As a neurodivergent who has aphantasia and can’t visualize, not like this at all. For me it’s more like, what I need is there when I need it (usually), not in a visual sense but a conceptual sense.

Copilot / LLM code completion feels like having a somewhat intelligent helper who can think faster than I can, however they have no understanding of how to actually code, but are good at mimicry.

So it’s helpful for saving time typing some stuff, and sometimes the absolutely weird suggestions make me think of other scenarios I should consider, but it’s not going to do the job itself.

mobo isn’t fast enough

Laughable. Those people at the store are just trying to get more money out of you.

Your Ryzen 3600 should be more than up to the task of being paired with a 3060. That being said, CPU heavy games will still be impacted, and simulation games like Cities Skylines tend to be CPU heavy, but if your current bottleneck is the GPU (and I have no doubt it is) then you’ll still see an improvement.

The CPU bottleneck would emerge later on, if ever, when there is more going on in the game. For example, a huge city. Which means if you reach a point where your game starts lagging, while my ideal, you could simply start over or maybe look for mods that help with it.

And yes, you can reuse components if compatible, and a 3060 will work with stuff for a while yet, so you’ll be fine. You should double check that your PSU (power supply) is up for the task though.

As for case constraints - if you can find the model of your case, it should just max GPU size.

I think I know why the pepe looks so happy.

If you’re using a VPN for privacy, downloading torrents, etc you should be using a solution with a killswitch that doesn’t allow traffic through if the VPN goes down.

They’re pretty insecure anyway, my current P14s Gen1 has a working fingerprint reader on Silverblue but I haven’t really used it.

I haven’t used a T14, just the E14 and the P14S.

New ish. My current Thinkpad is a P14s Gen 1 with a Ryzen 4750U 16GB of RAM, and it came with a 512GB SSD. I paid just under $300 for it on eBay and well worth the cost. I wouldn’t get anything that is still a TXXX variant anymore though (e g. T490), they simplified the product line. So T490 was replaced by the E14 Gen 1, and the P14s Gen 1 is an AMD variant.

Highly recommend. One thing worth noting though is to double check the fingerprint reader if you desire that, the E14 Gen 1 has a reader not compatible with Linux in a functional way. The P14s Gen 1 however does.

I personally haven’t noticed this, but I’ve also only used the software center to install an update once as opposed to just letting them update automatically. Are these the flatpak versions or installed at the system level?

I use Groupy by Stardock for this. It’s a neat little tool that lets you make pretty much any application into a tab by grouping them.

It isn’t open source nor free though, and I didn’t even realize there was a Groupy 2 until I searched it to get you a link. For something I use daily, it was worth it for $10.

DDG is suggesting it because it has no custom rules setup to sanitized indexed URLs. Most likely it just means the first link to the article their crawler found has the referral tag in it.

But the plot thickens.

On Bing if I search “TechRadar shufliada” I see a few similar results, including other websites with the same site in the URL.

If I go to the website I find in their about page it says:

The goal of “Shufliada” is simple: to make the Ukrainian language more visible on the Internet/Google search.

Which while it could be innocent, almost suggests they’re intentionally backlinking popular websites in hopes their link to it is what gets indexed. IMO the blame doesn’t lie with either DDG nor TechRadar.

Kids are stupid xD. My bike memory like that is our parents for one of those BMX bikes from a garage sale. The ones with the bars coming out the wheels for tricks.

My sister, brother, and I decided those were for carrying additional people, so obviously having one driving, one standing in the rear, and one standing on the front would be fun. But in our infinite wisdom we decided going down the concrete driveway was a bad idea, so we decided to go down the dirt hill (which had a steeper incline).

We had one tame wipeout and decided to go for round two. This time we made it to the bottom of the hill (where the trees and branches/debris are) and wiped out. It wasn’t pretty lol. No hospital but a few road rashes (or the dirt equivalent) and other various cuts and bruises led to us going inside to our mom for intervention.

She was not happy. The next day, the bike was gone.

The article is paywalled for me, so I don’t know if this topic is mentioned in the article, but I’m curious - is this a case of manual rules / filters to prevent the answer, or the training data containing so much disinformation that the LLM basically can’t make heads or tails of it?

For future reference, the two hour window is more of a suggestion not a requirement. Just last week I refunded a game with 4 hours.

I had a friend that played civ, he invited me to multiplayer. Little did I know, he plays against the hardest bots on a regular basis. I had only done like, two single player games.

I don’t play with him anymore.

For me personally I don’t have much control over my empathy. Sure I can look at someone and glean their emotional state based on conscious guessing, but my “affective empathy” as you put it, is more my brain subconsciously picking up on their emotional state and then sharing it.

For most emotions, including anger, it’s not targeted. Not until I actively participate in the emotion. It’s also not something that applies to everyone and every situation, with my own personal emotions easily overriding the empathetic emotions.

Of course, everyone experiences empathy their own way.

Holy crap I’m blind xD. I take it back, it does seem to portray the notion. Goes to show how subtle it is.

I don’t think it was designed to imply that. None of the language appears to steer the viewer to a specific conclusion, letting the viewer interpret it for themselves.

That being said, I would agree that the data itself represents both access to mental health care and culture (specifically, if that culture has a stigma against it).

I think some of the larger countries are not really useful in the dataset though. I’m curious how say, California and say, Alabama, would look in the dataset.

I’ve also had no negative experience with VPNs and Discord. This being said, I see two potential factors.

Accessing an account from a different IP and then changing the email on that account from that different IP can look like the account got hacked.

And the specific VPN being used could play a role (I’m not asking you to reveal that) - if the VPN is being used by bad actors, their IP ranges might be flagged as well.

Considering the problem with stolen accounts being used to send spam to their friends and the servers they’re in, I can see why there would be automated logic to rate limit actions. The only real issue I see here is the appeal / support process just not being there, going by your other message about a lack of success appealing.

I saw a few others, but the ones I looked at were basically instruct layers where you’d need to add your own parser. I didn’t find anything (in my 3 minutes of searching) that offers an openai chat completions endpoint, which is probably the main stopper.