adrian_b
The reason is that AMD could not make enough server CPUs to satisfy the demand for datacenters, due to limited production capacity at TSMC, so Intel, who makes server CPUs in their own fabs, could also sell everything they got.
So as you say, even the older fabs of Intel, which use the somewhat obsolete "Intel 3" CMOS manufacturing process, with which the old Intel Granite Rapids server CPUs are made, have been an important asset in these circumstances.
I also really hope that they will continue to design ARC GPUs with good FP64 performance, because they are the only GPU vendor with good FP64 throughput, which has even increased in the Battlemage generation, while all the other GPU vendors attempt to reduce more and more all the "professional" features of their "consumer" GPUs, in order to reserve them for overpriced "datacenter" GPUs, which are not affordable for small businesses or for individuals.
The hardware of the first Larrabee implementations was bad, being designed by some third-tier team in older technologies, without the resources used by the A-team that designed the Sandy Bridge CPU and its successors.
Nonetheless, after a couple of rebrandings the "Larrabee New Instructions" have become "AVX-512", and more recently "AVX10", and they are now the best part of the Intel-AMD instruction set.
Intel has tried to reserve the better instruction set for big spenders, but fortunately AMD has forced them to unify again their ISA, so starting next year, when Intel should launch the "Nova Lake" CPUs, Intel will have for the first time really competitive CPUs for personal computers, since the launch of AMD Zen 2 (mid 2019, so about 8 years before the launch of Intel Nova Lake).
Their sales of CPUs for personal computers dropped because the huge prices of memories discouraged everybody to make upgrades.
On the other hand, they sold every server CPU that they could produce, despite the fact that the AMD server CPUs are much better, but AMD also could not produce enough of them, so both AMD and Intel sold every server CPU that they could produce.
So their earnings in personal computers dropped, but their earnings in server CPUs have risen like never before, bringing a profit instead of the losses of the previous years (when they had to accept great discounts when selling server CPUs, to prevent their customers to buy better AMD server CPUs).
There is a bijection between these 2 spaces, so you can compute anything about a projective space using the "directions space" with an extra dimension, but nonetheless they are different things.
The projective method extends an N-dimensional space with a set of points at infinity that forms an (N-1)-dimensional space.
That means that a projective straight line has 1 point at infinity, a projective plane has 1 straight line at infinity, a 3-dimensional projective space has a plane at infinity and so on.
The other 2 methods are exemplified by the extended real number line and the extended complex plane.
The extended real number line has 2 points at infinity, not 1 point like a projective line. This method is seldom used in higher dimensions, but nonetheless it works in any dimension, i.e. a plane can be completed with a circle at infinity, a 3-dimensional space can be completed with a sphere at infinity, and so on.
The extended complex plane has only 1 point at infinity, not a straight line like a projective plane. This method can also be used for spaces with any number of dimensions, which are always completed with a single point at infinity, regardless of the number of dimensions. This method is used in the so-called inversive geometry.
While the other 2 methods of handling infinities are also useful in certain applications, the projective method of handling infinities is by far the most beautiful in my opinion, because it allows the unification of many concepts, simplifying a lot their handling. When I first learned projective geometry when young, it was quite a revelation. I consider only very few parts of mathematics and of sciences in general as beautiful as projective geometry.
When thinking about a problem, there is absolutely no need to think about the hack of homogeneous coordinates. That hack corresponds with a particular model of a projective space, where an N-dimensional projective space is mapped to an N-dimensional sphere embedded in an N+1 dimensional linear space, e.g. a projective plane is mapped to a sphere (with double cover).
When you make computations, homogeneous coordinates are the easiest, but when you think qualitatively it is easy to think directly in a projective plane or in a 3-dimensional projective space. E.g. in the latter you have an extra plane at infinity. Each point in that plane corresponds with a straight line in the 3D space, and also with any other straight line parallel to it. Each straight line at infinity corresponds with a plane in the 3D space, and also with all planes parallel to it. Each circle at infinity corresponds with a conical surface in the 3D space, and also with all other conical surfaces that have the same vertex angle and the axes parallel between themselves. And so on, any geometric figure at infinity will correspond with some surfaces or curves in the 3D space, which end at infinity into that geometric figure.
