PDA

View Full Version : Experiments, SB3 vs Sony CDP-X5000, Coax-SPDIF comparison.



maniac
2007-11-12, 07:01
Well, as I have said in another message that I had only recently acquired my SB3, it is quite interesting for me to see how much of a difference that SB3 will have when I compared to my prized SONY CDP-X5000.

To make things simple, I elected to compare both unit's coaxial output to a KECES DA-131 DAC (albit it is a tad bit in the test mule state.)

Equipments used:

Source:
Sony CDP-X5000
http://img146.imageshack.us/img146/3972/dscn2055no3.jpg

Squeezebox 3 with modified antenna, standard dual antenna, no matter what the hype was, just won't work with multiple rebar/concrete walls in the way.
http://img267.imageshack.us/img267/4396/dscn2052oc2.jpg

digital cable:
1 meter of Canare 75 coax, with eichmann bullet RCA on both end.
Same cable were used to hook up both devices.

DAC:
Test mule version of KECES DA-131, some of the cable used can be seen here.
http://img132.imageshack.us/img132/3934/dscn2053zl2.jpg

Analog cable:
a pair of enamel wire cable, also tipped with eichmann. Ugly, but sounds surprisingly good, can been seen in the DAC photo.

Preamp:
Slightly modified Proton AP400 PRO, using balanced out.

Analog Cable:
a pair of Monster Pro balanced cable, Neutrik XLR connector.

Power amp:
Ashly FET1500M * 2 (running in bridged mode, effectively working as a mono block). Modified to remove the DC blocking capacitor with silver wire. Massive amount of detail increase. At 3mm per cap, this is a steal.
http://img122.imageshack.us/img122/3457/dscn2050rd2.jpg
Yup, it looks old, but the inside were sparking new, quite a surprise when I first got it.

Speaker cable:
Weird tinned copper 12ga cable with CMC "wavy" type of banana connector

Speaker:
Acoustic Energy AE-120SE + SoundCare Superspike.


Well, that's about sums up what I have here.

maniac
2007-11-12, 07:13
First off, I played only ONE song, one of the more familiar song of mine. Somewhere in the vicinity of the heart, by Shenandoah and Alison Krauss.


First round, SB3 coax into DA-131.

The sound sounds relaxing, focused, and generally alright. Easy to listen to and relax with.


Second round, coax removed and installed on X5000.

The sound instantly gained massive amount of ambiance, you don't just hear the singer and band member anymore, you can almost hear the room that they are in. However, it sounded hollow for some reason??? could it be the WIFI interference? Well, I turned SB3 off with the remote, no effect... I went over and physically unplug the factory power supply, and what do you know...


Third round, with SB3's power removed.

The sound not only have ambiance, it also solid. Instead of feeling like a invisible entity singing, now I can almost feel a person is there in a room with a bit of echo. I would kinda classify this as the thickening of the mid range.


I guess that settles the question from people who are wondering why people like me can't leave well enough alone, because it isn't quite as good, plus there looks like a lot of room for enhancements.


Anyways, off to modding SB3... No attack on SB3 intended, after all, a second hand X5000 cost 2~3 times as much as SB3 new. plus I'm just about to start on modding SB3. To start off, some SMA adapter and some big antenna and antenna stand for starters. ;)

ar-t
2007-11-15, 20:31
That gizmo in the 2nd and 3rd pictures, I assume it is the "KECES DA-131", right?

Uh......that is NOT how to make a digital input. 2 pieces of wire, loosely coupled to an RCA jack won't cut it. My guess........return loss 10-12 dB.

Get some coax and a BNC jack. The impedance mismatch from what you have will limit your ability to do an accurate comparison.

Yes, you will hear a difference between the 2 units. However, it could easily lead to erroneous conclusions.

Pat

Deaf Cat
2007-11-18, 04:31
Have you thought of a linear psu for your SB?

Nice to actually see what is in the test you are doing :)

maniac
2007-11-25, 14:59
That gizmo in the 2nd and 3rd pictures, I assume it is the "KECES DA-131", right?

