Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Rambus (RMBS) - Eagle or Penguin -- Ignore unavailable to you. Want to Upgrade?

To: dumbmoney who wrote (31913)	10/10/1999 4:43:00 AM
From: Tenchusatsu	Respond to of 93625

Dumbmoney, <I wouldn't say insane, but...is there any end-user benefit provided by the new "hub" system? (aside from shaving a few pennies from the packaging cost)> Packaging is part of it. But HubLink also relieves the PCI bus. In the 440BX chipset, all south-bridge traffic had to be routed through PCI. Now in the 800-series of chipsets, south-bridge traffic has a higher bandwidth, more efficient interface available, which is the HubLink port. South-bridge traffic includes DMA transfers from the hard disk interface, USB traffic, Super I/O stuff like parallel and serial ports, BIOS off the FWH, etc. Of course, the demands of each south-bridge device isn't much. After all, how much bandwidth is a piddly USB device going to consume? But all those requests do add up. And the HubLink protocol is more efficient than PCI. (DMA requests from PCI typically look like this: "Is the data ready?" "No!" "Is the data ready?" "NO!" "Is the data ready?" "NO!!!") The only penalty of the new HubLink architecture is a longer latency for traffic to and from PCI devices, since the PCI is now hanging off the south bridge instead of the north bridge like in the 440BX. But the added latency is pretty insignificant, except in some contrived scenarios. Tenchusatsu