V-Link, Fireface, Dip Upsampler I don't hear a difference

[Sjef]

O.K. not a DIY topic but anayways.

The last couple of days I have spent a lot of tme comparing different playback options from my PC into my DEQX crossover/preamp.

I was playing for years with a RME fireface wich I bought not only because of it's firewire to spdif conversion but it also is a very good soundcard for use with measurments software, I'm a DIY guy after all.

Having read all the rave about async USB and specially about the Musical Fidelity V-Link I just bought one. Set it up in my system and guess what, I can't hear the slightest difference between the V-link and the Fireface, none, zip, nothing. Hmm, interesting. The fireface was using ASIO playback, the V-link standard windows XP playback wich according to everyone is much worse.

So I got curious. Is it just me, is it the deqx wich doesn't give a **** what it is fed with or what.

After a couple of hours of switching back and forth I pulled everything I had for computer playback out of the closet and started comparing stuff.

I have tried several different firewire cables, several different USB cables from standard printer cable to a quite expensive AudioQuest type. I have tried a DIP upsampler in between everything, Tried several spdif cables and aes/ebu cables from dirt cheap to way too expensive. Tried several software players with direct sound, kernel streaming, ASIO (don't have wasapi becuase off winXP) And at the end, Nothing and I mean nothing made any difference. Well actually there was one thing I tried wich made a slight difference and that was the difference between playback from my laptop or playback from my desktop computer.

What does this tell me. Again, it might be just the DEQX. I don't have another DAC around to compare and besides it's not easy to compare beacuse the DEQX is my crossover as well. Maybe I'm just totally deaf or maybe everyone else is super sensitive. For me it's clear, I'm not going to spend any time anymore trying to improve my PC audio playback. Changing a tweeter for instance makes a bazillion times more of a difference to the whole system.

the system I'm using

Shuttle XPC with external PSU
Intel E8200 processor, bla bla bla
RME Fireface, Musical Fidelity V-link, DIY PCM2706 converter, Nuforce uDAC (did not try those two yet) Monarchy Audio DIP Upsampler
DEQX PDC 2.6P with HDP3 upgrade (basicaly a HDP3 in the old PDC2.6p case)
bass: W-baffle with two JBL 2205H powered by UCD700
Mid: PHY-HP HB21-LB15 on open baffle powered by a Patek gainclone
high: Veravox 5s as dipole tweeter and ESS Air motion transformer powered by a Hiraga 20W classe A
and off coarse cabling mains conditioning etc etc

 

[Sjef]

Just measured the Fireface and the V-link spdif output with my scope. The V-link looks very very good. Has some rounded edge but no overshoot and very low noise. The Fireface looks worse but has a faster risetime but with overshoot so don't know which one is better. depends on the DAC used I guess.

Would like to try the difference into a typical NOS DAC because they are more sensitive to the input signal. From what I see from the measurements I guess the V-link could be the better one then. Anyways, my V-link is going back to the store.


I'm keeping the Fireface for a couple of reasond. first it has cost me a lot off money and It won't pay itsellf back. Second, the playback is unconditionally stable. Nomatter how heavily I use my PC it just plays on without a glitch, the V-link start to stutter then. Third, it's part of my measurment system.

 

[qusp]

yeah i'd say its the deqx as it resamples everything to 24/96 anyway so i'm not sure what you hoped to achieve by feeding it a cleaner bitperfect spdif input as it will be torn to pieces and put back together again using the local clock and digital filter and then torn apart again for multiple channels and EQ. same as your different playback softwares not quite sure what they could do feeding the deqx. there of course will be some some amount of the jitter residual passed along and added to by the deqx, but i would doubt it would be enough to be audible between units. you would have to use the external clock input on the deqx and perhaps run the same clock from the gear you were testing, that would probably mess with the xo and filters though.

but the fireface is a superb rock solid piece of hardware, the drivers on the RME gear are simply the best i have ever encountered.

 

[qusp]

if you tapped the i2s on the v-link and injected it after the crystal semi receiver you would stand a better chance of hearing the difference.

on that note...you even have an upsampler externally too and maybe in some of the tests up sampling in the software, so you went USB->async reclock->spdif->reclock and upsample (what rate?)->spdif->receiver and reclock possibly downsample to 24/96 but at least resample-> DAC->multichannel filter-> amp-> speakers........

thats a shmozzle mate, truly. how to hear jitter and clock edge differences from the very beginning of that seems an impossible task and i would have been skeptical if you HAD heard any differences between units

i'm not sure how the XO filters work in the DEQX, so there could be another set of resampling in there

 

[Sjef]

Hi Qusp,

I don't know how the DEQX handles it's input signal. I have asked the DEQX team a couple of times but they won't tell. The only thing they did tell is that it does not resample anything, not sure about any relocking though.

If you feed it with 44.1 kHz all the filters are handled in 44.1 kHz. It does not upsample to 96kHz.

I have tried the Monarhy Audio DIP upsampler but I don't hear a difference but that's because it upsamples to 96kHz with empty data. I do hear a difference with 24/96 recordings though. Upsampling does not increase resolution. Normally I only use the DIP to convert the toslink from my TV to AES/EBU so I can feed it to the DEQX. It's not being used in between the DEQX and the Fireface normally, just did that to compare.

The reason I was doing this whole experiment was because Steve Nugent from Emperical Audio told me that the DEQX was very sensitive to jitter at it's input and I would defenitly need one of his reclocker and expensive cables to cure the problem. Could be, but if this is true I would at least hear some difference between my "cheap" solutions. Someone else told me the other day that I needed a spdif cable wih cost at least 1/3 of the price of my DAC to hear a difference, what a *******

I guess a better clock inside the DEQX would do more, but that's only a guess since I don't know how it handles it's input signal. Would like to know though.

And yes, you are right about the Fireface. Rock solid, never let me down. I use it as a DAC preamp as well when I'm fooling around with passive loudspeakers.

An yes again, the DEQX messes up the bits pretty good, so speaking of bit perfect playback seems silly but if the source ain't right then the rest never will be. Funny thing is that I did hear a slight difference between my destop and my laptop. That might have to do with system grounding I guess.
I also hear a slightly bigger difference between either computer solution and the spdif output from my CD player. Not that any of the two is better then the other, just different.