![]() |
![]() |
#1 |
Jan 2003
North Carolina
3668 Posts |
![]()
I have a Dell Inspiron 5100 with a true P4, 2.53GHz, not an Mobile or a Centrino, etc.
Based on the documentation, I don't think the hw adjusts the clock speed when the room temperature rises (no a/c where I was staying out west for a week) The iteration times went up and down (nothing was running, I made sure of that) from a low of 0.083 up to 0.091. Can the clock itslef slow down as the ambient air temperature rises? |
![]() |
![]() |
![]() |
#2 |
Aug 2002
3·43·67 Posts |
![]()
No, it can only throttle the CPU, and if that happened your times would really slow down a lot...
The variation in iteration times you see might be due to things running in the background or maybe there is some paging/swapping going on... Set the output to 100,000 iterations per line (Or higher!) and see what you average... |
![]() |
![]() |
![]() |
#3 |
Aug 2002
2×101 Posts |
![]()
The throttling is in the chip itself I think, and it certainly should be in the P4. The first step down on my Celeron 466 is to 400, which is approximately the same percent that you're seeing as the variance on your iterations times.
|
![]() |
![]() |
![]() |
#4 | |
Aug 2002
3×43×67 Posts |
![]()
I think Trif might be right...
http://www.mersenneforum.org/viewtopic.php?p=2582#2582 Quote:
|
|
![]() |
![]() |
![]() |
#5 |
Jan 2003
North Carolina
2×3×41 Posts |
![]()
Thanks Trif/Xyzzy,
What I observed fits very well with the 12.5% off duty cycle. I have iterations set at 10,000 which displays the iteration times every 2hours+. I take all that to mean it switched from 100% down to 87.5% (or more) and back over a long period of time. That is really great that the P4 protects itself up to 135C. I wonder if the AMD XP1700 (and faster) do this? In terms of heat management, maybe the extra cost, in general, of a P4 is worth buying over the AMD (I'm an AMD fan, except for prime95 work, but nuggets like this may will sway me towards P4's). |
![]() |
![]() |
![]() |
#6 | |
Aug 2002
110010102 Posts |
![]() Quote:
So I'd go by the demands that will be made of the processor. If you need absolute thermal protection, buy the Intel. If you don't then the chance of having a freak accident with an Athlon and having to replace it doesn't outweigh the price differential that you'd have to pay for equivalent performance on the Intel (not counting SSE2 friendly apps like Prime95). |
|
![]() |
![]() |
![]() |
#7 | |
Apr 2003
California
1348 Posts |
![]() Quote:
|
|
![]() |
![]() |
![]() |
#8 | ||
Jan 2003
North Carolina
F616 Posts |
![]() Quote:
|
||
![]() |
![]() |
![]() |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
WTF ??? P-1 slowdown | PageFault | Software | 2 | 2012-03-01 21:53 |
XP 32-bit to Win7 64-bit slowdown | willmore | Software | 6 | 2009-11-07 05:33 |
Prime95 PM Slowdown? | Primeinator | Information & Answers | 10 | 2009-08-03 11:06 |
Mysterious Slowdown | MS63 | Hardware | 10 | 2005-12-03 23:38 |
WinXP P95 slowdown | sgrupp | Software | 21 | 2003-12-28 18:48 |