Vlyn @ Vlyn @lemmy.world

Posts

0

Comments

76

Joined

2 yr. ago

At this point you're just arguing to argue. Of course this is about the math.

This is Amdahl's law, it's always about the math:

https://upload.wikimedia.org/wikipedia/commons/thumb/e/ea/AmdahlsLaw.svg/1024px-AmdahlsLaw.svg.png

No one is telling students to use or not use parallelism, it depends on the workload. If your workload is highly sequential, multi-threading won't help you much, no matter how many cores you have. So you might be able to switch out the algorithm and go with a different one that accomplishes the same job. Or you re-order tasks and rethink how you're using the data you have available.

Practical example: The game Factorio. It has thousands of conveyor belts that have to move items in a deterministic way. As to not mess things up this part of the game ran on a single thread to calculate where everything landed (as belts can intersect, items can block each other and so on). With some clever tricks they rebuilt how it works, which allowed them to safely spread the workload over several cores (at least for groups of belts). Bit of a write-up here (under "Multithreaded belts").

Teaching software development involves teaching the theory. Without that you would have a difficult time to decide what can and what can't benefit from multi-threading. Absolutely no one says "never multi-thread!" or "always multi-thread!", if you had a teacher like that then they sucked.

Learning about Amdahl's law was a tiny part of my university course. A much bigger part was actually multi-threading programs, working around deadlocks, doing performance testing and so on. You're acting as if the teacher shows you Amdahl's law and then says "Obviously this means multi-threading isn't worth it, let's move on to the next topic".

You still don't get it. This is about algorithmic complexity.

Say you have an algorithm that has 90% that can be done in parallel, but you have 10% that can't. No matter how many cores you throw at it, be it 4, 10, or a billion, the 10% will be the slowest part that you can't optimize with more cores. So even with an unlimited amount of cores, your algorithm is still having to wait on the last 10% that runs on a single core.

Amdahl's law is simply about those 10% you can't speed up, no matter how many cores you have. It's a bottleneck.

There are algorithms you can't run in parallel, simply because the results depend on each other. For example in a cipher where you first calculate block A, then to calculate block B you rely on block A. You can't do block A and B at the same time, it's not possible. Yes, you can use multi-threading to calculate A, then do it again to calculate B, but overall you still have waiting times while you wait for each result, which means no matter how fast you get, you always have a minimum time that you'll need.

Throwing more hardware at this won't help, that's the entire point. It helps to a certain degree, but at some point the parts you can't run in parallel will hold you back. This obviously doesn't count for workloads that can be done 100% in parallel (like rendering where you can split the workload up without issues), Amdahl's law doesn't apply there as the amount of single-core work would be zero in the equation.

The whole thing is used in software development (I heard of Amdahl's law in my university class) to decide if it makes sense to multi-thread part of the application. If the work you do is too sequential then multi-threading won't give you much of a benefit (or makes it run worse, as you have to spin up threads and synchronize results).

That's really weird. But I can explain one value there:

You set it to 650 GB (1000 bytes), fdisk might show you GiB (1024 bytes) though. Which would be 634.8 GiB.

It's the same as you buying a 1 TB disk and you only get a 977 GB partition out of it.

No clue about your other apps though, sorry.

But you can easily test this yourself: Go and try to create a new Discord account without a phone number. Then try to use it.

I couldn't, Discord immediately forces me to add a phone number on login.

Maybe it's regional? Or they have different behavior based on your email address provider?

I gave up on having a second account.

There’s this Computer Science 101 concept called Amdahl’s Law that was taught wrong as a result of this - people insisted ‘more processors won’t work faster,’ when what it said was, ‘more processors do more work.’

You massacred my boy there. It doesn't say that at all. Amdahl's law is actually a formula how much speedup you can get by using more cores. Which boils down to: How many parts of your program can't be run in parallel? You can throw a billion cores at something, if you have a step in your algorithm that can't run in parallel.. that's going to be the part everything waits on.

Or copied:

Amdahl's law is a principle that states that the maximum potential improvement to the performance of a system is limited by the portion of the system that cannot be improved. In other words, the performance improvement of a system as a whole is limited by its bottlenecks.

You don't get it. Server settings are only one part of it, Discord also has their own account wide policies.

For example I had a second account I used on Discord. Verified email, used it for years on and off every now and then. Last time I tried to login? No longer possible, I need to add a valid phone number and verify it. I couldn't even get back into the account at all!

No way around it, if Discord flags your account for whatever reason they log you out and force you to add a phone number. All it might take is you pressing logout once and next time you try to login you're blocked. If even that.

