HAKAI playground

A DIY digital music streamer with exceptional performance

Moderator: Staff

tokenbrit
Very active member
Very active member
Posts: 2039
Joined: 2012-03-22 19:47
Location: New England

Re: HAKAI playground

Post by tokenbrit »

Fredrik, do you have a feel for 'typical' HAKAI burn-in time, if there is such a thing, based on a build with recommended components, from new?
User avatar
lejonklou
Administrator
Administrator
Posts: 6524
Joined: 2007-01-30 10:38
Location: Sweden
Contact:

Re: HAKAI playground

Post by lejonklou »

tokenbrit wrote:Fredrik, do you have a feel for 'typical' HAKAI burn-in time, if there is such a thing, based on a build with recommended components, from new?
I found that it takes weeks for the digital parts to settle fully. Often they sound great when brand new and then performance goes up and down, not unlike how a single soldered joint varies after it's just been made.

The 320 SSD is a special case, which I haven't fully understood the pattern of. It usually sounds great after it has been written to and then after a while (weeks/months, unclear!) it can go into a "grey"and boring state, where its performance in terms of sound approaches that of other SSD's (many of which don't seem to vary in performance). If it's then written to (unclear how big a part of the 320 needs to be rewritten), it appears to become "reset" and once again sounds magical.
User avatar
ThomasOK
Very active member
Very active member
Posts: 4358
Joined: 2007-02-02 18:41
Location: United States
Contact:

Re: HAKAI playground

Post by ThomasOK »

Hmmm. They are both pretty good but I find clip 1 more enjoyable on both tracks. On the guitar and voice I find a roundness of guitar tone and quality of playing and singing that sounds more real to me on clip 1, and the voice sounds a bit more constricted on clip 2. On the sax one (Jan Garbarek?) I can hear the small texture notes he puts in better and it has a better flow to all the instruments on clip 1. On clip 2 it is coarser and doesn't jell as well. So I'm definitely favoring 1 across the range. This is using the store iMac and its built in speakers. I did have to listen multiple times as I find sometimes there is a shift in sound on the second playing of the track. Might just be my ears getting used to the iMac or the piece.
Last edited by ThomasOK on 2018-09-18 23:52, edited 1 time in total.
The LP12 Whisperer
Manufacturer, Distributor, Retailer and above all lover of music.
Spannko
Very active member
Very active member
Posts: 2292
Joined: 2008-01-24 21:46
Location: North East of The Black Country, UK

Re: HAKAI playground

Post by Spannko »

beck wrote:This has reinforced my 1. clip preference. I wonder of this will end in a kind of source first/pre first discussion........
Mine too, and I thought the differences were bigger this time.

On the Leonard Cohen song the guitar sounds out of tune on clip 2 and the guitarist sounds like a better player on clip 1.
tokenbrit
Very active member
Very active member
Posts: 2039
Joined: 2012-03-22 19:47
Location: New England

Re: HAKAI playground

Post by tokenbrit »

I did like 1 but it sounded a bit congested on the 1st track, in particular... 2 had better separation, but a harshness... Then beck picked 1, so I couldn't ;)
matthias
Very active member
Very active member
Posts: 2092
Joined: 2007-12-25 16:47
Location: Germany

Re: HAKAI playground

Post by matthias »

Number one is more musical for me.

Matt
Matt

MBP / Exposure pre + power (both modified) / JBL3677
David Neel
Very active member
Very active member
Posts: 975
Joined: 2008-02-08 23:17
Location: The Magical Forest

Re: HAKAI playground

Post by David Neel »

The first time round was 1. J3455N-D3V and 2. N3050N-D2P. I reversed this for yesterday's clips. Each Hakai was identical except for the motherboard and memory (the N3050N-D2P requires DDR2) and the SSD was switched between them.

My thoughts in the room are: J3455 has better pitch stability and timing, both rock solid, but lacks emotion and engagement; N3050 carries much more emotion but frustrates with poor timing and tentative rhythm. There have been variations in both as they burn in, but these contrasts have been constant so far.

Thank you to everybody for the comments. Even though opinion was divided, the observations all related to what I hear in the room.

EDIT: a week later, the N3050 is sounding better and the timing/rhythm is improved. This may be partly due to further power strip/cable experimentation.
Last edited by David Neel on 2018-09-27 23:14, edited 4 times in total.
The search for knowledge is not nourished by certainty, but by a radical distrust in certainty
tokenbrit
Very active member
Very active member
Posts: 2039
Joined: 2012-03-22 19:47
Location: New England

Re: HAKAI playground

Post by tokenbrit »

Thanks 'David', you've definitely got me interested in an N3150 m'board now :)
Spannko
Very active member
Very active member
Posts: 2292
Joined: 2008-01-24 21:46
Location: North East of The Black Country, UK

Re: HAKAI playground

Post by Spannko »

David,

Thanks for going to the trouble of posting these videos.

