Well an interesting night of comparisons at home. I have had a number of things I wanted to test in my home system and three nights ago I finally had some time to do so. The things I tested are a bit wide-ranging so I thought I’d just make a new thread to put them in.

The listening was mostly to nail down some possibilities with my Harmoni racks and to evaluate some digital streaming items. First off, I was loaned a pair of Nordost Heimdall 2 Ethernet cables. These were both 3 meter cables with directional arrows and they sell for $999.99US each. So after warming up by listening to my LP12 on the top of the different Harmoni racks (more about this later) I put my KRDS/0 on a Mimer and swapped the Heimdall Enet cable from the switch to the KRDS compared with my standard Microconnect 3 meter cable. While I will say that the Heimdall wasn’t bad, being certainly more musical than the AudioQuest Ethernet cables I tested some time back, it was definitely not as musical as the Microconnect. I also tried two of the Heimdalls replacing the cables from the LS-NAS to the switch and from the switch to the KRDS and confirmed that two wrongs don’t make a right – it was even worse than with one swapped out. Just for fun I invited Debbie to give a listen to the comparison of just changing the switch to KRDS cable, which she did blind. She told me she liked cable two (the Microconnect) better, adding that when the violin came in on the Loreena McKennitt piece we were listening to she felt the second cable was quite a bit better. She admitted that the Heimdalls looked pretty but didn’t like what they did to the music. I found that they removed the joy in the piece, which was such as to make you want to get up and dance with the better cables. It was an achingly beautiful piece on the Microconnet cables, just another track on the Heimdalls. The differences in the flow, the perceived quality of playing by the musicians, and the note structure and harmonics were all negatively changed by the lesser cable.

Next up was an interesting new device we picked up called the Melco N1A digital music library. This is an NAS that is designed specifically for music streaming and has had a lot of attention paid to getting the best sound to your streamer or DAC. It also has a number of interesting features designed to make it more of a piece of Hi-Fi than a typical IT item. It has received some positive comments here and there, particularly in the UK, and since one of our distributors carries it I was able to get one in. There were two things in particular that I wanted to test: the claimed quality of music stored on the unit itself and also the claim that it will improve the sound of music stored on other devices when sent through it to the DS unit. The explanation for this is that there are some devices inside the unit that isolate all 8 lines on the Ethernet connection between the RJ45 “LAN” socket and the RJ45 “Player” socket. This isolation is supposed to remove noise from the network and improve the signal sent to the DS. So first I did A/B comparisons between music from my LS-NAS through the normal GS108Tv2 switch to the KRDS and with the Melco N1A inserted between the switch and DS using an additional Microconnect cable. Overall I was surprised at the result but not quite in the way Melco would hope I would be. I did not find the N1A being inserted to be a musical improvement, however, I was surprised how little difference it made. Honestly I would normally have expected putting another component and an additional cable in the circuit would have substantially worsened the sound unless it worked as described as a kind of filter and made it noticeably better. What it actually did was make it sound mostly the same but just a touch less tuneful. It was a hair easier to follow notes with the direct connection and the texture of the instruments came through a bit more clearly.

My second set of comparisons was with the same music off the N1A and the LS-NAS. Again I was surprised at how close these two were, especially as I ripped the CD directly to the LS-NAS whereas the same music was copied over to the N1A from the store ReadyNAS and who knows how it got on there in the first place. Now the difference here was bigger than when just running through the Melco but it was still relatively small. Same kind of differences – the music didn’t grab you quite as much, everything just a bit less musical. By the way, while doing this I also compared using Kazoo server vs. Songbox. Despite the fact that they are reported to be the same software with just a different name, I found the Songbox to be just a touch more tuneful. Pretty small difference but a difference nonetheless. Considering the work that went into the LS-NAS with the choice of all the components, orientation of the MB, picking the most musical internal SATA ports, torquing the unit, etc., I think it is fair to say that the N1A comes pretty close. For people who don’t want to go through the work of assembling an LS-NAS of their own I have no hesitation in recommending the Melco N1A as a reasonable choice for a server.

