Yummy new CPU/GPU combo

[Andy Green]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:
| On Thursday 05 June 2008 10:29:36 Andy Green wrote:
|> The Glamo offers normal async memory bus interface which we use, but it
|> has a bunch of timing constraints.  (There is a synchronous burst bus
|> mode that we don't use because the CPU doesn't support it and adding a
|> CPLD in there to translate will eat power and doesn't make sense.)
| What exactly do you mean by an async memory bus interface? Surely the bus
| has a clock to clock data in off the data bus?

Async busses use strobes like "chip select", "read" and so on instead of
an explicit clock.

| But it _should_ be an aweful lot faster than it currently is. I
understand the reasons
| why it's so slow, but these are hardware bugs. Surely SMedia are the
ones who should

When we figure it out we'll know.  It the meanwhile, it's OK.

|> However like I said when you hold the thing and use it, IMO it's
|> perfectly adequate.  The "August" Software Update ;-) has some cool
|> alpha and moving things going on that look nice, it's just not so bad as
|> you got the impression I believe.
|
| Have you watched the video which started this thread? Seen an iPhone
before?
| What about the latest Google android videos:
|
http://androidcommunity.com/first-live-images-of-fullscreen-android-demo-20080528/

I just jumped in to correct your wrong idea about graphic device
connectivity, not answer "is there a God?" :-)

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkhH5S8ACgkQOjLpvpq7dMpytQCfaKEEUEXfxWdZXpwoZBdZjAJ7
3RcAnA9G+6X1VyM5HvPz+M5tMwDoNgns
=LPl7
-----END PGP SIGNATURE-----