elpocko

8 ms·

91.
by elpocko 1y ago
Not impossible but you'd gonna have to do a bit more than that. Most people are ignorant, but not all of them. An experienced user can tell what model family was used from a bunch of generated images. Also, no one would believe a nobod
92.
by elpocko 1y ago
Show your math.
93.
by elpocko 1y ago
Where does this "10-50" figure come from? A quick google search says a H100 draws 700 W, while a 3080 draws 320 W. AI provides tangible value to businesses and private users beyond mere entertainment. We'll see how much power
94.
by elpocko 1y ago
>writing off a major new generator of atmospheric carbon is dangerously irresponsible. I agree: ignoring the carbon footprint of the gaming industry is irresponsible.
95.
by elpocko 1y ago
Okay but these tiny models are being used by people and businesses instead of GPT-4. My point was that they consume less energy per user than a rig used for gaming. I have no insight into how many GPT-4 users are served per GPU, but I would
96.
by elpocko 1y ago
Yes, one AI GPU uses more energy than one GPU used for gaming. The one AI GPU however is shared between a large number of users, while the one gaming GPU is used by one player. Apparently a single gaming GPU can be used to run an LLM that s
97.
by elpocko 1y ago
>100k+ GPU's, just for training. That's a tiny drop in the sea of the almost 2 billion PC gamers [0], and hundreds of millions of gaming consoles [1] in the world. Not to mention the energy required to manufacture all that hard
98.
by elpocko 1y ago
Thousands of users are being served by those GPUs. A gaming PC has one user. I can't believe I have to point this out on HN of all places.
99.
by elpocko 1y ago
Besides common sense, I can tell you about my kW h counter going brrr when playing games (400 W continuously, sometimes for hours on end) vs. running Stable Diffusion or Llama-whatever (400 W for 15 seconds every 3 minutes for an hour or tw
100.
by elpocko 1y ago
How many AI users are served using a single H100 per time, and how many gamers are served using a single 3080 per time? How many gamers are simultaneously running a 3080 or equivalent for their entertainment?
101.
by elpocko 1y ago
I hear people advocating for the construction of nuclear reactors every day. They don't mention gaming, just like they don't mention refrigerators or washing machines specifically. A gaming machine consumes the same amount of ener
102.
by elpocko 1y ago
There are many more millions of GPUs in computers and game consoles around the world burning electricity for your entertainment, for decades. The same class of devices. The environmental impact of having pretty pixels on your screens is at
103.
by elpocko 1y ago
I've been thinking about making a fantasy computer/console that uses an emulated 65C02 CPU, maybe a bit simplified -- every instruction only taking 1 cycle, for example. You could use all existing tools for the 6502, like the amaz
104.
by elpocko 1y ago
>very close to the metal >(pico8) But PICO-8 with its integrated tools and Lua programming is super high level, pretty far removed from "8bit". It's only 8 bit in aesthetics, entirely artificial and forced. Why not writ
105.
by elpocko 1y ago
You click "flag" when off-topic shit shows up. Others do it too. Shit goes away. No moderators needed. Except when they disagree with the popular vote, then they make you swallow, whether you want it or not.
106.
by elpocko 1y ago
Obviously. It was meant to be tongue-in-cheek. ;)
107.
by elpocko 1y ago
I wish LLMs were generally viewed as Eliza on steroids, a thing to generate plausible sounding text with, in places where we used primitive generators based on Markov models before. To implement smarter NPCs in games, and virtual chat partn
108.
by elpocko 1y ago
If I recall correctly, AmigaOS had a boot menu that would be triggered by holding down both mouse buttons at boot time. Not saying that's the reason for the oddity, I don't see why it would be useful to see a menu if you can'
109.
by elpocko 1y ago
>run any model that talks to openai (which is nearly all of them) What does that mean? I've never seen any locally run model talk to OpenAI, how and why would they? Do you mean running an inference server that provides an OpenAI-com
110.
by elpocko 1y ago
That's because it looks exactly like the CPC. Just a happy accident.
111.
by elpocko 1y ago
We boil the oceans much faster so you can play the latest Call of Duty.
112.
by elpocko 1y ago
Using the environmental impact of AI as an argument against AI is ridiculous because AI currently consumes at least an order of magnitude less energy than, say, the computer games industry, which is of no concern apparently.
113.
by elpocko 1y ago
Yes, because ...all hackers from all over the globe will ...uhm, suffocate and die if they don't want to deal with political drama in the US.
114.
by elpocko 1y ago
When you simply ask for "a picture of a cartoon car" and nothing else, these models will give you a bland and generic depiction of a cartoon car, something like the lowest common denominator among cartoon car images according to t
115.
by elpocko 1y ago
A lot of people are saying that a) AI generates slop that no one needs, and b) AI is putting human artists out of work. If the machine can do art that's indistinguishable from human art, and art is the soul of humanity, then the machin
116.
by elpocko 1y ago
If your work is of so low quality that an algorithm that only generates "shitty art" (by popular consensus) puts you out of work, then you don't deserve to get paid for your "art." It's Schrödinger's art:
117.
by elpocko 1y ago
Isn't this common knowledge? The statistical odds are always in favor of the casino, otherwise there would be no casinos. You can only win short-term, if you're lucky; long-term, the bank always wins.
118.
by elpocko 1y ago
What model(s) does this use? Is it built using open source models and tools? I couldn't find any mention of models in the FAQ (who even asked those questions?) or elsewhere. I doubt you trained your own diffusion model. The GitHub cat
119.
by elpocko 1y ago
I'm using Firefox with UBO and for quite a while now every YT video stops playing after 45 seconds with an error message. So I made a bookmarklet that opens the embed link instead, which does not error out, but some videos can't b
120.
by elpocko 1y ago
I think they've given up on this, it's ubiquitous. I've recently noticed that a CLIP model (a neural network for image understanding) gave me image descriptions using "it's" where "its" would have bee

More ›