Yes, I think the BT hub is just acting as a modem with the Airport as the router. (You can get a sense of my naivety here in my struggle with basic terminology. I really should spend more time getting to grips with the concepts.)
With a Switch in place the Airport would then simply take care of the Wi-Fi business and any new LAN attachments to non-Sooloos equipment (there are none at present but we may connect the desktop PC via LAN when a replacement arrives).
Presumably by separating out the Sooloos equipment to the extent that it is connected via a Switch (albeit which is in turn connected to the Airport and therefore to the wider network) this helps in some way to reduce RFI/EMI?
I will certainly try disconnecting that connection as Ian suggests and happy to report back. And then try to assess any contribution made by the GISO. My problem is finding the time. If I get round to ordering a Switch today I might get the chance to have a further play at the weekend.
You are correct in your understanding, to do Ian's test ensure your AEBS is connected to the BT HH when you power up all your Sooloos equipment, the reason I say this is that when I had the same setup as you my BT HH was the DHCP server (the one who gives out IP addresses) once everything is working disconnect the AEBS from the BT HH or disconnect the switch from the AEBS whichever is easier? As for your new desktop pc, you can connect wirelessly to the AEBS or wired to the AEBS, try to avoid going via the switch obviously!
Quick update: This isn't based scientific a/b testing, but purely on listening to the music - and I'm going to increase my score related to the imaging from my previous 10-15% to 25-30%, but with a slight concern that it doesn't appear quite as warm sounding as it did.
While I am aware this is very subjective, it does seem that it is improving the imaging substantially. I'm not sure whether the reduction in warmth is my imagination, but I seem to recall reading that Bob Stuart has said that increased jitter can lead to music sounding warmer (and, one presumes, v.v.).
The SOLE purpose of the ID41 is make sure the received Data is IDENTICAL to what was sent... and to act as a �buffer tank� to allow that all necessary data to be re-transmitted from the Core and then re-assembled [in the correct timing order] so that it matches IDENTICALLY with what the Core sent.
The ID41, despite its �1,000 cost, performs no other function within an 861... it doesn�t Upsample or Apodize... add any dither... or perform any other processing, other than ensuring that Identical data is passed from the Core to the 861.
If, even after multiple re-tranmissions, a time-out is exceeded, then the ID41 will simply ignore those �packets� and send the remaining data onto the 861.
This means that if a Data packet was not identical, then there would simply be a �blank� in the sound replay... which would be easily audible as a �dropout�.
And it is because we don�t hear these �dropouts� that we know that the data received at the ID41 and passed to the 861 is identical to the data transmitted from the Core
I didn't realise that! Would a regular network card be able to carry out that function (potentially i.e. with the correct drivers)? Is there anything from technical perspective that warrants the extra expense?
Also would the whole system benefit from this type of implementation i.e. from processor to speaker?
Possibly yes but it would create big issues to resolve like timing between speakers. Currently this is controlled via the system controller or master DSP (in a pair). Think of star wiring to a surround rig all sending un-timed packets of data and then reconstructing them into sound.
Its certainly been suggested in the past and I believe Meridian already considered it, albeit rejecting it due, as Nick suggests, to timing issues.
Still think that an audio endpoint in the master DSP would be good though - slave linked conventionally or a dedicated network that could be synchronized accurately. Then upstream devices would communicate via standard packet based network protocols rather than a synchronized audio stream subject that was subject to jitter.
Meridian owner since 1992 Prime & PSU, Focal Elear headphones, Roon (ROCK 8Gi5 NUC), Explorer 1 & 2, F80. 200/203, various Sonos (via Roon).
Best answer I can give you about the cost of the ID41 is an answer I received long ago from a Far East manufacturer that we approached about making a particular part for a client.
His paraphrased response went like "We can make you One for a million dollars... or we can make you a million of them for $1 each... either way, it'll cost you a million dollars".
A "network card" to fit to the proprietary back plane of an 861 would only sell in small volumes and therefore that will raise the Unit Price.
I would also re-iterate that when we're buying Meridian kit, a very high slice of what we're paying for is initial design, development testing, listening time, fine-tuning time etc., etc. until Meridian arrive at a Prototype they are happy with.
Then the prototype needs to be documented, validated and the the manufacting procedures tested to make sure that Production units match the expected performance that was achieved from the prototype.
Given the small number of units that Meridian sell, these large development costs cannot be simply amortised over high unit volumes, resulting in the high prices we see.
As to your other question, I do believe that Meridian are aware of the potential benefits that Ethernet transfers might offer when connecting Sources to DSP's.
As remarked by Nick & Ian though, the issue that needs to be overcome is Synchronization between 2 to 7 DSP's and will not be an easy nut to crackI think.
If there are synchronisation issues between speakers if fed using packet based bit perfect delivery, then this would also be the case for wireless systems which, in another thread, it was said that Meridian were looking at.
possibly, however, I thought that those devices shown in the new partner piece were single speaker/portable types.
Not saying it won't ever get resolved for >1 speaker but it's going to have to be really clever.
I know from various upgrades in recent years that the better the timing and jitter seems to get the better the stereo imaging and enjoyment. Anything that goes backwards from the high performance now available will be easily heard.
Quick update: This isn't based scientific a/b testing, but purely on listening to the music - and I'm going to increase my score related to the imaging from my previous 10-15% to 25-30%, but with a slight concern that it doesn't appear quite as warm sounding as it did.
While I am aware this is very subjective, it does seem that it is improving the imaging substantially. I'm not sure whether the reduction in warmth is my imagination, but I seem to recall reading that Bob Stuart has said that increased jitter can lead to music sounding warmer (and, one presumes, v.v.).
Let it swim at volume 1 for a couple of days. Turn off DSP speakers at the back if you want. To speedup if there is any burn-in required rather than brain adjusting to less noise.
In a week I will swap out the AQ Diamond RJ45 and see if anything is out of the ordinary when running on the Sooloos audio endpoint LAN.
DSP8000SE, DSP7200SEHC, DSP5200SE sides, DSP7200SE rears, 861v8, 210, Sim2 Ultranero 4, Stewart Cinecurve 2.35:1, XEIT CM-5E Anamorphic lens, Dune HD Duo Cinema 8K and Wireworld Platinum 48G HDMI -> UHD722 -> Shunyata Omega Ethernet. Shunyata/Furutech power-cables/filters. SR Master Fuse@861v8. Uptone Etherregen with 10MHz OCXO clock.
Thanks for the explanation. It's logical that when you add up the components in M kit that they are a smaller percentage of the final cost then compared to a similar item with a high turnover of sales.
I was interested from a technical point of view as to whether the component provided any other functionality or properties that you wouldn't find in a standard network card. e.g. it had been carefully selected not to introduce contamination into the system etc. Can I read that the only likely difference between this network card and another is the form factor?
I think there is more to it than a NIC as ID4x has XMOS chip on board (I assume for audio rendering?) and a maintenance port. Possibly NIC being a simplified analogy.
Google Search
|
Meridian Systems
|
Meridian Streaming
|
Restaurant
|
WiKi
|
Duncan's Meridian Info
|
Board Rules
|
Restaurant Rules