Uh......that is NOT how to make a digital input. 2 pieces of wire, loosely coupled to an RCA jack won't cut it. My guess........return loss 10-12 dB.

Get some coax and a BNC jack. The impedance mismatch from what you have will limit your ability to do an accurate comparison.

Yes, you will hear a difference between the 2 units. However, it could easily lead to erroneous conclusions.

Pat

Well, for the current version, it will stay roughly the same as shown on the photo, except the enclosure is much nicer on the actual one. We did thought about Coax, but that would really be a PITA to actually install it. Most of the better 75 ohm Coax are too thick and/or too stiff, which creates its own problems.

The next version will feature PCB mount connector, which will basically remove the hassle with RCA+wire. As for BNC, that could be considered as an option, but seeing the fact that 95% of the people who use SPDIF does not have BNC, it's not going to be a really useful option for most people.

I personally prefers AES/EBU whenever possible tho, one of the reason I got that X5000 is for the AES/EBU output. ;)

ar-t
2007-11-26, 10:28
We did thought about Coax, but that would really be a PITA to actually install it. Most of the better 75 ohm Coax are too thick and/or too stiff, which creates its own problems.

Nonsense. There are lots of ones to try. RG-59 is not the only type of 75 ohm coax.


The next version will feature PCB mount connector, which will basically remove the hassle with RCA+wire. As for BNC, that could be considered as an option, but seeing the fact that 95% of the people who use SPDIF does not have BNC, it's not going to be a really useful option for most people.

Unless they want it to sound good.


I personally prefers AES/EBU whenever possible tho, one of the reason I got that X5000 is for the AES/EBU output. ;)

Great.......use the one type of cable and connector that is worse than coax with an RCA jack. That makes a lot of sense.

SPDIF is RF. Accept that, or stop trying to improve it. You won't until you wrap your head around that concept.

NewBuyer
2007-12-06, 02:12
...Great.......use the one type of cable and connector that is worse than coax with an RCA jack...

I just gotta ask: What's wrong with AES/EBU for digital audio transmission?

ar-t
2007-12-07, 11:00
Pick up a 110 ohm cable in one hand, and an XLR connector in the other. Notice anything different?

The pin spacing on the XLR connector is much further apart that the spacing of the conductors. When you split the cable open wide enough for the conductors to fit into the XLR, the impedance has just gone to hell.

Actually, an XLR connector does measure to be 110 ohms on a TDR. (Which is really the only way to determine it.) However, there will large amounts of stray reactance because the cable and connector do not have the same physical dimensions, wrt cenrte-centre spacing.

There is no way that you can get a return loss anywhere near -30 dB. Yes, I bet some of my RF buddies could make some kludge to compensate for it. The average DIYer or audio engineer has no such abilities.

Again, this is an RF interface and must be approached and treated as such if one expects to get top drawer performance.

NewBuyer
2007-12-11, 03:21
Pick up a 110 ohm cable in one hand, and an XLR connector in the other. Notice anything different?

The pin spacing on the XLR connector is much further apart that the spacing of the conductors. When you split the cable open wide enough for the conductors to fit into the XLR, the impedance has just gone to hell.

Actually, an XLR connector does measure to be 110 ohms on a TDR. (Which is really the only way to determine it.) However, there will large amounts of stray reactance because the cable and connector do not have the same physical dimensions, wrt cenrte-centre spacing.

There is no way that you can get a return loss anywhere near -30 dB. Yes, I bet some of my RF buddies could make some kludge to compensate for it. The average DIYer or audio engineer has no such abilities.

Again, this is an RF interface and must be approached and treated as such if one expects to get top drawer performance.

Thanks for this info. I wonder why the pro-audio industry would tolerate such issues with the AES/EBU (XLR interface) for digital transmission, is it perhaps a cost concern?

ar-t
2007-12-12, 00:47
None of them are RF engineers. At least what they have is better than the original standard. Yes, I told the AES committee that designed it what was wrong. I suspect several others did as well. Doubt that they really cared what I thought.