InvestorsHub Logo
Followers 24
Posts 15456
Boards Moderated 0
Alias Born 12/30/2001

Re: fingolfen post# 1764

Thursday, 10/17/2002 11:32:18 AM

Thursday, October 17, 2002 11:32:18 AM

Post# of 151694
Fingolfen -

Interesting... hopefully your analysis is spot on... would be great for Intel. It does, however, underscore something that I've (and many others) have been saying. The K8 core isn't going to be the core that lets AMD "switch positions" with Intel (as many AMD supporters have indicated), but rather what simply keeps them in the game...

As others point out, I may be a bit agressive in my estimate but we'll just have to wait and see. One thing AMD is banking on for Opteron's success is bandwidth. But HyperTransport isn't quite what it appears. First, I believe it is 8b10b, meaning that 8bits are encoded into 10 bits. The purpose of 8b10b is to balance the energy on the signal lines and it encodes the data based on the running parity sum. Second, I believe it requires a CRC to verify data integrity. So 6.4GBps is in reality somewhat less. Intel on the other hand uses a shared bus which includes ECC so no bandwidth is lost. At 533MHz a 64bit bus has 8*533MHz=4.2GBps. At 667MHz the bandwidth would be 5.3GBps and at 800MHz it would be 6.4GBps and because no bandwidth is lost it would exceed HyperTransport's bandwidth. Additionally, Intel's architecture doesn't require a hop from node to node for processor communication and all the ensuing complexity and validation problems. So I believe that AMD's claim of superiority is somewhat of a myth.

Can anyone else expand on this or correct me if I've errored?

EP


Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent INTC News