PDA

View Full Version : Prosessor speed...


Roswell_NX
12-16-2004, 7:48 PM
my HP Vectra has a 500Mhz Celeron and the actual speed ranges from 500.11Mhz to about 500.27, but mostly stays at 500.19.

Does anyone know why that happens and what controlls it? My other computers' clock speed doesn't change as much, and never goes higher that the processor is supposed to be running at.

Is that normal, or do i have bad capacitors on my board? i had a PSU die recently.

Thank you.

Roswell

mickwish
12-16-2004, 8:01 PM
How are you gauging the speed? Most processes the monitor the speed will be influenced by other factors. Slower CPUs are more impacted by these processes than faster ones.

Depending on how it is monitoring the core CPU speed, and what OS it is running, how much RAM it has, etc, etc, it could well be normal. :)

Cheers
Mick

Roswell_NX
12-16-2004, 8:11 PM
i have a little program called WCPUID and it has a clock speed monitor. I'm running it under Win 98, with 128Mb RAM

Roswell

Shyguy
12-17-2004, 12:19 AM
I wouldn't worry any at all about that, from what i understand most processors run slightly slower or faster than their Rated speed, for example my XP2200+ is usually referred to running at 1.8GHz when in reality it only runs at 1795.5 MHz(133MHz * 13.5 Clock Multiplier). & usually most CPU's are capable of running slightly faster than their rated speed, the manufacturers just give them a slightly lower Rated speed, or rate them at the next lowest class below the actual speed it runs, so if your celly is running at those speeds, intel probably just stamped it a 500MHz celeron. Also one of the reasons so many CPU's can Overclock so well these days, they are capable of running faster than their rated speed, but "detuned" a bit & sold at that rated speed, From what i understand of course!

if I'm wrong, let me know?

Win2Kuser
12-17-2004, 2:35 AM
Makes sence to me, why would Intel/AMD etc have a different die process for each type of processor, just have the one main process for the highest of that class, and de-clock some for the lower classes.

Most CPU's have the same basic core, but are slightly modified to fit other classes. A prime example is the MP range, they share the same basic core as the XP class of CPU, and most (if not all) XP processors can be modified into MP chips.

As for Ros's question, certainly nothing to worry about, if it was fluctuating by >+-5Mhz then I would say a problem was beginning to accur probably in the PSU, but the amount you state is just normal power fluctuations on the incoming mains supply.

Axel
12-18-2004, 4:46 PM
Originally posted by Win2Kuser
why would Intel/AMD etc have a different die process for each type of processor, just have the one main process for the highest of that class, and de-clock some for the lower classes.
.

because they want to sell to all markets and to anyone at any price they can make a profit at. If everyone in the world could afford P4 3.6 gHz processors and they were given out with happy meals at McD's - then that is all you would see on the market.

But as the best publicly available full instruction set processors go for several hundred dollars, that put's the bottom half of the economic populace out of that market - but here we have 2.6gHz celeron processors that 6 million people can afford which handles the needs of but the most dire speed freaks and gamers out there so let's sell 6 million of those and make a few $$$ off the volume.

Also - hey - they tooled a production line for celerons because ther was a market - there still is a market - so lets keep that line going. As for the new dies and new processors, we have to fund R&D or go out of business when the next manufacturer comes up with a process that cuts another 4 microns off the imprintable transistor so you can fit another 400 transistors on the waffer and claim you have a faster processor than the next wise-apple coming up the ladder.

I just wish they'd spend more time on better storage media. I'd love a solid state hard drive with no moving parts over 3 GB costing less than $16,000.00 - imagine a bootable drive so fast that they'd have to actually code a pause so you can see the microsoft logo.

It would be in a PCI form factor in a slot that went between the agp slot and the processor similar to what used to be known as a "hard card"- You'd have an SATA hard drive for your swap file and data, but the OS and apps would ride on the hard card for speed and stability.

And I can imagine a dual and a quad processor design on one processor chip so you'd have the same as a server class multiprocessor environment in a single physical processor slot.

Next I want an external USB 2 device with bays of memory slots that you could put all of your FLASH memory sticks into to make your own little fully scalable static hard drive out of with whatever spare flash rom you had laying around.

sorry - went rambling on - it's christmas time and that's what I want - always something that don't yet exist, but I can fully imagine.