p_l

21 ms·

421.
by p_l 1y ago
Fun fact - if you handle DNT properly, you don't need to show the consent screen... because you're not doing anything requiring said consent.
422.
by p_l 1y ago
Guess what, those banners are still up because it's pretty hard to actually bring the banhammer. At best you have too small team working with huge backlog
423.
by p_l 1y ago
BMC - vendor neutral RedFish - Current DMTF standard replacing IPMI ALOM - sometimes encountered in the wild as part of iLO :O HMC - IBM on non-x86 systems (except OpenPower), not compatible with anything else RSA - used to be
424.
by p_l 1y ago
There's a terrorist attack, sorta, in latter book, but Red Mars was admittedly a legitimate military strike, as was destruction of Phobos.
425.
by p_l 1y ago
Was that still on the VMS-based blockbuster video system? Weird question, but I accidentally ended up with one of those in my hands that ran in probably non-blockbuster place from 1996 to 2000 :)
426.
by p_l 1y ago
Is it the OpenMP related one or another thing? I kinda lost track , this whole thread reminded me how hopeful I was to play with GPGPU with my then new X1600
427.
by p_l 1y ago
sshhhh, you're breaking the perfect invisible hand of market Germany wants to dogmatically push in EU grid The mandatory EDI platform to interact on German market is also a bit annoying, though it's in details theory is theoreti
428.
by p_l 1y ago
Germany is very weird in a certain way of total belief in power of free market. So you see, the market was supposed to correct that. But profit laid with cheap gas turbines to backstop wind and buying from france ;)
429.
by p_l 1y ago
Well, unless you invest non-trivial extra work, you're going to start showing up on grid monitoring and finally someone will drill down in problem looking for broken substation and find you doing exactly the things that caused the prob
430.
by p_l 1y ago
Heard lots of grumbling from an acquaintance in Germany that a big issue is, I quote, "Bavarians not wanting either overground nor underground power lines that would bring power from north to south, so at best we sell wind power from n
431.
by p_l 1y ago
Because curtailment is about the available pickups not being able to take the energy. If you had a sink for it, you wouldn't be paid for curtailment so long as it was send through the grid. I bet if you used an alternate local grid the
432.
by p_l 1y ago
Actual execution of curtailment depends on mechanism of individual turbine and various support equipment at provider. Essentially the "electricity market" is putting out orders for increasing or decreasing production (plus various
433.
by p_l 1y ago
Yeah, LLMs are easier to keep available ;)
434.
by p_l 1y ago
(I will use "GPU" because "3d accelerator" was very gaming PC oriented term predated by 3d graphic hardware for decade) Only in consumer market - which is why GeForce 256 release had the game devs with engines using GL s
435.
by p_l 1y ago
Which one? We first had the flurry of third party work (Brook, Lib Sh, etc), then we had AMD "Close to Metal" which was IIRC based on Brook, soon followed with dedicated cards, year later we got CUDA (also derived partially from B
436.
by p_l 1y ago
And yet CUDA has looked way better than ATi/AMD offerings in the same area despite ATi/AMD technically being first to deliver GPGPU (major difference is that CUDA arrived year later but supported everything from G80 up, and nicely
437.
by p_l 1y ago
The protocol is open so you can build software for other systems that uses it, whether it's GNOME or macOS
438.
by p_l 1y ago
It exists partially because of ossification of protocols that killed SCTP. As in, we wouldn't have to implement it on top of UDP
439.
by p_l 1y ago
AVX-512 size is explicitly a derivative of P54C having 64 byte cache line in fact
440.
by p_l 1y ago
There was no locking (at least not necessarily), it was a shared resource that allowed programs on multiple computers to utilize together (also major use case for RAMsan where I worked with them - it was not about not being able to add memo
441.
by p_l 1y ago
I think it was implied to me that every operand of ADDP6 could have been indirected, and all the resulting 3 ranges all could straddle pages.
442.
by p_l 1y ago
Main cases I've seen with mainframes involved network-attached ram disks (actually, even earliest S/360 could share a disk device between two mainframes, so...) A fast scratch pad that can be shared between multiple machines can b
443.
by p_l 1y ago
I might be mistaken then, but I recall reading something about most extreme decode on VAX going into ~522 bytes. What I am more certain of was complaints about possibly ending with maaaany TLB lookups (and pagetable walks) for certain &quot
444.
by p_l 1y ago
few classes above. Only ISA where I've seen a single instruction span two memory pages despite being page aligned
445.
by p_l 1y ago
Yes, it's not possible to do it the same way, because what made x86 successful in their application is that x86 was remarkably RISC-y in actual behaviour compared to 68k or VAX. The main reason for that is that x86 code, outside of bei
446.
by p_l 1y ago
It was often used to supplement memory available in cheaper ways or otherwise more flexible. For example many hardware solutions allowed connecting more RAM than otherwise possible to be accessed by main bus, or at lower cost than the main
447.
by p_l 1y ago
Historical, but also there was a bunch of physical ram drives - RAMsan, for example, sold DRAM-based (with battery backup) appliances connected by fiber channel - they were used for all kinds of tasks but often as very fast scratch space fo
448.
by p_l 1y ago
Jobs dropped such suggestions after he was informed Keynote would get hit back
449.
by p_l 1y ago
I mean, at its most basic, it's a feature from even the earliest compilers for C. Not sure if Borland or MS shipped big fat symbol tables for all hardware registers of an IBM PC though?
450.
by p_l 1y ago
Some of the older autonomous challenges involving flight I remember a lot of contestants had QNX on board vs. let's say VxWorks or others. I also guess QNX having more point-to-point IPC/RPC fits some stuff better without having t

More ›