Print

Print


Hi Tom, Konrad, and David,

I thought that you might appreciate some input from my world which is 
both Recording on the Mac Platform and Broadcast on PCs.

First a correction.  AES/EBE type I is balanced and 110 ohms.  Type 
II is unbalanced on 75 ohm coax.  Type 1 is very useful for long 
runs, and really doesn't seem to care much about impedance.  I, and 
many others have used Belden 8451, an analog balanced cable, for AES 
3 runs up to 100', and found it quite reliable.

More and more broadcast studios are going to Cat 5 for audio cabling. 
I've been running Cat 5 within studios and control rooms for the last 
20 months, running into Logitek Jetstreams in each room with Cat 6 
for networking between the rooms.  None of the problems that you 
might expect from unshielded cable have reared their ugly heads. 
Tightly twisted pairs seem to work very well, even without shielding. 
This is not surprising as this was the way that radio studios were 
wired fifty years ago.

SPIDF is limited largely by the current capability of SPIDIF hardware 
drivers.  Still, it has proven itself very reliable on runs of up to 
5 Meters, even in environments  rich in RF noise.  It is probably 
reliable beyond that, but why take a chance when a SPIDF to AES BALUN 
is cheap and effective.

When wearing my recording hat, I have made extensive use of Firewire 
400.  I make a point of buying good cables, generally from Other 
World Computing, a Mac specific vendor.  In a pinch I just look for 
the largest diameter cables, on the frequently true premise that 
large diameter conductors limit line losses.  That does seem to be 
true for digital as well as analog cabling.

I couldn't agree more about keeping audio outside of the RF rich 
environment  of computers.  External boxes for audio interfaces are 
always preferable.  In my small studio I use a MOTU 896, modified for 
better analog audio by Black Lion Audio for Jazz recordings.  For on 
location two track classical recordings, I have used a Digigram VX 
pocket PCMCIA card in an older MAC laptop.  Unfortunately Digigram 
has not made a card for newer Macs and PCs, as they feel that the 
newer card slots do not provide enough current for good mac preamps.

  I'm still  looking for a new two channel interface for a newer 
MacBook Pro that I have finally purchased.  I'm seriously looking at 
an Apogee Duet 2 for the task.  I'm somewhat put of by the the fact 
that they went from Firwire to USB 2 for the interface to the 
computer, due largely to bad experiences with USB 1 in the past.  USB 
2 is faster of course, but still not faster than even Firewire 400, 
due to the overhead that Konrad mentions.

Lastly, broadcast, at least radio broadcast, has stuck, with XP Pro 
as the OS of choice for PCs.  General unstability for audio work and 
problems with drivers is the reason, possibly along with the 
conservative nature of broadcast engineers.  Also in this world, the 
Lynx cards have largely replaced the CardDeluxe for two channel 
applications.  RME cards seem more oriented to mult-channel 
applications.

Just my $.02.  Take it for what it's worth.

