Weiss AFI1 , Anybody using it ? Feedback?

[ISedlacek]

Wait 2-3 days, it is already on the way to me ...

[Adebar]

Originally Posted by rentedroom

I was looking to buy this firewire interface for Sonic Studio . Anyone have any experience witht the Weiss AFI1?

If it had 24 channels AES/EBU than it would be really interesting. For higher sample rates than 48k Adat is ... hmm, not my choice.

[livingstone]

got it,
great stability and sound !

[ISedlacek]

Originally Posted by livingstone

got it,
great stability and sound !

Just wonder how a simple straight digital AES EBU interface (without any convertor etc.) could have a "sound" of its own, that is distinguishable from the "sound" of other digital interfaces

In a review on this device it was also written"

"Sound Quality

The best I heard from any sound card, period "

( Weiss AFI1 firewire interface | Design w Sound )

[just.sounds]

When the converters are slaved it does!

But some algorithms could cause some harm.
In some interfaces audio goes through digital volume control and sample rate conversion.
Lots of mistakes to be made there.

Also crosstalk of clocks through earthing could have some effect.

[livingstone]

Originally Posted by ISedlacek

Just wonder how a simple straight digital AES EBU interface (without any convertor etc.) could have a "sound" of its own, that is distinguishable from the "sound" of other digital interfaces

In a review on this device it was also written"

"Sound Quality

The best I heard from any sound card, period "

( Weiss AFI1 firewire interface | Design w Sound )

Ivo, i was running my ad/da with a metric halo 2882 before and now with the weiss and i can tell you there's a clear difference,
once i tryed to clock the prism with it and it's still like this since !

[ISedlacek]

Originally Posted by livingstone

Ivo, i was running my ad/da with a metric halo 2882 before and now with the weiss and i can tell you there's a clear difference,
once i tryed to clock the prism with it and it's still like this since !

It means that you set Weiss as master clock ? So far with my RME AES 32 I have Forssell MADC set as master clock ... I am curious whether there will be any difference ...
But would different master clocks project anything into the recorded sound ?

How do you feel about the stability/bandwith of Weiss ? Do you work on MAc or PC ? What is the load you give it (number of channels) ? No drops, issues etc. ?

Mine should arrive hopefully tomorrow

[zenmastering]

Apparently, clock stability is important in digital audio, hence comments like Fredrik's (which I agree with.)

I suppose if you have a DAC that's impervious to such problems, it will be less of an issue. But anything that's in the signal path should be up to the task, including simple digital interfaces.

They aren't all perfect.

Graemme

Originally Posted by ISedlacek

It means that you set Weiss as master clock ? So far with my RME AES 32 I have Forssell MADC set as master clock ... I am curious whether there will be any difference ...
But would different master clocks project anything into the recorded sound ?

How do you feel about the stability/bandwith of Weiss ? Do you work on MAc or PC ? What is the load you give it (number of channels) ? No drops, issues etc. ?

Mine should arrive hopefully tomorrow

[ISedlacek]

Two days ago it arrived ... I installed it on my MacBook Pro (in OSX first). It was quite smooth, but first it requires some manual software settings (master clock, frequency, mode etc.). With Cubase under Leopard it worked well.

Then I switched to XP under bootcamp (my main intended area for working in Samplitude) and some troubles started. After proper installation, it just kept saying "driver error, device connection timeout". Simply did not work, although connecting and disconnecting the device was recognised. I contacted Mr.Weiss, tried also the new firmware but no help. I was told the bootcamp may be the reason (not very optimistic)

Then I installed it to my old Acer laptop. There it was installed, worked with Samplitude but when recording, some audible drops (lost ASIO buffers) were happening now and then (a bit weak CPU, RAM and tuning in general).

After that I got the following detailed message from Mr. Weiss' colleagues:

"We have seen a number of issues when using Boot Camp on Mac hardware.

High DPC latencies
Hotfixes from Microsoft should be installed
Incompatibilities with the LSI chipset.


DPC

The boot camp drivers and utilities introduce extreme DPC latencies,
specifically the Keyboard Manager (kbdmgr process) and the WiFi drivers.

When using time-sensitive software and drivers such as with audio
recording, stop the kbdmgr process and disable the WiFi adapter.


Hotfix

Macintosh computers often have an S800 1394B FireWire interface, and the
Microsoft drivers do not handle this correctly without installing a
hotfix (SP2) or making a registry change (SP3). The result without this
is that the driver always uses the interface at the S100 speed, which is
not adequate for high channel-count audio applications.

Windows XP SP2:

- Install the hotfix mentioned in the linked article below. If you add
another OHCI card later you must either re-run the hotfix or manually
fix the SidSpeed field for the new adapter. The first time you run the
hot-fix it will actually update the drivers as well, which fixes other
problems.

XP SP3:

- The SP2 hotfix will not run on SP3, and actually on SP3 the OHCI
drivers are fine. You do however still need to add the SidSpeed manually
if you are using an OHCI card based on a 1394b chip. The article below
explains where to add the registry entry.


See this article for details:
When using 1394 audio devices, users may experience "low resource" "device timeout" or other errors.

After you install the Windows XP SP2update, if you add a host controller for a new 1394 device to your computer, you must add or modify the SidSpeed entry in the Windows registry for the new 1394 host controller.

The hotfix: Performance of 1394 devices may decrease after you install Windows XP Service Pack 2 fixes the same problem on XP SP2 but the problem was reintroduced in SP3.

Note: Microsoft has assured us that they are working on a hotfix for this issue in SP3, but has not released it yet. Please check the Microsoft support website in case it has been released since this article was posted.

Otherwise, SP3 users can follow the instructions for manually changing the relevant registry setting in the KB885222 article above.



Incompatibilities

Some Apple computers, particularly the new MacBook Pro (from the 2nd
half of 2008) include an LSI (Agere) chipset which is not compatible
with our drivers:

Currently, the only workaround is to use an ExpressCard not based on the
LSI chip set. Like this one (http://www.byteccusa.com/product/xP-car ...
t-ecFW.htm)

We are currently working with LSI and Microsoft to solve this problem.

Note that our drivers work fine when running OS X on these computers."



After applying the mentioned hotfix, the installation of the device under XP bootcamp was suddenly completed (before the installation of the MIDI IO was missing) and the device was fully there and working, also recognised by Samplitude. Additionally I also stopped wifi, keyboard manager, IR, ethernet etc. in the XP device manager. The XP were already tuned to music as per the known tuning procedures.

It all resulted in grand finale: now everythig is working perfect, smooth and without any glitches. I even did test recording of 4 channels in 96k for more than half an hour - just perfect without any problem.

I mention all this in case someone faces a similar situation to show the successful way how to perfectly solve it.

Overall it works great now and for me it is not only a perfect solution for mobile recordings with MacBook Pro but also a complete and full power "studio backup" in case the studio computer would fail (it is a good feeling).

(BTW - after unpacking I was surprised that it does not have rackmounting possibility, but soon I found it was carefully packed separately and just needed a simple installation, which I easily did).

Forgot to mention that I am the lucky one who still has Texas Instrument firewire chipset on his MBP