Meridian Unplugged - Powered by MrTechGuy.co.uk

I've done a bit of listening, but unfortunately I didn't end up comparing what I meant to! However, the result is still valid and very relevant.

For some stoopid reason, I thought I was comparing:

a) Transporter 44.1k/raw --coax--> Radiance --HDMI--> HD621 --SmartLink-->G68
b) Transporter 88.2k/sox#3 --coax--> G68

This is, of course, nonsense as the Transporter was outputting the same simultaneously to both paths, but stick with me...

Bear in mind that I *thought* that I was comparing sox#3 vs. the long path including Meridian apodizing though the HD621. I was pre-disposed to *want* the sox upsampling to sound better.

I used the introductions to a number of songs as it was easy to simultaneously change source and rewind to the beginning of each song. Material was:

- Clive Carroll - The Red Guitar - March and the Messenger (acoustic guitar)
- Duffy - Rockferry - Warwick Avenue
- Eagles - Hotel California - Hotel California
- Gerry O'Connor - No Place Like Home - Bag of Spuds / Copper Plate (Reels) (banjo)
- Paul McCartney - Memory Almost Full - Dance Tonight (vocals + mandolin)

With the caveat that I hate A-B'ing ('cos it does my head in), I felt that the Transporter direct to the G68 was superior with every one of the above tracks. Specifically:

- Voices seemed clearer and less edgy.
- Acoustic instruments seemed "tighter".
- There was better separation of instruments.
- Bass seemed to have more impact.
- I was drawn to secondary melodies that I'd not noticed before (ever).

There were definitely differences, and I kept asking myself: Is the sox#3 apodizing really better than the HD621 path or am I just hoping that it is. In the end, I was convinced enough to come and write this post.


And then it hit me.


The Transporter was outputting sox#3 upsampled audio to both paths. I was actually comparing:

a) Transporter 88.2k/sox#3 --coax--> Radiance --HDMI--> HD621 --SmartLink-->G68
b) Transporter 88.2k/sox#3 --coax--> G68

I think this makes the result all the more interesting and valuable because the only difference between the two paths is our good old friend jitter.


My observations:

1) The Transporter is almost certainly the lowest jitter source I have ever had in my system.
2) HDMI is known to be awful from a jitter perspective.
3) This was the first time I'd listened to the Transporter direct into my G68
4) The improvements I describe are virtually a textbook definition of the effects of jitter on audio reproduction.
5) Given I didn't think I was comparing jitter in sources, I'm even more inclined to believe my results!


My conclusions:

1) This is the most obvious demonstration of the effects of jitter that I have yet heard.
2) There must be a significant difference in jitter between the two audio paths for it to be so readily apparent.
3) HDMI is simply bad. The HD621, or more correctly the Radiance + HD621, should not be used as an apodizer for CD or server-based audio.
4) The Transporter is damn good!

I guess that's not entirely unexpected, but it is a bit of a bombshell. Incidentally, in absolute terms, sox#3 direct to the G68 sounded quite superb.

Which leads me, rather embarrassingly, to any claims that I may have made regarding the efficiacy of the HD621 in my system. I'm pretty sure I heard some improvements on a number of tracks in terms of spaciousness. I was using my Oppo BDP-83 at the time and I didn't A-B. The improvements could have been some other change in my system that I'd made since I played the tracks in question. For the time being, I'm going to eat humble pie and retract any claim I may have made previously about the HD621's apodizing sounding superior to non-apodizing.

To be fair it would be cool if your head could withstand doing your first intended test.

Also someone asked about using Sox "offline" so you could build a few tracks and easily A/B. Have your figured that out in you process?

Be nice to analyze using Audacity as well, which seems like what you were doing there a few posts back?

Again, great work and appreciate someones honesty saying they changed their opinion on something. I've certainly done that a few times myself. Perhaps the Oppo was helped by the HD621 and you assessment may have still been correct. One thing for sure, it sounds like your happy with the Transporter.

By the way (not wanting to take this off topic) I think if you are A/B ing and do hear a difference (ideally blindly) it's probably valid. If you hear no difference then you don't know if it's the "learning effect" obscuring the difference.

So it's not clear yet if your happy with the Sox stuff or not

Looking forward to more reports.

Hi Joel
Interesting work for sure.

Two observations that I'm sure you've considered

1) That Jitter still seems to be an important factor, even though the both Data Streams were in the High Speed 88k format. It obviously becomes even more important for a 44k source. Which leads me to.............

