<<
I feel bad for all the poor sapps who got stuck with the orginal 845 chipset. They could have had this performance for the same price one month ago if Intel had renegotiated their contract with Rambus earlier. >>
Unfortunately this is just another one of those urban legends. There was no such thing as any "renegotiated" contract with RAMBUST, and any contract they have ever had has never prevented Intel from supporting other memory types. The FACT is that their contract required Intel to support RDRAM on a minimum of 20% of Intel's _total_ chipsets produced, in exchange for which Intel received options on RAMBUST stock (at a price that previously was very favorable for Intel, but currently worth far less due to RAMBUST's stock price flameout, that is, IF Intel hasn't already exercised them, in part or whole).
As for the new, not renegotiated contract ...
Intel and RAMBUST renewed their original licensing pact, plus added:
1.) Intel gets access to RAMBUST's entire patent portfolio for five years
2.) RAMBUST gets patent licenses from Intel for memory interface technology
3.) Intel will pay RAMBUST $10M per quarter (for the five years of the pact) as a licensing fee
As for the subject mobo of this thread ... how could they mess up an otherwise nice design with, as well noted by the review site, such poor locations for PS connector and IDE connectors?
<<
If you are looking for the most powerful P4 platform available, have no objections to Rambus? behavior, and money is no object, the P4+i850 remains the fastest combination you can buy. >>
That VH website, as well as most others, tickles me. While the objections to RAMBUST's antics are understandable, when are people going to finally take the blinders off and fess up to the fact that for a typical system (256MB), the cost "penalty", if it can even be called that anymore, is down to ~ 2 to 3% of total system cost. Sure, you can be tricky and compare just the RAM costs and it looks like anywhere from a 25-50% cost penalty, but when you look at the sum totals for all components in the two (ie. RDRAM vs. non-RDRAM) systems, that extra $30-40 memory cost turns in to a coupla percent overall. A good analogy would be the current rally in tech stocks since the Sept. lows. On paper it looks impressive, 30-50% gains were easily attainable, but that is because the prices for the base of comparison had sunk so low.
But wait ... maybe all these reviewers are actually helping ... if they can keep deceiving enough people from using RDRAM because of price, demand for RDRAM will shrink even more, stockpiles will increase even further, and RDRAM will end up being cheaper than both SDRAM and DDR SDRAM. On top of that, the fact that this would further erode RAMBUST's ever-shrinking revenues would probably please a bunch of people too.
<<
I'm still glad about my choice with the ECS P4S5A. Not only is it NOT a via-chipset, it is a P4 (478) with support for DDR 333 ram. >>
Yes ... nice ... but ... the way the price of DDR is rising, it soon could cost more than any other type of memory. Demand is gonna skyrocket even more, and that nasty old rule of supply and demand may very well bite us hard in the pocket.
