Jump to content

We thought we knew streaming


Recommended Posts

1 hour ago, peter@57m said:

FYI in case you haven't seen it

It’s a gas ring!

  • Haha 3
Link to comment
Share on other sites

FPGAs would offer Linn a great deal of flexibility. In theory you could model the Katalyst with FPGA code and this code is software updateable. Two very prominent products that use the FPGA approach are the PS Audio Directstream DAC and the Chord M-Scaler. In the case of the Directstream, it gets a new software update every so often that can totally change the sound of the DAC. If you do not like it after application, simply roll back (assuming the ability to do so is baked in). 

Now, if I looked at this correctly, a full stack Chord system with all the expensive bits would be less cash than what we are speculating here. So the move the FPGA is not driving the costs we have discussed even if it uses multiple FPGA chips. Expensive casework however could drive costs as well as how many units can realistically be sold.

More good news. Many of us have good sounding systems as they are and while we like to get new toys we most likely do not need new toys. 

  • Like 2
  • Upvote 4
Link to comment
Share on other sites

8 hours ago, Paulssurround said:

I am wondering what the benefits are of going from a DAC to an FPGA architecture, in terms of sound quality?

It could be none or it could be enormous. It depends on how well they code the FPGA and do the necessary physical changes to keep it quiet. Effectively, Linn would sit down and decide on what are the requirements for the perfect DAC, then they need to have tested those requirements, they follow this up by writing the specifications for the perfect DAC and cross check that if the DAC has those specs it will meet all the requirements, then you translate the specification into the hardware description and use a simulator to demonstrate the hardware will meet the specification. There are various Hardware Description Languages available. 

The only important thing is that there are no constraints imposed at the outset. The production DAC can perform exactly as Linn wants it to. Of course, Linn must not go beyond the capacity of the FPGA, but this is very unlikely and if they did they could simply use multiple FPGAs. 

The other aspect is it is not fixed in stone. It is Field Programmable - so if Linn makes an even better DAC next year, owners can download and install it. No hardware is hurt in the process. 

Having said all of that using FPGAs to create DACs and associated DSP is not exactly new. Lots of companies do it. Their implementations differ and no doubt Linn’s will be significantly different. There is no technical reason I can think of that would stop Linn adopting its FPGA DAC architecture across all its ranges. Chord Electronics has done this for years: the higher up the range the more accurate its reconstruction is, but it is mostly done in the DSP (filters) FPGA rather than the DAC FPGA. 

Edited by Nestor Turton
  • Like 5
  • Upvote 2
Link to comment
Share on other sites

8 minutes ago, Nestor Turton said:

It could be none or it could be enormous. It depends on how well they code the FPGA and do the necessary physical changes to keep it quiet. Effectively, Linn would sit down and decide on what the requirements for the perfect DAC, then they need to have tested those requirement, they follow this up by writing the specifications for the perfect DAC and cross check that if the DAC has those specs it will meet all the requirements, then you translate the specification into the hardware description and use a simulator to demonstrate the hardware will meet the specification. There are various Hardware Description Languages available. 

The only important thing is that there are no constraints impose at the outset. The production DAC can perform exactly as Linn wants it to. Of course, Linn must not go beyond the capacity of the FPGA, but this is very unlikely and if they did they could simply use multiple FPGAs. 

The other aspect is it is not fixed in stone. It is Field Programmable - so if Linn makes an even better DAC next year, owners can download and install it. No hardware is hurt in the process. 

Having said all of that using FPGAs to create DACs and associated DSP is not exactly new. Lots of companies do it. Their implementations differ and no doubt Linn’s will be significantly different. There is no technical reason I can think of that would stop Linn adopting its FPGA DAC architecture across all its ranges. Chord Electronics has done this for years: the higher up the range the more accurate its reconstruction is, but it is mostly done in the DSP (filters) FPGA rather than the DAC FPGA. 

There's nothing like having a subject matter expert on your team. Thanks!

  • Like 2
Link to comment
Share on other sites

1 hour ago, macfan said:

Now, if I looked at this correctly, a full stack Chord system with all the expensive bits would be less cash than what we are speculating here.

I have a Chord Choral system: Blu2 DAVE Etude and it cost less than half what is being speculated here. I am not aware that the Pulse Array DAC architecture differs between DAVE and Chord’e entry level Mojo. Though DAVE is 10x the price. DAVE has much better build with superb casework and power supplies. 