In addition I should mention that the N1A has several other attractive attributes that might make it a good choice for a number of people. First off it is really quite easy to use as it pretty much sets itself up when you turn it on (although you should have it check for updates). Along with that you can just turn it off from the front panel power switch whenever you want and turn it back on and it just works! This is not recommended procedure with most NAS units but the Melco was designed for it. It does seem to be pretty bulletproof overall as I have turned it off and back on several times and it just comes up and works. If our store ReadyNAS loses power and comes back up it re-indexes the whole library causing all playlists stored on control devices to be useless. Not on the Melco – it just continues to work fine (although my LS-NAS also doesn’t have this re-indexing problem). Another interesting aspect of this unit is that it has three USB ports on the back one of which is intended to feed a DAC. If you hook a DAC to it the N1A can act as a network interface turning your DAC into a UPnP renderer. It can also download high-res music directly although currently only from a couple of sites. The other two USB ports are for backup and expansion (beyond the 2-2TB HDDs installed). I have to say for $1999US the Melco N1A seems a pretty reasonable no muss/no fuss storage solution with pretty good flexibility. It would be really nice if they updated it so you could just plug a USB CD/DVD ROM drive and rip directly to it but I expect that would require a fair bit of programming.

Enough of the digital stuff (although I must say that a fully torqued KRDS/0 on a Mimer shelf can reproduce some pretty beautiful music). On to the Harmoni racks. I had a couple of questions I wanted to answer for myself about the Harmoni racks and the Mimers specifically and I was able to answer them. I started out with two pretty close to identical racks. Both had four shelves with a reference bottom with the best damping I have found so far (a specific size foam block cut from the foam in Linn Trampolin boxes) and with two Tors above that. Both had a Mimer for the top shelf but one started as a Mimer K and the other a Mimer 3 but with the same damping as the Mimer K. All the wooden shelves were Sitka Spruce – the standard wood I am using in the US. So I start by moving the LP12 between the Mimer 3 and the Mimer K and, unsurprisingly, the Mimer K is more musical, easier to follow notes, better flow, a readily noticeable improvement. So I take the 8mm spruce shelf from under the Mimer 3 and substitute a 6mm one as normally used under a Mimer K. Interesting! Now the two Mimers are very close – close enough that I have to listen to a couple of things to make sure which I felt was better, as did Debbie. It did turn out that the Mimer K is a bit more tuneful but not a lot. What this menas to me is that if you’re buying new you want to get the Mimer K for your sources (and maybe others depending on your budget) but if you already own Mimer 3 you really should upgrade the lower shelf to the 6mm spruce – it will take you most of the way to Mimer K at a pretty reasonable price. Just to verify that the difference was the Mimer K vs. the Mimer 3 and not some other difference in the two rracks, I then swapped the Mimer 3 metal shelf for another Mimer K. This brought the tops of both racks up to pretty much the same level. There was still a slight difference but that could be accounted for by difference in rack position or the change in cable dressing moving the LP12 from one rack to the other. But the differences were quite small and certainly enough to demonstrate the superiority of Mimer K over Mimer 3 with the thinner spruce shelf. Finally I have to say don’t underestimate the power of the Mimer! Just for the fun of it when I was doing the digital testing I tried the KRDS/0 on the second from bottom Tor shelf vs. on top of the Melco N1A (turned off) on the Mimer K. (Note that the Melco N1A is about 15.5 lbs so a fair bit heavier than a Tundra 2.) No question, the KRDS on the Melco on the Mimer K was easily significantly more musical than the KRDS was directly on the Tor two levels down. Certainly by a bigger amount than the LS-NAS vs. the Melco!

So my recommendations are get a Mimer K under your source components (in the past I have tested it under Sagatun Mono 1.1s and it improved them too, but not to the level it did to the sources). If you already have Mimer 3 I wouldn’t worry about replacing the metalwork but do get yourself 6mm spruce shelves for them and the right damping. Also, if you have Mimer shelves make sure you drill out the holes to 10mm and keep the posts centered in the holes when tightening things down so the metal post doesn’t directly touch the Mimer shelf. This is crucial to getting the Mimer to perform properly as without it a Tor will be more musical. I should also mention that the precise amount of damping on a Reference K bottom makes a definite difference in the musicality of the rack so if you can do a Reference K bottom indeed do so and get the damping right.

