Best overclocker and advice

[StormPC]

Yep...common in A64 S754 systems and is the reason S754 systems have trouble running at PC3200 speeds with all three slots populated.

[Rave]

Struggling to run with 3 or more sticks is a timing problem, it's nothing to do with "the shear volume of data its trying to process". It's common to all DDR platforms, a P4 system will struggle just as much as an Athlon XP or A64 system to run 3+ sticks at more than 200MHz. You need buffering of some sort to run more. That's why Intel is developing their 'fully buffered' memory, and it's one of the reasons why Opterons need registered RAM.

Rich :¬)

[StormPC]

Struggling to run with 3 or more sticks is a timing problem, it's nothing to do with "the shear volume of data its trying to process". It's common to all DDR platforms, a P4 system will struggle just as much as an Athlon XP or A64 system to run 3+ sticks at more than 200MHz. You need buffering of some sort to run more. That's why Intel is developing their 'fully buffered' memory, and it's one of the reasons why Opterons need registered RAM.

Rich :¬)

Um...no that's not true at all. You can't compare the memory controllers in the S754, S939 and S940 A64s because even though they are all on-die they are designed to work very differently. Each has it's own set of problems and limitations. The P4 does NOT have it's own memory controller and is 100% chipset dependant. Because of this the way memory is controlled varies greatly from motherboard to motherboard based on the chipset and implementation thereof by the manufacturer.

Either explain to us the nature of this "timing" problem that you believe is present in ALL platforms (man, talk about oversimplification) or stop pretending to know what you do not.

[Rave]

Um...no that's not true at all. You can't compare the
memory controllers in the S754, S939 and S940 A64s because even though
they are all on-die they are designed to work very differently.

Um, so what are the differences? Explain them to me, since you're so confident that I'm wrong. The S940 (Opteron) processors may have a slightly different controller since they need registered RAM (although I doubt it), but the controllers on S754 and S939 are identical save that the one on the S754 processors is only 64 bits wide. The actual silicon is (I'm 99% sure) identical, the only difference is in the packaging. Do you really think that AMD have similar but slightly different silicon for all four of their K8 processors? Of course they don't, the difference is only in how the chips are packaged up.

Each has it's own set of problems and limitations. The P4 does NOT have it's own memory controller and is 100% chipset dependant. Because of this the way memory is controlled varies greatly from motherboard to motherboard based on the chipset and implementation thereof by the manufacturer.

Um....no it doesn't. DDR is a standard, and so the way that DDR memory is controlled is the same no matter how the controller is implemented. Do you think that RAM modules have a little sensor on them that checks whether they're on a P4, XP or A64 system and then adjusts their settings appropriately? I can assure you that they don't.

Either explain to us the nature of this "timing" problem that you believe is present in ALL platforms (man, talk about oversimplification) or stop pretending to know what you do not.

Have a read of these then; he goes into the limitations of DDR (and DDR2) memory in order to make the case for the advantages of the FB-DIMM platform:

http://www.theinquirer.net/?article=15167
http://www.theinquirer.net/?article=15189
http://www.theinquirer.net/?article=15214

Rich :¬)

[StormPC]

By the things you say you obviously have more issues than experience. I am not a doctor so I don't think I can help you. This is the last response you get from me.

Good luck to you!

[schmunk]

This is the last response you get from me.

Thank Jeebus for that...

[Rave]

By the things you say you obviously have more issues than experience. I am not a doctor so I don't think I can help you.

Right, so is Charlie Demerjian wrong? Or would you rather insult me than accept that- unlike you- I have some evidence to back up my assertions? It's got to be one or the other, which is it?

This is the last response you get from me.

Suits me. On the other hand, if you keep offering people rubbish advice, I'll continue to criticise you.

Rich :¬)