If a picture is worth 1,000 words, these videos have to be worth at least 10,000, if not more!
tokenbrit
Very active member
Very active member
Posts: 2039
Joined: 2012-03-22 19:47
Location: New England

Re: HAKAI playground

Post by tokenbrit »

lejonklou wrote:
tokenbrit wrote:Fredrik, do you have a feel for 'typical' HAKAI burn-in time, if there is such a thing, based on a build with recommended components, from new?
I found that it takes weeks for the digital parts to settle fully. Often they sound great when brand new and then performance goes up and down, not unlike how a single soldered joint varies after it's just been made...
Thanks Fredrik. Out of interest, are the GA-N3050N qualities (& weakness(es)) captured in the recent clips recognisable & consistent, at all, based on your experience with the 3050 leading up to the original hardware recommendations?
Also, are the differences here, between N3050 and J3455, similar to or greater than those you've experienced between 3050, 3150, & 3160?

Big thanks to David Neel we have got a good sense of the strengths of each of the Gigabyte N3050 & J3455 mobos .. at least for his 2 examples. We haven't yet heard the 3150 or 3160, so it would be good to get an idea of how the clips compare with wider experience and, if poss, how the 3150 & 3160 might compare.
David Neel
Very active member
Very active member
Posts: 975
Joined: 2008-02-08 23:17
Location: The Magical Forest

Re: HAKAI playground

Post by David Neel »

It may, or it may not, be relevant that the D2P version of the 3050 uses ddr2 memory, whereas boards tested and recommended by Fredrik all use ddr3 memory.
The search for knowledge is not nourished by certainty, but by a radical distrust in certainty
User avatar
lejonklou
Administrator
Administrator
Posts: 6524
Joined: 2007-01-30 10:38
Location: Sweden
Contact:

Re: HAKAI playground

Post by lejonklou »

tokenbrit wrote:Thanks Fredrik. Out of interest, are the GA-N3050N qualities (& weakness(es)) captured in the recent clips recognisable & consistent, at all, based on your experience with the 3050 leading up to the original hardware recommendations?
Also, are the differences here, between N3050 and J3455, similar to or greater than those you've experienced between 3050, 3150, & 3160?
I don't feel that clips capture the true qualities that you experience in the room. Clips are a fantastic tool for comparisons - with some limitations and potential traps - and they can also bring a general feeling of whether the system performs well or not. But they are on a different level than experiencing the differences live, in the room.

Regarding the motherboards, there are multiple differences between them, so it's not as simple as specifying the processor. I've had great 3050-equipped motherboards and I've had really terrible ones. So what we're evaluating are very complex designs, with multiple variables. And this is the challenge in the HAKAI project - it's all down to trial and error, testing one motherboard at a time to find the currently best sounding one.
User avatar
lejonklou
Administrator
Administrator
Posts: 6524
Joined: 2007-01-30 10:38
Location: Sweden
Contact:

Re: HAKAI playground

Post by lejonklou »

David Neel wrote:It may, or it may not, be relevant that the D2P version of the 3050 uses ddr2 memory, whereas boards tested and recommended by Fredrik all use ddr3 memory.
I think that this is relatively unimportant.
Charlie1
Very active member
Very active member
Posts: 4831
Joined: 2007-12-11 00:30
Location: UK

Re: HAKAI playground

Post by Charlie1 »

I couldn't get on with the first track but enjoyed both with the second track. I genuinely hadn't seen tokenbrit and Fredrik's posts and had decided I liked them both. I was aware of them being different, but couldn't really add much detail as to why. If pushed, then I have a slight preference for #1.
tokenbrit
Very active member
Very active member
Posts: 2039
Joined: 2012-03-22 19:47
Location: New England

Re: HAKAI playground

Post by tokenbrit »

lejonklou wrote:
tokenbrit wrote:Thanks Fredrik. Out of interest, are the GA-N3050N qualities (& weakness(es)) captured in the recent clips recognisable & consistent, at all, based on your experience with the 3050 leading up to the original hardware recommendations?
Also, are the differences here, between N3050 and J3455, similar to or greater than those you've experienced between 3050, 3150, & 3160?
I don't feel that clips capture the true qualities that you experience in the room. Clips are a fantastic tool for comparisons - with some limitations and potential traps - and they can also bring a general feeling of whether the system performs well or not. But they are on a different level than experiencing the differences live, in the room.