So an interesting evening of experiments that let me know that I don't have to mess with my digital side (except to get a Mimer under the KRDS/0). Plus some interesting findings on the Harmoni racks that clears up some of my questions - although there are always more questions to answer.

Wow.

That is one post crammed full of useful information!

Thank you Thomas!!

Great info ThomasOK!

How would you say the Melco compares to the best of the off-the-shelf NASes?

Interesting! Thanks for sharing!

I have a QNAP hs210 on the recommendation of Cymbiosis. It's purpose built to look Hifi and for music/media streaming. It sounds very good and all for £120 from ebuyer.

Hi Tom, thanks for posting your findings on both the racks & the Melco N1A

How close do you think the Melco and the LS-NAS would be if a track was ripped to both in the same way, i.e. straight from preferred Samsung SH-224BB DVD drive to the NAS storage, even if that's over the network rather than via a USB enclosure?

I still have Intel 320 SSDs, Kingston RAM, Seasonic power supply, and elusive DeltaCo SATA cables to track down if I were to build an LS-NAS, then it would need to go on a separate rack... Alternatively, I could buy an N1A, put it on a shelf with the rest of the system, and treat it like a piece of hifi equipment. The latter approach has its attraction, but only if the results would be close to an optimised LS-NAS, or ahead of an LS-NAS without all of the best components.

ThomasOK,

Your review of the Melco has really raised my interest. Other reviews (linked from the Melco website) seem to concur with your findings too. There's only one concern for me, and that's Martin Colloms' comments about the Melco producing the best "sounds" that he's heard from a streamer, but he felt that the Melco made the music loose its mojo by upsetting the timing and slowing things down a bit. MC reviewed the expensive Melco and his comments are very typical of expensive "HiFi".

Have you heard the difference, or have any knowledge about how the two compare from a TuneDem perspective?

One more question re Melco N1A

Does it have to be situated close to the KDS - i have been told by the local linn dealer that it needs to be linked via a short cable to the KDS.

My network switch is upstairs and routed via long cat 5 to the room downstairs with the KDS and hi Fi.

I would prefer to put the N1a upstairs next to the switch / router etc as the cabinet is full of Monos and the KDS

Hello all! With the busyness leading up to the Linn musical evening we had here on the twelfth, followed by my nasty cold that came on the day after, it has taken me a while to get back here and try to answer your questions.

Spannko, I really have little experience of off the shelf NAS units and have had almost no opportunity for direct comparisons. I can tell you that the Melco at work sounds a little bit more tuneful to me than the ReadyNAS/NV+ with Seagate drives that we have had as our main server for a few years now. On the other hand, TJ and Craig listened to the Melco vs. the ReadyNAS and didn't feel there was any significant difference. However, I find the store to be relatively useless for comparing NAS devices. The main reason is that the store is wired with a mile or so of some blue basic CAT6 cabling. This runs from the outlets in all the rooms to a patch panel and then to a pair of switches in the store room interconnected with a bunch of black generic CAT6 cables of various lengths. The switches are pro NetGear 24 port units (I did have some influence on that) but I have no idea about the quality of the cabling and certainly direction was not paid attention to. The morning after we had our musical evening I requested TJ give a listen to the Nordost Ethernet cables as we have to get them back to Nordost. We compared them to some really basic, unshielded CAT5 cables going from the AEDSM to the 530s and the Nordost were noticeably better to both he and I. Considering how much better the Microconnect cables were than the Nordost, the Nordost winning this comparison shows how bad the CAT5 wiring we tend to use to connect up things, including the input to the AEDSM and the connections to the Melco, are. Hence having to take things home to do any really valid comparisons of network devices.

tokenbrit, this is very hard to say as I haven't been able to make that kind of comparison. I did find out that the music on the Melco was downloaded from TJ's notebook which is his main source for music at home. However, they were transferred from another HD so although they were both ALAC files I don't know what generation the Melco ones are. Over time I might get a chance to haul the Melco home and try more experiments but I wouldn't hold my breath waiting for them if I were you. I have more rack experiments and a bunch of interesting tests involving speakers coming up so the Melco is a ways back in line.

