Mac Mini to SuperNait: Hiface EVO Vs Halide Bridge
Posted by: winkyincanada on 31 December 2010
I understand these to be considered an upgrade in the way I connect my Mac Mini to my SuperNait (currently S/PDIF via optical Toslink).
They are similarly priced. Would someone offer a view as to the relative sonic merits, please?
They are similarly priced. Would someone offer a view as to the relative sonic merits, please?
Posted on: 31 December 2010 by Tog
Hailde Bridge removes some of the harshness that I can hear when attached via Optichord - now sounds very close to streamed.
Great value - could be bettered by Wavelength but at twice the price.
Tog
Great value - could be bettered by Wavelength but at twice the price.
Tog
Posted on: 31 December 2010 by badlyread
quote:now sounds very close to streamed
So why not stream if it is better?
Happy New Year
Neil
Posted on: 31 December 2010 by winkyincanada
Streaming isn't an option I'd consider at this stage. I hate all that network guff with a passion. UPnP? What a joke. NAS is the tool of the devil.
No, my files are local on my Mac Mini that connects directly to the SuperNait. Super reliable.
No, my files are local on my Mac Mini that connects directly to the SuperNait. Super reliable.
Posted on: 31 December 2010 by jadip
I am interested in this question as well! I now have a Mac mini, a Christmas present. It is connected to my supernait via toslink. Sounds lovely, more analogue, less glare than streaming through the Apple TV (from a PC in another room) which is also connected by toslink to the Supernait. I am interested/curious about USB to SPDIF converters. Some say it provides a terrific improvement in SQ, others don't perceive a difference! What do others say???
Posted on: 31 December 2010 by Marc0209
quote:Originally posted by jadip:
I am interested in this question as well! I now have a Mac mini, a Christmas present. It is connected to my supernait via toslink. Sounds lovely, more analogue, less glare than streaming through the Apple TV (from a PC in another room) which is also connected by toslink to the Supernait. I am interested/curious about USB to SPDIF converters. Some say it provides a terrific improvement in SQ, others don't perceive a difference! What do others say???
Out of curiosity are you using the ATV2, or the original? I run an ATV2 in my SN via toslink. I find it to be a little less resolute, especially in the area of micro-dynamics as compared to running my cd player dig out via coax into the SN dac.
Posted on: 31 December 2010 by Right Wing
quote:Originally posted by jadip:
I am interested in this question as well! I now have a Mac mini, a Christmas present. It is connected to my supernait via toslink. Sounds lovely, more analogue, less glare than streaming through the Apple TV (from a PC in another room) which is also connected by toslink to the Supernait. I am interested/curious about USB to SPDIF converters. Some say it provides a terrific improvement in SQ, others don't perceive a difference! What do others say???
get a hiface. for £100 or so, it takes it up another level - absolute bargain IME. I am yet to hear someone disliking one - although I am sure that the critics will be out there, naturally. I use one on a supernait/mac and i works a treat. I have also tried it in other systems, same story applied.
IME, you can buy one and return it for a full refund if you're not happy with it. go for it !
Posted on: 31 December 2010 by aht
At the risk of repeating myself, the HiFace is good, but the EVO is much better. More expensive, sure, but still very good VFM, IMHO.
Posted on: 01 January 2011 by Tog
And the Halide Bridge is IMHO even better.
Tog
Tog
Posted on: 01 January 2011 by Jack
Of course the other good thing with the Halide is that you don't need to purchase another interconnect e.g. DC1
Posted on: 01 January 2011 by jon h
going to get a halide -- will connect between macmini server and the ndac
Posted on: 01 January 2011 by DHT
I purchased a Halide Bridge from the US whilst I waited for the EVO, sold the Bridge as soon as I heard the EVo in my system,no contest, the HB was smooth but just lacked dynamics and resolution.
Posted on: 01 January 2011 by jon h
but it requires drivers. Prefer something which is native OS supported for drivers.
Posted on: 01 January 2011 by jerryct
maybe you should have a look at the new wavelengths: WaveLink HS 24/192 USB to SPDIF converter. it support 192khz/24bit with native drivers on mac os x (Class 2 USB Audio) greater or equal than 10.6.4.
(halide "only" supports up to 96khz)
jerry
(halide "only" supports up to 96khz)
jerry
Posted on: 01 January 2011 by jon h
800 quid is very spendy for this functionality.
Posted on: 01 January 2011 by DHT
quote:Originally posted by jon honeyball:
but it requires drivers. Prefer something which is native OS supported for drivers.
Why?The Bridge being limited to 24/96 was a bit of a pain as well.
Posted on: 01 January 2011 by jon h
over the years i have had more than enough grief with custom drivers from small time developers who never really get things debugged or keep up with OS releases or etc etc
im not suggesting this applies to this particular vendor. Just I prefer OS-supplied drivers wherever possible. They tend to have several orders of magnitude more testing than 3rd party items.
im not suggesting this applies to this particular vendor. Just I prefer OS-supplied drivers wherever possible. They tend to have several orders of magnitude more testing than 3rd party items.
Posted on: 01 January 2011 by Guido Fawkes
Anybody tried the KingRex UC192, which uses a isochronous design. The adaptive clock generator is for synchronous mode of audio bit stream. The clock frequency is adjusted automatically to synchronize with the host computer. The adjustments are done in a manner that limit clock variation to a minimum impact to the audio performance. The Master clock generator is a 1ppm TCXO with proprietary drivers lower jitter.
Thing is when you start adding things like the Wavelength converter then the price for MacMini/Converter starts to rival that for the nServe ... and at the level I might just as well go for the nServe and get Naim support. AFIK Wavelength does not have any dealer support in in this country.