Bob Cham



>Hi Konrad:
>
>This is very helpful. Thank you. My thinking was apparently based on 
>previous generations of technology.
>
>My DAW is circa 2009 and is somewhat old-school in that it's a 
>Windows XP (not emulated, real-deal XP Pro), uses PCI interfaces (a 
>linked pair of CardDeluxe) and all the external drives are Firewire. 
>It's rock-stable and works fine. When it dies, I will need to 
>rethink this whole MO.
>
>As far as Firewire cables and connectors, I've never had a problem 
>BUT all my stuff is wired up once and never touched after that. I 
>can see your point when it's a multi-user setting, but it seems like 
>the USB connectors and cables would be just as bad. Agree that "in 
>the box" Firewire cables can be quite cheapo. I've always used 
>Belkin after-market cables, which have thicker insultion and seem to 
>be well-constructed and well-shielded.
>
>In general, your statement about keeping audio outside the box is 
>very sensible. CardDeluxe happen to be well-build and have always 
>run dead-quiet in my systems, but I've had varying levels of noise 
>and hash with Soundblaster, Ensoniq and M-Audio PCI cards over the 
>years. It seems more dependent on specific computers' internal 
>environments than specific cards; a card that is noisy in PC 1 will 
>run just fine in PC 2, etc. I assume a Lynx or RME PCI card would be 
>designed and built as well as a CardDeluxe and would thus run 
>dead-quiet in any or most PC's. But, as I said, next time out I need 
>to rethink all of this in light of advanced technology.
>
>-- Tom Fine
>
>----- Original Message ----- From: "Konrad Strauss" 
><[log in to unmask]>
>To: <[log in to unmask]>
>Sent: Wednesday, August 10, 2011 9:46 PM
>Subject: Re: [ARSCLIST] digital input sound card recommendation
>
>>My problem with firewire is the cables and connectors. Especially the
>>connector which is easily dislodged, not a good thing when recording
>>audio. As Bob Olhsson mentioned, the build quality of cables is generally
>>poor. Both firewire and USB are capable of isochronous data transfer so
>>there is no reason why firewire should perform better. However Firewire is
>>peer-to-peer whereas USB requires the host to arbitrate traffic and since
>>USB will also have input devices (mouse and keyboard) attached, I suppose
>>it's possible that this could interfere with streaming audio. But if you
>>do the math, both Firewire and USB should be able to handle over 80
>>channels of 96-24 audio input/output, so there should be enough overhead
>>to fit a couple of mouse clicks in-between the audio stream. (For the
>>record real-world transfer rates are about about 25-30% below the stated
>>max)
>>
>>In general I feel it is a bad idea to have AD conversion in the computer -
>>it is full of EM interference and clock chatter. Try this experiment: take
>>a piece of 24 gauge wire and wrap a half dozen tight loops ~1" diameter in
>>the middle. Solder the ends to pin 1 & 2 of an XLR and plug it into a mic
>>input of a mixer, crank up the gain and hold the loops next to your
>>computer - all that crap you hear is also being picked up by your ADC. An
>>outboard unit will be better shielded and isolated.
>>
>>--
>>Konrad Strauss
>>Director of Recording Arts, Professor of Music
>>Indiana University Jacobs School of Music
>>http://mypage.iu.edu/~kstrauss
>>http://www.music.indiana.edu/departments/academic/recording-arts/
>>
>>
>>
>>
>>
>>
>>On 8/10/11 8:16 AM, "Tom Fine" <[log in to unmask]> wrote:
>>
>>Hi Konrad:
>>
>>Thanks for the information and clarification.
>>
>>So your preferred system is an internal AES/EBU card and do all the ADC
>>and DAC through external
>>units not integrated with the computer, so the computer is more a "record
>>and playback head" than a
>>full-fledged "tape machine"? I've thought about going this way with my
>>next DAW, not using the
>>computer for any analog interface, either way, but using it as a storage,
>>processing and editing
>>system all in the digital world.
>>
>>What is it about Firewire cables and connectors that you find to be
>>"dodgy"? I thought Firewire was
>>the more reliable interface for constant-stream data like digital audio
>>and video, that USB was
>>designed more for bursts of data like moving files between hard drives or
>>emptying photos off a
>>camera or sending a document to a printer.
>>
>>-- Tom Fine
>>
