RachelF

8 ms·

1.
by RachelF 9d ago
This is not a great idea. Pre-2000 GPS time server firmwares do no handle leap seconds properly.
2.
by RachelF 14d ago
For those who can't afford RTX 6000's you can unlock around 20% increased card to card speed on consumer GPUs using this library: https://github.com/aikitoria/open-gpu-kernel-modules The hardware supports it,
3.
by RachelF 15d ago
Similar statistics for Australia: https://www.roymorgan.com/findings/10192-australian-unemploy... It all depends on the definition of "Unemployment". Most governments do not count you as unemployed if you&#x2
4.
by RachelF 15d ago
Very true. The problem is mainstream business media hasn't asked the questions that Zitron has. So, much like current US politics, we're left with hype on both sides. That's all that gets the clicks/attention, and little
5.
by RachelF 15d ago
The fact is there is BullSh_t and hype on both sides. Both Zitron and AI execs have good points, but both are trying to sell you something. The murky truth lies between the two extremes. IMHO, having Zitron around is a counter to the AI lea
6.
by RachelF 16d ago
Dan Ariely's fraud is pretty widespread. He sold his research to an insurance company, who should have known better: https://www.lemonade.com/blog/oh-behave/
7.
by RachelF 16d ago
Looks good. However, an RF question: Both CC1314 and LR2021 are sub 1GHz components, are you not working outside their spec at 1.090GHz?
8.
by RachelF 18d ago
This thing has been in the wild since 2024, so it had a good run. It looks pretty bespoke, so there are likely other variants. Also targeting the anti-virus executable is a smart way to hide. I don't think this backdoor was cheap to de
9.
by RachelF 21d ago
Yes, this is ironic - it was the tinkerers using CUDA on cheap graphics cards that made Nvidia stuff useful in constrained academic environments. True about Intel and ECC, but AMD now does similar things, even with their consumer CPUs and c
10.
by RachelF 21d ago
I guess we may have to start paying to download models. Or perhaps they will start throttling downloads for free users. I don't know what they business case is, it might be to shut them down: I suspect good free models on local hardwar
11.
by RachelF 21d ago
Linus Torvalds: 'Nvidia has been the single worst company we've ever dealt with' His other comments on Nvidia often contain expletives.
12.
by RachelF 24d ago
Economists would be wise to read this. The world economy is now a rather complex system, and many of the safeguards from previous generations have been removed in the interests of "efficiency".
13.
by RachelF 24d ago
One more trend I'd like to reverse is the always on cellular data connection, reporting back to the car manufacturer where you are and how you are driving.
14.
by RachelF 27d ago
Remember it was MIT that dobbed him in. MIT then helped the criminal prosecution even after JSTOR settled with him. https://www.eff.org/deeplinks/2013/07/mit-aarons-swartz-case...
15.
by RachelF 28d ago
Don't give them ideas!
16.
by RachelF 29d ago
True, I have to agree with you. The AI giants might be investing a huge amount of money in generation 1 technology. There might be a much better way to do it just around the corner. They might know this and thus the hurry to IPO. A rough an
17.
by RachelF 1mo ago
I have a bad feeling about this. US companies have spent hundreds of billions on their models, and they are not much better than the cheaper open Chinese models. Perhaps they can out-compete them. If not there will be increasing calls to li
18.
by RachelF 1mo ago
To me, the amazing thing is that we now have local models that rival the reasoning of high end models from about a year ago. I hope this trend continues.
19.
by RachelF 1mo ago
So true. The levels of indirection from a pointer to an actual DRAM chip address are insane. Then there's the electrical bus: DDR5 runs so fast it need channel characterisation (sorta like the old model dial up sounds) on the lines bet
20.
by RachelF 1mo ago
Try Kagi translate to LinkedIn https://translate.kagi.com/?from=en&to=linkedin
21.
by RachelF 1mo ago
To me the most interesting thing here is that AI did it, albeit with very good prompting. Claude AI just created something in a few hours, perhaps, that teams of intelligent developers at the time struggled to do in years of effort.
22.
by RachelF 1mo ago
Many. They did an overly generous self examination a few weeks ago. Generally, they are a trailing indicator, like someone appearing on the Forbes millionaire under 30 list (Theranos, FTX). However, they are not always wrong. Think of Econo
23.
by RachelF 1mo ago
It is amazing how war accelerates technology development. Radar and the Manhattan project are cases in point, probably squashing 20 years of normal research into 3. Drones in Ukraine are a contemporary example. However, when the war is over
24.
by RachelF 1mo ago
>As soon as things get in range for motivated amateurs, progress skyrockets. So true. The reverse is also true - when greedy companies overprice their initial release so that it is out of range of the enthusiastic hobbyist they stall pro
25.
by RachelF 1mo ago
The BBC "Secret War" series and RV Jones' books are superb. I would recommend also reading his "Reflections on Intelligence" - it was written later and is slightly more philosophical. By that time Jones had been bur
26.
by RachelF 2mo ago
They should not call it "age verification" - it is actually identity verification.
27.
by RachelF 2mo ago
Flowers had to begin building Colossus out of his own pocket. The British government eventually repaid him, but not in full.
28.
by RachelF 2mo ago
Dolby was just pre-emphasis of the higher frequencies before they were written to the tape. It will sound the same with 8 bits. 8bit quantization will give you 48dB dynamic range (best case). Call it 36dB practical, around the same as an av
29.
by RachelF 2mo ago
The original timedemo was how fast Sopwith game from 1984 crashed the plane: https://en.wikipedia.org/wiki/Sopwith_(video_game)
30.
by RachelF 2mo ago
Benchmarks are like the SP500 index or IQ measurements: They distil multiple variables into one number. In doing so you lose details, but do get a useful measure. Yes, they are imperfect, but they do broadly measure how fast a processor is,

More ›