• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Bug in AMD EPYC "Rome" Processors Puts Them to Sleep After 34 Months of Uptime

Not sure you mean “errata” here, but this editor is wonky as hell
a bug in CPU's is known as errata, from the latin erratum meaning a printing error

It's in the first line of the news
1686198170550.png



It's an unexpected bug but also non-critical, since it's very very easy to plan something as simple as yearly maintenance, let alone 3 yearly
 
Right… the news piece refers to an error in documentation, not a processing error in and of itself, but you bothered to look up the Latin so I’m sure you know :)

 
It's semantics :)

A flaw in a processor design is also designated an erratum in tech jargon. It may or may not be correctable, and their severity can range from low to extreme, usually errata which are low severity such as this one are documented but no fix is issued or planned. In these cases, the chipmaker documents the problem and suggests a workaround.

Occasionally higher severity issues are also not fixed (such as Milan's USB stack reset problem), or the fixes come at a performance penalty (eg. Intel's fixes for speculative execution exploits). Very rarely, erratum of extreme severity result in designs being cancelled or permanently recalled. These rarely get past ES stage, but it can occur.
 
Right… the news piece refers to an error in documentation, not a processing error in and of itself, but you bothered to look up the Latin so I’m sure you know :)

You're arguing about quite an odd thing.
It's used correctly in the news article, because that's what it is.
1686201169701.png
 
It's semantics :)

A flaw in a processor design is also designated an erratum in tech jargon. It may or may not be correctable, and their severity can range from low to extreme, usually errata which are low severity such as this one are documented but no fix is issued or planned. In these cases, the chipmaker documents the problem and suggests a workaround.

Occasionally higher severity issues are also not fixed (such as Milan's USB stack reset problem), or the fixes come at a performance penalty (eg. Intel's fixes for speculative execution exploits). Very rarely, erratum of extreme severity result in designs being cancelled or permanently recalled. These rarely get past ES stage, but it can occur.
You’re still referring to documentation, or an appendix, not the processor’s error itself :)
You're arguing about quite an odd thing.
It's used correctly in the news article, because that's what it is.
View attachment 299863
Like I said, the article used the term correctly (so is Intel) :)

I don’t really want to do this again (see our debate about what a router is) — I agree to disagree, because who cares? Have a good night!
 
Last edited:
I don’t really want to do this again (see our debate about what a router is) — I agree to disagree, because who cares? Have a good night!
Oh of course. Sorry, forgot that if you get the last word in and say goodnight it means you're correct. Of course.
That thread became a dumpster fire and useless with all the useless arguing over what you call something, vs what it's actually called.

Maybe work on that.
 
No I’ll play at any hour, I was just offering you some outs and reminding you what happened the last time we got into “semantics” :)

Errata/erratum are not a “bug” in an electronic product, they are/it is an appendix to the documentation of said product.
 
Last edited:
No I’ll play at any hour, I was just offering you some outs and reminding you what happened the last time we got into “semantics” :)

Errata/erratum are not a “bug” in an electronic product, they are/it is an appendix to the documentation of said product.
Hopefully not adding fuel to the fire but to my understanding, in the specific context of technology, hardware/software erratum/errata is accepted lexicon for published acknowledgment of having or finding a "bug" or "error".
 
No I’ll play at any hour, I was just offering you some outs and reminding you what happened the last time we got into “semantics” :)

Errata/erratum are not a “bug” in an electronic product, they are/it is an appendix to the documentation of said product.
It's not 'playing' when you're told to stop derailing threads by a moderator and you think that's entertaining.
enoughs enough.
 
Back
Top