[iFi audio]

  iSilencer
iReducer
iReductor
iEM

 
Hi,

 

 
You need to post over here:
 
http://www.head-fi.org/t/695086/ifi-nano-ican-name-that-attenuator-competition-nano-ican-or-1-head-fi-rated-micro-ican-up-for-grabs

 

[Watcherq]

Radical idea: instead of using one of the existing DAC chips, how about writing your own using FPGA like what Chord Electronics is doing...

 

[obsidyen]

  Radical idea: instead of using one of the existing DAC chips, how about writing your own using FPGA like what Chord Electronics is doing...

That's the best idea on this topic. That said, it requires top class electronic engineers. What Chord Electronics have done is really, really hard.

 

[john57]

DP

 

[paulraulerson]

After listening to the iDDSD Nano for about a month, I am utterly sold on what you guys can do. It sounds absolutely delightful. I already have my credit card hot and ready for the iDSD Mini. 
 
So given that, what I *really* want is a iDSD/HDMI PrePro with about four HDMI input ports on it and one output port, as well as a USB input, and a great headphone amp. You might already be halfway there.  
 
It is a great nuisance to maintain two systems (okay, feed the output from a normal AVR to the HT input on a normal preamp) just to have good sound. Give me something that sounds as good as the iDSD and so long as it costs under $10K, it's mine.  
 
Nobody has really cracked that nut for the average person or the audiophile yet. Or worse, we few audiophiles who also listen to a lot of music! All most people really want is two channel sound anyway.  
 
-Paul 
 
Grin- build a remote and put a good headphone amp in too- for late night and private listening!  

 

[Watcherq]

 
That's the best idea on this topic. That said, it requires top class electronic engineers. What Chord Electronics have done is really, really hard.

I'm sure iFi engineers can match them   After reading what they did to the XMOS processor on the other thread, I've no doubt they are capable of it.
 
Given that Hugo is using a Xilinx Spartan-6 FPGA chip, iFi can use a higher end chip, like the Kintex-7 on the mini which has far less constrains compared to the Hugo in terms of battery life (none on the mini) and heat dissipation (much more on the mini).  An alternative is to combo it with boards like those from Parallella.

 

[obsidyen]

  I'm sure iFi engineers can match them   After reading what they did to the XMOS processor on the other thread, I've no doubt they are capable of it.
 
Given that Hugo is using a Xilinx Spartan-6 FPGA chip, iFi can use a higher end chip, like the Kintex-7 on the mini which has far less constrains compared to the Hugo in terms of battery life (none on the mini) and heat dissipation (much more on the mini).  An alternative is to combo it with boards like those from Parallella.

 
Sure, iFi have AMR's support. That said if what Chord did were easy, everyone else would have done it. Using a top line ESS or TI chip is much easier and cheaper, you don't deal with R&D. iFi would have to R&D for the best sound with their custom chip and that would take time and money. If they did it though, they'd be leagues ahead of other companies. There's a reason why a small portable amp/dac sounds better than most full size desktop amps out there.
 
That chip indeed seems better and pricier. I guess Hugo would be much more expensive if they used it. I wonder how much that chip costs.

 

[kugino]

I don't know much about FPGA but from the little I read about it, i like what's possible with it. I think the lynx Hilo uses FPGA as well...IF iFi went this route with the mini, wouldn't it then run closer to $2000-3000? Then we're getting into AMR territory. I'd imagine them implementing this in some AMR product first and then porting that tech to the ifi line of products.

 

[Watcherq]

   
Sure, iFi have AMR's support. That said if what Chord did were easy, everyone else would have done it. Using a top line ESS or TI chip is much easier and cheaper, you don't deal with R&D. iFi would have to R&D for the best sound with their custom chip and that would take time and money. If they did it though, they'd be leagues ahead of other companies. There's a reason why a small portable amp/dac sounds better than most full size desktop amps out there.
 
That chip indeed seems better and pricier. I guess Hugo would be much more expensive if they used it. I wonder how much that chip costs.

It may be straight forward to use existing DAC chips to do it, but OTOH, they have to figure out the tricks to the chips.  As for having a basic DAC on FPGA, there are already plenty of research and grad-school papers on them.  The advantages of using FPGA are that the manufacturer can implement all sorts of tricks and filters.  Hell, they can even sell custom filters and DSP and launch a third-party market with it.
 
