Normally, I'd agree with this last statement -- by the time DDR systems show up in volume (if ever), there won't be any profit to be made on DDR, either....
Ever get the feeling that RDRAM and DDR are solutions looking for a problem, or killer app? If broadband streaming data was thought to be the killer app, it may be a long time coming. Dvorak has an interesting article in the current issue of PC Mag,
"The Myth of Broadband." (June 12).
Clip, "I have a megabit line into my home office, and when I view a streaming video feed, I still get the herky-jerky 20-Kbs stream. The true advantage of broadband is realized only on FTP sites or peer-to-peer, where downloading is optimized for speed."
Those performance results aren't surprising. If you compare the topology, those connected by cable are on a LAN, and those by telephone are on a "Star".
In a Star configuration the user sees the full bandwidth his connection is capable of making.
Those on coax cable are sharing already limited bandwidth, (the BW available beyond the TV programming). The users sharing the LAN to the local head-end are impacted by the number of users actively on-line. Basically, the amount of BW available to a user is inversely proportional to the number of users concurrently downloading. And being a LAN topology, there are issues of contention.
Until fibre is widely available, I think the problem can only get worse.
But, JMO's |
