[Official Topic] AMD APU Kabini, Temash, Kaveri e... Carrizo

[Fottemberg]

http://vr-zone.com/articles/new-confirm ... 47455.html

Boh, sono voci che non mi convincono del tutto ...

[Mitch]

...e ancora VR-Zone (che se la canta e se la suona):

Update Aug 6: The AMD PR office in Taiwan said that there is no change to Kaveri’s schedule. It is not clear what that means.

This isn’t Kaveri’s first delay. By all accounts, the chip was supposed to be out by now but it was a no-show on the market after this year’s CES. Richland, which is overclocked Trinity by any other name, is now the stopgap to push the company through 2013.

AMD saying one thing to the press and another thing to OEM partners behind closed doors is not a tactic a company trying to win back confidence of investors should be engaging in. Investor newsletters and analysts had many favorable things to say about AMD as the company racked up wins with its APUs and got its intellectual property into the next-generation of game consoles.

Paper launches, product delays and not being truthful with investors can quickly erode confidence that was earned through design wins. At Computex, Intel’s big keynote was an admission that it was late to the APU game and was now playing catchup. In the David and Goliath narrative that has perpetually been AMD vs. Intel, this has the making of a well formed rock about to be flung.

Kaveri’s delay is a big deal not because the chip will be revolutionary upon launch — try asking a developer how to code effectively in hUMA as extracting parallelism in existing code is hard — but because it shows that AMD hasn’t outgrown its former self. CEO Rory Read has done some amazing things reforming the company, but it looks like you can’t un-teach an old dog its usual tricks.

http://vr-zone.com/articles/hold-for-pu ... 49389.html

[Fottemberg]

Insomma, lo sapremo solo quando sarà effettivamente uscito se Kaveri è posticipato o meno.

[Mitch]

Almeno il paper launch va fatto per la fine dell'anno, ne va della credibilità dell'azienda.
Se poi AMD ha cambiato totalmente i piani ed intende uscire anche con un Kaveri a 4 moduli (8 cluster) allora potrei anche accettare un ritardo di 6~7 mesi

[Fottemberg]

Così l'Am3+ diventa inutile.

[Sasha]

Così l'Am3+ diventa inutile.

a me non dispiacerebbe affatto
Un solo socket per dominarli e nel buio incatenarli

[Fottemberg]

[Fottemberg]

http://vr-zone.com/articles/amd-confirm ... 50308.html

Ufficiale, peccato. Kaveri Mobile ad inizio 2014 (febbraio), Kaveri Desktop in seguito (metà anno?).

[Mitch]

E' il contrario, Q1 desktop e H1 mobile ... consoliamoci con il fatto che le preview usciranno molto prima

[Sasha]

ah, confidiamo in Tom's Hardware

... o avete tra le mani già qualcosa voi?