Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
to11mtm
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
51 ms
·
751.
▲
by
to11mtm
4y ago
Jitting is useful for things like dynamic loading/dispatch and simplifying distribution of libraries around it. FBOW it's worth remembering the 'cruft' and legacy around .NET's Architecture designs and choices, even
752.
▲
by
to11mtm
4y ago
I more or less agree. I have two exceptions that come to mind... My Galaxy J7 (Original version) was a lovely little device that still works, aside from lack of updates from the mfg. It wasn't fancy or fast, but it was cheap and has be
753.
▲
by
to11mtm
4y ago
> It seems that nobody cares at Google. My dystopian response is that they know if you're going 'all in' on their branded phone you are either geeky enough to deal with it, or drank too much kool-aid to care.
754.
▲
by
to11mtm
4y ago
Oh they know. I've had call issues (that included ability to make emergency calls) since the G1 days. Oftentimes it is a diceroll between the firmware of the MFG and the version of android. One of the things I miss most about WinPho is
755.
▲
by
to11mtm
4y ago
This is, in some ways, the unseen benefit of 'patterns' in software. [^0] Outside of certain problem domains or extreme requirements, the basic patterns work well enough for a variety of tasks with the right mixing, you usually ge
756.
▲
by
to11mtm
4y ago
> This can be a contentious subject because there's a lot of nuance and the right answer is often context-dependent. But I personally think that the Java style of "we must absolutely protect the library user from themselves and
757.
▲
by
to11mtm
4y ago
That's a fair assessment although perhaps not giving AMD enough credit. IMO GloFo's spin-off worked out very poorly for AMD in the short term, but long term it let them partially-leapfrog Intel much as they had done 20 years prior
758.
▲
by
to11mtm
4y ago
> The article notes some advantages of being able to use different processes for different wafers, but there's not much more on that. It might be helpful if you wanted different fabs for memory and compute, or for flash type memory.
759.
▲
by
to11mtm
4y ago
> But I wonder how accessible it is for regular TSMC customers. TSMC is part of the UCIe (Universal Chiplet Interconnect) consortium, so I'd assume they have some capability. But the other members are ARM, Intel, AMD, Qualcomm, and
760.
▲
by
to11mtm
4y ago
Yeah... Back in the day, as another comment mentioned, the PPro had a 'chiplet' style configuration where The CPU and Cache were on the same chip but separate dies. The problem with this was the CPU and Cache had to be bonded to t
761.
▲
by
to11mtm
4y ago
Wellllll... Ryzen is only 6% behind for multi-core PPW, but The M1 appears to still have a huge advantage in single-core PPW. I have a half dozen thoughts on this, but the foremost is; Apple keeping memory on chip, is likely providing a hug
762.
▲
by
to11mtm
4y ago
IIRC they have something like that but different in the works, 'Software Defined Silicon' (SDSi) is what it's called.
763.
▲
by
to11mtm
4y ago
I've had a couple... Two stand out. One was my glorious AXP1800 + Gigabyte GA-7DXR. Ok so it -started- glitching eventually, but that was because capacitor plague and not the AMD-ness. But that was a workhorse machine of mine through m
764.
▲
by
to11mtm
4y ago
There's two groups of people I can think of that do this: - People who like to frequently change cars, not so frequent as to lease but still prefers to let it go between 50-100KMi. (Contrast folks who sell past 100k or drive cars into
765.
▲
by
to11mtm
4y ago
> Third, disc brakes and a quick release on the front wheel is a known safety hazard? A staggering number of bikes have that setup. If it’s truly as dangerous as the plaintiffs say, literally hundreds of thousands of bikes will need to b
766.
▲
by
to11mtm
4y ago
> I'm obviously cherry picking, but if an applicant asks me a series of questions akin to "what source control tool do you use and why" I'm going to begin doubting that this person can focus on actually shipping produ
767.
▲
by
to11mtm
4y ago
> I've seen teams spend a lot of time in low/no-code tools and either it grows more complex than actual code or they resort to the escape hatch (node that executes user defined code) and the visual tool basically becomes a cont
768.
▲
by
to11mtm
4y ago
> It could even be a relatively fake example and I think it would still land if your audience has had experience exploring large codebases The problem is this can get pervaded quickly. In most extreme cases, the 'example' is li
769.
▲
by
to11mtm
4y ago
> I haven't had much problem spending the older ones Generally speaking you have to take it, within a certain point of reason. I will say that in my time in retail, I've seen 50s or 100s that looked 'weird' to me but
770.
▲
by
to11mtm
4y ago
I'm a little surprised in ASML's pivot here TBH. Just earlier this year they were complaining that Chinese companies were ripping off their IP [0] and IIRC even within the last 2-3 years they were themselves hesitant to send bleed
771.
▲
by
to11mtm
4y ago
I mean... in the nightmare case... I honestly don't know if you can 're-clean' a clean room if it's flooded with the right things. And the right things might just be seawater. (IDK. Not a fab engineer by any stretch.)
772.
▲
by
to11mtm
4y ago
Well TSMC does do the 'state of the art' yes. I'm guessing they have some less modern stuff. Maybe . But. BUT. UMC is kinda an unsung player here to some extent. They are another Taiwanese Fab, and do a lot of the older proc
773.
▲
by
to11mtm
4y ago
> Even military hardware uses very outdated chips. I could be wrong but I thought that some of the ML stuff for military is causing more demand on the high end?
774.
▲
by
to11mtm
4y ago
> He also said they usually left much more daring things to safer places and preferred to keep the air show maneuvers less risky since crashes are not a good marketing plan. Makes sense. Fancy maneuvers at lower altitudes (i.e. Air shows
775.
▲
by
to11mtm
4y ago
(Not a DB Engineer but poke at enough different DBs to armchair semi-confidently) In the case of Postgres I'd expect to see it indeed be mostly about random updates to leaves. I suppose Partitioning could come into play but I'm no
776.
▲
by
to11mtm
4y ago
Not all. AFAIK the spray-in bedliner is a physical component. Tonneau cover, depends on the model. Mine, there was a slight delay, but FWIW was a 'rolling' (i.e. no electronics) style. Mud flaps, pretty sure are not electronic. As
777.
▲
by
to11mtm
4y ago
This is the future we have brought :`) Data analytics and JIT often leads to an invisible cost to the lower class. I remember the days where my parents would go to the meat store and make some random thing off schedule . I remember when I
778.
▲
by
to11mtm
4y ago
DRAM. IIRC TSMC's 135MBit 5nm example is 79.8mm^2, although that's got other logic. In the abstract, a 0.021 square-micrometer-per-bit size [1] says you'd need about 21mm^2 for a gigabit (base 10) of 5nm SRAM, without other l
779.
▲
by
to11mtm
4y ago
There's a fun curve on this to be sure. If I had to guess, Servers would not go any further than some sort of memory-backplane where the memory for multiple channels was integrated onto a single PCB. Even then, IIRC hot-swapping of mem
780.
▲
by
to11mtm
4y ago
I mean it kinda depends to an extent. If I had to distinguish Taco bell from, say, McDonalds, I would say that Taco bell has the advantage of being filling but not putting you in a food coma. One of it's biggest unsung heroes is the be
More ›