Processor P-states and power management

[Solar]

Performance can be measured in a multitude of ways. Instructions / second. Response time. Requests served / second.

But you claimed this:

In general, peformance in a typical embedded systems means that things get done in the timeframe expected, and that the user perceives the system as responsive. Often, neither throughput nor latency means anything. As in the examples I gave earlier with animations and sound. The only thing that means anything is if these things are complete in the expected timeframe. It is no advantage whatsoever if they complete before they are expected to complete, as this only adds idle-time to the processor.

That statement is simply ridiculous, and I say that in full awareness of your claimed track record in the business. Again, you are contradicting yourself even within the same paragraph.

A "responsive system" is a system reacting to messages in a timely manner, i.e. low latency. Then you say that "latency does not mean anything", which either means you're contradicting yourself or have another screwed definition of a standard computing term.

A system rendering an animation with a lower % of available computing power has more resources to either 1) be "more responsive" (i.e., not queueing messages in favor of rendering the next frame in time, but reacting right away), 2) doing the same job with less resources (i.e. on a lower-rated CPU, i.e. cheaper), or 3) do something else that was not on the "must have" list, but does a good job sweetening the deal for the customer (e.g., not only doing the financial transaction in the allocated time, but also displaying an on-screen animation instead of a simple static message).

A real-time system meeting its deadlines as expected is not about "performance", but about "performance / job = sufficient". Once your system meets its deadline, no further performance is needed, other than with regard to 2) or 3) above. But that doesn't mean that the definition of the term "performance" changes in any way, as you claimed. Your definition of the term "performance" is simply and objectively wrong.

But go ahead, I am sure you will again weasel your way around admitting to anything even closely resembling an admission that you are anything less than perfect.

[Combuster]

Then you should have learned that there is almost always more than one way to do everything.

THERE ARE 846 POSTS THAT SAY ITS NOT TRUE

Get. Out. Now.

[rdos]

I won't answer to any more OT posts here.

[Combuster]

Thank you.