monocasa

6 ms·

91.
by monocasa 2mo ago
TSMC wasn't the major player until relatively recently. SuperMicro, Foxconn, Asus, MSI, Gigabyte, etc. have existed for quite some time and are all Taiwanese.
92.
by monocasa 2mo ago
> You're right! That sequence does brick Voyager 2's CCS. Let's try ....
93.
by monocasa 2mo ago
It says in the rules > Trapped/emulated/virtualized instructions may only time the trap, not the handler. But I feel like that 12ms write to an ACPI IO port at current leaderboard position 8 is probably trapping to SMM and bein
94.
by monocasa 2mo ago
Well, in more formal terms, cosmology has the 'cosmological principle'[0], which is a conjecture that at large enough scales, the universe is universally isotropic and homogenous wrt the distribution of matter. Basically that it
95.
by monocasa 2mo ago
I think that's more a result of our incomplete/inconsistent scanning.
96.
by monocasa 2mo ago
To be fair, the 6502 is very dense, and very hand optimized. Modern systems tend to use "standard cells" for logic which is a lot more digestible.
97.
by monocasa 2mo ago
The main pirating route at the time (a device (oh which there were several) that sat under the N64 and had either an Iomega Zip drive or a CD drive for holding ROMs), had a port for inserting a real cartridge to use its lockout chip.
98.
by monocasa 2mo ago
Yeah, the full SDK leaked decades ago when the system was still economically relevant to Nintendo.
99.
by monocasa 2mo ago
FWIW, interpreted languages living on top of a native code engine has been a pretty common theme throughout. It goes back to early PC days (when PC meant IBM, Apple II, etc), where it was common that engine would be written in ASM, but it
100.
by monocasa 2mo ago
IDK, a lot of the emulators I've seen and a couple I've written will visit that. Not everything is built on simple traces, but a lot of the time will translate more complex graphs at a time.
101.
by monocasa 2mo ago
Giving my payment info Oracle feels like giving matches to an arson.
102.
by monocasa 2mo ago
> took a day - who knows why I bet they want to see the test payment work it's way through the system, which given the number of midnight batch jobs that still exist in banks means waiting a day.
103.
by monocasa 2mo ago
I mean, every major arch has crypto specific acceleration instructions.
104.
by monocasa 2mo ago
If anything, Mach style IPC has gotten worse because of the greater overhead if kernel/user transitions. Spectre really did a number on context switching overhead.
105.
by monocasa 2mo ago
Previous related discussion: https://news.ycombinator.com/item?id=49137060
106.
by monocasa 2mo ago
Yes, but even farther back. Apparently the Pentium Pro team was forked off to work on Itanium in the mid 90s. At that point a GPU was at best just the rasterizer and ROP phases from an acceleration perspective.
107.
by monocasa 2mo ago
It apparently works with https. Weird. Quite a journey to figure that out. https://courses.cs.washington.edu/courses/csep590a/06au/proj...
108.
by monocasa 2mo ago
So I feel like I need to write a blog post about this, but succinctly I think there's a good argument to be made that the issue wasn't the compiler despite popular wisdom. I don't even think it was the nature of unpredictabl
109.
by monocasa 2mo ago
There's more to open source than reproducibility. For instance introspection which is even more important with weights since detection of backdoors in models is NP hard IIRC.
110.
by monocasa 3mo ago
It's also mapped into the address space so there's very little extra latency in grabbing the embedding as opposed to something like nvme that will have to setup a command list, submit it to the drive's microcontroller, wait f
111.
by monocasa 3mo ago
Depending on where you slice the model up, it can be not a whole lot of data. For instance each transformer block outputs a single vector in an embedding space. I can see that being cheaper to bitbang with PIO than to actually compute. The
112.
by monocasa 3mo ago
You might be interested in IBM's archival digital storage on microfilm format from the early 1960s. https://en.wikipedia.org/wiki/IBM_1360
113.
by monocasa 3mo ago
The output is a probability distribution for all potential tokens. Then a "temperature" is applied to weight the sampling randomly (unless the temperature is zero in which case the stack can be deterministic and simply the highes
114.
by monocasa 3mo ago
Or just perform a form of distillation, where you don't actually change the hyperparameters, but maybe shift around the embeddings or something. You could even have another model watch the distillation process to check for goofy backdo
115.
by monocasa 3mo ago
I'm a security researcher. "Garbage" isn't the same thing as random noise. I have looked at the entropy graphs of encrypted data commonly. It's one of the first things I look at when looking at a new firmware image
116.
by monocasa 3mo ago
You didn't generally need to manage the ALAT as a programmer. You can, in the same way that you can manually manage reservation station port residency as a programmer in an OoO core to get maximum throughput, but neither of those are r
117.
by monocasa 3mo ago
If there's a difference visible, that's a weakness in the encryption algorithm. Encrypted data should be statistically indistinct from random noise. It's an information theory thing. Everything in the ciphertext should be si
118.
by monocasa 3mo ago
Once again, Itanium isn't a normal, simple in order core that stalls on loads. It has a big table called the ALAT to allow you to start loads as soon as you know the address, and then finalize them later when you're out of other
119.
by monocasa 3mo ago
I've heard about mathematicians going through kind of the same thing when they get a weird proof that ends up being right from some weird source or themselves. Which is fair, they get inundated with kooky proofs from amateurs all the t
120.
by monocasa 3mo ago
OoO cores have load stalls too after preemptions. Even something like a Apple M core basically has to stall if the load has to go out to the memory controller. The goal is to move far enough ahead in the stream that you can at least issue

More ›