A lot of CVTs have virtual gears for that even, although others like the prius only have B mode.

Those are super rare though.

There can even just be manufacturing defects the FR4 weave that mess up SI that you might want to check for as a QA step on the assembly line. For hi volume that gets slow or expensive

For example: You add a dithering signal which can be processed out. If the signal has the right properties (for example, random but evenly distributed noise bounded to one LSB), you can then average out multiple samples to get more effective resolution than the ADC has. The additional number of bits scales something like 2^n samples, although if you don’t take sufficient samples this mainly just reduces your SNR. It also requires a periodic input.

However you can also pull similar tricks in the time domain or using simultaneous sampling with multiple ADCs. You can also interleave slower ADCs with a phase shift. This produces stitching artifacts unless you average them out though because ADCs generally are not well matched at the limits. You can bin or calibrate this out somewhat if you can characterize the error.

You can do a similar thing in the frequency domain if the ADC sample window is narrow enough but it has arguably the worst artifacts. Lo-pass the first ADC around N/2. The second ADC use a bandpass from N/2 upto N. The third is N upto 3N/2 etc… but the fourier transform will have a bunch of junk at the stitching points.

Or you can take the sampling scope approach using a fast but low sample rate ADC and many triggers.

I’ve seen most of these done on commercial instruments if you dig into the settings. Some of them you can see in normal operation (like the stitching in the frequency domain).

But I think the other poster was suggesting the first case applies - if you think about it there are certain periodic signals you can add instead of a random signal. That has the advantage of limiting SNR degradation and can also be filtered out easier/ detected i n the data.

Traces were highly predictable if you pressed it just gently past the detent. But if you did it too slowly (restricting the detent force) or used too much follow through (adding to the detent force) the result was very chaotic. Sometimes you could get the bounce period to extend by 10x if you took enough samples.

And if there’s too much current or other problems i saw decent ground and supply bounce too. That definitely makes it toucher to measure the behavior- in some respects a diff or current probe gave even more interesting results.

IDK, but I wouldn’t even want to debug circuits for this thing.

If you instead restore network access after it’s been offline long enough - depending on the exact process it will still accept the old password. Entering the old password isn’t enough to trigger domain check in. However, if I recall correctly entering an incorrect password will cause the login window to hang for 30+ seconds while it attempts to perform such a check in to see if your password changed in the interim. This will usually fail - but not always.

It’s probably bad behavior but it’s probably configurable in the domain settings. But it makes the user experience terrible because logging in gets super slow, because domain syncs in azure/ Active Directory are super slow.

The device itself usually has shielding, capacitors, transient suppressors, etc… as well as usually designed to make a poor antenna so on it’s own it will be affected much less than when charging

Surge protectors do work, mind you - but only for weaker storms or pulses coming in from the outside power lines. Just by physically being separated from the final device they are limited in how much they can protect from direct coupling

The entire rest of the government could be free and we would still be screwed if we don’t make cuts where it counts.

Early processors were typically 1 or lower. Modern stuff is all superscalar piplined and out of order and can do way more than you’d expect. Not to mention SIMD operations and other technologies. Branch prediction is probably better on the new chips too.

And with more RAM and cache algorithms can be chosen with different tradeoffs for less instructions.

16 cores at 4ghz was a thing like a decade ago - chips today might have the same specs but are definitely far faster.

Wireless tho so you’ll have to get over that one. Not a lot of options with usbc wired - unless you look at the custom or boutique space most wired keyboards have permanent cables. Battery easily lasts a year.

It’s kinda a weird list of requirements imo - most mac users are exclusively on wireless options. And even the wired ones only have usbc if they’re mechanical with like a teensy or something. Not that you’re wrong to want it but its unusual.

Not to every company per se but it’s been commonplace well probably for as long as business itself has been.

Just an example - nothing that happened in wolf of wall street was original to them - just the getting famous for being caught part. And that was only a few decades ago.

The defense and finance industries are famous for that sort of thing. I’m sure it’s pervasive elsewhere too.

There’s nothing special about software or tech or clouds that makes schmoozing impossible.

And if it’s resistive leakage, that’s almost a watt of dissipation - also huge and probably a short in a cable and potentially a fire risk.

5mA is a massive threshold for well designed devices

Wonder if they’re related?

But the problem is that over those ten years google docs mostly feels like it’s been enshittified. Today’s office suites all suck I just want office 2007 or the google docs beta experience back - those were snappy and obvious how to use. Even basic stuff like linear regression in excel is terrible UX these days

And so mostly they don’t care to push it and if they do it’s not a material over modern office anyway.

Could also be affecting the analog circuitry more if the droop is too bad or its browning out. That could be loss of gain, SNR, etc.. that could cause packet loss and retransmissions