InvestorsHub Logo

mas

Followers 13
Posts 14434
Boards Moderated 0
Alias Born 01/08/2004

mas

Re: calbiker post# 26238

Sunday, 02/15/2004 8:54:30 AM

Sunday, February 15, 2004 8:54:30 AM

Post# of 97833
'
Keep looking. You're on record as lying. BTW, calbiker was on record as stating the TbredB was maxed out at 2700 or 2800+, long before it occurred. And that's exactly where it did max out. It's a power issue, not the short pipeline. The fact that overclockers were getting to higher frequencies shows it's not the pipeline. It could scale higher, but power was the problem.
'

Firstly you did say it and you are only spared being shown it now because of Aces crap lack of search facility as I am being forced to use Google to hunt your garbage down. However I do have it backed up somewhere and I will dig it out.

Yes the great Calbiker said the 2800+ was maxed out power-wise it does not make it true though. So you are relying now on overclockers now to back up your statements, how 'technical' :). The pipeline critical timing path is dependent on voltage and broad temperature gradients i.e. 5-10c steps. So when an overclocker cools his cpu to below freezing and pumps 2V through his processor he is decreasing the critical path timing limit which allows for greater clock speed. The 2800+ like all the Thoroughbreds and Bartons will *not* hit their 85c limit before running out of clockspeed headroom and thus are *not* Power limited.

The Prescott however is an example of a processor that *is* currently Power limited as it will hit its throttling limit (> 70c) before running out of clockspeed headroom. These are such basic fundamental principles of computer architecture that for you to deny them and peddle some other pseudo-science babble shows how dangerous a little knowledge can be. Stick to being a retired EE and leave CPU/Chipset architecture to the professionals who actually have to make their products work rather than blow hot air on forums.

p.s. so Eric Bron is a fool ? LOL, yeah right !
The great Calbiker also argued ad-infinitum that an IHS will *decrease* the temperature of a cpu as opposed to a basic cpu/heatsink interface. Shame that EVERY known instance of removing a 'heatspreader' from a P4 or K8 has shown a drop in temperature and boy were some of his wacky theories sure fun reading on this topic for a laugh. :)

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