Jump to content
Computer Audiophile
The Computer Audiophile

The tragedy of FireWire: Collaborative tech torpedoed by corporations

Rate this topic

Recommended Posts

gmgraves   

 

15 hours ago, The Computer Audiophile said:

Interesting article about FireWire. I wonder how many of us still use FireWire DACs. 

 

https://arstechnica.com/gadgets/2017/06/the-rise-and-fall-of-firewire-the-standard-everyone-couldnt-quite-agree-on/

 

I have a FireWire ADC. I've always thought that it worked better than USB 2.0 for as a recording interface.

Share this post


Link to post
Share on other sites

Yup, Firewire was always a better technology than USB, but was doomed by wrong decisions. Not only for music, but also for plain data transfer. I certainly hope Thunderbolt won't go that same way...

Share this post


Link to post
Share on other sites
firedog   

FW was better than USB until a few years ago, when some of the newer USB tech came on the market. I used FW till I tried a few USB "thingys" and they made USB equal or better than FW.

Share this post


Link to post
Share on other sites
esldude   
18 minutes ago, firedog said:

FW was better than USB until a few years ago, when some of the newer USB tech came on the market. I used FW till I tried a few USB "thingys" and they made USB equal or better than FW.

I concur.  Even with USB 2.0 rarely is there really an advantage of Firewire anymore. 

 

I do have a TC Impact Twin Firewire ADC/DAC.  Rarely use it now.  Would let it go really cheap for such a good device. 

Share this post


Link to post
Share on other sites
On 6/24/2017 at 9:48 AM, The Computer Audiophile said:

Interesting article about FireWire. I wonder how many of us still use FireWire DACs. 

 

https://arstechnica.com/gadgets/2017/06/the-rise-and-fall-of-firewire-the-standard-everyone-couldnt-quite-agree-on/

 

  Not currently using one, but still have a couple. I never have tried one of the USB dongles, but FW just seemed to offer better sonics than USB. From an Apogee Duet, TC Impact Twin, and several KRK Ergo's I have been very pleased to stick with FW.

  Bought a highly specced USB dac with dual 9018 chips, clocks, and naked resistors. The KRK Ergo with some board work easily bested it. Maybe it was the USB implimentation.

  I saw an update from Metric Halo a long time ago where it said they could get better performance from USB than FireWire. A newer version of their interfaces was right around the corner with USB buss and no FW. They may have been speaking too soon. Now they are working on a TB3. 

Share this post


Link to post
Share on other sites
gmgraves   
On 6/24/2017 at 11:19 PM, foodfiend said:

Yup, Firewire was always a better technology than USB, but was doomed by wrong decisions. Not only for music, but also for plain data transfer. I certainly hope Thunderbolt won't go that same way...

 

When High-definition TV was brand-new (circa Y2K), I had a Panasonic 1080 X 1920 high-definition VHS recorder that worked in conjunction with a Panasonic HD "tuner" to allow one to record over-the-air HD broadcasts. It had component outputs (PbPrY) to interface to an HD "ready" TV and a FireWire interface between the tuner and the VHS recorder. The digital playback of the HD recording played the HD tapes back through the tuner to interface with the TV. So the only connection between the recorder and the "monitor" was through the FireWire cable between the tuner and the recorder. BTW, the recorder didn't record video, it only recorded the HDTV data stream. Alas, a voltage surge (caused by a squirrel meeting his maker across the neighborhood pole transformer) sent my mains voltage from the normal 120 vac to over 240 volts early one Sunday morning! It fried my alarm clock, the HDTV tuner, I lost a Phillips video surround processor, a studio JVC S-VHS recorder and a C-band satellite receiver! The electric company compensated me for the losses but I was unable to replace the HDTV VHS system, as Panasonic had stopped making it. :(

Share this post


Link to post
Share on other sites
gmgraves   
On 6/24/2017 at 11:55 PM, esldude said:

I concur.  Even with USB 2.0 rarely is there really an advantage of Firewire anymore. 

 

 

I respectfully disagree. FireWire doesn't require the shenanigans and workarounds that USB has to use in order to get it to work as an acceptable audio transfer medium between computer and DAC (or ADC). Even with Gordon Rankin's "Streamlength" Asynchronous technology, USB still has too much jitter. It's not a matter of bandwidth, it's the way USB handles traffic that is inferior to FireWire.

