Why number no go up
Why number no go up
Falling into your wing while paragliding is called 'gift wrapping' and turns you into a dirt torpedo pic.twitter.com/oQFKsVISkI
— Mental Videos (@MentalVids) March 15, 2023
Why number no go up
Falling into your wing while paragliding is called 'gift wrapping' and turns you into a dirt torpedo pic.twitter.com/oQFKsVISkI
— Mental Videos (@MentalVids) March 15, 2023
Proof that God still loves us somehow
Invention is dead.
Silicone limit, wait for graphite processors
Wouldn't just about any metal work, at least in theory?
NTA, but graphene (not the Android OS) is the best element on earth for conducting electricity.
and it can't be formed consistently and reproducibly enough in production to make CPUs or really anything useful out of.
>Silicone limit
damn, no more big-titted bimbos?
Made me kek, but unironically when I was little and learnt for the first time about Silicon Valley my first thought was not on technology.
Because physics exists.
So they stopped making them faster, and started making them "wider" (more cores, more execution units, more instructions, etc).
Someone explain to me GHz and why higher isn't always better
Kill yourself
Because your graph cuts off a decade ago.
All else being equal, it is. There's tricks that let you get higher clock speeds at the cost of performance, though. The maximum clock speed of a single silicon transistor is well over 300 GHz, but you're not going to be doing very much with a single transistor.
Because once you get past a certain number they stop being gigahertz and go into more of a gigahurts range.
Physical limitation on the material. Too high frequencies are not stable on any CPUs. Look at what overclockers have to do to reach stable 5GHz.
This is also the main reason why back in 2004 manufacturers started thinking about multi-core architectures.
>not stable
what causes the instability?
is it a matter of using sturdier materials?
I'll explain
Frequency is nothing more than a voltage signal that goes from a state to another, usually from 0V to ~1.5V in modern CPUs.
CPU clocks have to do that fast enough for internal component to actually detect this change in the signal, but to go from 0 to 1.5 and viceversa it needs some time, even if VERY little (talking about nanoseconds).
So if you want to have 4GHz, it means the clock has to change its state 4billions times per seconds, which means a period (frequency goes from 0 to 1.5 and then back to 0) can't last more than 0,00000000025. This means that the higher the frequency the shorter the time the frequency need.
But metals have some time to discharge electricity. you go short enough and you don't see the differences in the voltage change anymore, hence the instability.
my off the shelf desktop cpu runs at a stable 5.2ghz without overclocking
>i did 104 in a 100 zone
>the police didn't get me im so leet
The fact that you thought this was postworthy is proof of how hard the 5GHz wall is.
It's almost like Moore's law ended in 2008, right around the time of Lehman bros crash.
World never recovered from 2008 financial crisis and is in regression ever since.
2008 is just the consequences of the 60s and 70s and in the US *that* was all caused by WW2.
2008 was George Bush thinking he could turn Mexicans into Republicans by putting them into houses they couldn't afford, and wasting resources and attention on Iraq.
and WW2 is just an extension of the problems caused by the first guy to take a shit in a urinal in the late 1800s
israelite
It's funny the Republicans are *still* trying to do this. That was half of Trump's platform.
What the HELL is that Y scale RETARD