They are directly derived from the feminine Greek word "andreia", meaning "courage" or "virility", which had been derived from "andros", as supposedly a quality typical for males.
E.g. Ezra was Esdras in Greek and in Hebrew it probably ended in "-ahu", while Elisha was Eliseus in Greek and in Hebrew it ended in ayin after the "-a-".
So the original grammatical gender was feminine, hence the -a termination.
Many of the most important syscalls belong to this class, but most of them already have io_uring alternatives. Many other syscalls are not of this kind, so for them there is no reason to replace the traditional syscall interface.
For the syscalls appropriate for io_uring the essential characteristic is that their completions can occur in a different order than the submissions of their requests, which is why all previous methods of handling them were inefficient, because they required the user program to waste time with superfluous activities, instead of just picking the completion messages from a queue.
Where this out-of-order execution is not needed, invoking the syscalls in the same way as you invoke any user function is fine.
The initial implementation of io_uring lacked the alternatives for many important syscalls, but since then most of the holes have been filled.
Besides more obvious syscalls of the read/write type, which are used for files accessed in the classic way, in order to reach maximum performance with memory-mapped files the old "madvise" syscall must be replaced with IORING_OP_MADVISE, used with Linux-specific advices (MADV_COLD, MADV_PAGEOUT, MADV_POPULATE_READ, MADV_POPULATE_WRITE).
Therefore even the US Army has not been better prepared than supermarket chains.
Where you really need white light with a continuous spectrum, there are very few alternatives to arc lamps.
Some time ago there were published some interesting research results, which suggested that it is possible for incandescent lamps to make a comeback, this time at a efficiency comparable with white LEDs. That could be achieved with a carbon filament in a gas mixture that prevents evaporation, where all the inner walls of the lamp are very good mirrors for infrared light, but the front window of the lamp has an optical filter that passes visible light, while reflecting infrared light.
Such an incandescent lamp could have a much better efficiency than an arc lamp, and also a longer life (the very low efficiency of the traditional incandescent lamps is due to most of the energy being lost as infrared light).
It remains to be seen whether that research can lead to commercial products.
If they do not use a multiple LED structure but they still have a fluorescent conversion layer, the layer must be very thin and deposited on the top surface of the LED. It could consist of "quantum dots", like in some TV panels, i.e. very small crystals which emit light of a color depending on the composition and size of the nanocrystal when illuminated with ultraviolet or blue light.
However, "quantum dot" is a term that has already been used for a long time in marketing with the intention to suggest a superior technology. If they were using this, I expect that they would have bragged about it.
So from the wording of the press release I think that a multiple LED structure is more likely than a "quantum dot" conversion layer, even if the latter would provide both a better spectrum coverage and superior aging characteristics (by avoiding a color hue shift in old LEDs).
But the press release does not have any information about this, so the peaks could also be narrow, resulting in a poor CRI.
I had been obese for a long time, for more than a decade. During that time I did many failed attempts to lose weight, but eventually I did it in the right way and after about a year I had a weight more than 1.5 times lower than before.
That was more than a decade ago, and since then I kept whichever weight I desired. Nonetheless, to avoid gaining weight again, it was essential that I switched to eating only food that I cook myself from raw ingredients.
With the food that I cook myself, I plan the amount that I would eat and after eating that quantity I am satiated and I am not hungry during all day (I eat only twice daily, a breakfast after waking up and then the second meal in the evening).
On the other hand, if I eat food from a restaurant or ready-to-eat food from a supermarket, I am not satiated even after eating much greater quantities, and I become hungry again after a few hours.
The food that I cook myself has a greater percentage of protein and fat than typical for commercial food, only moderate salt, and most of it has no added sugar, but beyond that I have no clue where is the difference.
More protein and fat and less salt and sugar is likely to matter, but it cannot be a complete explanation, because there is a great difference even between the bread that I bake and commercial bread. My bread contains nothing else except flour, water, salt and yeast, but it is very tasty and after eating it I am satiated, while after eating a greater amount of commercial bread I would still want to eat more.