As for the price, it would really depend on the quantities ordered and the other negotiation stuff.  Point of reference: I participated in the Adapteva Kickstarter.  They were selling the Zedboard which is a combination of a dual core A-9 ARM processor with the Artix-7 (successor to the Spartan-6) and Parallella's (the owner of the Kickstarter) own custom parallel chip for a princely sum of $200 (see http://www.adapteva.com/announcements/the-parallella-board-uncovered/).  By stripping out stuff that they don't need, I think they can keep the cost down.

 

[JWahl]

  Radical idea: instead of using one of the existing DAC chips, how about writing your own using FPGA like what Chord Electronics is doing...

 

   
Sure, iFi have AMR's support. That said if what Chord did were easy, everyone else would have done it. Using a top line ESS or TI chip is much easier and cheaper, you don't deal with R&D. iFi would have to R&D for the best sound with their custom chip and that would take time and money. If they did it though, they'd be leagues ahead of other companies. There's a reason why a small portable amp/dac sounds better than most full size desktop amps out there.
 
That chip indeed seems better and pricier. I guess Hugo would be much more expensive if they used it. I wonder how much that chip costs.

 
Yes, although it would seem like a good idea, it really does take a person or people who are very experienced in both programming and digital audio.  Not to say that iFi doesn't have those resources, they do have someone that does their xmos programming, but it appears they're strength is in knowledge of analog signals and the most cost effective approach is to apply that expertise using an already well performing off the shelf chip.  
 
Even Resonessence Labs, who is very experienced in the above, uses the same Xilinx chip, in conjunction with ESS Dac chips, for custom filters and other functions.  
 
I've recently finished a long term test of the iFi iDAC and iCan and have been quite impressed with them.  I also recently bought a Resonessence Labs Concero HP which IMO just edges out the iFi stack on it's own, at a slightly higher price.  I hope to eventually try the iDSD Mini and I hope to try the Concero HP as a USB Converter to test the difference.  Though I suspect the built in USB of the Mini iDSD will be very good on it's own. 

 

[Watcherq]

I agree more or less with the above posts.  However, iFI seems to want to break into new markets with the iFI brand and honestly, before they came along, there was nothing in the performance + form factor of the micro devices (eg iTube, iCan, etc).  They should continue to try pushing the edge with using FPGA since using them instead of standard DAC can be a competitive advantage.
 
Since they know the XMOS USB interface very well, they should continue to use it.  Keep the FPGA to render the D/A conversion, filters and DSP functions.  In the case of filters and DSP, they can even make them customizable/loadable, like the FIR filters in Foobar2k and DSP like in JRiver, etc.  This would allow unrivalled customization and personalization.  Of course, product stratification should come in.   May be the nano range allows no customization but the micro range allows for loadable filters and mini range, full customization.  They can move from lower end to higher end FPGAs in order to have more programmability, higher order noise shaper, and more taps (as per Chord's terminology), etc in order to sound "better".
 
The reason I participated in the Kickstarter I mentioned above was to experiment with the programming environment is to test out using the FPGA + special processor to be Proof of Concept (ignoring the noise in the circuitry) to be a music player.

 

[Watcherq]

Ok, my Kickstart board that I was talking about finally arrive last week.  Below are the pictures of it.  Note that it as 2 microUSB, a full size network port, microSD slot below and a micro HDMI port, I think.  This baby has a Zynq 7020 chip (the black one in the middle) that has a duo-core ARM and the Artix-7 FPGA (the next gen after the Spartan-6 on the Hugos).  The silvery chip is the proprietary co-processor that has 16 cores.  It has the other stuff for programming like GPIO on it too.  All these for $200 for a production run of a few thousand (less than 5000).
 

 
The actual size is actually a lot smaller than expected.  As seen below, it is barely larger than the iFI iDAC nano and the Fiio X3 next to it and all three are sitting on top of my Lenovo Helix which is the size of an A4 sheet of paper.  I'm sure in an actual production, a board that does the DAC-only functionalities will be smaller and cheaper.

 

[EraserXIV]

Curious about the power supply that is going to be used in the mini, from what I can gather it seems like it will be using an external power brick of some sort. Will it be of the linear or switching variety? Any other details you can divulge about it would be great too (noise, etc).

 

[EraserXIV]

iFi audio Any news on my previous question about the power supply?

 

[Vartan]

Will Mini iDAC come with integrated iPurifier?