Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[methylatedspirit]

And even then, is there a way to stress all the parts of the brain, sorta like a brain version of Prime95, to measure the brain's peak power consumption?

There's probably some drug cocktail or another that could manage it or something fairly close, sure.

Only problem is I'm pretty sure people just die before peak stress is actually achieved. Human brains don't exactly like being particularly overclocked, more or less. You can do it but the brainmeats don't work very well under too much pressure.

(I know this is a bit late, but since the previous post mentioned it...)

That's the equivalent of overclocking a Threadripper, running Prime95, and then wondering why there's smoke coming from the CPU, PSU, and motherboard after a few minutes. It's an extreme upper bound. I was thinking of something less... violent? Less "pouring LN2 on CPU", more "running at stock settings". At that point, the problem then becomes: what task will consume the most power? Social interaction? Math? Being tested on something? Is there a universal "hard task" for humans?

I dunno, I'm just guessing here. I don't know anything about these things.

[Naturegirl1999]

I watched the part I thought about. It was a misinterpretation.

How does one write binary code and have it executed by the computer? Like how we used to do before coding languages became a thing?

[Iduno]

I watched the part I thought about. It was a misinterpretation.

How does one write binary code and have it executed by the computer? Like how we used to do before coding languages became a thing?

They were still teaching Assembly and Binary when I was still in school. I assume it's the same now.

[dragdeler]

-

[Eschar]

You would toggle a bunch of switches to select a memory address in the computer, then toggle another set to select what value/instruction to place there, then some button to write that instruction in. Then you move on the the next memory address.

Disclaimer: I have no experience with this, it's just what I've read.

[Reelya]

Or punch cards.

In the early days people would also memorize the machine instruction codes. For example on 8-bit machines home users generally wouldn't have an assembler, you'd use commands to write numeric values straight into memory and could write your own machine code that way.

[methylatedspirit]

What is the oldest CPU that can run DF? "Run" here means it can generate a world (any world, any size, any parameters), and it can do a 1x1 embark. I've seen someone do it on an 800 MHz Coppermine (Pentium III-era) Celeron with 180 MB of RAM (dunno what version, but seems to be 40.xx?), so I'm wondering if it's possible to go even older. Does DF require any CPU instructions that aren't present in something older, or would it be theoretically possible to go all the way back to the OG Pentium or even a 486?

(I don't know myself, but I'm suspecting finding enough RAM to play DF with is going to be a struggle if one were to go this far back, considering that 180MB seems to be the absolute bare minimum to do a 1x1 embark on a tiny world with 5 years of history. Even that needed some careful trimming to get the OS to use as little memory as possible.)

[dragdeler]

-

[ZBridges]

In Destiny lore, if the Light is a force that is embodied by The Traveler, does the Darkness have a similar manifestation?

[methylatedspirit]

Is square (and more generally, rectangle) packing in a circle, like, a thing? Has that been well studied like its converse, circle packing within a square/rectangle? I'm asking because part of the process of making semiconductor chips is cutting square/rectangular dies out of a circular wafer, so I'm sure there's a decent practical reason to do research on such a thing. Is it actually a studied problem, or is it considered so trivial that it couldn't be studied?

All I can find about this is within a section within this page, and that deals with finding the minimal radius of circle needed to pack n unit squares into a circle, and the source it cites doesn't even say that are those necessarily the optimal packing for n > 2. I'm thinking of a related problem, which is finding the number of dies per wafer (an approximation is given in the link), given the dimensions of the die and the radius of the wafer, which should be fairly easy to convert to math terms (replace "die" with "rectangle" and "wafer" with "circle", and you're most of the way there). It doesn't seem to be a very popular problem, if anyone's ever studied it. It's a "computationally complex problem with no analytical solution" according to Wikipedia (no inline citation to go with that), but surely at least one person's at least tried or proved it as such. Am I not looking hard enough?

[bloop_bleep]

Well, given that silicon dies are rather small and wafers are rather large, I don't think the wastage from using the trivial solution is large enough to really justify studying using anything else.

[Reelya]

Is square (and more generally, rectangle) packing in a circle, like, a thing? Has that been well studied like its converse, circle packing within a square/rectangle? I'm asking because part of the process of making semiconductor chips is cutting square/rectangular dies out of a circular wafer, so I'm sure there's a decent practical reason to do research on such a thing. Is it actually a studied problem, or is it considered so trivial that it couldn't be studied?

All I can find about this is within a section within this page, and that deals with finding the minimal radius of circle needed to pack n unit squares into a circle

The issue is that you're not looking at the optimization of the entire process of creating chips there, just the optimization of one element. One issue you're not considering is that as well as packing a lot of elements onto the die, they need to be easy to cut apart, so you also want to minimize the number of cuts needed.



With a square grid you can just feed it through a slicer, rotate it then feed it through a different slicer. Sure, they could stagger or rotate the elements to cram more in, but how is the cutting machine going to actually work then, and would it cause more problems than it solves?

Having everything being square / grid oriented would just massively simplify every step of the entire operation. Materials themselves are cheap.

[bloop_bleep]

Yes, materials are cheap, especially if the excess can be recovered, which I suspect it could be.

[methylatedspirit]

What you're saying is that it's impractical and utterly pointless to do optimal packing on dies in a wafer. I get it. The thing that bothers me is that seemingly no-one looked at that, and decided to think about the abstract version of that, with rectangles and circles and making a bunch of simplifying assumptions. I know that's basically entering mathematical fairy tale land, but since when has practicality stopped mathematicians? What seems to be known about square/rectangle packing in a circle (a related problem) is sparse compared to its converse. You'd think that someone, after thinking about circles in squares, they'd start considering squares in circles. It's not exactly a giant leap.

I dunno, I just see an interesting problem that not many have tackled. I don't think I could be the one to start; I don't know enough to even know where to begin.

[Reelya]

EDIT: didn't read your point well enough.

Packing research goes back a long way.

https://link.springer.com/article/10.1007/s10898-018-0711-5

This paper considers the task of finding the smallest circle into which one can pack a fixed number of non-overlapping unit squares that are free to rotate. Due to the rotation angles, the packing of unit squares into a container is considerably harder to solve than their circle packing counterparts. Therefore, optimal arrangements were so far proved to be optimal only for one or two unit squares. By a computer-assisted method based on interval arithmetic techniques, we solve the case of three squares and find rigorous enclosures for every optimal arrangement of this problem.

So it seems that they needed computers to prove the arrangements were optimal. Hence why the papers proving them are all recent.