These 2 syscalls have a history typical for Linux syscalls, where frequently new syscall variants make completely obsolete their predecessors, but the old syscalls must still remain supported forever by the newer Linux kernels, to allow running legacy programs.
"pipe" has been obsoleted by "pipe2" (on 2008-10-09), which in turn has been obsoleted by "IORING_OP_PIPE".
"open" has been obsoleted by "openat", which in turn has been obsoleted by "IORING_OP_OPENAT".
Most users do not invoke Linux syscalls directly, but they use the standard C library, which has frequently shielded the end user from the evolution of the Linux syscalls. For instance, you might use the functions "open()" and "pipe()" in your C program, but the C library might invoke instead of those the more modern syscalls "openat" and "pipe2", after adjusting their parameters.
Moreover, for a large number of traditional Linux syscalls (including the many variants of read/write/send/recv) there are now better alternatives that are invoked as subcommands of a single syscall: io_uring_enter.
Nonetheless, about one third of those syscalls are obsolete (being replaced by newer improved syscalls), even if a large number of programs are still using them, and about one other third of them are rarely used in the environments familiar to most Linux users, like personal computers and small servers.
The obsolete syscalls inflate a lot the number of Linux syscalls, because there are a lot of syscalls that were originally defined with an inadequate set of parameters, then they were fixed by adding parameters or changing the meaning of the existing parameters. There are many syscalls were even 3 or 4 attempts were needed until reaching an acceptable definition for their parameters.
New programs should use only the newest improved syscalls, but the documentation of the Linux syscalls in various places is usually not up-to-date itself, so a lot of people are still using older syscalls, without being aware that they are suboptimal. So the obsolete syscalls will have to be supported forever by the Linux kernels (though in most cases when an obsolete syscall is invoked it just fills the parameters with adjusted values and then it jumps to the current syscall).
However, this does not mean that other required technologies, like for transportation or for protection against dust, cannot be solved before.
I also agree that recycling technologies are actually more important than spacecraft technologies, because they are equally applicable on Earth, not only on the path to other planets or in bases established there.
I do not believe that interplanetary travel will be possible for humans otherwise than in spaceships with artificial gravity produced by rotation.
This not only would avoid the health problems (the experiments on mice in the ISS imply that a gravity as high as 2/3 of that of Earth might be necessary for preventing harmful effects during long-term missions) but it would also simplify many of the recycling technologies, because handling fluids and powders is difficult in weightlessness. It would also simplify the growing of plants or fungi.
In big cities there are a lot of people who, at least for a big part of the year, spend all their lives only inside their home apartment and inside their work office.
If someone had been willing to spend many times more money on a base in Antarctica, it could have been built to have a much greater interior space and with much more comfortable facilities, so that for the people living inside it there would not have been significant differences between living there and living in some big city of the Earth, especially now, when with modern communications they can speak face-to-face with anyone else on the Earth, at any time.
These 2 tasks can be solved with only the current knowledge, but they depend on the solution of many technological problems that require only a lot of money and of work.
Only a fraction of the huge amounts of money that have been invested now in AI would suffice to solve these problems, if a serious effort would be done towards such goals.
Whether it is useful to spend so many resources on such goals is a different question, but developing the technologies for an almost-closed self-sustaining environment is something that certainly would also generate by-products that would be useful on Earth, not only in interplanetary expeditions.
A very good CRI does not guarantee a very good color reproduction, but a poor CRI guarantees a bad color reproduction.
That could be done if the LED actually contains a multiple structure combining red, green and blue LEDs, but in the past such a combined LED had the disadvantage that in time the efficiency of the 3 components degraded at a different rate, which resulted in a color shift.
For now, the luminous efficiency of this LED is quite poor, unsuitable for lighting but only for indicator LEDs or display applications, but they claim that for such small LEDs even the traditional LEDs had poor efficiency and they hope to improve the efficiency later to more acceptable levels.
Modern internal-combustion engines have 35% to 40% efficiency, with even efficiencies of over 50% for the very big 2-stroke Diesel engines used in ships.
Nonetheless that remains about half of the efficiency of electric vehicles and the energy efficiency of the latter can be greatly increased over this ratio by the recovery of the kinetic or gravitational energy, when braking or descending.