jorticka

It goes deeper.

People learn over time that complex web apps work badly in Firefox, because developers mostly test in Chrome.

So they don't even bother trying it in Firefox.

RocksDB supports hashing at multiple levels (key, value, files) because Meta also realized the importance of integrity. It also supports verifying them in bulk.

Presumably filesystems built over rocksdb also support this.

What crazy thing will they think next?

Pineapple on pizza?

So stop paying then.

Today you can use it for 90 or 180 days.

But then you download a new one.

You couldn't sell a package as a cat toy.

Only if made by Apple.

It was mostly an old-school kind of wargames.

There was an article about a giant real 3D map in a room with all the commanders there discussing what each one has to do, contingencies, ...

The reason for the counter-offensive failure were multiple and complex, good or bad software would not have changed the result significantly.

I wasn't talking about dual channel, but something else. There are a number of ways to map linear memory space to multiple individual memory chips.

The recent Ukraine war shows that soldiers lifes are cheap - according to commanders.

So many soldiers on both sides died because of really dumb commander decisions, missing kit, political needs, that worrying about CPU errors is truly way way down the list.

This is for cheap tags.

Nike can afford a $1 crypto tag.

If it's consistent and persistent, wouldn't that classify as broken hardware requiring device change?

Even with 3 chips, if one is permanently wrong you are then left with only 2 working ones so no redundancy is left for further degradation.

> just compute 3x

That might be difficult if CPU is broken. How are you sure you actually computed 3 times if you can't trust the logic.

It would be easy for competition to reverse engineer the physical layout.

I think it's "secret" because nobody really cares.

Do you care for example how the CPU maps the linear memory space across physical RAM sticks and chips? I recently discovered that there multiple alternatives and BIOS settings which give you some control over this.

Like all stress tests, linpack will find some errors, but not all.

I had memory stability issues which would immediatly show under Prime95 (less than 1 minute) but pass hours of Linpack.

If you're not hard realtime 2 is enough, you just redo the computation.