adisakp.bsky.social
Computer Programmer. Foodie. Anime Fan.
USA. Midwesterner
107 posts
105 followers
95 following
Regular Contributor
Active Commenter
comment in response to
post
Grok: This reflects the extreme concentration of wealth, where the top 0.1% (~133,200 households) hold as much as the bottom ~106.6 million households.
2/2
comment in response to
post
I asked Grok what % of Americans would you need to equal to the wealth of the top 0.1%.
Grok: Approximately 80% of the bottom Americans (by household) own the same amount of wealth (~$22 trillion) as the top 0.1%, based on Q3 2024 Federal Reserve data and interpolation of wealth distribution.
1/2
comment in response to
post
Bonchon is the gateway to spicy Asian food. The have the fried chicken that’s so crunchy the skin shatters audibly when you bite it — and their spicy is actually hot enough to make your lips tingle.
comment in response to
post
Interesting... Meta AI (Facebook / Llama 4) got the correct answer to the same question.
comment in response to
post
For what it's worth, ChatGPT also got the answer wrong
chatgpt.com/share/67fd95...
comment in response to
post
Figment is the mascot of the Imagination Pavilion at EPCOT at Disney World
comment in response to
post
Could be Figment?
comment in response to
post
When I was doing a lot of work with ffmpeg, I made A LOT of batch scripts.
comment in response to
post
It’s a good arc… definitely some heavy stuff goes down later in the arc.
comment in response to
post
Same
comment in response to
post
38 years and I don’t know how it works. LOL.
comment in response to
post
LOL
comment in response to
post
Hardware bugs are crazy. Compiler bugs and low level system software bugs are a pain in the butt too.
comment in response to
post
What keyboard brand? That’s likely the “FN Lock” key and it will change the state of your number / function keys. It’s like a CAPSLOCK key for FN. Although if it’s next to the trackpad if might do something else.
comment in response to
post
I found a lightning cable today while looking for a USB-C cable.
comment in response to
post
Yeah sorry. This only works for watching Hulu on the Chrome Browser right now. I hate the obscuring fades that all the streaming apps add though :-(
comment in response to
post
Looks useful.
comment in response to
post
If the “fade-to-black during show” is a different element, I could probably modify the plugin to remove it if you can identify the name of the element.
comment in response to
post
Source code for the plugin is here:
github.com/adisak/Hulu-...
comment in response to
post
What device are you watching on? I was having a similar issue with their fade scrims while watching anime and I wrote a chrome plugin to reduce the fade effect.
chromewebstore.google.com/detail/hulu-...
comment in response to
post
The trim on dry aging does feel a little wasteful but I use fat for tallow and pellicle and extra bones for stock.
comment in response to
post
And this is the resulting prime rib roast
comment in response to
post
Interesting that it’s just incorrect metadata. Thanks
comment in response to
post
Totally valid use case. Noise cancelling tech is awesome - especially active (vs passive) noise cancellation. I always put mine in the gym to decrease the volume of the loud music they play there.
comment in response to
post
Of course the drawback to wired adapters is carrying them around (and for some people, the wires) but you can just leave the adapter attached to the end of your headphone cables. I don’t recommend headphones with the usb-c adapter already as part of the cable because that part is prone to failure.
comment in response to
post
As a bonus, for wired adapters, you don’t have to worry about charging your headphones and you don’t have to worry about losing your AirPods (some of the models are notorious for easily falling out of people’s ears)
comment in response to
post
It’s simple though - just responding to a single request box once when you first open the case near the phone - and it works flawlessly. If this is a barrier to entry for you, they do have adapters for usb-c to headphones (or lightning to headphones for previous generations of iPhones).
comment in response to
post
AirPods are ridiculously easy to set up. Just open the case near your phone and they ask to pair - it takes about 10 seconds. After that, they just work. There’s a reason Apple went with their own technology for pairing rather than straight Bluetooth for AirPods.
comment in response to
post
It adds a feature similar to the ‘ incbin “binary.file” ’ directive that is common in assembly language to C++. When you “embed” a file in C++ is will initialize an array variable to the binary contents of that file at compile time.
comment in response to
post
Check the amount of physical memory mapped post malloc and then after you run your “memory overflow” test. If those differ by a significant amount it might be dynamic physical page mapping to virtual memory reservations for the heap memory.
comment in response to
post
Is there a way to check how much memory is committed (physically mapped) to a process in MacOS? They might be using an allocator that does a single virtual memory reservation and dynamically assigns pages on access for their clib allocator.
comment in response to
post
Embedded systems almost never zero memory in release build of standard clib because of performance reasons. You’d almost certainly have to have a custom allocator or map malloc to calloc manually to get that behavior.
comment in response to
post
You might be using a custom allocator that has non-standard behavior. But if you’re expecting language standards defined behavior, don’t assume 0x00. As I stated, you’re also going to get a debug fill pattern (0xCD or 0xA5 or something else) from many debug clib versions.
comment in response to
post
The the OS zeroing memory during physical memory mapping to virtual memory is VERY DIFFERENT from the clib allocator zeroing memory. The OS is preventing cross-process data leaks. When the clib allocator returns a block without a mmap in a release build it usually won’t clear the block first.
comment in response to
post
Now if you’re in a release build (where clib malloc() won’t fill with memory patterns) AND you allocate a block that requests a mmap operation from the OS, that will return zeroed memory on *most* OS’s as the OS itself zeros memory for security reasons when mapping physical memory to virtual memory.
comment in response to
post
I’m not sure what you mean by “most”. The clib allocator isn’t required to nor will zero memory for allocs for performance reasons in release builds. Neither does dlmalloc or many other common allocators. Also, don’t assume zeroed memory - MS debug libs fill heap with 0xCD while FreeBSD uses 0xA5.
comment in response to
post
Some systems do use that access exception to map physical memory to the virtual addresses on the fly but it’s much higher performance to map in larger blocks. (2/2)
comment in response to
post
The memory they were accessing was probably physically mapped. Accessing unmapped memory will cause an exception (even for virtually mapped addresses without physical store). (1/2)
comment in response to
post
Please please please let me know what you think if you do watch it :-)
comment in response to
post
Video game consoles has faster “large pages” that can be 1MB or 2MB in size. PC’s have large pages too but PC allocators typically use 4K mapped backstore (at least on Windows) due to limitations with VM page files.
mmap’ing per alloc becomes very slow and wasteful for smaller allocations.