How to display the current CPU speed?

That actually makes no sense. The speed varies according to load whether it's on battery or AC.

When it comes down to it, it never runs at the same speed, battery or not. So any "current speed" would depend on the load at the time it's measured.

My time today is finished, I have to travel again.

Last remark: your CPU is running at 2.2GHz, determined by the clock. It can processing slower by not using all cycles. There is difference between cycles used or not. The speed is 2.2GHz whatever you do. And that for the layman may seem lower "speed" when a lot of CPU cycles are not used.

Indeed C, and "speed" varying or not is irrelevant, and that is implicitly what you said. Unlike others who "claim" that speed is subject to load, which it is not. Throughput in terms of calculations made is subject to cycles used i.e. load.

Ciao, see you later when I am back.

Lex

Interesting Drew. I guess I always assumed the AC adapter would supply enough power. I wonder if it's still true with today's laptops. I don't wonder enough to pull a battery however.

Thanks for the links.

Back again after 1050 km by car.

I wrote: "Last remark: your CPU is running at 2.2GHz, determined by the clock. It can processing slower by not using all cycles. There is difference between cycles used or not. The speed is 2.2GHz whatever you do. And that for the layman may seem lower "speed" when a lot of CPU cycles are not used. " To which I have not anything to add. Dialabrain wrote to me in this thread:

"I'll try to explain this for you. Frequency = Cycles Per Second. Therefore, a CPU running less cycles per second is running at a lower frequency. See?"

Which demonstrates that he does not know about the hardware workings: the speed in GHz does not vary and the number of cycles per second does not vary: even if the CPU is idling there still are the same number of cycles thus the same speed, 2.2GHz in your case. The clock still ticks at the same speed. But the idling CPU is not using energy ( not using the cycles for calculations) and does not heat up (almost).

Dialabrain suggests that there are only "used" cycles. If that were true, his "calculation would be right: but the bus could not transport the used cycles and the CPU used energy is lost and nothing comes out. Summary: speed is not the same as load. And CPU speed (and bus speed are constant, unless there is "overclocking" possible in the system, which is not the case in your mac.

Sorry but you are 100% wrong. The reference clock speed is 2.2GHz in this case which is the maximum number of cycles per second this particular CPU can run. I didn't bother looking to see if it can be "overclocked". But it does NOT run at a constant 2.2GHz per second. At idle it's more likely to be running at 800MHz-1.4GHz.

One reason Intel and AMD use CPU Throttling is to use less energy to preserve battery life and to generate less heat.

Hopefully this will end your arguments…

CPU Clock Speed

From the link…

Dynamic Clock Speed Adjustments

Modern CPUs also aren’t fixed at a single speed, particularly laptop, smartphone, tablet, and other mobile CPUs where power efficiency and heat production are major concerns. Instead, the CPU runs at a slower speed when idle (or when you’re not doing too much) and a faster speed under load. The CPU dynamically increases and decreases its speed when needed. When doing something demanding, the CPU will increase its clock rate, get the work done as quickly as possible, and get back to the slower clock rate that allows it to save more power.

Not at all, too simple and only for layman. It is more complex when there are more cores than one, but in this article speed and load are mingled together too. CPU does not heat up when unused cycles pass. For energy or heating reasons it is not necessary to lower the frequency, the unused cycles just pass by. A CPU is not an independent item in a computer and a varying clock speed would not work because it is the bus speed that must be kept.

After you share your insight with Intel and AMD, You might want to let Apple know their Technical Note is flawed and there's no reason, according to you, that they throttle the CPU speeds of laptops when the battery is removed.

http://web.archive.org/web/20081204024530/http://support.apple.com/kb/HT2332

p.s. Don't forget to let them know they can't throttle the CPU.

I am back available again.

Quickly scanned your proof articles. None said anything about "slowing down" in terms of speed. They only talked about turbo possibilities. And saving power when not fully loaded.

The Apple article is three lines: When the battery is removed the CPU will/may go lower, you should not do that because you will/may loose information.

The CPU is one of the components that has to communicate to the other components in the system. These other system components are connected to the bus: the medium that transports all the communication. The process is discrete ("digital") and needs a stable clock. This clock is - must- be at the same frequency as the all other components or information is lost. When a CPU will slow down it can only work when the bus (and the other components slow down to exactly the same speed, thus that could be only in discrete steps and needs an enormous quality of precision. A step already would mean loss of information (unless the step speed is infinitely high). I know a little bit about these things (although it is old knowledge) because I was leading the development of medical equipment and needed computer control instead of old electronics. I was not allowed to use "computers" from other firms, and "my" firm did not have thier computers not ready. So I had my department design and produce a "technical" computer: it was at the time of the Dec PDP 3 and newer, and the one we developed was at that level.

The problem with these discussions is that one is inclined to think analogical instead of "discrete".

I only posted a link to one article. You can't even get that right. I also quoted a paragraph from that article which clearly states the CPU increases and decreases in speed.

FWIW, the DEC-PDP 3 never existed. It was a paper design only. I'm done with you. You are clearly delusional.

Edit: actually M.I.T did have a PDP-3.

Going on: of course things are even more complicated because there are now more cores in a CPU, and all components around a bus are more complicated and much faster. But the principle remains the same : clcok speed must be exactly the same everywhere around the bus, otherwise you loose information (check bits or not), and one speed change requires exactly the same speed change at every point in the around-the-bus-system (which is almost impossible).

I must apologise that I am so stubborn about not accepting sayings that CPU "can run at slower speeds" without explaining and then producing articles that are saying different things than is implied, when you read carefully. And that I am compromising perhaps this thread by seducing replies that are not really contributing to an answer on the question of the OP. By explaining about bus, cpu, caches, memory etc I hope that you realise that it is (almost) impossible to measure actual CPU and or bus speeds: it requires knowing the exact clock speeds, and it requires hope that everything works perfectly.

My few cents.

intel power gadget

Then there's nothing.