Microsoft - Carbon negative by 2030?

Jan 17, 2020 13 Replies

On my drive home, I heard a quick piece on the radio about Microsoft claiming to be on-track to "Carbon negative" by 2030. And they're going to spend a Billion dollars to get there. (Wow, a whole Billion for a company that size!)



Link:

formatting link



My immediate thought was: Yeah, if Microsoft really wanted to be "Carbon negative", they would have built a better operating system! (Maybe I'm just getting cynical?)



Personally, I can't count the hours wasted (and energy burned) trying to get various Microsoft products to work. I'll bet the Carbon wasted just on periodic updates totally swamps any so-called savings they will ever achieve!



Your thoughts? Should make for an interesting "back-of-the-napkin" calculation.


No carbon in their distribution disks? No paper used at all? No black toner in their printers or copiers?

Not how it works; you can scavenge CO2 from the atmosphere, and pump it into suitable geological formations, at a few hundred dollars a ton. So, it's not zero usage required, it's zero net contribution to atmosphere.

Black toner is magnetite, not carbon black. There's enough to bend a sheet of printed output with a strong magnet.

Not to mention the 10^26 (*) CPU cycles wasted doing useless things (like polling) when it should have been in a wait state.

(*) finger-in-air estimate: a billion machines, 40 years, 1GHz

If M$ really wanted to be "carbon negative",they would lay off every programmer except the best three. Would make for a lot less drivers and guzzlene used. Also would lend to a very large improvement in quality of software and perhaps compatibility of products afterward. Would increase their profit by scads (massive decrease of expenditure in wages etc).

While Win85 used a busy loop in the null task instead of using a low power wait for interrupt instruction loop when no work was done. At least all WinNT versions used proper wait instructions and hence the power consumption was low when idle.

Mediocre PC hardware that executed Win 3 without problems did not work well when upgraded to Win85, not even after memory update. The computer crashed after being idle for a while and was then asked to do some work. With the original wait loop the current consumption fell during idle periods and when actual work was needed, the CPU current consumption increased rapidly. If the voltage regulation was bad, the CPU crashed.

To "solve" this problem, MS replaced the wait loop with a busy loop and the power consumption remains constantly high all the time and the voltages did not fluctuate :-).

This created a market of "power saver" programs for better motherboards, which executed a low priority wait loop process just above null task priority, which prevented the execution to fall down to the null task busy loop, thus saving power during idle periods.

They're intending to use equal amounts of carbon and anti-carbon, the details are still being worked on.

Cheers Clive

Except it is almost never really idle.

I run Windows (various NT versions - no longer interested in the old

16-bit versions) in virtualisation, so I see all the cycles it actually uses - which is far beyond the ones that Windows tools will tell you about. Some versions idle (with "nothing" happening) at above 30% CPU usage.

CH

Yeah, otherwise known as Windows 1.0.

Rick C. - Get 1,000 miles of free Supercharging - Tesla referral code - https://ts.la/richard11209

Never heard of it; do you not mean Win95 (and Win95 SE)?

.... so you're saying Microsoft actually has THREE good programmers?!

I think he's making the point that so many Microsoft products have problems. ...including the digit "9" not working on his Microsoft keyboard! :)

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required