well I could give you a solution, generally with x86_64 architecture they use a lot more wattage than macbooks, m-chip SOC’s (system-on-chip) utilize about 30W of energy whereas just a modern x86_64 CPU utilize 15W. which means you have a 15W overhead for your GPU and memory generally speaking.

So the entire reason your getting less battery life is OS required applications for it to function, and you. So if you minimize the amount of wattage (ideally building a linux system from scratch) you can optimize it to consume less resources.

I did this with my personal laptop, installed arch and mangoWM, didn’t even bother with a display manager or network manager (still use iwctl). on idle it uses about 600 MB, and I’ve beaten the m1. my point is not to compare or benchmark the macbook, but to just show you that you can maximize battery life with a little tinkering. So long as you are comfortable doing it.

I have used pop_os and cosmic DE it should be noted that is a beta version of pop_is, which means there are plenty of bugs, which means there are still a lot of optimizations. the fact you could get 12 hour battery is kinda surprising especially with a nvidia GPU.

so people reasonings are, that if you don’t use all memory then it’s just going to waste. which is true, but you also need to take into account the applications I use, ~10 GB of memory is used when i run vivado/vitis synthesis and bitstream.

the second thing i was going to add is, Do you know what also makes ram redundant? when the laptop is fucking dead. Which is ultimately why I’m trying to optimize it resource wise to extend battery life.

that said, i’ve optimized it to where I can beat the m1 macbook on idle. again, not something that is a benchmark, but to say an engineering student beat a trillion dollar company. I continue to smear that in apple fanboys faces as just a massive “fuck you.”

ontop of other user comments where it boils down to trust in the maintainer and code reviewers of the project, another reason depends on the use case that you plan on using your Linux system. for example, if I were to setup a nextcloud server, i’d generally go with alpine for it’s lightweight design, or Debian for it’s stability. I wouldn’t necessarily use Kali Linux, because with those features it also uses a lot of resources for it to function, and I don’t need that for a server.

in terms of my personal device I generally build those from the tty and add other modules (like DE, utils, etc…) to give it more functionality. Much like my servers I like to have my laptop optimized – take as minimal resources as possible – which is a rather controversial take after seeing users bash at me that I’m not taking advantage of all my memory. anyways, I don’t think there is a specific distro that has everything that I want. I want a system that works that doesn’t use 2 GB from the DE alone and that is accomplished by adding the modules myself.

I don’t trust any other sub-distro other than myself. I generally go with one of the corporate base tty installer (arch) and I build the system from that. I’m not going to switch to say cachy OS for it’s aesthetics, or ease of use, I couldn’t really care less.

I guess you didn’t read my last part, humans are deterministic by nature; they draw trends based on patterns. if you wanted to win with one ship just place that ship in a place that is completely counterintuitive to what they expect. Having one ship does have strategic implications, that it narrows the possibility of being hit. Combine that with counterintuitive placement of said ship and your opponent is as good a player as a fish.

so does probability play a role in battleship? yes, it removes the statistical likeliness of being hit (ignoring the premise of humans drawing generalizations by pattern recognition). It is drastically less likely to hit one ship than 5, all you need to do is place that one ship in a position that is counterintuitive to the trend to remove any biases of your opponent who has played several games already.

what are you going on about, that probability depends on the number of turns you get? last I checked it, no it does not depend on that.

the argument I’m trying to make is it’s statistically less likely to hit that 5 point battleship (probability is 5%) compared to the 17 points in ships spread abitrarily (probability is 17℅). IT DOES NOT DEPEND ON THE NUMBER OF TURNS IT TAKES TO SINK ALL 5 BATTLESHIP’S. The probability does depend after the first event has occurred (where you’ve hit one of those pieces). Whether you get a hit on your first turn or the first hit in 30 turns, it does not matter. probability does not depend on the number of turns it takes.

technically speaking the probability calculated at the beginning is when the number of trials (theoretically) approach to infinity. The 1/6 probability to role a six sided die on a number (for example, lets say 6) is after doing trial and error of rolling a 6. Now whether you get the 6 on the first try or the 100th try these are approximations of the 1/6 th probability. As you do more and more trials it does eventually converge to the 1/6 th probability.

So yes, technically speaking someone could hit the standalone stacked battleship in one try, probability and statistics doesn’t concern that. So if that concerns you, psychologically manipulate your openent by putting your battleship in the first row.

I’ll add to this, it’s not just about finding the initial probability since the probability for event B given A, or event C given A and B, these would increase the liklihood of being found. I couldn’t tell you exactly what the probability is but say for example event B given A, it would be 1/n for n being the number of legal spaces around A. So if the ship was in the middle of the board it would have the smallest probability of being found but if you’re along a wall, there’s only 3 possible legal spaces so the probability increases to 33.3℅, and if A was sandwitched between two walls, the probability of B is 50℅.

so if there’s a moral to this story, assert dominance and put all your pieces in the center.

hold on there, it actually does have strategic advantage from a statistical perspective.

the basic notion is that for a probability of event A to occur it is proportional to the area of the event; so if larger area, larger probability, smaller area, smaller probability.

if we take that idea and apply the same basis to battleship you could say the probability as the sum of each probability of each point which is 0℅ if we span the area to infinity.

practically speaking, this is not true as you can’t span to an infinite scale, but you could say that the probability of hitting a point is 1℅ since battleship is a 10 x 10 grid so the probability is just 1/(10 * 10) = 0.01. Then the probability gets more complicated since you are being asked what is the probability of the second, third, fourth, etc… point being hit given that initial probability. the probability grows dependent on the first point being hit.

I’m sure there is a way to find algorithmically an optimal method to finding in what location are the best positions in battleship, but generally speaking, no, there are worse conditions that have a higher probability of being guessed

here’s another one if you use chatGPT to solve the first one. Anyone who is fully optimized should be able to solve analytically shrodingers second order partial differential equation.

start with those problems, then we’ll talk if you are fully optimized.

find all possible eigenvalues and eigenvectors for the following 3 x 3 square matrix [1/0 1/0 1/0; 1/0 1/0 1/0; 1/0 1/0 1/0]

prostitution and sex trafficking. But besides that not much…

oh wait I forgot, he doesn’t support the independence of women; just sex trafficking.

you could approximate it using Taylor expansions, although this generally isn’t a rapidly convergent series. You might take a fancy for some other numerical method that would get really precise digits really quickly…

loves to see the world burn

I think having both the indentation and the braces surrounding code is much better than just indentation or braces, since it removes any ambiguity of what code goes to what if, for, or while loop/condition. we already have to memorize precedence to define ambiguous expressions. Why have another set of rules that could otherwise be prevented with braces and indentation?

a dolphin poisoned by pollution and carbon monoxide gas.

yes, and I couldn’t really care what it was intended to be used for. clearly I could still use it as a WiFi utility posts installation.

As per the arch wiki it says “iwd is a wireless dameon for Linux written by Intel.” “The core goal of the project is to optimize resources utilization by not depending on external libraries.”

the point of why I’m using it is for that exact reason. it does the same thing every other WiFi GUI/TUI out there. unless there is a vulnerability/security risk of using iwd, I couldn’t really care less about the other options, even if that means manually configuring enterprise and public networks.

hippity hoppity this meme is now my profile picture

reactions check out.

you do know he was an engineering major?

mangoWM. my workload requires a lot of resources like vivado/vitis. the more resources I have available the better. I’ve no where near customized my WM (FFS, I’m still using iwctl to configure WiFi), that said it does run great, using about 560 MB on idle.