2) Your previous test of the effectiveness of the HD621 using the Oppo as a Source, all related to a 44k data stream??? It's possible that your previous observations may still hold as the COMBINED Jitter reduction of the HD621 and then G68 may be more effective on that stream

3) As a follow on to your HD621 comments, when it's receiving an 88k data stream, it is obviously doing nothing of value, except for Jitter Reduction. If it was a 44k Stream from the Low jitter Transport (or even more importantly, a likely higher jitter Oppo / SB3 / Sonos / Apple TV, then the HD621 has the opportunity to improve the Data Stream by BOTH Jitter Reduction and APODIZING??? So again, your previous observations may still hold true????


EDIT: Simplifying. The reason for putting a Music Source thru an HD621 to begin with is to Apodize. It obviously wasn't doing that in the above test with the 88k stream. Likewise, the HD621 may also add "more value" with high jitter sources typically tested thru a HD621

The sox stuff is clearly good enough for me to determine when it has been degraded by jitter. However, future comparisons of upsampling algorithms must be made with the Transporter connected directly to the G68; that much is obvious.

One thing I noticed is that Duffy's vocals on Warwick Avenue were particularly clean and lacking in stridency - really nice, in fact - but there are two factors in play: sox upsampling and direct connection of the Transporter to the G68.

Agreed. I'm reasonably convinced that the jitter of the Transporter is *really* low. Which is really cool.

Joel
So you think it's POSSIBLE that your previous observations of the HD621 benefits may still hold true???

I'm pretty sure sure you understand where I'm coming from, but I think it's best if the conclusions of THIS test doesn't become "Apodizing in the HD621 doesn't really work or isn't of benefit", to the many others here who will take your comments above as Read.

The HD621 didn't Apodize in your test above..........Which is obviously the Only reason for having it in the chain at all.........

It's possible, but as I said I think it's best to retract any claim I may have made. The true answer could fall either way.

Edit: I agree with your edit above!

George, it was me that asked about how to make a FLAC file that has already been through SOX. I don't know how to do that but it would be the best way to test both ways quickly.

I've always had concerns about the need to faff around with extra boxes to get apodizing into my system hence why I have not yet gone with a HD621. 99.9% of my music is currently delivered via SPDIF and I don't want to go through 2 more boxes to get to the G68....it can't be a good thing.

Of course it could fall either way and I'm not trying to force your opinion here.

Your first test (months back) used Apodizing in the HD621 and you obseved a difference you liked. Your test today did Not use Apodizing in the HD621, so it's essentially an Apple and Oranges comparison and IMHO, withdrawing your previous observations is not valid as they MIGHT still hold.

I'm already on record here as saying that the HD621 did not significantly improve a 44k LPCM Stream from an 800 player (which would have the equivalent, or better, low jitter characteristics of the Transporter)

Likewise, an 800 producing an MHR 88k stream sounded significantly better when DIRECT into an 861v4 via SPDIF, than when trying to force an 88k PCM Stream thru a HD621

BUT, the HD621 did improve things when playing a Sonos and Apple Mac thru HDMI.........

As ever, Source Jitter matters............even at 88k..........

Simplifying the above......

** The HD621 did NOT help an 800 Player (with Upsampling On or Off)

** The HD621 DID help the Sonos, SB3 and Apple Mac players that I put thru it

I.e. A complete Performance Statement about the HD621 is only valid when stating what Source and Data Stream it was tested with

Keep in mind a few things.

VK was observing from memory and didn't do A/B when he fist got the HD621.

If i'm hearing correctly, from Memory, VK is currently happiest with Transporter & Sox direct to G68 over anything.

The fact that HD621 makes it worse AT ALL (which I think is VK's main point here and a bit of a shocker) means the HD621 has a long hill to climb even if he has HD621 do the upsampling/apodizing compared to Sox.

Hi George
I FULLY understand your points and also Joel's.........

So I'm really writing this more for others reading than for ye......

All I'm saying is that Statements and Tests needs to compare Apples to Apples

The HD621 didn't Apodize in Joel's most recent test............His Source was 88k...........Most people's Sources here are 44k capable only............Likewise, his Transporter is likely much lower Jitter than most people's Sources.........Therefore, people shouldn't use Joel's comments above to decide on the potential benefits of an HD621...........As ever, only a Home Loaner will help decide it for THEIR system

Also worth emphasising that it's NOT the HD621 which is making things "Worse".............It's the getting the Data Stream into HDMI (and transmitting over HDMI) that is the most likely cause

An Additional important point worth bearing in mind.........Joel's system of Transporter to Lumagen to HD621 introduces another, possibly higher Jitter Input and Output Interfaces on the Lumagen.

This MAY make things worse than if someone were playing DIRECT from an Oppo HDMI or Apple HDMI.........

Summary: Apples to Apples and Oranges to Oranges..........Extrapolating can cause misunderstandingd