That's an old account then. New ones don't work without. And at some point they suddenly force you to add a number out of nowhere.

If you want to use Discord nowadays you need a valid phone number, that's it.

Either you agree with it and use their service, or you don't and delete your account with all your data.

They use this to battle spam accounts (with quite some success, been over a year since I've been contacted by scammers), there is no good way around it.

It doesn't really matter that it was her in this image. When you put "professional" into it then you can expect something along these results:

https://www.google.com/search?q=professional+woman

And overall in I'd say.. 7 out of 10 images this is a white woman in a Google search. So the probability is high that the training data also has a bias towards that.

Someone in the original lemmy.nz post said they did the exact same thing, same image, same prompt, and it turned her Indian. So if you have very wide training data the result would be rather "random". Or you have very narrow training data and the result will always be looking similar.

Grab an app focused on an Asian audience with beauty filters for example and it will turn a white person into an Asian one. But no one complains there that the app is racist.

You are on lemmy.world, the person I responded to is on lemmy.ml.

Same here, I started with lemmy.ml and that was nice (as lemmynsfw.com was defederated). But their random slur filter was too annoying, especially when the rest of the fediverse sees the actual words.

So I switched to lemmy.world, but I probably already have 50+ NSFW communities in my block filter and they keep coming :(

You're on lemmy.ml, which defederated from lemmynsfw.com (where most of the porn comes from). I was on lemmy.ml too, which was nice, but their slur filter is annoying as fuck. For example if I type 'bitch' you'll not see it, you'll see 'removed' on lemmy.ml (while the entire rest of the fediverse sees the original word). If you want to see the word, follow the lemmy.world instance link (rainbow icon).

Random censorship annoyed me so much, I switched to lemmy.world. But now 'All' is flooded by porn communities I have to block :-/

You only get good quality if you use the right model, the right keywords, the right negative prompt, the right settings, .. and then it can still be pure luck.

If you see a high quality AI image that actually looks good (not just parts of it, but the whole composition) then someone probably spent hours with fine-tuning and someone else spent weeks to customize the model.

And even if you're good at that, you'll never get exactly the image you had in your mind. Especially as most models are heavily biased (You can create a portrait of a busty beautiful woman, but the second one you create probably has a very similar face).

This might get better relatively fast, but right now AI art is not a replacement for good artists. Especially if you need more than one image with consistency between them.

It's more like a superpowered Photoshop where you can mess around with and get cool results, just that instead of filters or a magic stamp you generate the entire image.

Super cool tech, but of course artists feel threatened. Except the popular ones who already drown in commissions.

You forgot a massive step in-between: Digital art / Photoshop.

Which already vastly sped up art creation and made it easier (when you can just use special brushes instead of having to spend hours doing a pattern by hand).

And even though it's a lot easier, you still need artists to produce proper products. Good artists and designers will keep their jobs in the foreseeable future, while more simple one-shot works can be done by AI.

If you have a basic understanding how AI works then this argument doesn't hold much water.

Let's take the human approach: I'm going to look at all the works of popular painters to learn their styles. Then I grab my painting tools and create similar works.

No credit there, I still used all those other works as input and created by own based on them.

With AI it's the same, just in a much bigger capacity. If you ask AI to redraw the Mona Lisa you won't get a 1:1 copy out, because the original doesn't exist in the trained model, it's just statistics.

Same as if you tell a human to recreate the painting, no matter how good they are, they'll never be able to perfectly reproduce the original work.

Can you use any tools to give you relieve? The most basic one would be a walking stick for support (if your arms work well). A foldable walker you can also sit on works even better, but might be too much of a hassle.

Or you take it really far and see if something like that is commercially available by now: https://futurism.com/the-chairless-chair-allows-you-to-sit-anywhere

Your post is lacking details: Where did you post your review? What community?

The only ones able to remove your post are the instance admins (pick an instance you trust) and the community mods (pick a community where you agree with their rules).

If you got your content removed here then it was either wildly inappropriate or you broke community rules.

Honestly that's more user friendly than 9 out of 10 application forms I've run into.

The best way for me to avoid this mess for now has always been an email with my pdf files attached.

Grim Dawn was my Diablo 2 fix. It runs decently, has a dark atmosphere (though if you want story and lore you'll have to read a lot in-game) and cool items.

Also a massive world, especially with the DLCs. Definitely worth playing in my opinion :)

You get nearly free unlimited respecs, the only thing you can't undo is your 2 classes you choose.

Ah well, I have plenty of uses for my salary. Though I'm a software developer, so that's more like ITish.

I also run my own mail server with a self-learning spam filter, so I know how easy it is to mess that one up.