Hi Scumbria; I'd forgot that KZNerd was connected that way to Rambus. His posting series is great, but you will save bandwidth if you replace the "http://www.siliconinvestor.com/readmsg.aspx?msgid=" with "reply-" in your script output. I know that this makes your post no longer exportable out of SI, but for that many links, who cares.
KZNerd starts out as a Rambus long, but was incredibly level headed. For instance, like me, he said that PixelFusion was likely to go under. He says "no" in answer to the question "Can Rmbus use their patents to attack other dram technology?". #reply-11070748
He turned on Rambus early in September 1999, when he realized that the ramp wasn't going to plan:
Bilow, Sept 3, 1999
By the way, I think that the rambus fiasco will damage both INTC and DELL and, of course, eventually, RMBS, but the primary people getting hurt here are the users. Rumor has it that INTC upper management forced the RMBS decision on to the engineers, rather than engineers suggested it as a great idea. Several years ago, I posted something on this thread to the effect that there was no way INTC would ever tie its product line to a single memory technology. Too much risk of allocation and/or high prices.
KZNerd, in reply
Yes, I still don't understand why Intel has been so intent on ramming Rambus home so quickly. Why didn't they just provide an extra year of overlap by fully supporting sdrams until Rambus was proven ready for prime time. The stakes they are playing with are so huge! It is like as if they decided in 1997 that Merced would be ready in 1999 and stopped x86 work in 1998. #reply-11149607
Here's his reply to a long post of mine: #reply-11159938
After the announcement of the 4i RDRAM version, on Sept 6, 1999, he wrote: "As a rmbs long that quote by Tate has me really worried. If they are considering such a big change then they must be really worried too." #reply-11171358 also #reply-11182455
He agreed with us about the power consumption issues: #reply-11207719 . This is a great post: " No way, John. Rambus system power is much higher. They acknowledged ths to me several times. They make their claim by imagining some sdram system with enough extra chips operating in order to match their bandwidth and then counting the power of all the extra chips. Of course, no one would ever build such a system. ... But the Rambus claim of half the power is about as close to a baldfaced lie as you can get." #reply-11210114 . Dave B thought we were wrong: #reply-11207862 , but funny thing. It's mid 2001 and I still don't see any portable RDRAM based PCs.
He knew about the industry word on RDRAM: Sept 10, 1999 " I often think about the many people at the dram companies who privately denigrate Rambus and Rdram but who are unable to speak out publicly. I even know engineers at Intel who criticize the Rambus decision but would never dare say so publicly. You'd be surprised at how Compaq had been planning on following Intel's direction and switching entirely to Rambus but has reassessed that decision due to all the problems with Rambus and now may not intro any Rambus products this year. I think of Dell who always follows Intel and wonder why they are not switching 100% to Rambus instead of only 50%. (Maybe they believe Jay Bell's benchmark report -- just kidding!)
So now we come to Sony. If you looked closely at the Sony PlayStation II and talked to its architects and designers like I have you'd soon come to the conclusion that they just threw everything that sounded whizzy into it with no regard for cost or risk. That is why I've predicted on this thread that PSII will ramp to volume long after it is being claimed it will. #reply-11208088
-- Carl
P.S. One of my favorite posts of my own, about infinity, mucus and bugs: #reply-11358405 |
