Neywiny

17 ms·

421.
by Neywiny 10mo ago
Maybe they're starting to realize that the first step to fixing a problem is admitting you have one? Feels safer to say "look we didn't do well and we're going to not do well for a bit, but we'll get back on our fee
422.
by Neywiny 10mo ago
What pattern in the data shows that's what's being measured? I would expect to see basically 0 latency between adjacent "cores" then since L1 is shared per thread?
423.
by Neywiny 10mo ago
I don't understand on one of the later graphs the core to core latency for strix halo goes out to 32 cores but he says only has 16 cores?
424.
by Neywiny 10mo ago
1. Be not good at or a fan of git when coding 2. Be not good at or a fan of git when committing Not sure what the disconnect is. Now if it were vibecoded, I wouldn't be surprised. But benefit of the doubt
425.
by Neywiny 10mo ago
Hard disagree. Initial commit was 6k LOC. Author could've spent years before committing. Ill advised but not impossible.
426.
by Neywiny 10mo ago
For moms, not for women, but sure
427.
by Neywiny 10mo ago
Really wish the cost of the drives also went down. I'd imagine a lot of good could happen if it was easier for people to choose the next tier of price/capacity/performance trade. Mostly so it'd make HDDs cheaper for me
428.
by Neywiny 10mo ago
Works just fine with mine
429.
by Neywiny 10mo ago
This is what I keep hearing. "You just need something more agentic" "if you had the context length you could've fixed that" etc etc. yeah sure. I'll believe it when I see it. For me it's parsing 3000 page
430.
by Neywiny 10mo ago
Micron isn't pulling out of the market. They discounted Crucial. Those are very different things. They pulled out of the direct to consumer market, not DRAM
431.
by Neywiny 10mo ago
I spent some time reading the later chapters which are what I'm fuzziest on. It's nice to have the python code there. I do overall think this is a good resource. However, the hand-waviness of the specifics gets to me. The author e
432.
by Neywiny 10mo ago
You know what, fair point. I know because my fans go haywire (except that one time my fans decided to just not work until I ran some incantations. The joys of Linux). If you're passively cooled, you get no feedback on CPU load. Makes s
433.
by Neywiny 10mo ago
Seems like good work. From what I've heard developing on MacOS has not gotten easier over the years. I do question the point, though. I suppose there's merit in knowing if your configuration causes thermal throttling, but what are
434.
by Neywiny 10mo ago
I use https://github.com/graphitemaster/incbin . It mostly works (had to make some mods when I needed more section magic to happen). Nothing I'm on supports #embed yet but here's to hoping.
435.
by Neywiny 10mo ago
I would also argue it shouldn't be a gif. It's nice that it shows the command is fast I guess but it's one command that's still visible in the final frame. Not as bandwidth efficient and agreed I can't read it all i
436.
by Neywiny 10mo ago
No, it's really not that good a feature and turning it off improves my experience so I don't care if they're the only ones with it. That said, if it's part of the open source, when better. And even if it was, I can'
437.
by Neywiny 10mo ago
While I agree that new feature adoption is hard, changing tab completion is a bit hardcore. I agree that a different key bind, maybe right arrow key, or shift tab, or something would have been better.
438.
by Neywiny 10mo ago
I've had to take a beat to find the right words as all the frustration in the issue ticket impacted me too which has left a very bad taste in my mouth after being initially curious and open to the new feature. I think you're doing
439.
by Neywiny 10mo ago
I self host. I also have no paying customers and negligible compute needs. It's free minus the power cost which is, again, negligible. If it was on docker on my main computer instead of 18 year old junk, it would probably be even less.
440.
by Neywiny 10mo ago
It did, yes. On an architecture without bit field extracts.
441.
by Neywiny 10mo ago
Even worse, the windows drivers know they're knock offs and let you use them for a bit and then error out. It's the perfect lesson. They give you that high of "yes, I got one over on the big guys, we're cruising now&quot
442.
by Neywiny 10mo ago
I'm once again surprised at GCC being slower than clang. I would have thought that GCC, which had a 20? year head start would've made faster code. And yet, occasionally I look into the assembly and go "what are you doing?&quo
443.
by Neywiny 10mo ago
I think it depends on your skillset and goals. If you want to run other peoples' software and compilers and whatnot, it doesn't hurt to implement to an existing spec. If you're going to write your own or need a lot of mods, m
444.
by Neywiny 10mo ago
You need to choose another field if it takes you 6 months to hello world
445.
by Neywiny 10mo ago
I'm all for your endeavor, but didn't see a device support list on your front page. Clicked the first of 2 links in your sidebar (docs) and got a 404. I'm not saying it's telling that your issues page works when your doc
446.
by Neywiny 10mo ago
I think there will always be vendor lock in. The same way there have been architectural differences between Intel and AMD's x86, or even stuff like one specific chip/family tanking performance because one instruction was implement
447.
by Neywiny 10mo ago
Not for anything mid to higher range, but I believe there's open source tooling for some of the older Lattice and Xilinx parts. I would say for me it's not as big a deal as on the software side, because each vendor's hardware
448.
by Neywiny 10mo ago
I do the vast majority of my work on xilinx and it's easiest to just use the built in simulator. It's free and supports vhdl and verilog. Most support just one. For lattice and microchip work I use what the tool provides which is
449.
by Neywiny 10mo ago
I recommend getting started like the author did: simulation first, then FPGA. Honestly FPGA will take you very far. I always get a kick out of being able to design my own SoC. "Hmmm I need 9 separate I2C ports... Ok, copy block, paste
450.
by Neywiny 10mo ago
That's a very nebulous metric. Usec of overhead depends on a lot of runtime things and a lot of hardware options and design that I'm just not privy to

More ›