What computer front end is worthy to run the Scarlatti DAC?
- Thread starter caesar
- Start date
Bruce B
WBF Founding Member, Pro Audio Production Member
Bruce, do you have any experiences with the many various computer sound cards that you can share?
Bruce B
WBF Founding Member, Pro Audio Production Member
I've used many so called "soundcards". I love the cards I use now. They are Mykerinos cards using the Pyramix MassCore system. Unfortunately very expensive.
I've used M-audio 24/96, 24/192, Lynx Two-B and AES16, RME Hammerfall, Digidesign PT HD cards and Soundscape Mixtreme cards. For the average consumer, it's hard to beat a Lynx AES16 PCI card with the latest drivers. For folks that don't want to go inside the computer, get a decent USB or Firewire DAC and keep it simple. For the PC, I'd recommend going FW to a Weiss DAC and for MAC I'd go FW out to a MH DAC. Don't make it over complicated. We've had so many clocking issues, driver issues, compatibility issues and what not. Gary has a great no-nonsense server he builds. At RMAF we tried a MAC mini going out USB using very hi-rez files. Stay tuned for more info.
I've used M-audio 24/96, 24/192, Lynx Two-B and AES16, RME Hammerfall, Digidesign PT HD cards and Soundscape Mixtreme cards. For the average consumer, it's hard to beat a Lynx AES16 PCI card with the latest drivers. For folks that don't want to go inside the computer, get a decent USB or Firewire DAC and keep it simple. For the PC, I'd recommend going FW to a Weiss DAC and for MAC I'd go FW out to a MH DAC. Don't make it over complicated. We've had so many clocking issues, driver issues, compatibility issues and what not. Gary has a great no-nonsense server he builds. At RMAF we tried a MAC mini going out USB using very hi-rez files. Stay tuned for more info.
garylkoh
WBF Technical Expert (Speakers & Audio Equipment)
The cookbook for building the server is available in the latest Newsletter on my website. As far as I know, it's been built (or being built) 9 times already. We also discuss the server in the PNWAS thread in this forum.
Since the Scarlatti is a DAC, no need for the Weiss DAC. The Weiss INT202 has the dual-wire AES output that will enable up to 192kHz sampling on the Scarlatti.
Since the Scarlatti is a DAC, no need for the Weiss DAC. The Weiss INT202 has the dual-wire AES output that will enable up to 192kHz sampling on the Scarlatti.
Hi Bruce,
if going the modern Mac route with the Weiss, is it worth considering their firewire product instead of USB - Weiss Int202?
I am wondering if this is the most ideal way to connect to an existing DAC for a home user as it takes the firewire output from the mac and outputs s/pdif, while possibly using Amarra as well.
One other thing trying to get an answer from others in general is; how much CPU overhead-cycles on a laptop is taken up by associated audio resources (such as Amarra-drivers-codecs-etc)?
I would say knowing this makes it easier to spec a laptop, in theory if the overhead and cycles are low then a very low spec Mac can be used, as long as its firewire or USB is the up to date implementation.
I am thinking more from a heat dissipation (and fan noise) perspective rather than cost, for spec of the laptop.
Edit:
Doh I missed Gary mentioned the Int202 as well hehe, but interested to know if firewire is still the ideal or best way for connecting a laptop.
Thanks
Orb
garylkoh
WBF Technical Expert (Speakers & Audio Equipment)
Just for the heck of it, and because I hear rumours that the new Mac Mini is the best sounding Mac for a music server, I can ever find the time, the next music server I build will be based around the Mac Mini and Amarra, and then I'll see if I can tune the heck out of it.
You need very little CPU power. In my server, selecting the slowest possible processor (two reasons - heat dissipation and not to stress the power supply) we run at less than 3% CPU even with 24/192 files. HDCD decoding of the 24/176.4 HRx files can bring this up 1%.
You need very little CPU power. In my server, selecting the slowest possible processor (two reasons - heat dissipation and not to stress the power supply) we run at less than 3% CPU even with 24/192 files. HDCD decoding of the 24/176.4 HRx files can bring this up 1%.
Gary
Loking at the low computing power required for music playing. Why don't we see more atom-based system using a stripped-down Windows or better Linux?
On the subject of Interface at first take Firewire is a much bettr interface than USB for music (and other stuffs as well). I don't know if one would really hear the differences but it remaisn FireWire is superior to USB...
Loking at the low computing power required for music playing. Why don't we see more atom-based system using a stripped-down Windows or better Linux?
On the subject of Interface at first take Firewire is a much bettr interface than USB for music (and other stuffs as well). I don't know if one would really hear the differences but it remaisn FireWire is superior to USB...
