Sep 25, 2014 at 2:29 AM Post #46 of 167
  I have been trying out some high end DAC's recently and that's bought me to ask the question of myself and others what is the best method, FPGA or more traditional off the shelf chips.

 
Off-the-shelf. The FPGA DACs are all hype and no substance. The task of a modern DAC is too complex to be properly done solo by one guy with a bunch of odd theories and a Xilinx.
 
Sep 25, 2014 at 8:21 AM Post #47 of 167
I haven't heard it with its ASRC disabled but when enabled its shrill sounding metallic trebles are unbearable to my ears....I'm big enough to upsample to 384kHz should I want to, TYVM.
 
Sep 25, 2014 at 12:15 PM Post #48 of 167
   
Off-the-shelf. The FPGA DACs are all hype and no substance. The task of a modern DAC is too complex to be properly done solo by one guy with a bunch of odd theories and a Xilinx.

Of course it is feasible. Do you think Rob Watt did it in one week? 
rolleyes.gif

 
The theory is right, perfect reconstruction can be achieved with infinite length sinc reconstruction filter. Anything below that is an approximation. The more taps, the closest to 'perfect' reconstruction you get.
 
Now, it's obviously not the only recipe for good sound. But the theory does not lie.
 
Sep 25, 2014 at 5:14 PM Post #49 of 167
I haven't heard it with its ASRC disabled but when enabled its shrill sounding metallic trebles are unbearable to my ears....I'm big enough to upsample to 384kHz should I want to, TYVM.


Sorry not sure which chip you are referring to ?
 
Sep 27, 2014 at 1:03 PM Post #50 of 167
The best Sabre DAC I've heard was definetly the Apogee Symphony. That said, there are tons of Wolfson and Cirrus Logic based DAC's out there that sound amazing too.
 
Sep 27, 2014 at 7:07 PM Post #51 of 167
 
  I have been trying out some high end DAC's recently and that's bought me to ask the question of myself and others what is the best method, FPGA or more traditional off the shelf chips.

 
Off-the-shelf. The FPGA DACs are all hype and no substance. The task of a modern DAC is too complex to be properly done solo by one guy with a bunch of odd theories and a Xilinx.

 
FPGAs aren't used as DACs, they are custom programmed to do the filtering etc. before the digital data is sent to the DAC.
The companies that use them tend to be those with engineers that have been designing DACs for decades that don't want to be limited to whatever filters and options that are built into the DA chips. 
 
Back on topic though, I've found less than stellar Sabre-based DACs make the music somewhat "flat" and dull. The best results I had with a Sabre DAC, ironically was with one that had that very problem, the Calyx DAC 24/192. Using a better power source than USB power (which it uses) and an Audiophilleo 1, the results were very good.
 
Sep 28, 2014 at 11:01 AM Post #53 of 167
The Invicta Mirus uses FPGA and a pair of 9018s. It was designed by the same people behind the 9018 chip so implementation should be optimal.
 
I haven't heard it yet but apparently it sounds phenomenal eclipsing the FPGA only Hugo.
 
Unfortunately it's out of my budget range.
 
Sep 28, 2014 at 6:17 PM Post #54 of 167
  The Invicta Mirus uses FPGA and a pair of 9018s. It was designed by the same people behind the 9018 chip so implementation should be optimal.
 
I haven't heard it yet but apparently it sounds phenomenal eclipsing the FPGA only Hugo.
 
Unfortunately it's out of my budget range.


I have heard the Mirus in my system against the Hugo, it did some things better, not all. I really wanted it to be the solution for me. It did not have the natural analog like ease that the Hugo has or the sense of air. It did some things better, others on a par, but it did not engage me the same way as Hugo. My feeling, and I could be wrong is that the analog stages were not all that they could be. The Kalliope should present the Sabre 9018 chip in its bests light. 
 
I am beginning to think that Hugo is going to be sitting on my desk for a while. I am going to try the Gyhopn Kalliope in the next couple of weeks. Unfortunately the dealer has sold the demo unit, so I will have to wait for a new unit to ship. I know the dealer very well and I have owned high end DAC's like the Audio Note DAC 5 signature, I am told the Kalliope makes that sound poor, so I have high expectations.
 
The real comparison would be between the Nagra HD (FPGA), the new Chord QBD 76  (when thats released) (FPGA) and the Kalliope (Sabre 9018). Rarified products indeed.
 
Sep 28, 2014 at 7:19 PM Post #55 of 167
I thought the Hugo was at the very limit of what I would(could) ever spend on a DAC. The Nagra, Kalliope, etc fall into the unattainable dream gear category.
 
Oct 18, 2014 at 9:21 PM Post #56 of 167
Speak that truth brother.
 
Considering all the components that an electric current has to channel through before it becomes a signal, the thickness of the cord makes little difference in the end.
 
Withstanding the mains voltage and amperage is all a power cord needs to do.
 
Look at what McIntosh provides it's dacs for a power source.
 
Oct 19, 2014 at 1:26 AM Post #57 of 167
I note that the new Gryphon Kalliope DAC uses two ES9018s, and it retails for $26k.
 
Nov 18, 2014 at 4:55 PM Post #58 of 167
  I note that the new Gryphon Kalliope DAC uses two ES9018s, and it retails for $26k.

 
The cost is partly a feature of their brand and target audience economics, but also probably because they paid a lot of attention to things that matter: power supplies, audio section, filtering, clock functionality. If specific attention is paid to these and the attention is good and/or innovative, great results can be achieved.
 
That said, I much prefer the sound of a well implemented 1704K (using Lite Audio LT-1, DAC 83 and I2S) to my experiences of the Sabre chipsets thusfar (Auralic Vega, Oppo 105, PS Audio) - but I guess that's personal.
 
Aside from the charms of NOS DACs, the only other 'reasonably' priced DAC that has my attention currently, is the forthcoming Schiit Yggdrasil.
 
Nov 18, 2014 at 5:23 PM Post #59 of 167
The Schiit Yggdrasil and Bottlehead DAC both seem to have a great approach to handling sampling and filters... I think DACs in the next two years should change quite a bit as these ideas trickle down. The ES9018 is great, but if an engineer knows what he's doing, he could make another chip surpass it. And if that same engineer is told to use the buzzword compliant ES9018 and isn't comfortable with it... problems ensue.
 
Nov 18, 2014 at 5:43 PM Post #60 of 167
Listened to the Kalliope at length and liked it, you could hear its refinement, agree with the comments about its intended market, it had that Gryphon house sound.
 
Did everything well, sounded processed and as though it was just doing its job, it didn't involve me emotionally.
 

Users who are viewing this thread

Back
Top