[tremor] Fw: Algorithm complexity

[Michael Smith]

On Monday 17 March 2003 23:52, Julien Margetts wrote:
> Sorry if I posted this more than once :)
>
>
> ----- Original Message -----
> From: Julien Margetts
> To: tremor at xiph.org
> Sent: Monday, March 17, 2003 12:46 PM
> Subject: Algorithm complexity
>
>
> Hi There,
>     I am interested in using the Tremor codec in an ARM based audio codec
> ASIC (which will support many other audio compression standards also)
>
> Is there any information available on processor / RAM / ROM requirement for
> Encode and Decode operations? Obviously these will vary from platform to
> platform, but ball park figures would be useful.
>

Tremor was originally written to support realtime decode (for cd-rate audio - 
44.1 kHz/stereo) on a 74 MHz ARM core of some sort. It's since been optimised 
a fair bit, and probably requires less than this now.

RAM requirements are relatively high - it could be done in very little space, 
but the current Tremor decoder uses more than the minimum - possibly as much 
as a couple of hundred kB, I think. I'm not entirely certain about that, 
though.

ROM requirements I really don't know about, but should not be unreasonably 
large.

> In particular, is the overall complexity of the decode algorithm of a
> higher, lower or similiar complexity to MP3 for example.

Computational complexity is slightly lower (I'm not sure how the current 
implementation compares - it's similar, I think, to typical mp3 decoder 
implementations.). Memory requirements are somewhat higher.

Mike

--- >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.