I think the big thing with Unreal is the vast majority of games are closed source. It's already only used for games, as opposed to asking questions about general-purpose programming, but there is also less training data.
100K at 7% (market rate adjusted for inflation) is 1.6M after 40 years, so that would be 1.6M in today's buying power. Pretty good by the vast majority of peoples' standards.
What happens to stock market when everyone is given 100k to invest? The inflation adjustment is likely understated as you’re basing on historical factors and this is large macro shock to the system that needs to be adjusted for as well.
I also still don’t think this helps most people during the bulk of their lives. Most people in this cohort don’t save now, having knowledge of a more secure retirement wouldn’t change much for them. It’s not like they can turn off 401k contributions and have extra pocket money today.
I think there is an important third category beyond "avatar of techno-optimism" vs "advocate of caution". Artists typically oppose GenAI because the models are trained on huge swaths of copyrighted works. They certainly aren't supporters of the tech, and they aren't concerned about the risks of a new, not-entirely-understood technology either. They think it's outright stealing, copyright infringement, etc.
I'm not too sure what "MCP-based methodology" is, but Jan-nano-128k is a small model specifically designed to be able to answer in-depth questions accurately via tool-use (researching in a provided document or searching the web).
It outperforms those other models, which are not using tools, thanks to the tool use and specificity.
Because it is only 4B parameters, it is naturally terrible at other things I believe-it's not designed for them and doesn't have enough parameters.
In hindsight, "MCP-based methodology" likely refers to its tool-use.
I mean, Clear Linux was the leader in the vast majority of Linux benchmarks, to my knowledge. So much so that even AMD used it in their advertised benchmarks for CPU releases because of the performance advantage.
I think it was quite successful, and I doubt they are shuttering it because they don't see the value in it, but because of overall lackluster company performance and the new CEO cutting costs/the workforce aggressively.
I don't think there will be one, a company would need to commit to salaried devs. What would the value-added proposition be for them that they can't get by using any other distro out there?
The problem is that Clear Linux did a lot of tweaking in their packaging to get good performance, up to and including actual code patching IIRC, so it would be a nontrivial ongoing effort to continue that work.
The abuse of AI here blows my mind. Not just the use of AI to try to find a vulnerability in a widely-used repo, but the complete ignorance when using the AI.
"hey chat, give this in a nice way so I reply on hackerone with this comment" is not language used naturally. It virtually never precedes high-quality conversation between humans so you aren't going to get that. You would only say this when prompting an LLM (poorly at that) so you are activating weights encoding information from LLM slop in the training data.
The most common GPU per the steam survey is the RTX 3060 at 170W TDP. A huge % of users have cards near or below this TDP. The SXM H100 has a 700W TDP, and will spend far more of its life at or near that value.
Given an average ~8 hours of work/school and ~8 hours of sleep, gaming GPUs likely don't use anywhere near as much power. Plus, even when they are on, they will probably idle near 30W-60W for a lot of time spent browsing the web or watching videos.
There are more gaming GPUs in existence right now, but the number of AI chips is likely closing that gap rapidly.
And of course, what is that energy being used for? People playing games are typically having fun, bonding with friends, or engaging in social behavior. A huge amount of AI is illegally trained on copyrighted works without license to use them, causing significant harm to various fields. Plus the deluge of AI slop bogging down the internet, social media, forums, image/art-hosting sites, search, and more.
I think it will be a while before modern generative AI is even close to providing value in aggregate.
Hadn't heard of this eGPU dock before, but it seems to support USB 4 as well. I wonder why they only mention USB 3. USB 3 is 5Gb/s, but USB 4 gets you to 40Gb/s, near HDMI 2.1.
I think that if the main goal here was speed, they'd just be using Thunderbolt instead of creating this solution that seemingly only uses regular USB3, and I'm very interested to hear how this works exactly.
What are the sociological factors that separate old school from new school?
To me, I'd summarize the situation something like: The modern internet is one of continuous popularity contests, engagement farming, and status wars. The old internet was one of authentic sharing, rambling, and candid conversations.
But what are the causes driving those effects?
I suspect one key dividing line is the importance of feedback metrics. Likes, upvotes, downvotes, shares, whatever.
Imagine going to a party where whenever you said something, everyone briefly did a thumbs-up or thumbs-down motion to indicate how much they liked it before the conversation continued. Obviously, that's a bastardized way to party. But it's our whole world online. We compete for popularity, and we copy popular behaviors, in search of an attention-farming fixed point.
I also think the size of the community matters a lot. I remember in the earliest days of reddit, noticing the same usernames over and over made it feel like more of a community.
Modern "social media" is not really "social"; it should really be called "DIY broadcast media" in my view. A key clue here is despite our brave "social" world, the concept of an "online friend" is considerably diminished relative to what it once was. You tend to be either smothered with attention, or totally ignored. I prefer chilling out over fighting to get a scrap of interaction.
BTW here's a fun old-school guide to internet culture: https://www.flamewarriorsguide.com/ (I don't think it describes the modern internet very well)
itter and prosh are not offline.
I can wrote my text offline and exange data. I cant send my text to my friend for publish (I dont have internet, he have)
I cant print my text on paper like qr-code
No, for taxes, married filing separately is a different category than single. Married filing separately results in a higher tax bill than married filing jointly for most couples.
There are places in the tax code where Single != Married Filing Separately.
One pertinent example is that Washington State's capital gains tax applies after $270k per single person, per married couple filing jointly OR split in half for married filing separately. Which could be a theoretical $18.9k/year difference in taxes.
