MELCO

[Rufus McDufus]

Actually want to say "Congratulations" and "Thank you" . I truly humbly appreciate the efforts put in by G and R. read all posts. took two glass full of wine .

So its
N1Z, X100 at the top of the line followed by
N1A, N100.

Couple of questions:
1) I wonder if Cached SSD mattered for SQ, then N100 better than N1A because N100 does have 120G SSD which is more than sufficient for audio files?
           - i m sure about from SSD there are other component to snub the noise, to keep power supply as quite as possible for separate sections in N1Z etc. but SSD might contribute too.
2) If there is SSD NAS replaced with HDD then it should also improve for N1A ?

SSD cache is fairly important for the Melco and I think the cache is pretty big on the SSD disks provided with the device. For actual storage N1Z has 2x 512G SSDs which unfortunately isn't enough for my needs as I have just over that. I'm not sure if further space is possible inside with larger capacity disks or more disks. There is external expandability which is pretty funky because it's simply a case of plugging an external disk in and the whole process of expansion (or in fact contraction if there's sufficient space on remaining disks) is invislble to the user.    I'm not aware if they only recommend the same Buffalo SSDs for expansion - that would seem to make sense but I don't know if they're actually available separately.

Putting regular consumer SSDs in the N1A may not necessary result in better sound quality, mainly because their controllers have the life-extending & speed-improving software built-in which is also (apparently) detrimental to sound quality.

[ogs]

TCP/IP optimizations? You can optizime all you want, but that won't give you better audio quality.
TCP/IP transfers packets of data. Nothing else.

And the whole SSD vs HDD is still vague to me.
A player will use a buffer (in memory) to play the audio file.
So whether it's placed on a SSD or HDD or USB stick or SD card won't make a difference.
It's all played from the buffer.

Weust, I am sure you where present at the listening sessions then?
Many years ago I said to myself "no chance there is any difference on USB cables. It's only packet data" Today we know better..

[Rufus McDufus]

TCP/IP optimizations? You can optizime all you want, but that won't give you better audio quality.
TCP/IP transfers packets of data. Nothing else.

And the whole SSD vs HDD is still vague to me.
A player will use a buffer (in memory) to play the audio file.
So whether it's placed on a SSD or HDD or USB stick or SD card won't make a difference.
It's all played from the buffer.

Weust, I am sure you where present at the listening sessions then?
Many years ago I said to myself "no chance there is any difference on USB cables. It's only packet data" Today we know better..

You can certainly lower the load requirement on the target player by reducing the computation necessary to reassemble the TCP packets.
[edit] Yes it's questionable whether it has a noticeable effect, but the end result of all these optimsations sound pretty good to me.

[weust]

When a proper NIC is used, the controller handles that, not the CPU.

And no, I will not waste time and money on that.
And by that this forum from now on as well.

[Womaz]

Great post, very informative. As a new 200 owner I find this type of post invaluable.

[Kunter]

Thanks for the excellent write up guys.

I understand you did not have time to try the MELCO's as NAS replacements to the TotalDAC server. Any plans to do so? With their attention to detail, I think it must provide some improvements and may be put the TotalDAC sound above the N1Z with USB to Devialet. Although the total price would be ***..

Cheers

We did plan to do this but the major issue was with the Totaldac. It's rather inflexible regarding device naming. There were a few other issues too and it was all taking rather a long time and eating into listening time. It would have been interesting to hear though.  [edit] - oh yes, we had to write a new OS image for the Totaldac with it enabled as a UPnP renderer and we encountered a few problems.
N1A/Totaldac vs N1Z could be interesting too as the price is about the same.

How about Melco - > Totaldac USB Input - > Devialet AES ?

[Jwg1749]

HEALTH WARNING: this post is slightly philosophical and a bit off topic.

These questions about the optimization of digital transports are interesting.

No doubt clever IT people can argue meaningfully about whether e.g. CPU drain could have any impact on data transfer and so on. I don’t know enough about it to have a view.

What makes me sceptical about it isn’t the IT side, it’s the human side. One would surely want to weigh up the (well attested) inability of the human auditory system to make fine discriminations against the (well attested) power of the human mind to imagine things. It’s this mismatch between perceptual insensitivity and mental sensitivity that makes me sceptical.

It doesn't make me a pessimist though. People sometimes say that expectation bias implies people are stupid. I look at it the other way: the fact that we can experience expectation bias shows how subtle and complex our minds are.

Sorry for going off topic.

Matt

[Charlemagne]

It's pretty clear to me. Having NO SPEAKERS is putting you in a philosophical mindset. ;-)

[johnsap]

Does MELCO give less details in comparison to AIR ?
I found Moon to be less detailed than AIR ?
Is it the same with MELCO ?

[Confused]

There is no doubt that expectation bias exists, many truly scientific experiments have demonstrated the phenomena. However, there have been many times I have auditioned audio kit and the final conclusions have been very different to my initial expectation. Possibly this indicates that I am immune to expectation bias, or possibly it indicates I am slightly insane.