JustSomeGuy1
About
- Banned
- Username
- JustSomeGuy1
- Joined
- Visits
- 60
- Last Active
- Roles
- member
- Points
- 1,172
- Badges
- 1
- Posts
- 330
Reactions
-
Alleged M4 benchmarks verify Apple's iPad Pro performance claims
AppleInsider said:The baseline M-series processor makes small gains with each generation, but the more important upgrades aren't tested by Geekbench.Small gains?!? Are you nuts? It's looking like 15-20%. That a *monster* number to put up. It instantly makes the M4 the fastest consumer CPU in the world, single core. (To the extent you accept GB6 as reasonable, which... is reasonable.)
You think that's holy? Wait until it ships in the M4Max, or the Studio M4whatever. If they are consistent those have a good shot at passing 4000.tht said:3767? Holy Shit!
-
The history -- and triumph -- of Arm and Apple Silicon
AppleInsider said:This is why Apple's chips usually match or beat rival systems even when the rivals have more RAM, more processor cores, and other seeming advantages. Apple can integrate software innovations with hardware features -- such as optimizing battery life -- on a level almost nobody else can.[...]It can't last forever, but Apple's ability to design chips perfectly customized to its own hardware has proven to be a major advantage.In a truly mediocre article, this stands out as a glaring giant pile of foolishness. The fact that many others have said it before doesn't excuse this. You should know better.This "integrating software and hardware" nonsense is exactly that: nonsense. Apple isn't doing better in benchmarks because of some mythological synergy. Instead, there are two well-understood reasons Apple's performance is better.Apple is willing to spend more on silicon. They often use more transistors than others to get the "same" job done. This will result in either better performance, or better energy efficiency. For example, look at the display controllers on the Mx chips. They are multiple times the size of the display controllers on x86 chips, and because of that, you get only two of them on Mx chips (more on Pro/Max), whereas low end x86 will have 3 or 4. But you know what you get in exchange? Low power use. Like, a LOT lower.More silicon also translates to better performance. They spent transistors on giant caches, ROBs, a thousand other things large and small, and in exchange got CPU cores that embarrassed the rest of the industry. They used large memory busses to get big bandwidth, again pushing performance. There are many more examples.It's true they've had other advantages at various times- they often have a process node advantage. But even without that, everything I've said is still true.I did say two reasons, didn't I? I did. The first is bigger silicon budget, but the second is, they're just better at this. At least for the moment.And BTW, the comparison with the Oryon (Snapdragon X Elite) does not cover Qualcomm in glory, once you look at the details. The M3 still beats that chip in most ways, including single-core performance, despite a clock disadvantage.AppleInsider said:[...]now Qualcomm's very best consumer-grade desktop and laptop chips struggle to break even with Apple Silicon at the same power-sipping levels.If by "struggle to break even" you mean "fail abysmally", then sure.
-
Ugreen DXP8800 Plus network attached storage review: Good hardware, beta software
You would have to be nuts to use this, now or ever, except maybe on an air-gapped network (no internet connectivity).Like all the others, this is surely running Linux. (Or possibly a BSD?) And like all the others, it's going to have security vulnerabilities that are discovered over time.At least with Synology and QNAP you have a reasonable expectation that they'll stay on top of that and issue timely updates. (Maybe, you hope, cross your fingers.) The chance that this vendor will do a competent job with this over the long term is, unfortunately, very low. I say that not because I know them, but because that's true on average for all vendors. Perhaps Ugreen will show that they are the exception, over time, but I'd want at LEAST five years of track record before trying them.The one possible exception would be if you plan to run a different OS from the start.
-
Does Apple's platform need to be opened up?
Oops. I wrote "Writing about evolution is *hard*, even when you really do understand it." And sure enough I left out something that matters."Variation necessarily leads to selection" - not always. You can have variation in alleles that has no effect at all on fitness. In that case, you get "genetic drift", where the differences just sort of wander through the generations subject to chance, and not economic forces. This isn't very common, though it's probably not rare. AFAIK this isn't settled because a number of traits that were thought to be irrelevant to fitness and thus subject to genetic drift were eventually discovered to have fitness repercussions after all.Oh, and also... none of this necessarily applies to humans any more. Natural selection on humans is at the moment overwhelmed by our intentional selection, and that trend will continue to accelerate sharply as long as society doesn't collapse. Though... such a collapse could be due to a strong selection event (say, meteor strike or plague), so the end of human evolutionary history isn't sealed yet. -
Does Apple's platform need to be opened up?
AppleInsider said:We can learn a lot about how to do things by considering the greatest technology ever produced on Earth: life. The genetic variety, adaptions, and innovations of life are not adapted purely to be low cost and efficient. Life has developed over millenniums of natural selection to be fit for survival: suitability for use.Leaving aside historical errors (which really don't matter), and the parts of the article I haven't gotten to yet (it's even longer than your usual), this deserves correction.Writing about evolution is *hard*, even when you really do understand it. Variation necessarily leads to selection, and selection happen on one basis: "fitness", which as a term of art is defined to be a measure of successful reproduction- though it's a little complex, because that includes descendants. So for example a variation that leads to lots of children who are all sterile would have zero fitness.People are often confused by anthropomorphising nature. They'll say things like "Evolution takes the long view! Sure it selects for traits that aren't low-cost or optimized, because sometimes species need higher-cost 'features' to survive adverse times. So you get traits that aren't normally economical because every X generations, any individual without that trait gets wiped out by a natural disaster." But... evolution *doesn't* take the long view. It really does only care about *right now*. Every generation in which that natural disaster doesn't occur, animals with that survival trait are selected *against* in favor of animals that don't have it, because it's more efficient not to have it. That's why, by the time that disaster comes around again, most of the species lacks it, and they die off... leaving every survivor with the trait, which will then take many generations without disaster to again become uncommon. (Or maybe, it's entirely lost, and that's how that species goes extinct.)So when you write "not adapted purely to be low cost and efficient", you can do some handwaving about what you mean by "low cost" and "efficient". Life doesn't always optimize for low-cost and efficient reproduction. But for whatever reproductive strategy a given species is taking (which can be very high-cost child rearing, or very low cost "lay eggs and wander off") evolution will always produce, over generations, the most efficient and lowest cost realization of that strategy.There are of course tons of interesting factors that can produce surprising results. For example, the development of a bimodal ESS in a given species, so you have two different strategies, and thus traits, both being selected for simultaneously. But for each of those, you get maximum efficiency.You can also have a species trapped in a "local maximum", which is the most efficient state you can get given the maximum amount of variation possible in a single generation. Evolutionary optimization isn't guaranteed to develop the most low-cost and efficient possible solution, just the most low-cost and efficient one available given the variation available.Maybe the shortest way to say this is, while there are tons of interesting things that happen due to evolution, you will never see it do anything that *doesn't* drive lower cost or efficiency, assuming you're measuring the only thing that matters, fitness.AppleInsider said:[...]we are ignoring the incredible power of genetics to deliver life forms that aren't just efficient, they are wondrous, and usher in new worlds of possibility that wouldn't exist if life decided at some point that there should be no exciting future, just an increasingly efficient status quo.That's wrong twice. It's not the status quo if it's "increasingly efficient", and that's exactly what evolution does. What it doesn't do is "decide" anything, or care about the future. The future just happens. Evolution only does "now".Now if you want to talk about what humans are doing with genetics, that's a different story. We now have a group defining DNA data storage standards. Science Fiction becoming Science Fact in real time. So cool.