Link to comment
Share on other sites

5 hours ago, macfan said:

FPGAs would offer Linn a great deal of flexibility. In theory you could model the Katalyst with FPGA code and this code is software updateable. Two very prominent products that use the FPGA approach are the PS Audio Directstream DAC and the Chord M-Scaler. In the case of the Directstream, it gets a new software update every so often that can totally change the sound of the DAC. If you do not like it after application, simply roll back (assuming the ability to do so is baked in). 

Now, if I looked at this correctly, a full stack Chord system with all the expensive bits would be less cash than what we are speculating here. So the move the FPGA is not driving the costs we have discussed even if it uses multiple FPGA chips. Expensive casework however could drive costs as well as how many units can realistically be sold.

More good news. Many of us have good sounding systems as they are and while we like to get new toys we most likely do not need new toys. 

4 hours ago, Nestor Turton said:

It could be none or it could be enormous. It depends on how well they code the FPGA and do the necessary physical changes to keep it quiet. Effectively, Linn would sit down and decide on what are the requirements for the perfect DAC, then they need to have tested those requirements, they follow this up by writing the specifications for the perfect DAC and cross check that if the DAC has those specs it will meet all the requirements, then you translate the specification into the hardware description and use a simulator to demonstrate the hardware will meet the specification. There are various Hardware Description Languages available. 

The only important thing is that there are no constraints imposed at the outset. The production DAC can perform exactly as Linn wants it to. Of course, Linn must not go beyond the capacity of the FPGA, but this is very unlikely and if they did they could simply use multiple FPGAs. 

The other aspect is it is not fixed in stone. It is Field Programmable - so if Linn makes an even better DAC next year, owners can download and install it. No hardware is hurt in the process. 

Having said all of that using FPGAs to create DACs and associated DSP is not exactly new. Lots of companies do it. Their implementations differ and no doubt Linn’s will be significantly different. There is no technical reason I can think of that would stop Linn adopting its FPGA DAC architecture across all its ranges. Chord Electronics has done this for years: the higher up the range the more accurate its reconstruction is, but it is mostly done in the DSP (filters) FPGA rather than the DAC FPGA. 

Brilliant!!!

Thank you for the explanation of how FPGA’s work and how Linn would benefit from them.

Link to comment
Share on other sites

TBH I must confess to being a little disappointed with what we know of Linn’s klimax plans at this time. Hopefully the release info will fill in the gaps as just moving to FPGA DACs is a little old hat. As has been mentioned a number of other manufacturers use this architecture and is not necessarily a game changer in its own right - although you could argue that it’s the minimum you might expect at Linn’s price point. Developments in digital sound seemed to have moved on to network considerations such as; audiophile switches, music servers and various reclocking components. Now I haven’t investigated these myself but it does seem to be the direction that the high end is moving and being embraced by the digital enthusiast. It might have been interesting if Linn had embraced some of these aspects and offered products that were ‘additional’ to our systems as well just being ‘replacement’ products. However it could be that within these new products that Linn have encompassed all of the Super clocking / reclocking across inputs and exact outputs that render the requirement of audiophile switches and separate clicking mechanisms superfluous from a digital sound quality perspective. We shall have to wait and see!

  • Like 1
Link to comment
Share on other sites

This thread has become much like a battle between philosophers and theologians to prove God’s existence. The amount of wild speculation based on very little empirical evidence is highly amusing. The skullduggery so for is impressive: but are the results enough to get bent out of shape over? 😂😂😂

  • Like 4
  • Haha 2
Link to comment
Share on other sites

4 hours ago, Eldarboy said:

This thread has become much like a battle between philosophers and theologians to prove God’s existence. The amount of wild speculation based on very little empirical evidence is highly amusing. The skullduggery so for is impressive: but are the results enough to get bent out of shape over? 😂😂😂

We don’t know enough to be pessimistic, so I think positive 

  • Like 2
  • Upvote 2
Link to comment
Share on other sites

59 minutes ago, Paulssurround said:

We don’t know enough to be pessimistic, so I think positive 

I’m positive I can’t afford it! 😁

  • Upvote 4
Link to comment
Share on other sites

5 hours ago, SeeSaw said:

Developments in digital sound seemed to have moved on to network considerations such as; audiophile switches, music servers and various reclocking components.