Bruce B
WBF Founding Member, Pro Audio Production Member
I would still stick with USB for the ease of implementation. There are only 2 FW DAC's that I've heard that I would recommend, the Weiss and MH. Prism makes the Orpheus, but it's overkill for most folks. There are a slew of USB DAC's that will do really hi-rez. Even the Playback will do USB.
garylkoh
WBF Technical Expert (Speakers & Audio Equipment)
I think that the trick for good sound might be the low CPU utilization. I tried building an atom-based one and it didn't sound as good. In my server, most of the Windows services are turned off.
I don't think that Firewire is inherently better than USB. Both have their advantages and disadvantages. Some of the current async USB are technically very good. Take this one as an example:
http://www.antelopeaudio.com/en/products_zodiacplus.html
Until I can get the chance to hear them in a system I am familiar with, I can't give an opinion one way or the other. But I love the idea of being able to link a Rubidium Clock with it!! One CES, I had the chance to demo with the Esoteric P-03/D-03 combination. I was floored by how much better it was with the atomic clock.
Since caesar wants to know what computer to use for his Scarlatti DAC, I would recommend that he builds my server, and uses the Weiss INT202 to link between the computer and his Scarlatti. The alternative would be Mac + Amarra + INT202. I have not heard any USB interfaces that are as transparent as the INT202.
Gary/Bruce thanks.
Bruce, if you already have the DAC or digital amp though without USB, would you get the Weiss Int202 that is basically a firewire-to-s/dif box or asynch USB solution?
From what I understand there seems more complexity with USB and clocking/jitter, but I can definitely see its appeal if looking for a more universal solution and greater choice of DACs.
And yeah I am wondering how long before even more lite laptops along the atom/more powerful ipad type solution.
Thanks
Orb
Bruce, if you already have the DAC or digital amp though without USB, would you get the Weiss Int202 that is basically a firewire-to-s/dif box or asynch USB solution?
From what I understand there seems more complexity with USB and clocking/jitter, but I can definitely see its appeal if looking for a more universal solution and greater choice of DACs.
And yeah I am wondering how long before even more lite laptops along the atom/more powerful ipad type solution.
Thanks
Orb
Bruce B
WBF Founding Member, Pro Audio Production Member
Why put anything between the computer and DAC? You want to keep it simple. Stick a Lynx AES16 card in the computer and connect directly via AES/EBU all the way up to 192 and be done with it!
garylkoh
WBF Technical Expert (Speakers & Audio Equipment)
I don't know how good a job the Scarlatti does isolating the ground of the AES/EBU inputs, and I don't know how good the Lynx card electrically isolates. Hence, the noisy power plane inside the server might contaminate the rest of his system. The one time I tried using the Lynx card, I was using a Classe DAC1, and just unplugging the AES/EBU connector improved sound of my system while playing LPs. Hence my hesitation to recommend this solution.
I know that a lot of the expensive pro stuff that you use have to do a good job electrically or else the studio would be a nightmare of ground noise. Since the Scarlatti comes from pro background, and if it does a good job of electrical isolation, then the Lynx card will certainly be the way to go and be the simplest solution.
Thanks for the replies so far. Someone suggested Mac Mini + Amarra + firewire + script to give amarra priority. Any thoughts on this
What's the purpose of the Weiss INT 202?
Going to hear some live music tonight to put things in perspective
What's the purpose of the Weiss INT 202?
Going to hear some live music tonight to put things in perspective
Its just a way to connect a PC/laptop to your existing DAC using firewire interface, without the additional complications that USB can throw up (isolation-jitter-etc).
http://www.weiss-highend.ch/int202/index.html
But if you have the dCS setup then USB should be the ideal route with the Scarlatti Upsampler, this makes even more sense if you have the transport and DAC.
Cheers
Orb
http://www.weiss-highend.ch/int202/index.html
But if you have the dCS setup then USB should be the ideal route with the Scarlatti Upsampler, this makes even more sense if you have the transport and DAC.
Cheers
Orb
Garyllkoh and BruceB (and all other digital experts of the forum, surely)
Since you refer to 192kHz digital, what do you feel about this post concerning 96kHz versus 192 kHz?
http://www.soundstagehifi.com/index...-vs-192khz&catid=57:reader-feedback&Itemid=24
Since you refer to 192kHz digital, what do you feel about this post concerning 96kHz versus 192 kHz?
http://www.soundstagehifi.com/index...-vs-192khz&catid=57:reader-feedback&Itemid=24
Isolation: if you connect 2 boxes with USB or with Firewire, you make an electrical connection between the 2. No difference between these protocols on this aspect.
Jitter: depending on the implementation, both protocols can generate more or less input jitter.
http://www.nanophon.com/audio/1394_sampling_jitter.pdf
Both protocols can be set to Isochronous mode, the needed amount of bandwidth is reserved for the audio device.