Spannko again, I have not heard the (much) more expensive Melco unit - the N1Z. Currently the US distributor has not decided whether they will offer that unit in the US. Considering it is close to four times the price and has less storage, I can understand that it might be a difficult sell. I don't know at this point if they have even listened to it. The specs for it read like a lot of audiophile gear: dual power supplies for different sections, special "audiophile" grade caps, special "audio grade" SSD drives, etc. so it could well be that it is a more audiophile sound than a musical one. If I ever get a chance to listen to one I will report but at this point I don't know that the distributor even has one in the country.

u252agz, I hadn't heard this but it wouldn't surprise me as we have found the 3m Microconnect cables to be the most musical with increases or decreases in length reducing musicality somewhat (although 5m and 10m are still not that bad). I only tried it with the 3m lengths I normally use with the switch I have near my Hi-Fi, which has the LS-NAS and DS connected to it and a long cable back to my router, WAP and main computer. So I can't say for sure what effect long Ethernet cables would have on it.

Thanks Thomas, comparing lots of different NAS's isn't easy, but if the Melco's more tuneful than the NV+, it's got to be worth an audition.

Someone told me(I think sound organisation in York) that the Melco N1a is being modified with Software so as to allow direct ripping from an attached usb cd drive.

Will make it much more attractive to technophobes like me.

u252agz wrote:Someone told me(I think sound organisation in York) that the Melco N1a is being modified with Software so as to allow direct ripping from an attached usb cd drive.

Will make it much more attractive to technophobes like me.

That makes it an even better proposition, provided the ripper produces a musical file.

Hi u252agz, or may I call you "u"?

I have owned a Melco for a year, I think, now, and put one of the first threads anywhere on this forum about it.
Forgetting the debate about whether digital processing affects audio sound in any way (It's certainly a bit of a lost cause on the Linn forum), it is a useful piece of kit, and makes life simple.
I also heard a while ago that they were going to produce, essentially, a plug-in CD-ripping drive. Which would be great, because you could do this in your listening room.
My hifi stack is now very simple - a Melco N1A, An ADS, and a Tundra 2. All very neat and self contained on an isoblue stack.

Those of us who are a bit fussy about our music libraries, and particularly consistency of useful metadata, often - in fact nearly always - use something like mp3Tag to change the meta data to conform with existing stock. Just take for example - as a genre, can you define, Rock, Hard Rock, Rock/Pop Pop/Rock and Pop? I call all this Pop. Is Bob Dylan Rock or Folk?
Which means making changes nearly every time I add an album. Actually it is part of the fun. Just like cataloguing your LPs. Also before your digital library is any size at all you have to overcome stuff like "The Temperance Seven", "The Temperence Seven", "Temperance Seven", and "Temperance Seven, The".

The point of this post is to simply say, yes direct ripping, not least to able to play a friend's CD when they visit, would be useful, the Melco development may be of limited value. It is not hard to set up a PC or laptop somewhere in the house to be able to access the stuff on your Melco just as an external drive. And using something like MP3tag is vital and fun, as I have suggested.

You also need the use of an external computer for backing up your audio collection to another hard disk drive(s). You cannot put all your eggs into the one Melco basket, in case of failure or theft.

If you are local to York and want to pm me we could arrange a further chat. Or I am sure one of the guys at Sound org would help you set up your network.

PS the Melco will work with Minimserver - you would have to get on to its creator, Simon, to discuss using it. That means you can convert FLACs to WAVs before the file reaches your streamer. Some folk say they sound better - I do not, on my system at least. Also you can convert 16Bit files to 24Bit on the fly, also allegedly improving the sound. (Not a true conversion of course - cannot get more detail than you have already in a file, but by adding eight zeros to the bottom allegedly makes digestion by your DAC easier - to be clumsily non-technical).

Sorry to have rambled......

Cheers
Donuk sunny downtown York

Tom, getting back to the racking. Why not put a Mimer K shelf under every peice of kit. Surely this would just explode the capabilites?

