Seriously? I have recently been working on my personal programming projects on my ThinkPad from 2014. That thing also has 8 GB RAM. It’s slow, but that’s only because the dedicated video card is no longer supported by NVIDIA. I was totally able to run PyCharm, my program (which was hungry for ram), and Firefox with quite some tabs open without any issues. And most people will be doing more basic stuff on this than what I was doing. Browsing around, editing some documents, viewing some photos. I’m not sure how heavy MacOS is, but I’ll assume it’s more like Linux than Windows. You can do a lot with 8GB if your OS isn’t gobbling up resources to spy on you, show you ads, or run some useless AI shit you didn’t ask for.

I agree that it’s not a lot, but this laptop is not meant for people who need to do more than what I mentioned, putting much more RAM in there would just creep up the price without really offering anything.

Note that I’m not an Apple fan or anything, I’ve never even used anything from Apple.

Too little too late. I’m already over to Linux now. Shit’s been going downhill even before this whole AI craze went off the rails. I hope Microsoft Windows crashes and burns

For some issues, especially related to programming and Linux, I feel like I kinda have to at this point. Google seems to have become useless, and DDG was never great to begin with but is arguably better than Google now. I’ve had some very obscure issues that I spent quite some time searching for, only to drop it into ChatGPT and get a link to some random forum post that discusses it. The biggest one was a Linux kernel regression that was posted on the same day in the Arch Linux forums somewhere. Despite having a hunch about what it could be and searching/struggling for over an hour, I couldn’t find anything. ChatGPT then managed to link me the post (and a suggested fix: switching to LTS kernel) in less than minute.

For general purpose search tho, hell no. If I want to know factual data that’s easy to find I’ll rely on the good old search engine. And even if I have to use an LLM, I don’t really trust it unless it gives me links to the information or I can verify that what it says is true.

Damn what the hell. I’m Dutch and I pay less than €12 euros. For that I get 100 call minutes/messages and 6 GB. Neither of which I ever really get close to. Calling is mostly just in case of emergency or when we need to align something right now. And data is always useful because I definitely do not have free wifi anywhere. Do you never leave a big city or something? If I slip with my bike on an icy road and break something in the middle of nowhere it’s kinda good to have some way of communicating.

Yeah I’m on a 5700 or something with 64 GB of DDR4 and I don’t see any issues with it yet. I went that route because I could keep my mobo and it was also way cheaper than DDR5 back then. I figured I’d rather have more older RAM than a full system upgrade.

Is RAM already at a price that it’s worth considering selling one of those sticks? :3

Yeah I’m on a 5700 or something with 64 GB of DDR4 and I don’t see any issues with it yet. I went that route because I could keep my mobo and it was also way cheaper than DDR5 back then. I figured I’d rather have more older RAM than a full system upgrade.

Is RAM already at a price that it’s worth considering selling one of those sticks? :3

You’re probably aware, but Bitwig studio runs natively on Linux. And tools like yabridge can allow you to run many VST plugins as well. Though it remains a bit of a hassle compared to Windows. I’ve for instance lost access to a NI plugin because their new all-in-one installer/verification program won’t work on Linux

Fun fact, this loop is kinda how one of the generative ML algorithms works. This algorithm is called Generative Adversarial Networks or GAN.

You have a so-called Generator neural network G that generates something (usually images) from random noise and a Discriminator neural network D that can take images (or whatever you’re generating) as input and outputs whether this is real or fake (not actually in a binary way, but as a continuous value). D is trained on images from G, which should be classified as fake, and real images from a dataset that should be classified as real. G is trained to generate images from random noise vectors that fool D into thinking they’re real. D is, like most neural networks, essentially just a mathematical function so you can just compute how to adjust the generated image to make it appear more real using derivatives.

In the perfect case these 2 networks battle until they reach peak performance. In practice you usually need to do some extra shit to prevent the whole situation from crashing and burning. What often happens, for instance, is that D becomes so good that it doesn’t provide any useful feedback anymore. It sees the generated images as 100% fake, meaning there’s no longer an obvious way to alter the generated image to make it seem more real.

Sorry for the infodump :3

Yeah that’s a bit much tbh. Personally I find the train to be way too loud usually. Other people talking, train noises, maybe a screaming baby or something. I’ll just listen to some music and stare out of the window (if it isn’t dark).

I want to use it but I keep forgetting it exists. Something like this should just be accessible via button in the UI so no-one misses it

Recently my parents showed me some stuff they saw on Facebook. It was all just AI slop, rage bait, advertisements. Seriously, there was barely anything useful on it. They were very avoidant of acknowledging it tho. “But there’s also fun stuff on it”. They were constantly wondering whether what they were reading was real, yet it did plant seeds in their brains. It’s even influencing their politics, I had to have a whole discussion with them (traditionally centre left voters) about how our left wing parties didn’t want to let an unstoppable horde of immigrants into our country. And why voting for a centre right guy because “he looks pretty competent” will also fuck with poor people and important topics like abortion.

