Okay, I've been reading this Illustrated Guide to PC Hardware I downloaded, which is pretty good in explaining how all the different pieces of hardware work and stuff. When Im done (page 91 of 240 right now) though Ill have to read it again as its written by a person not a company so some of the info jumps aorund a bit, and at soem points you may not understand something that was said in a chapter until a few chapters ahead. Keep in mind this guide is a bit old. Probably from about 1998, so Im also asking this to check to see if any technology has drastically changed....

When reading about proccessors, it explained that proccessors used to work at one speed, back around the early '90's. Like if a processor was 25mhz, that was both the speed it proccessed all data internally, and the speed it sent the data out through the system bus. It then said that the problem with this was that as proccessor technology advanced, the external speed was becomming too much for the system bus. The system bus couldnt keep up, so I would imagine [it didnt explain this. this is what i came up with, so correct me if im wrong] this meant that the proccessor may cause problems with the external compenents, including thigns like extra heat perhaps.

So it then said that [intel i believe] introduced doubling clocking, which allowed the proccessor to run at its high proccessing speed, but it would regulate all data leaving the proccessor, to run at the speed of the system bus, therefore not cuasing any problems that would come about from the proccessor sending out data at its highest speed.

If I udnerstood correctly, [like I said sometimes it gets a bit confusing] it said that on newer proccessors, the internal clock speed is more then doubled, and youll get an internal clock speed of 3, or 4, or 4.5 now. Ide assume that means that the proccessor may have a capability to proccess at 600mhz, but if the system bus can only handle 100mhz at a time, and the clocking speed is 4.5, the internal speed would be 450mhz. Please tell me if Im wrong! [also, I would like to know what clock speeds proccessor run at today. Like I said, this thing is like 4 years old]

Now it didn't explain this, but I would imagine this is where overclocking comes in? Ive always known OF overclocking but never really cared to read much about so this is all a bit of a presumption made form what I read. So if your proccessor had a capability of 600mhz [yeah, kinda crappy today but its an easy speed to work with!] but was regulated to 100mhz on the system bus, and had a clocking speed of 4.5, you'de be stuck only using 450mhz of a potential 600mhz. So to utilize as much power as possible, people will pverclock proccessors. Basically turning up the internal clock speed, from 4.5 to 6. Therefore with 100mhz on the system bus, the proccessor wil clock up internally to 600mhz! Am I at least on the right track?

Also, it didnt explain a few other small thigns, and I was just wondering if someone could help clear these questions up as well:

1.f you do overclock a proccessor too much, is there a point where it will eventually cut off, and simply not run at those speeds, [a sort of safety feature] or will it just always try to run at whatever speed you tell it, pretty much self destructing if overlocked too much. [over heating I'de imagine would be the biggest priblem]

2. How would you go about finding [without some sort of manual for the proccessor] thigns like the proccessors default clock speed, maximum speed [before doing thingske getting too hot] and system bus speed? Ide imagine the system bus speed would be somewhere in the MB manual?

3. If a proccessor is said to run at 1.3 ghz [like mine] is that the POTENTIAL speed [like maximum before over heating] or the speed it will run on most MB's now a days, with its default clock speed. For the secodn too be true, Ide imagine that most MB's would have to run at roughly the same system bus speed. If so, what is that?

And finally, this one doesn't pertain to proccessors, but it still kind of confuses me:

4. When reading aboout I/O buses, I found myself a bit confused with what it said in the PCI section. It said that PCI was 32 bits wide, therefore Ide asume being able to handle a full clock tick, as opposed to ISA which is only 16 bits wide, so can only handle half a clock tick at a time, os each tick must eb spread out amungst two. It also said that though PCI only runs at 32 bits, it functions as if 64 bits wide. Now the confusing part...

It then goes and says that under optimal conditions it runs at 32 bits, but sometimes needs two clock ticks to send out a full 32 bit piece of data. Well call me stupid, but why the hell does it say it functions as if a 64 bit if it only even proccesses at 32 bit in optimal conditions, and often needs two ticks to send one ticks worth of data?

The only explination I can come up with is that because it said that PCI has a buffer, that the buffer can accept data at between 32 and 64 bits. Then the PCI bus would send the data to the expansions at 32 bits or somewhat less. This makes a bit of sense, because companies always do what they can to advertise their product at great. This way when PCI was first introduced, they could say it could 'accept data' at 64 bits per tick [which too me sounds unseless, seeing as how if a tick is only 32 bits, you truly dont need twice the speed, though I guess it has better potential for runnign at optimal speed then...] seeing as how that would be the speed the buffer accepted at, though the bus itself could only really distribute the data at 32 bits or less....

This could be WAAAAAAAAAY off and I think it is, but Ive got nothing else really!

SO, I hope soemone can help me out with a few of my inquiries!

Your post was copiously long and well thought out so here you go. (I would tell you to use the spell check, but it doesn't work)

I'm almost certain you're right about overclocking. That's how I temporarily booted my 800mhz celeron to something over 1GHz.
There was a limit- when I went too far, what happened was the computer simply didn't boot, and when I reset it, the BIOS popped up and said "naughty naughty, you've been a bad boy. I'm running in safe mode now till you fix me up". So I think on most new-ish motherboards the BIOS has a safety feature that does that.

The clock speed of a processor normally is with the guff that comes with the processor or motherboard. Motherboard site/docs would tell you all its guff.

My celeron runs on a 100MHz bus speed.
My dad's pIII runs on a 133MHz bus speed.
My mate's AMD runs on a 266MHz bus speed.
I've heard of AMD's running on 333MHz as well.
My other mate's P4 runs on 400MHz bus.
My mum's new p4 runs on 533MHz bus.

I don't know squat about PCI or ISA guff, but you're probably right. Search on google if it bothers you that much.

quote:

---

If a proccessor is said to run at 1.3 ghz [like mine] is that the POTENTIAL speed [like maximum before over heating] or the speed it will run on most MB's now a days, with its default clock speed

---

1.3 is a definate "safe-speed" for your CPU, but not the maximum "safe speed", or the maximum speed it will go before your computer won't boot . Yours will probably run without problems at 1.4 or 1.5 (possibly higher), but the CPU manufacturer decided to sell it at 1.3 because there is a very low chance that it will encounter problems at that speed.

To overclock stuff, sometimes all you have to do is change settings in the bios, other times you have to change jumper settings on the motherboard (sometimes both). With newer computers it's (usually) just a matter of changing stuff in the BIOS. Anyway, the stuff you've written on overclocking seems right to me (I'm no expert, but it seems right )

Well, good thing you mentioned the cpu ratio being in the bios on new MB's. I was lost here trying to find the jumper settings in the manual! Didn't even think to check the bios, but there it was!

oh, and just some useless FYI, I don't plan on overclocking my cpu. [well, I might if I buy some more fans, but Im not to worried about it] This is all just for info!