barbegal

This simulates the chip but do I still have to write the simulation of my actual board which often has equally complex ICs on it and mixed analog and digital?

There is some evidence that Gabapentin can assist with benzodiazepine withdrawal https://ispub.com/IJPHARM/4/2/3283#

Without a human to benchmark against it's really tough to gauge how good these models are vs how good the task definitions and existing codebases are.

My intuition from the example full instructions are that the tasks are poorly specified which results in ~60% failures due to bad assumptions and missing requirements.

18 O enriched water is available to buy but the pricing suggests it's quite expensive to separate 16 O from 18 O. And you get much less variability in the ratio of isotopes from mixing compared to separating a single isotope as much as possible.

Interesting but I can't deal with the AI generated prose. Sentences like "This was not a garment for walking between a carriage and a doorway. It was ..." show me no human has proof read or edited the text.

Yeah I couldn't get past so many issues in the AI generated illustrations. Not useful at all when they are completely wrong.

Yes I would imagine lots of those type of services would be vulnerable if they hadn't updated to the latest kernel versions.

Yes you can pin it to core 1 whilst pinning all other tasks to core 0. Then will never be interrupted or preempted (except by interrupts created on core 1)

The actual research paper shows it's pure nonsense

https://cdn.skoda-storyboard.com/2026/04/Skoda-DuoBell-Resea...

As expected ANC headphones cancel less noise at low frequencies so I guess the 780Hz is a trade off between high enough frequency to be a bell and low enough frequency to get attenuated a little bit less than high frequencies.

The research paper is pretty poor quality and this is mainly a marketing exercise.

I don't understand why people get so hung up on Chrome using so much memory. A lot of this memory is "discardable" so will get dropped when the system is under memory pressure and the amount of memory allocated for this type of usage will depend on how much memory your system has available. If Chrome is using lots of memory then it's almost always because your system has lots of available memory. It allows the browser to cache large images and video assets that would otherwise have to be re-downloaded over the internet.

Does the KV cache really grow to use more memory than the model weights? The reduction in overall RAM relies on the KV cache being a substantial proportion of the memory usage but with very large models I can't see how that holds true.

This series of graphs https://www.bmj.com/content/bmj/387/bmj-2024-082194/F1.large... shows that whilst those two professions are at the bottom of the distribution they are not particularly outlying and cherry picking of those professions has occurred. The statistical analysis should have adjusted for picking the best 2 occupations of the 443 in the study. That would likely show very little statistical significance.

Total receipts were over 11,000 so more like 100 hours or around $2000 so a similar price to the LLM.

This is good work. I wish branch predictor were better reverse engineered so CPU simulation could be improved. It would be much better to be able to accurately predict how software will work on other processors in software simulation rather than having to go out and buy hardware to test on (which is the way we still have to do things in 2026)

The memory overhead is fairly significant it uses between 1.5 and 3 times the space of the data stored.

For a real spinning top over engineered https://youtu.be/QLTsxXNekVE?si=S31kpZQHiYlUSedx

Only 2 students actually used an LLM in his exam, one well and one poorly so I'm not sure there is much you can draw from this experience.

In my experience LLMs can significantly speed up the process of solving exam questions. They can surface relevant material I don't know about, they can remember how other similar problems are solved a lot better than I can and they can check for any mistakes in my answer. Yes when you get into very niche areas they start to fail (and often in a misleading way) but if you run through practise papers at all you can tell this and either avoid using the LLM or do some fine tuning on past papers.

An interesting idea but in theory just three correct pass codes and some brute force will reveal the secret key so you'd have to be very careful about only inputting the pass code to sites that you trust well.

It's definitely computable on a piece of paper and reasonably secure against replay attacks.

The energy comes from the metal electrodes not the potato. Potato is just an electrolyte carrying current between the cathode and anode.

https://newsletterhunt.com/emails/215242 seems to be a copy

Congratulations! Works amazingly well.

Does it generate enough revenue to be self sustaining?

That 90,000 traces did take 225 hours to capture so it is truly a huge amount of data and not a trivial attack.

Thanks, good to know that I'm wrong!

Is it just me or does this have a familiar "edited by Chat GPT" feel to it? I can only take this chatty writing style in moderation but it seems everyone is using Chat GPT to edit their work in the same way.

Yeah as far as I can see all the companies that study EV batteries and provide degradation reports etc. all do so by using the data from the manufacturer. I would trust data about battery degradation a lot more if the data came from an independent data logger, logging voltage and current.

The ForTwo has two seats and no truck bed so it can have enough space for a crash structure in front of the driver. On top of that its NCAP rating expired in 2021 and it is no longer being sold. The next generation of the ForTwo is likely to be longer in order to improve safety (if it ever makes it into production)

There are smaller vehicles being sold but the distance from the front bumper to the driver's legs is much longer because they don't have a bed taking up space at the back.

The reason why we don't generally have vehicles this small any more is because they don't pass crash tests so I'm wondering how this fares in a crash test. I can't see any way this could be sold in Europe unless there's some very clever engineering to make the front end more resilient in a crash.

Assuming you could get these to 10% efficiency (which is theoretically possible) and a phone needs 0.2W of energy to function then you would need a source capable of supplying 2W of energy (of which 1.8W would be dissipated as heat). The phone would be fairly hot all the time but 2W could be dissipated without it overheating in most environments. Strontium 90 generates 0.95 W/g so in theory a few grams of strontium 90 would be enough to power your phone for many decades (the half life is 28 years). But if someone were to accidentally put such a phone into an insulating material it might overheat and become a dangerous radioactive mess!

Not comparable to a real battery any time soon based on the paper https://pubs.rsc.org/en/content/articlelanding/2025/cc/d4cc0...

> The perovskite betavoltaic cell achieved impressive parameters, including a short-circuit current density of 15.01 nA cm−2, an open-circuit voltage of 2.75 mV, and an energy conversion efficiency of 1.83%, all of which represent significant improvements over previous works.