[alsa-devel] Infrasonic Quartet

[Vedran Miletić]

2008/11/12 Pavel Hofman <pavel.hofman at insite.cz>:
> Vedran Miletić napsal(a):
>>
>> 2008/11/12 Pavel Hofman <pavel.hofman at insite.cz>:
>>>
>>> Vedran Miletić wrote:
>>>>
>>>> Auzen started to distribute one cool new Envy24HT-S card, and it looks
>>>> really interesting to me. Specs are awesome, and it also has word
>>>> clock on card!
>>>> http://www.auzentech.com/site/products/infra_quartet.php
>>>> http://www.infra-sonic.com/site/p_PCI_quartet.php
>>>> http://www.newegg.com/Product/Product.aspx?Item=N82E16829156007
>>>>
>>>> According to envy24.svobodno.com, it has:
>>>> Infrasonic Quartet      0x3030  0x4953
>>>> Envy24HT-S - supported by ALSA
>>>> AK4620B (ADC / DAC) - not sure, probably not supported?
>>>> AK4112 - not sure either?
>>>>
>>>> However it has Xilinx on card in similar way as ESI Juli@ does, so
>>>> it's probably not a standard Envy24HT-S card.
>>>> Any ideas if this card could be supported?
>>>>
>>> According to the block diagram in the User's Manual the Xilinx plays a
>>> major
>>> role. The FPGA provides even more functions than in Juli.
>>>
>>> Without some information from the manufacturer (even if under NDA) the
>>> driver will be rather difficult to write. The FPGA is controlled by
>>> Envy's
>>> GPIOs and their signals would have to be electrically analyzed under
>>> Windows
>>> driver - a pretty complicated task. It is very different from a regular
>>> card
>>> with well-documented codecs only where you just need to trace codec pins
>>> to
>>> corresponding GPIOs.
>>>
>>> Pavel.
>>>
>>
>> First off, thanks for answering. I didn't expect anyone to answer,
>> since this card isn't this common yet, so it surprised me that it
>> caught someone's interest. Good :)
>>
>> Well, IMHO, this card is as far as Envy24HT-S can go. Sure you can
>> have more inputs/outputs, but at 4x4 and without breakout box that's
>> it, in terms of features. So, I didn't expect the design to be simple
>> at all. Not to mention unsupported DAC/ADC and DIT.
>>
>> Yet, assuming we get the specifications from Infrasonic/Auzen, would
>> you or anyone else be interested in coding a driver?
>
> Well, I could work on the driver, it would be similar to Juli. ice1724.c is
> mostly ready for the proprietary clock model. The aforementioned chips have
> complete documentation publicly available. If I had the card physically to
> trace connection and test (I am no Takashi the wizard :) ) and some
> documentation available, it should be viable.
>
> Honestly I am not going to purchase it myself, I have no other use for it
> plus PCI is almost history now. Unfortunately I have not met any PCI-express
> card with quality 44.1kHz playback (i.e. dedicated crystal clock).
>
> Pavel.
>

I do agree that PCI is almost history nowadays. Better said, it should
be. However, I see no viable replacement for Envy24 cards on PCI-E.
Oxygen and X-Fi just don't cut it. Anything else?

Of course, when it comes to Envy on PCI-E there is Tokyo Style S010
DCS-SEV24, however that card doesn't support ASIO and since it uses
PCI <--> PCI-E bridge it's performance (in low latency operation) will
probably suck. And also it's not aimed at producers, and it doesn't
even support ASIO in Windows. I believe there are no MIDI ports
either.

Therefore, out of all those "new" Envy24-based cards (since that's
basically what's available in this price range), Quartet stands out as
basically the best producer-oriented solution out there. Even more
than that - this is by far the best Envy24 card I have seen so far.
Almost like there is nothing to add :-) It's not available where I
live (Croatia), but it might be in the future since Auzen started to
distribute it, that's why I was interested in possibility of ALSA
driver for it appearing.

Well, I will contact Infrasonic in the coming days and let you know
how it went. Fingers crossed.

-- 
Vedran Miletić