GTA04 Block V4 / risk and possibility

Carsten Haitzler (The Rasterman) raster at
Fri May 2 10:20:12 CEST 2008

On Fri, 02 May 2008 09:04:40 +0100 Andy Green <andy at> babbled:

> Hash: SHA1
> Somebody in the thread at some point said:
> | On Thu, 1 May 2008 19:34:28 -0300 Werner Almesberger <werner at>
> | babbled:
> |
> | btw. a lot of the mpu need seems to stem fro our sleep and resume. i just
> | double checked - the nokia tablets (770/800/8100 don't suspend. they
> just go
> I am reminded of one of the best bits of Die Walkure, the ultimate wife
> nagging scene:
> ''...Nichts lerntest du,
> wollt' ich dich lehren,
> was nie du erkennen kannst,
> eh' nicht ertagte die Tat.
> Stets Gewohntes
> nur magst du verstehn:
> doch was noch nie sich traf,
> danach trachtet mein Sinn. ...''
> MSP430 and aggressive suspend are not dependent on each other, we can do
> some kind of aggressive suspend even without it.

no they aren't - but the msp430 is billed as the solution, which imho it isn't
- all the tasks you say you want to do on the mpu in practice will need the cpu
to wake up anyway to do them. my point is that there is an argument for the mpu
which in itself is stating all kinds of uses that are not useful. those should
not be part of the argument for it. touchscreen gesture recognition,
accelerometer handling, led blinking (can be done elsewhere), button lag
covering. arguments for it should be ones with weight.

> Where you see that nobody else did it yet as evidence we shouldn't do
> it, I see a great opportunity to beat their battery life significantly.

i see it as adding complexity to development, hardware design, interaction with
hardware (as you need to specially program the mpu to get a benefit - or you do
it the normal way with  the cpu and thus get no benefit). i just heard there is
gcc for it this morning which makes me much less nervous. i still see it as
unnecessary with minimal practical value. :)

> - -Andy
> Version: GnuPG v1.4.9 (GNU/Linux)
> Comment: Using GnuPG with Fedora -
> iEYEARECAAYFAkgayxgACgkQOjLpvpq7dMqpZACcD0pZ+45Q9j5lzDM7B9e+pzGS
> ALsAnRp+R65irGXSzM02Q4xSZA5TgBSZ
> =Gxfd

Carsten Haitzler (The Rasterman) <raster at>

More information about the hardware mailing list