If the sources accounts are correct, that's a huge boost for AMD.
Last week AMD confirmed that IBM, with its DB2 database, was one of its partners.
And the same sources are optimistic about AMD winning deals with both Sin Microsystems and the Dell Corporation, although there's no signatures on dotted lines yet.
High powered sales people are visiting Sun on a regular basis, and there's a lot of interest in the Opteron, as we reported here some months ago.
Sun, apparently, is most interested in the kind of operations that Hypertransport can allow between the AMD 64-bit CPUs. Dell, however, is more problematical for AMD. The rivers of bad blood between it and AMD requires a bridge to be made at the highest levels. While there's a great deal of interest within Dell, it's the conditions of a deal, and their past relationship, that is creating the difficulties.
Dell also has a guaranteed lowest price from its old partner Intel, an arrangement that AMD would find hard to breach given the volumes of machines the PC manufacturer ships.
And more technical details have also emerged about the Hammer microprocessors themselves. The source claimed that the family will manage to max out at between 80 and 90 watts.
If this report is true, it should be contrasted with future Intel Pentium 4 designs next year - Taiwanese manufacturers are currently being given a 120W max for future clock speeds, although that could change when a 90 nanometer Prescott becomes available in Q4 next year.
The core of the Athlon 64 (Clawhammer) and the Opteron (Sledgehammer) are similar, but they're surrounded by different supporting logic, and by the hypertransport protocols. Hypertransport allows channels to be split which could have important implications for both memory channels and CPU-to-CPU communications.
As for memory support, it appears that AMD is more interested about DDR-II than DDR-400, which as we revealed last week Intel will now support. That doesn't mean AMD couldn't support DDR-400 but it appears to think DDR-II is a brighter prospect for the future.
As we suspected, earlier problems with Athlon this year are due to a speed path problem with the Palomino core - a similar problem that AMD had with the Thoroughbred A and B cores. It was this type of problem - now solved - that caused the delays we were reporting here in September and October. µ
Sign up for INQbot – a weekly roundup of the best from the INQ