AernaLingus

joined 2 years ago
[–] AernaLingus@hexbear.net 61 points 1 year ago (3 children)

"Owning a car = freedom"

"You need a big truck/SUV to haul things" (it's just a coincidence that people drove much smaller cars before a multibillion dollar deluge of advertising)

"It's consumers' responsibility to reduce plastic pollution by recycling, and recycling is effective" (whoever came up with this one belongs in the PR scumfuck hall of fame)

[–] AernaLingus@hexbear.net 39 points 1 year ago

Original Phoronix article which has all the individual benchmarks—weird that they didn't link to it

[–] AernaLingus@hexbear.net 12 points 1 year ago (1 children)

Ah, the halcyon days of DeepDream

[–] AernaLingus@hexbear.net 13 points 1 year ago (2 children)

showing a maverick side

Supporting the status quo = maverick

[–] AernaLingus@hexbear.net 3 points 1 year ago (1 children)

Totally beside the point, but is it just me or is that article written really strangely? Overall it has a kinda high school newspaper vibe, but in particular the attempts at humor are both awkward and numerous.

The information itself is good, though, so thank you for sharing!

[–] AernaLingus@hexbear.net 11 points 1 year ago

Archive.org version since HTTPS isn't available/working on the original site

[–] AernaLingus@hexbear.net 4 points 1 year ago

Full res (source)

The description:

Someone programming a website in HTML. But also a photo suitable for hackers. ;)

[–] AernaLingus@hexbear.net 17 points 1 year ago

There's a variable that contains the number of cores (called cpus) which is hardcoded to max out at 8, but it doesn't mean that cores aren't utilized beyond 8 cores--it just means that the scheduling scaling factor will not change in either the linear or logarithmic case once you go above that number:

code snippet

/*
 * Increase the granularity value when there are more CPUs,
 * because with more CPUs the 'effective latency' as visible
 * to users decreases. But the relationship is not linear,
 * so pick a second-best guess by going with the log2 of the
 * number of CPUs.
 *
 * This idea comes from the SD scheduler of Con Kolivas:
 */
static unsigned int get_update_sysctl_factor(void)
{
	unsigned int cpus = min_t(unsigned int, num_online_cpus(), 8);
	unsigned int factor;

	switch (sysctl_sched_tunable_scaling) {
	case SCHED_TUNABLESCALING_NONE:
		factor = 1;
		break;
	case SCHED_TUNABLESCALING_LINEAR:
		factor = cpus;
		break;
	case SCHED_TUNABLESCALING_LOG:
	default:
		factor = 1 + ilog2(cpus);
		break;
	}

	return factor;
}

The core claim is this:

It’s problematic that the kernel was hardcoded to a maximum of 8 cores (scaling factor of 4). It can’t be good to reschedule hundreds of tasks every few milliseconds, maybe on a different core, maybe on a different die. It can’t be good for performance and cache locality.

On this point, I have no idea (hope someone more knowledgeable will weigh in). But I'd say the headline is misleading at best.

 

This is one of my favorite videos of all time, perhaps even surpassing Watch for Rolling Rocks in 0.5x A Presses (although obviously not as iconic). It just perfectly scratches that itch of someone using a combination of technical skill and lateral thinking to complete an absurd video game challenge, all presented by an even-keeled narrator with clear visualizations and a bit of humor sprinkled in. I think I've watched it half a dozen times at this point--it's my "comfort food" for when I'm having trouble sleeping.

Kind of amazing that the very first video the creator has put out is such a banger, not just in terms of content but in terms of production value. He did say that he'd be making another video, so I'll be looking forward to that--tough act to follow, though, not unlike Watch for Rolling Rocks.

Can anyone recommend any channels/videos along the same lines? I've devoured the following:

view more: next ›