Posted on: 01 January 2011 by Tog
They don't and you need to remember that their production volumes are relatively small in comparison to Naim. Gordon Rankin was there at the birth of computer audio with the Wavelength Brick and is a champion of the USB Dac. His designs are hugely respected and the best comparison in quality would be to compare his valve dacs to Lobb shoes.
Buying even one of his mid-range dacs starts to make Naim gear like the nDac look inexpensive.
Availability is limited in Europe - but there is a dealer in Brooklyn if you know anyone flying to New York.
Tog
Buying even one of his mid-range dacs starts to make Naim gear like the nDac look inexpensive.
Availability is limited in Europe - but there is a dealer in Brooklyn if you know anyone flying to New York.
Tog
Posted on: 01 January 2011 by Guido Fawkes
No intention of doing that - Naim makes a DAC and it sounds great so that bit is sorted in my system.quote:Buying even one of his mid-range dacs starts to make Naim gear like the nDac look inexpensive.
I currently use a hi-Face so will probably go the EVO route as there is reasonable UK support for the product. There are lots of interesting designs around, but UK support is essential IMHO.
Posted on: 01 January 2011 by winkyincanada
quote:Originally posted by jon honeyball:
going to get a halide -- will connect between macmini server and the ndac
Just ordered my Halide.
Posted on: 01 January 2011 by Tog
One cable, no boxes
Sweet.
Tog
Sweet.
Tog
Posted on: 02 January 2011 by jerryct
quote:Originally posted by jon honeyball:
800 quid is very spendy for this functionality.
yes, that is absolutely right and you can go even higher if you look at the offramp from empirical audio. it is also "only" a usb-spdif converter but you can select between 3 different clocks and then it goes above 2000 USD. (which now uses the hiface implementation of async usb with offramp 4)
The idea behind these X-to-spdif Converters is getting a better, low jitter master clock for generating the spdif output (instead of using the computer clock).
So the first criterion of such a converter should be async usb or firewire.
Using a better master clock should also give better results for the supernait (as i think it uses a PLL) but i wonder where to stop when you have a ndac with the RAM buffer?
(i think there must be a minimum tolerance of the incoming spdif clock at which the naim dacs buffering mechanism do not need to switch between his 10 fixed frequencies to prevent data under- or overrun)
jerry
Posted on: 02 January 2011 by likesmusic
Surely if you have an nDAC and the Sync light goes on, that should be all there is to worry about - it is managing to buffer the incoming data stream and can reclock it, job done.
And the new macmini has an s/pdif output and costs less than some of these exotic USB/spdif convertor cables, so why not just get a new macmini and swap it in in place of the old one? No drivers, power supplies, wiring tangles ...
And the new macmini has an s/pdif output and costs less than some of these exotic USB/spdif convertor cables, so why not just get a new macmini and swap it in in place of the old one? No drivers, power supplies, wiring tangles ...
Posted on: 02 January 2011 by jerryct
in another thread there was made a good differentiation between "push" and "pull" data.
As SPDIF does not have a flow control it pushes the data to the dac with one clock and the dac clocks it out with another clock which are not synchronized, the dac must synchronize to the incoming data to keep the buffer in average half full. the dac is doing this by choosing between his fixed clocks to prevent data under- or overrun. This is a continuous process. And this is audible.
Thus i think in having a clock in the sender which is of the same quality as the two clocks for the audio frequencies in the dac gives the best results.
As SPDIF does not have a flow control it pushes the data to the dac with one clock and the dac clocks it out with another clock which are not synchronized, the dac must synchronize to the incoming data to keep the buffer in average half full. the dac is doing this by choosing between his fixed clocks to prevent data under- or overrun. This is a continuous process. And this is audible.
Thus i think in having a clock in the sender which is of the same quality as the two clocks for the audio frequencies in the dac gives the best results.
Posted on: 02 January 2011 by likesmusic
jerryct - I agree with you about the difference between "pull" and "push" systems; "pull" systems seem to me at least to have many advantages.
However, although s/pdif is a "push" system it seems to me that the intention of the Naim DAC is to mitigate many of the effects of such a system, particularly the jitter caused by retrieving the clock signal from the data stream. It does this by buffering the data, then choosing one of a set of clocks to re-clock the data in such a way as to keep the buffer from under or overrun.
According to Naim this is NOT a continuous process.
If you read page 7 of the Naim White Paper you will see that they state that:
" if the master clock is changed too quickly this can cause audible artefacts, ie introduce jitter. The SHARC DSP monitors the rate at which the RAM buffer is either filling or emptying and changes the clock frequency only if the buffer is going to either overflow or underflow. This way the incoming clock jitter is completely isolated from the NDX master clock. When the system has settled, it will only modify the master clock frequency every 10 to 15 minutes"
One (tiny) change in clock master frequency every 10 or 15 minutes is NOT a continuous process!
And are you saying you can hear this micro-event? What does it sound like?!
However, although s/pdif is a "push" system it seems to me that the intention of the Naim DAC is to mitigate many of the effects of such a system, particularly the jitter caused by retrieving the clock signal from the data stream. It does this by buffering the data, then choosing one of a set of clocks to re-clock the data in such a way as to keep the buffer from under or overrun.
According to Naim this is NOT a continuous process.
If you read page 7 of the Naim White Paper you will see that they state that:
" if the master clock is changed too quickly this can cause audible artefacts, ie introduce jitter. The SHARC DSP monitors the rate at which the RAM buffer is either filling or emptying and changes the clock frequency only if the buffer is going to either overflow or underflow. This way the incoming clock jitter is completely isolated from the NDX master clock. When the system has settled, it will only modify the master clock frequency every 10 to 15 minutes"
One (tiny) change in clock master frequency every 10 or 15 minutes is NOT a continuous process!
And are you saying you can hear this micro-event? What does it sound like?!