Aren’t they more old hat than FPGAs? High quality switches were around last century. 
I very much doubt Linn feels it could make better network switches than HP, Juniper, Cisco or NetGear.
Or they could make a better music server than a Mac mini

If they could I think they would - was there not a Linn music server in the early days of streaming? 
How many folk use Kazoo server? 

FPGA DACs can be innovative though. If Linn feels it needs to do so. 
PS Audio transcodes everything to DSD, Chord uses PCM with a million samples. 
What makes a perfect DAC depends on the philosophy of who is creating it. 

However, Linn may have something entirely different planned. 

  • Like 2
Link to comment
Share on other sites

15 minutes ago, Nestor Turton said:

Aren’t they more old hat than FPGAs? High quality switches were around last century. 
I very much doubt Linn feels it could make better network switches than HP, Juniper, Cisco or NetGear.
Or they could make a better music server than a Mac mini

If they could I think they would - was there not a Linn music server in the early days of streaming? 
How many folk use Kazoo server? 

FPGA DACs can be innovative though. If Linn feels it needs to do so. 
PS Audio transcodes everything to DSD, Chord uses PCM with a million samples. 
What makes a perfect DAC depends on the philosophy of who is creating it. 

However, Linn may have something entirely different planned. 

17 hours ago, akamatsu said:

There's nothing like having a subject matter expert on your team. Thanks!

I'll say it again. :)

  • Like 1
Link to comment
Share on other sites

19 hours ago, akamatsu said:

There's nothing like having a subject matter expert on your team. Thanks!

Now I’m really baffled. I’d always assumed you could only contribute to a forum if you weren’t expert in anything!

I’ll really have to rethink this...

’troll

  • Like 1
  • Haha 8
Link to comment
Share on other sites

Clocking has come up in the discussion and it is a very important topic in digital audio. For example, the company Antelope Audio had  a line of DACs that used an external word clock input. This particular clock called the Antelope Audio 10M Rubidium Atomic Clock was said to be accurate to 1 second in 1,000 years. Teac also made a similar product called the CG-10M which also is a external clock. One of the main points in favor of these external clocks is the ability to have a power supply and environment with a stable temperature to support highly accurate clock signals not generally achievable from clocks that are generated within a single enclosure. 

The same approach can be said for power supplies as well. While elaborate, one could see a system where you have a DAC/Streamer in one box, the clock in another, and a external power supply that provides very clean stable DC to various devices. Consider a Linn system where you the external power supply supplying power to not only the clock and the DAC/Streamer but individual components within the streamer such as the digital, analog, and network components separately. You could also design this power supply as a step up from Radikal to power an LP12 and preamp. 

This is just an example of what could be done but would rather expensive, especially if you milled the enclosure out of a solid block of aluminum or some other exotic metal. Fun to speculate but think of all the fun one can have coming up with names to these new components. This is sort of like the F1 silly season where we just make stuff up till all is revealed. 

  • Like 1
  • Upvote 1
Link to comment
Share on other sites

15 minutes ago, macfan said:

Clocking has come up in the discussion and it is a very important topic in digital audio. For example, the company Antelope Audio had  a line of DACs that used an external word clock input. This particular clock called the Antelope Audio 10M Rubidium Atomic Clock was said to be accurate to 1 second in 1,000 years. Teac also made a similar product called the CG-10M which also is a external clock. One of the main points in favor of these external clocks is the ability to have a power supply and environment with a stable temperature to support highly accurate clock signals not generally achievable from clocks that are generated within a single enclosure. 

The same approach can be said for power supplies as well. While elaborate, one could see a system where you have a DAC/Streamer in one box, the clock in another, and a external power supply that provides very clean stable DC to various devices. Consider a Linn system where you the external power supply supplying power to not only the clock and the DAC/Streamer but individual components within the streamer such as the digital, analog, and network components separately. You could also design this power supply as a step up from Radikal to power an LP12 and preamp. 

This is just an example of what could be done but would rather expensive, especially if you milled the enclosure out of a solid block of aluminum or some other exotic metal. Fun to speculate but think of all the fun one can have coming up with names to these new components. This is sort of like the F1 silly season where we just make stuff up till all is revealed. 

Klock? :D

  • Haha 4
Link to comment
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...