Regarding the motherboards, there are multiple differences between them, so it's not as simple as specifying the processor. I've had great 3050-equipped motherboards and I've had really terrible ones. So what we're evaluating are very complex designs, with multiple variables. And this is the challenge in the HAKAI project - it's all down to trial and error, testing one motherboard at a time to find the currently best sounding one.
Understood that in room will sound much better. I was trying to ask whether the clips showed the relative qualities, but that would be relative to the GA-J3455, and you may not have heard one yourself to have completed the same comparison, in-room, that 'David Neel' was sharing in the clips...
As for my references to 3050, 3150, & 3160, I only meant the Gigabyte ultra durable range of mini-itx boards that you recommended - I wasn't specifying the processor, generally, to include other mini-itx boards with the same cpu.
Since I tended to prefer the GA-N3050N board in the comparisons against the GA-J3455, I'm hoping to get a GA-N3150N-D3V to build my own HAKAI. Fingers crossed on getting a good one, and finding the rest of the parts necessary...
beck
Very active member
Very active member
Posts: 2752
Joined: 2012-10-22 22:25

Re: HAKAI playground

Post by beck »

David Neel wrote:The first time round was 1. J3455N-D3V and 2. N3050N-D2P. I reversed this for yesterday's clips. Each Hakai was identical except for the motherboard and memory (the N3050N-D2P requires DDR2) and the SSD was switched between them.

My thoughts in the room are: J3455 has better pitch stability and timing, both rock solid, but lacks emotion and engagement; N3050 carries much more emotion but frustrates with poor timing and tentative rhythm. There have been variations in both as they burn in, but these contrasts have been constant so far.

Thank you to everybody for the comments. Even though opinion was divided, the observations all related to what I hear in the room.
Thank you for the feedback David. It is really important for people like me who would like to get better at this. I still think the clips are valuable and tells us a lot more than a decription in words could do even though it can be hard to interpret what we hear.
Playing cd’s…………
Azazello
Very active member
Very active member
Posts: 630
Joined: 2007-01-30 21:59
Location: Stockholm, Sweden

Re: HAKAI playground

Post by Azazello »

User avatar
lejonklou
Administrator
Administrator
Posts: 6524
Joined: 2007-01-30 10:38
Location: Sweden
Contact:

Re: HAKAI playground

Post by lejonklou »

First clip is better!

By quite a margin I'd say.
beck
Very active member
Very active member
Posts: 2752
Joined: 2012-10-22 22:25

Re: HAKAI playground

Post by beck »

Not going to comment on the clips untill I have a proper opinion on them using a PA system for the listening. I understand if you do not want to wait a couple of days more for my answer! :-)
Playing cd’s…………
Azazello
Very active member
Very active member
Posts: 630
Joined: 2007-01-30 21:59
Location: Stockholm, Sweden

Re: HAKAI playground

Post by Azazello »

I'm not convinced that would even be better than just using iphone headphones.
Azazello
Very active member
Very active member
Posts: 630
Joined: 2007-01-30 21:59
Location: Stockholm, Sweden

Re: HAKAI playground

Post by Azazello »

Since no one else seemed interested in guessing; no surprises the first Clip is HAKAI NAS, the second Qnap.
User avatar
lejonklou
Administrator
Administrator
Posts: 6524
Joined: 2007-01-30 10:38
Location: Sweden
Contact:

Re: HAKAI playground

Post by lejonklou »

Azazello wrote:Since no one else seemed interested in guessing; no surprises the first Clip is HAKAI NAS, the second Qnap.
That's fantastic!

(I did harbour a slight fear that number one, which I found substantially more fun to listen to, would be Qnap).
Charlie1
Very active member
Very active member
Posts: 4831
Joined: 2007-12-11 00:30
Location: UK

Re: HAKAI playground

Post by Charlie1 »

I was blocked from that website and only have one machine where I can easily deactivate the security, which was never the one I had at the time. Will try later.
Charlie1
Very active member
Very active member
Posts: 4831
Joined: 2007-12-11 00:30
Location: UK

Re: HAKAI playground

Post by Charlie1 »

I get pro's and con's to each. The first one seems more in-tune, particularly the guitar chords at the start. The second one seems to have better rhythm, probably more noticeable as the track progresses. I prefer clip 1 overall as the (comparatively) out of tune sound of 2 is the more distracting for me. This seems to be a regular theme for me nowadays - i.e. superior tune vs superior rhythm. Now I'm aware of it, I seem to notice it a lot, such as some integrated amps I bought recently. Maybe I should start another thread.
User avatar
Tendaberry
Very active member
Very active member
Posts: 982
Joined: 2010-08-30 16:08
Location: Hamburg

Re: HAKAI playground

Post by Tendaberry »

lejonklou wrote:First clip is better!
By quite a margin I'd say.
I agree, clip 1 has a much more natural swing to it. Clip 2 sounds stilted in comparison.
Now how can I build a Hakai NAS with 4 TB storage space???
Post Reply