Share this post


Link to post
Share on other sites
5 hours ago, gmgraves said:

 

I respectfully disagree. FireWire doesn't require the shenanigans and workarounds that USB has to use in order to get it to work as an acceptable audio transfer medium between computer and DAC (or ADC). Even with Gordon Rankin's "Streamlength" Asynchronous technology, USB still has too much jitter. It's not a matter of bandwidth, it's the way USB handles traffic that is inferior to FireWire.

 I agree George. BJ Burkhalter of Metric Halo told me that every FW interface he has seen streams in Asynchronous protocol. He said it is native. 

  But the huge advantage USB has over FW is that people are still releasing USB buss equipment. I do not know the date of the last FW connected product release. But with FW ports only a memory on new computers there will never be another. 

  My Mini has both FW800 and TB busses. The TB connection with a TB/FW adapter yields better sonics than straight FW. 

  So if someone has a FW interface they like, try on a computer with Thunderbolt buss and the Apple adapter. This will extend the usable life of the interface also. Plus maybe be better performance.

  Far as I know I am the only poster here using a Thunderbolt connected interface. It is an intermediate step. Waiting for a four channel interface with TB3 connectivity. 

  

Share this post


Link to post
Share on other sites
jabbr   
7 hours ago, gmgraves said:

 

I respectfully disagree. FireWire doesn't require the shenanigans and workarounds that USB has to use in order to get it to work as an acceptable audio transfer medium between computer and DAC (or ADC). Even with Gordon Rankin's "Streamlength" Asynchronous technology, USB still has too much jitter. It's not a matter of bandwidth, it's the way USB handles traffic that is inferior to FireWire.

In reasonable DAC implementations, the FireWire, USB or Ethernet packets are accepted using the interface clock  and then transferred to a FIFO. The bits are clocked out if the FIFO using the DAC clock. If there are differences in jitter at the DAC latch level, there is another mechanism at play.

Share this post


Link to post
Share on other sites
11 hours ago, jabbr said:

In reasonable DAC implementations, the FireWire, USB or Ethernet packets are accepted using the interface clock  and then transferred to a FIFO. The bits are clocked out if the FIFO using the DAC clock. If there are differences in jitter at the DAC latch level, there is another mechanism at play.

 

Hi jabbr - Was this true of the early adaptive USB designs?

Share this post


Link to post
Share on other sites
jabbr   
25 minutes ago, The Computer Audiophile said:

 

Hi jabbr - Was this true of the early adaptive USB designs?

Haha ... well I carefully qualified my statement with "reasonable DAC implementations" :) 

 

...but the async USB solves only one class of problem, even if the clock of data coming into a DAC is exactly the same to the nth degree as the clock of the data inside the DAC, there remains the problem that the phase of the clocks is not synchronized and if you try to latch at a signal transition you get garbage (so the clock crossing FIFO is still needed). @Wavelength is obviously the real expert, but what I am saying here should not be controversial ... its really textbook FPGA stuff etc.

 

Its hard to know exactly what is done behind the scenes. People might do stupid stuff that then sounds suboptimal... there are, however, widely known rules of how to cross clock domains, as I said: clock into FIFO and clock out of FIFO and FIFO is made of dual ported memory.

 

So why should interfaces still be an apparently big issue? Are we all in a large collective hallucination as some in the hard core "objectivist" camp believe? 

 

I personally believe either of two things:

1) unreasonable implementations are more common than we think

2) there is some other as yet incompletely characterized factor e.g. ways that incomplete domain isolation allow "pollution" from the outside to creep into the DAC clock...

 

(I personally suspect the latter ...)

Share this post


Link to post
Share on other sites
4 minutes ago, jabbr said:

unreasonable implementations are more common than we think

 

It's amazing how common these "unreasonable" implementations are in the high end marketplace. Some manufacturers that are well respected and have huge followings online of people singing their praises, are actually cobbling things together and moving boxes out the door. I can't name names, but the behind-the-scenes stories are crazy.

Share this post


Link to post
Share on other sites
DavidL   

Still using an RME Fireface 400 as the audio interface for my Mac. Performance is fine so feel no urge to replace it. I had to buy a Belkin break-out box to continue using Firewire when I replaced my old Mac Pro with a 5K iMac tho.

I no longer use the DAC in the FF400 - I bought a Chord Hugo which is definitely superior.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×