The FlatEarther wrote:Tom, getting back to the racking. Why not put a Mimer K shelf under every peice of kit. Surely this would just explode the capabilites?

Hi Flat, I didn't realize I never answered this. Yes, I believe Mimer Ks under everything is musically the ideal way to go. I wasn't sure for a while if there was a disadvantage in using more than one or two on a rack but in conferring with others I have found that 3 or even 4 Mimer Ks on a rack still work very well as long as you stick with a Reference bottom - the metal shelves are not a good choice for the bottom shelf.

I do intend to end up with Mimer Ks under all components eventually but have other projects that are currently taking preference.

While I am at it I noticed that I also failed to make an important update on my findings in this thread. In some additional listening, after I changed my ATCs to passive and started using the Tundra Mono 2a amps, I concluded that the damping pad supplied by Anders is musically superior to the foam from the Trampolin boxes. Something which Fredrik had confirmed to me that he had found as well. I am in the process of trying to find the same thickness of fiberfill material locally so I can make another pad for my second rack.

What's the latest news re: the USB ripping drive software updatefor the Melco?

ThomasOK wrote:
While I am at it I noticed that I also failed to make an important update on my findings in this thread. In some additional listening, after I changed my ATCs to passive and started using the Tundra Mono 2a amps, I concluded that the damping pad supplied by Anders is musically superior to the foam from the Trampolin boxes. .

How did the passive ATCs compare to the active ? and did you mean Mono 2 ( without the a)

Interesting that you have changed your mind after switching to passive amplification. Was this a surprise?

Spannko wrote:What's the latest news re: the USB ripping drive software updatefor the Melco?

I am reliably informed that version 3.0, which includes this feature, is out being tested in beta form, and will be with us any day.
Don sunny downtown York

donuk wrote:

Spannko wrote:What's the latest news re: the USB ripping drive software updatefor the Melco?

I am reliably informed that version 3.0, which includes this feature, is out being tested in beta form, and will be with us any day.
Don sunny downtown York

Thanks Don.

To spannko, my understanding is also that the software for ripping is in testing. I think it will be a great addition. Also they just announced the Melco N1ZH unit in the US, which is a different configuration than the original N1Z and less expensive to boot. It has the dual power supplies and "audiophile" caps with the thicker aluminum front panel. But instead of having 1TB of SSD storage it has 6TB of HD capacity. I don't know if that is the reason but instead of being about 4 times the price of the N1A it is only double at $3995. Don't know if the changes really improve the musicality or not. Should be interesting.

To u252agz, I think I wrote about this somewhere else but I definitely prefer the ATC100s passive driven by the Tundra Mono 2a units over the same speakers with the ATC active amp/crossover packs. Indeed me saying this brought a well known forum crasher here to tell us all that my finding was impossible. Also, it is indeed the Tundra Mono 2a of which I have the first pair. Some of the same changes that were made to the Tundra 2 to make it the a version were also necessary for the same reasons- the new RCA connectors, etc. However, there weren't as many changes as some had already been incorporated into the Mono 2 design so I expect the differences might be smaller, but I haven't tested that theory yet.

It IS interesting that I changed my mind on the damping and I didn't expect that as I had done the comparison before a few times. However, there is simply more information, and more importantly, more tuneful and musical information coming through from the Tundra Mono 2as than through the ATC amp pack so all changes become clearer. On the other hand that is somewhat academic since the JBLs are significantly better yet as many have heard from the clips!

Thanks for the update Thomas!

Regarding the JBL's, they going to be even better. 4 months of 24/7 breaking in and last week my pair took a performance jump to a previously unknown level. Great! Followed by a few days bad performance where not only the music left the room but also ALL bass. And I mean ALL...
All variations are bigger on these babies it seems.

When fully settled in - likely wow!!

So Thomas, be sure you give them som heavy beating and play music constantly. And the return will be fantastic :)

I have never heard an active system which bettered its passive version musically. Yes active sounds more impressive, detailed and dynamic but passive always semms to have better musical cohesion using tunedem.