Re: Fwd: Re: Differences between constructive anddestructivecriticism

[Hiten Pandya <hmp@xxxxxxxxxxxxx>]

EM1897@xxxxxxx wrote:
> Processor speeds have "hit a wall", which means that they
> can't get much faster. So the best way to get more 
> performance is to increase support for MP. My point was
> that current 'BSD OSes don't utilize MP very well, so 
> the best price/performance is with UP. I find the empirical
> evidence somewhat undeniable, but apparently the notion
> is irritating to people. Why do I care if other people
> waste their money? That, I can't answer. 

	As for current BSD's not utilising MP very well; why, that is
	stating the obvious, again.  We, at DragonFly are trying to
	change this current dillema, and also bring other new things
	into the OS.

	Have YOU ever compared performance of a P4 and an Opteron system,
	both with PCI-X?  If so, you provide numbers, and then we'll see
	where we all stand.

> The "other" point, was that if I were to accept the premise
> Hope that clarifies things.


	EM,

	You are stating the obvious, countless times and trying to
	find and issue in something where one is not present.  No
	one is saying that the PCI bus is faster or equal in speed
	to PCI-X.

	PCI-X is something that has not undergone market acceptance
	yet, and that is what's meant by "bleeding edge".  Not many
	people have PCI-X in their systems.  AND before you say that
	your installed client base whatever is using it then you
	must think that you are at the centre of the Solar system.

	Having a fast bus is not a strange requirement and you are
	probably taking their words out of context.  But since this
	is hardware, I don't see what DragonFly can do about it
	apart from optimising drivers.  Having a fast bus is a user
	requirement and it has (more or less) shit to do with the
	operating system.

	On a side note, this doesn't apply to you alone; let us keep
	derogatory comments about someone or their work methods out
	of this and other lists.

		-Hiten
		Hiten Pandya