Your view is totally fine, but I guess you’re not understanding why people do this. I’m a millennial, around 30. Personally I buy CDs, I buy vinyl, and I even have some stuff on tape. I’ve also recently picked up film photography and among my friends it’s common nowadays to bring some 2000-2010 digicams.

So why? flac is perfect, and streaming services stream whatever high-quality music you’d ever want to play. Film is expensive, and digicams are often way more shit than whatever a modern smartphone that’s already in your pocket can do.

Personally I’ve become bored by perfection, overwhelmed by choice, and frustrated with the lack of owning anything. When I play a physical album I sit down for it, I am focused on the music. I cannot easily choose the music, I’ll just have to accept the order of the album. There are way fewer choices to overwhelm me. Likewise, with film photography, it feels simpler in a way. You shoot a few images in a go, because film isn’t cheap, and you’ll only get to see them weeks later when the roll is developed. No pressure of the perfect shot, no insane resolution to show any imperfection. And mistakes just happen, because you cannot see what you’re doing, so you just have to accept them. Digitally you can just take 20 pictures and take the best one.

So back to music. Why would one prefer vinyl or tape over CD? As a life-long CD collector, I wondered the same thing a few years ago. But when artists that I enjoy started skipping CD releases in favor of vinyl I hopped in, invested in a shit vinyl player, and didn’t really get it. Sure it had a character, but it wasn’t great in any way. After some more research I found out that it was probably just the vinyl player (please don’t get some cheap shit for a 100 bucks with a red unbranded needle). I invested in an Audiotechnica LP70XBT, and oh boy did stuff improve. I finally get it. The sound is gorgeous, though not necessarily better or worse than CD imo. It’s a bit warmer, with detailed bass but less clinical high end. And I love the whole tactile experience of it. Older vinyl definitely sounds worse than modern CD quality though.

I think it’s the whole experience that people enjoy. Putting the vinyl or cassette in the player, having something move and, as if it were magic, suddenly there’s music. With a slightly different character that differentiates it from the clean and clinical sound of high quality digital audio. Modern digital audio is great and definitely has its place, but at times it can feel sterile, too perfect. The crackles and warmth of vinyl, the grain and slightly off colours of photographic film, they feel like they have more personality. They stem from a time where the imperfections of the medium still kinda hid the imperfections of the artist.

(Okay this turned into quite a ramble but I hope there’s something useful in there :3 )

It’s horrendously slow, unusable imo. With the larger DeepSeek distilled models I tried that didn’t fit into VRAM you could easily wait 5 minutes until it was done writing its essay. Compared to just a few seconds when it does. Bit that’s with a RTX 3070 Ti, not something the average ChatGPT user has lying around probably.

Personally I also ran some distilled versions of DeepSeek locally, though I’d imagine that isn’t really possible for most people.

Better than nothing I guess. Obviously it’s a privacy nightmare. But therapy is hard to reach nowadays and I’ve noticed that many men are reluctant to make that step. It’d be preferable if they did, but if ChatGPT can at least give an outlet for the emotions then it might just save a few people. Seeing men demolish themselves because they’re too ashamed to seek help is something I’ve unfortunately seen quite often. Even though I’m aware of this I’ve still waited till it was way too late because I subconsciously didn’t want to give in to the “weakness”. I hate that men are conditioned this way, it costs lives.

Call me an optimist, but I still hold the hope that we can one day do better as humanity than we do now. Humanity has become a “better” species throughout its existence overall. Even a hundred years ago we were much more horrible and brutal than we are now. The current trend is not great, with climate change and far-right grifters taking control. But I hold hope that in the end this is but a blip on the radar. Horrible for us now, but in the grand scheme of things not something that will end humanity. It might in the worst case set us back a few hundred years.

Imo it’s a good thing tho. Spreading our civilization across multiple planets is the only way to guarantee long long term success. Obviously we should also fix the climate change issue (and many others). But still, being spread across the solar system would give our species redundancy. An extinction event on earth like a large meteor strike would no longer be the end.

Yeah that’d explain some stuff. Happy to have switched to Linux

I’ve switched a few months ago. Plenty of issues, but none of them major enough anymore to go back again. All games I play regularly apart from Assetto Corsa work, and AC should also be fixable as far as I read.

Just a few days ago I got the first racing sim working properly with my wheel. AC doesn’t start yet and Automobilista 2 does not match the irl steering wheel movements (this also happened on Windows sometimes), but ACC worked without much of an issue.

For music production I also got most of my setup working. I’m having a lot of issues with opening my old projects, but I wasn’t actively working on them anyways and with some effort I can get them back. Still some issues like the Vital synth CLAP version crashing when the window is opened and the Splice sample thingy not allowing drag and drop, but we’ll get there. For new projects it’s mostly workable.

Basically everything else I need just works. Games, photography stuff, everyday programs, and obviously my programming stuff because it was already on Linux.

I use Arch by the way :3

Looks like it. It’s a complete fever dream graph. I really don’t get how someone can use an image like that. Personally I don’t really like AI art anyways, but I could somewhat understand it as a sort of “filler” image to make your article a bit more interesting. But a graph that is supposed to convey actual information? No idea why anyone would AI gen that without checking