db in the case of the caps. The ultrasonic noise of the 4395 or 96 is so low that they sound great and are completely reliable with no analog filter. I have never read of anyone trying their V out dacs direct and then going back to active.

 

[doug20]

The problem with what you have done is people reading this and then drawing the wrong conclusions about a piece of equipment.

regards
Trevor

His tests are controlled. Maybe not to the degree you like, its not black and white. Either do it right or nothing at all. He controlled many of the expectation variables that exist. The conclusions are far more valid then any really subjective tests.

Are you complaining because of the test or because it does not suit your opinion on the products?? His conclusions match what I have done in a simple blind test. Of course it was almost impossible for me to remove all of the "hiss" from when the DCX was used so I kept switching during music (when the hiss isnt heard).

I own the DCX and the MiniDSP. I thought about modding the DCX but Im not a subjective audiophile so I have always thought its a waste of $$$.

I have always wanted the DEQX for its brickwall filters and linear phase correction.

 

[doug20]

It's a little OT, but can you expand on that? The DCX has both delay and phase controls. What does the DEQX do that the DCX does not? I have not used the DEQX, so am curious.

Im definitely interested in that discussion.

The DEQX has been my "I want that" product for many years now. I just can't get a great price on a used one.

From all my reading its really the linear phase correction that it offers. I can not see what else it does over the MiniDSP except 300dB slope filters but the MiniDSP can do 96dB slopes so that is pretty steep already.

I have read that Trinnov does linear phase correction so my next goal is to find a Pre/Pro with that functionality and use it with MiniDSP.

THe DCX is only going to be used in prototyping XOs or testing drivers. Im happy to have removed that device from my signal chain.

 

[MurphyUHF10]

Have you ever heard a DCX running it's dac chips direct out? I have found that any of the the V out dac chips I have tried, AK4393,4395,4396, CS43122,4397,4398 sound best by cap coupling one leg or doing a transformer balanced to single ended conversion directly to the output connector. It makes a nice improvement in sonics versus the over complicated stock DCX output scheme (and also in many other more purist opamp schemes such as with the Big Dac Board) and has the added benefit of lowering the output level out of the DCX by about 12db in the case of the caps. The ultrasonic noise of the 4395 or 96 is so low that they sound great and are completely reliable with no analog filter. I have never read of anyone trying their V out dacs direct and then going back to active.

Hi Sendler,

No I haven't heard the DCX running its output straight off the DAC chip,, but I have heard that it's very good sound!! I was considering transformer coupled straight off the DAC chips (from the US) but it wasn't available in 'kit' version..

I would be interested in doing a direct out mod.. What is "Cap coupling 1 leg"?? Can you offer more information about this direct-out mod? As I would be interested to give it a go

Cheers
M

 

[sendler]

Cap coupling

Hi Sendler,

No I haven't heard the DCX running its output straight off the DAC chip,, but I have heard that it's very good sound!! I was considering transformer coupled straight off the DAC chips (from the US) but it wasn't available in 'kit' version..

I would be interested in doing a direct out mod.. What is "Cap coupling 1 leg"?? Can you offer more information about this direct-out mod? As I would be interested to give it a go

Cheers
M

Just steal the signal for each channel from the big ribbon wires numbers 3 through 14 starting counting with the red wire. The signals are 6-, 6+, 5- ect in that order. If you are sure that the next thing they connect to, such as your amps or volume controls, has a blocking cap on the input to block the 2.5vdc then you can just wire these wires up with nothing else. Use only the + signals for each channel and leave the - unused for RCA connections. If you want to make it safe for all unknown equipment that it may plug into, you need to block the dc with a cap, the best I ahve found for this is the Dayton foil but any top of the line cap from 2-4uf will do, or use both plus and minus signals into a transformer since they have 0vdc between them. The AK4395 and 4396 have such low ultrasonic noise that they don't need any analog ultrasonic filter. The stock 4393 chips can also sound good with no filter even though it's noise is much higher. I prefer to run with no filter to avoid going active and to keep the output impedance to the minimum which is drastically increased with the use of a passive filter. This is old news and goes back 4-5 years now.
.
http://www.diyaudio.com/forums/digital-line-level/15943-behringer-dcx2496-digital-x-over.html
.
http://www.diyaudio.com/forums/digi...5-4396-ultrasonic-noise-spectrum-graphic.html
.
DCX2496 : DCX2496
.

 

[MurphyUHF10]

Just steal the signal for each channel from the big ribbon wires numbers 3 through 14 starting counting with the red wire. The signals are 6-, 6+, 5- ect in that order. If you are sure that the next thing they connect to, such as your amps or volume controls, has a blocking cap on the input to block the 2.5vdc then you can just wire these wires up with nothing else. Use only the + signals for each channel and leave the - unused for RCA connections. If you want to make it safe for all unknown equipment that it may plug into, you need to block the dc with a cap, the best I ahve found for this is the Dayton foil but any top of the line cap from 2-4uf will do, or use both plus and minus signals into a transformer since they have 0vdc between them. The AK4395 and 4396 have such low ultrasonic noise that they don't need any analog ultrasonic filter. The stock 4393 chips can also sound good with no filter even though it's noise is much higher. I prefer to run with no filter to avoid going active and to keep the output impedance to the minimum which is drastically increased with the use of a passive filter. This is old news and goes back 4-5 years now.
.
http://www.diyaudio.com/forums/digital-line-level/15943-behringer-dcx2496-digital-x-over.html
.
http://www.diyaudio.com/forums/digi...5-4396-ultrasonic-noise-spectrum-graphic.html
.
DCX2496 : DCX2496
.

Thanks Sendler!! Sounds easy enough,, just a 2 to 4 uF Cap to block DC...

I have 3x AK4396's, but I haven't put them in yet. It's the only mod I haven't completed yet :-( They're probably a bit too small for me as I haven't got any proper smd re-work tools..

Thanks for the links too.. I have looked in the first thread you mentioned, but it's a bit hard to catch up with all those posts!!

I had heard about the DCX2496 tech club on Yahoo,, so thanks for the prompt.. I might join up!!

Cheers..

 

[fzaad]

This test sound pretty fair but why use normalization. Why not just compare in and out signals. All this test shows is that guys behind those product knows what they are doing (or guessed right) Seriously science behind this is developed about 200 years ago but that doesn't make it any easier to make consumer product.

 

[doug20]

Seriously science behind this is developed about 200 years ago

Really? What was developed in 1810 again?

 

[paulspencer]

fzaad, because a little thing like different levels is enough to throw the result off. The slightly louder version tricks the ear and you think that it sounds slightly better. It's a psychoacoustic trick that we don't detect with a small level difference. It's not quite so bad in an instant switch where you can tell the level changes.