/dev/chat

« first day (3948 days earlier) ← previous day next day → last day (989 days later) »

2:10 PM

When my computer fan is louder that can be a good thing and a bad thing at the same time, a quiet fan does not have a clear enough meaning. I need RPM on the fan to know.

Faheem Mitha

@prosody-GabVereableContext A louder computer fan is generally a bad thing, assuming it's regularly louder. It might need cleaning or replacement.

prosody-Gab Vereable Context

And sometimes a fan is too old or too new you know...

I know what you mean. Depends. Depends on which machine, kind of like body language but computer language too.

I do not overclock the hardware but the software sometimes changes the fan speed alone by itself. Kind of like a heart rate.

I know when Audacity is running a 1-hour mp3.

Just by listening.

Faheem Mitha

If it's louder only occasionally, that might be OK.

terdon

@prosody-GabVereableContext Good. That means it's working. It becomes louder because it is speeding up. If it didn't speed up when doing something CPU-intensive, that would indicate a problem. If it is speeding up when you aren't doing something CPU-intensive, that also indicates a problem.

prosody-Gab Vereable Context

I can not describe a car if it is in front of me. :P

Talk about a car engine and I do not know what you are talking about.

prosody-Gab Vereable Context

2:49 PM

If I understand correctly, at some level at or beyond bytecode there are physics ramifications to even a computer design. Some programming decisions are about the actual processor physically.
But once I pass the theoretical angle for cars, I still do not really know why an engine revvs, I have the cartoon version in my head of a car, I can not identify vehicles. Yet I can tell when my computer fan is sending me a message, and I can grok the fancy language in HTOP (with the fan speed, a health check) almost enough to make an educated guess what's happening at any given time. Like reading the…

Faheem Mitha

3:35 PM

@ibuprofen I was hoping you could tell me how you came by this graphic. I have memory problems here, and it might be helpful to see how much memory individual tabs were grabbing. Thanks in advance.

And what is the total memory usage for the tab? The very first line?

I see you answered this in the next line, i.e.

Jul 28 at 13:26, by ibuprofen

about:memory in FF

But how is that then for one tab?

top is reporting weird numbers. Like 37.6G for chromium. But I only have 16GB total.

2 hours later…

ToxicFrog

5:27 PM

@FaheemMitha is that VIRT, RES, or SHR?

Faheem Mitha

@ToxicFrog VIRT, sorry. I meant to mention that.

ToxicFrog

VIRT is virtual address space, i.e. the total amount of address space the program can access including memory-mapped files and devices, shared memory, etc.

RES (RSS, resident set size) is the amount of actual physical RAM being taken up the program; roughly this is the size of all of its executable code + the size of its data segments + the size of its stack + any memory it's dynamically allocated and not yet freed

SHR is the amount of memory it shares with other processes (including dylibs and shms)

So, e.g., a program that uses mmap() to map in lots of big files is going to have a very high VIRT but may have quite a small RSS.

Faheem Mitha

5:44 PM

@ToxicFrog Still seems surprising that VIRT is higher than actual RAM.

ToxicFrog

Why?

You've got 16 exabytes of address space to play with, might as well mmap() in everything and let the OS deal with paging it in and out for you.

Faheem Mitha

6:47 PM

@ToxicFrog So this doesn't map in any way to actual physical RAM?

Faheem Mitha

7:18 PM

I was using gist-paste a little while back, and I got a warning email, probably from Github, saying that some token was obsolete or something. I tried to update it per their instructions, but now gist-paste is broken. It's quite aggravating.

So I now need to fix it, on top of everything else I'm doing?

ToxicFrog

7:38 PM

@FaheemMitha no, just to address space.

Faheem Mitha

@ToxicFrog OK. I was not aware.

« first day (3948 days earlier) ← previous day next day → last day (989 days later) »

all rooms