[tremor] Physical implementation of tremor

[Segher Boessenkool]

Brendan Moran wrote:
> 
>>>> External busses are expensive, too.
>>>
>>> Depends on how much I/O you start with.
>>
>>
>> It is the actual bus that is expensive, not the amount of
>> traffic on it.  Expensive as in "money", and expensive as
>> in "Watts" -- expensive as in "slow" is no real concern for
>> this kind of devices (the bus is certainly not slow compared
>> to the cpu).
> 
> I did realise what you meant.  I was simply refering to the number of 
> I/O pins available on most FPGAs (believe me they have a lot).  Now, as 

Oh, FPGAs...  That changes matters.

> to expensive in Watts, that is still a problem, but all transitions in a 
> semiconductor are expensive in Watts.

Yes, but i/o is not "in" the semiconductor component.  Line drivers
take a _lot_ of power.

 > If a little I/O saves enough
> processing cycles, it becomes cheaper.

Yes -- but note the "if".

> Roughly what is the maximum memory requirement for decoding a Vorbis 
> file with Tremor, anyway?

Basically unlimited, with the current Vorbis spec.  I believe
the spec will be changed soon, to better accomodate embedded
decoders...

<p>Segher

<p>--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'tremor-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.