USB can do asynchronous transfer in isochronous mode, Firewire can do the same.
The big difference between the two is that there is a standard for USB audio; class1 (24/96 max) and class 2 (24/192 max).
As there is no standard for Firewire audio, each manufacturer has to develop his own driver at the PC side. This make a Firewire interface more expensive.
As a consequence, the number of Firewire DAC’s is small as only the pro-companies do have the budget and the resource to develop it.
As far as I could judge, there is no reason to assume one protocol is superior to the other as far as playback is concerned.
As usual, you won’t hear the differences in protocol but you do hear the quality of the implementation
http://thewelltemperedcomputer.com/Intro/SQ/USB_FireWire.htm
Hi Vincent,
isolation is still critical in terms of implementation for both firewire and USB, look to Charles Hansen for his views on USB, and consider that firewire has made it into the pro world in same way as we see XLR compared to RCA in audio consumer.
I have worked on the architecture of USB (at a manufacturer engineering level) on microsoft based hardware solutions and feel it is anything but standard (again depends upon what you deem standard beyond the low level interface-communication but in same context firewire IS following standards from IEEE), this can be seen looking at Microsoft's own architecture documents.
The additional problem is not all audio USB solutions are using the same transfer mode.
If needed I guess we could discuss this in further detail, with others piping in and provide relevant articles and documents but not sure if its ideal here or a new thread.
Edit:
But I do think the long term future of firewire could be under threat and USB will become the only universal solution, and current solutions are about how either are implemented (some great USB products such as the Ayre and also firewire such as those by Daniel Weiss).
An example of one company that is fresh to audio and so has different perspective is that of Devialet, and for them they recommend firewire over USB due to their lack of confidence for most current USB solutions, now these are ex-Nortel R&D engineers with no axe to grind with the firewire vs USB debate from those who have a vested interested in either standard (not meaning you and me here
).
Still, as I said USB would be the way I would go if I had the transport and DAC from dCS by using their Upsampler.
Cheers
Orb
isolation is still critical in terms of implementation for both firewire and USB, look to Charles Hansen for his views on USB, and consider that firewire has made it into the pro world in same way as we see XLR compared to RCA in audio consumer.
I have worked on the architecture of USB (at a manufacturer engineering level) on microsoft based hardware solutions and feel it is anything but standard (again depends upon what you deem standard beyond the low level interface-communication but in same context firewire IS following standards from IEEE), this can be seen looking at Microsoft's own architecture documents.
The additional problem is not all audio USB solutions are using the same transfer mode.
If needed I guess we could discuss this in further detail, with others piping in and provide relevant articles and documents but not sure if its ideal here or a new thread.
Edit:
But I do think the long term future of firewire could be under threat and USB will become the only universal solution, and current solutions are about how either are implemented (some great USB products such as the Ayre and also firewire such as those by Daniel Weiss).
An example of one company that is fresh to audio and so has different perspective is that of Devialet, and for them they recommend firewire over USB due to their lack of confidence for most current USB solutions, now these are ex-Nortel R&D engineers with no axe to grind with the firewire vs USB debate from those who have a vested interested in either standard (not meaning you and me here
).Still, as I said USB would be the way I would go if I had the transport and DAC from dCS by using their Upsampler.
Cheers
Orb
Last edited:
Hi Orb
A discussion involving people with you’re level of knowledge about the pro and cons of both protocols would be very interesting indeed.
As a typical armchair engineer I’m inclined to say that as far as playback is concerned, we don’t have much reason to prefer one over the other.
Up to now I haven’t seen any evidence that the “sound quality” of Firewire is better than USB. Still there a couple of people who firmly believe the SQ of Firewire is superior
A discussion involving people with you’re level of knowledge about the pro and cons of both protocols would be very interesting indeed.
As a typical armchair engineer I’m inclined to say that as far as playback is concerned, we don’t have much reason to prefer one over the other.
Up to now I haven’t seen any evidence that the “sound quality” of Firewire is better than USB. Still there a couple of people who firmly believe the SQ of Firewire is superior
garylkoh
WBF Technical Expert (Speakers & Audio Equipment)
He is referring to the Benchmark DAC..... if he is not, then he should qualify his statement. The Benchmark does not use PLLs to reduce jitter. It uses a variable oversampling ratio to do it. If I recall right, the Benchmark upsamples all input to 110kHz.
That is also why I've found that various jitter-reducing interfaces (like the Monarchy DIP and Genesis Digital Lens) and even different cables sound different on the Benchmark. If it is truely "jitter immune" at up to 96kHz as their marketing literature claim, then different digital cables should not sound different.
I'm not saying that the Benchmark is not a good DAC - I still have one - it's just not as perfect as they claim.
Similar threads