>>----- Original Message -----
>>From: "Konrad Strauss" <[log in to unmask]>
>>To: <[log in to unmask]>
>>Sent: Tuesday, August 09, 2011 9:30 PM
>>Subject: Re: [ARSCLIST] digital input sound card recommendation
>>
>>>+1 on the Lynx cards. We have six AES16 that have been in service for
>>>four
>>>years now. As with all computer systems they require tweaking from time
>>>to
>>>time as the OS is updated, but I've found maintenance to be pretty
>>>straight-forward.
>>>
>>>To answer a couple of Tom's points below, jitter is not a issue in
>>>digital-to-digital data flow. This is called interface jitter, and as
>>>long
>>>as the samples arrive at their destination intact no harm is done. (The
>>>he
>>>other kind of jitter is called sampling jitter, it happens during ADC and
>>>can cause major problems) So Firewire and USB interfaces are not
>>>necessarily bad, it really depends on the implementation. Personally, I
>>>dislike the cables and connectors, especially Firewire. The whole system
>>>seems dodgy to me. You can also run into problems if you want to use
>>>Firewire/USB DAC and ADC units from different manufacturers, often this
>>>is
>>>impossible because of incompatible drivers. I much prefer to stick with
>>>AES.
>>>
>>>The interaction between audio and video subsystems is probably caused by
>>>an IRQ conflics. My memory is a little hazy since I haven't had to deal
>>>with this for a few years, but occasionally PCI slots will share the same
>>>IRQ address so if the audio and video card are in slots that share,
>>>glitches can be put into the audio stream when minimizing windows or a
>>>monitor going to sleep. Usually the recommendation is that if this
>>>happens
>>>try swapping PCI slots. I think that in Windows 7 you can manually assign
>>>IRQ addresses but like I said I haven't had to deal with this for a while
>>>so I may be misremembering.
>>>
>>>--
>>>Konrad Strauss
>>>Director of Recording Arts, Professor of Music
>>>Indiana University Jacobs School of Music
>>>http://mypage.iu.edu/~kstrauss
>>>http://www.music.indiana.edu/departments/academic/recording-arts/
>>>
>>>
>>>
>>>
>>>
>>>
>>>On 8/9/11 1:46 PM, "Tom Fine" <[log in to unmask]> wrote:
>>>
>>>Hi David:
>>>
>>>That's a real bummer. Those are my front-line ADC's too.
>>>
>>>There has been much progress in recent years with external ADC's.
>>>However,
>>>I keep reading articles
>>>and papers about the USB connection not being ideal due to jitter and
>>>latency issues. And yet,
>>>firewire keeps getting de-emphasized by hardware manufacturers. I'm also
>>>not sure what role video
>>>circuitry plays in a modern Windows system. Even thru XP, there was some
>>>interaction between digital
>>>audio and the video subsystem, I'm not expert enough to discuss details
>>>but there are articles and
>>>white papers out there.
>>>
>>>If you're doing your analog-to-digital conversion externally and then
>>>"recording" an AES/EBU stream,
>>>it seems like you could use a very simple interface, but again I'm not
>>>sure how much clocking and
>>>latency matter in that setup.
>>>
>>>I've had some CardDeluxe cards since 1999 and they still work fine. Why
>>>not just recycle them as you
>>>upgrade computers, or are you building out more workstations and now need
>>>a new way to skin the cat?
>>>
>>>-- Tom Fine
>>>
>>>----- Original Message -----
>>>From: "David Seubert" <[log in to unmask]>
>>>To: <[log in to unmask]>
>>>Sent: Tuesday, August 09, 2011 1:38 PM
>>>Subject: [ARSCLIST] digital input sound card recommendation
>>>
>>>>This discussion has come up repeatedly over the years but I just learned
>>>>that the Digital Audio
>>>>Labs CardDelux has been discontinued. I'm not sure why they have been
>>>>discontinued, but we got a
>>>>call from the manufacturer who said that they are "obsolete." We have
>>>>long used their cards (we
>>>>are a PC shop) and I need to spec out another PC-based audio
>>>>workstation. I use the CardDelux for
>>>>the AES/EBU digital inputs.
>>>>
>>>>What are people using for digital input from an external ADC these days?
>>>>Is there an equivalent
>>>>sound card to the DAL card with AES/EBU digital inputs? I know some ADCs
>>>>offer firewire/USB
>>>>outputs, but I haven't really explored that.
>>>>
>>>>Please send any suggestions to me on or off list.
>>>>
>>>>David Seubert
>>>>UCSB


-- 
Bob Cham
K-TRUE
90.1 HD 2 & ktru.org
Houston