• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

AMD and Intel

S

SidewinderX

2[H]4U
Joined
Mar 30, 2004
Messages
2,645
No, this isn't what you're thinking, I don't want to know which one is better. (I put this in General Hardware, because there isn't a generic "processor forum")

Ok, I have always been cuious about this, and I would greatly apreciate it if someone could either explain in detial, or link me to an article that explains the differnece between AMD and Intel processors (structurally). IE why a 3Ghz Intel chip is comparable to a much lower clocked AMD processor.


I know the general differences... but the most technical thing That I know about it is that for every "cycle" and Intel processor does 4 instructions, and and AMD processors does 6, which is why the lower clocked AMD processors have similar performance.


But obviously, I don't know the real terms and whatnot, and I'd like to.


Thanks in advance!
 
Since no one else has taken this, Ill explain as best I understand. A processor performs work on something. Many programs/devices need work to be done for them in order to function properly. The problem was only would thing could have work done on it at once and this caused the others to have to wait. So they invented the pipeline. Instead of having the work done at one time they split it into steps. The same amount of work is done but they do it in a predefined number of steps which is the length of the pipeline. Say Prog A needed work to be done, it could start the work at step one, perform a small amount of the work and then move him to step 2 where a little more of the work is done, and so on until all the work was done. All the stages take the same amount of time to complete. The benefit to this is that after Prog a moved to step 2 if Prog B needed work done, it could begin at step 1. This way the processor is performing work on many items. The drawbacks are if somewhere along the steps if say the result of the small amount of work caused the Prog A to need to change dynamically or branch, the pipeline would have to be cleared and all data in it would have to start over. Intel with the Pentium 4 line opted for a very long pipeline, I think 25 stages, while AMDs use 15 I believe. Pentium 4 can then do more work at once giving them higher megahertz ratings, but when programs branch, they loose more data. Ex. An AMD and a Pentium have a program branch at the last step, reseting everything. The Intel now have 25 pieces lost but the AMD only has to redo 15. Intels have become very good at predicting program branching and they know raw megaherrz speed sales to the average consumer.

This is how I understand it, It could possibly be slightly wrong/misled. Im gonna review on my processor understanding and reupdate this. Hope it helps.
 
RancidWAnnaRIot said:
where are these pipelines??? in the CPUs ALU?
Click to expand...

Yes.
They are packed full of magic blue smoke.

Luck.......... :D
u=Tigerbiten.gif
 
You must log in or register to reply here.
Back
Top