As opposed to the CCP, which is essentially what the Chinese model companies map to, who are only interested in freedom and liberty for humanity right?
Seems like the inverse? Select a prebuilt PC from Origin, Alienware, Cyberpower, etc and it independently prices each part of the build so you can see if you are getting ripped off(you are).
I certainly feel that I wasn't ripped off when in Feb this year, HP sold me this Omen desktop (https://www.hp.com/gb-en/shop/products/desktops/omen-35l-gam...) with a 9800x3d, rtx5080, 2TB SSD and 128GB?! of Kingston fury ram for £2700... with four years interest free credit...
it's now selling for £2300 more than when I bought it...
I didn't even want 128GB of ram... it just came with the pre-built spec I chose... I feel like they sold me the GPU and half the ram, and then threw in the rest for free! it's my deal of the decade...
half wish I'd stretched for the one with the 5090 now... :)
GPU and ram prices have certainly gone up since then, but they were also were high even then. I put that spec into pcpartpicker the day I bought it and was was over £3800... I think they'd set the prices the year before when they built the PCs and didn't update them...
not just you, all of us. the thing is that there will be more demand on what is available on the market: make sure to check and have at least some reference. Seems that hardware pricing will become more so like commodity/stock market.
Well, whether or not its a "good deal" depends on how much you value the OEMs efforts to select, test, assemble, and support the parts. I say this as some one who has been building computers for myself and others since I was like 17, and working on them longer- and I have charged a fair amount of money for these services. I think what I charged was fair, too.
there is a difference between a "gourmet" service and mass assembled PCs for sure. We now focus on scoring the run of the mill computers sold by big e-tailers.
Well test an example for yourself. I did and around 25% of the total price was for their "efforts". I doubt you would have charged as much to plug components into a Mobo...
I think the point is that we can be oppressed by governments or corporations using this super human tool just as well as we can be oppressed by skynet. Presumably the only way we could protect ourselves is with our own un-neutered AI. Or something.
I have tried dozens of different writing skills and my own prompt ideas, listened to podcasts with engineers from both major american LLM companies, actually read the prompting guides from anthropic for the models I use, and of course, as if it needs to be said, I use the latest and greatest frontier models, for difficult work.
Can you give an example of what you wanted explained, the LLM explanation and why it is bad?
Obviously if you don't understand an idea you aren't really in a position to evaluate how good an explanation of it is. I have a background in Physics research and find models like Astra to be excellent at explaining concepts at a level for an advanced practitioner. It also has done very well, from what I can tell, for areas I am less knowledgable in. So I am confused by statements like yours that directly contradict my experience and the subjective experiences of many others.
> Can you give an example of what you wanted explained
Something I do dozens of times a day:
"Please explain what this PR does, I haven't read it"
it will make me 20 pages of gibberish about caches, correctness checking, race conditions, github action runners, tradeoffs between a bunch of invented terminology internal to the PR without defining the terms- you'd have to have already read the code to know what it was referring to
without ya know, explaining it. I ask my colleague and he says something like "oh it deletes records from such-and-such a system when they aren't needed anymore after a merge" and I know exactly what he means. A single well composed sentence was all that was needed.
You can ask the agent for a single sentence explanation and it will write something like "linked candidate records are trimmed when canonical record determination window is passed", none of which is terminology any of us use, some of which is defined in the new PR and some of which is made up. I can't emphasise enough, I use these things EVERY DAY. it knows what I work on, it knows how I talk. There is no earthly reason it should be this bad given all the skills and advance and context it has. As mentioned, I constantly work to correct its shortcomings in this area but its like trying to hold water in a sieve, it just cannot do it.
"What does this PR do?"
LLM: "Oh, it specifies the kernel window over the cache candidate list to classify possible invalidated swap identifiers"
Colleague: "It removes duplicates in the target_uuid_cache table in the db"
The agent has full context of what systems the code talks to, our terminology sheet etc. It just pathologically cannot simplify things down. If you work with them enough you learn that there are some words it reaches for a lot in certain circumstances, like it loves "canonical" in relation to merging things. But this is more like learning to communicate with a deeply autistic person who struggles to adjust their speaking to be relevant to their audience rather than talking with a "good communicator"
>A more rational society would acknowledge that higher education is a good thing, and therefore try to make it affordable, not try to undermine it entirely.
This assumes you believe that graduates are indeed gaining skills or beliefs that contribute to the betterment of society. Part of the reaction is a backlash to DEI/Woke and its a true statement that many recent graduates are not in fact contributing to but actively wishing for or working towards the downfall of US Society.
I think that influence is a handful of classes at best. Unless you’re suggesting that every university in the United States has no educational value anymore, and has been replaced with a Fox News caricature?
I think this maybe used to be true but no longer. With the right harness extremely prompts like "find all the vulnerabilities in php" are very effective. No fancy prompting required.
Well yeah, but that's because the harness is iteratively using the model to massage the prompt and feed it back. It's not magic; prompting is the only interface it has, so "fancy prompting" is what it automates for you.
>It's really refreshing to put some cold hard numbers on this stuff, hopefully this quells some of the "AI is unprofitable everyone is spending a dollar to make fifty cents" arguments.
Inference has always been profitable. The problem is that this profit shrinks as your model becomes less competitive. The labs are in a constant race to release the best model to maintain inference profit. Deepseek is not training their own models at the scale of OpenAI and Anthropic and mostly sustains off their efforts via distillation.
>Even if some parts of the community care about how a tool was made – Most don't, they just want something that fulfill the goal.
Yes, thats the majority of humanity.
>Now even my hobby programming was overtaken by AI, something I did for love and expected no pay out of it whatsoever, which is... depressing, to say the least.
So your hobby was getting adulation and respect or writing software? Only one of those has changed.
reply