The GHz isn't the main issue; it is latency. We were stuck at 3-3.5 GHz for a long while with dual core being super expensive.
Now we're at 6-6.2 GHz in turbo mode. Can we go faster?Likely, but I don't see 12 GHz within forseable future. Instead, internal parallelism within a core have increased significantly. The bottleneck is cache reload and flushing from/to RAM. Highly optimized programs /may/ fit in cache (assuming they don't get swapped out due to task switching) but most don't since they use large frameworks and/or are interpreted or runs on Windows that since dawn of time always take 2-3% CPU.
But there's a lot of programs that are unable to utilize multiple cores, so even if there's 20 cores available, 19 are unused and the last one is limited.
Lower voltage allows for somewhat faster clock speeds, allowing faster execution. Microcode based instruction sets (x64) benefit the most from internal parallelism due to the ability to begin an instruction before the previous has completed, while ARM CPUs generally split the entire instructions into independent parts that all are executed simultaneously, but there's a limit on how much that can be parallelized given a certain RAM bus width. Increasing RAM bus width, expecially external RAM bus width is likely where the biggest gain is.
* Politically, since the USA is no longer a reliable partner.
* Strategically, since the Gripen is designed specifically for defence against russia in an arctic environment from rual highways, while F35 is an offensive plane mostly designed for use in warmer climate from fixed air bases.
There used to be a wealth tax and land value tax in Sweden (aka world champion of taxation), but they were abolished simply because being taxed yearly on an fixed asset doesn't mean you have the liquidity to pay the taxes from your income. They had also caused the some of the wealthiest people in the UK to be Swedes, IKEA being Swiss and Dutch, and a lot of other movements of capital to other countries. All in all, wealth and value tax are a big loss of income for the government.
But why can't we just say "2% over a billion, 1% over a million; 50% if you choose to move your assets out of the country". It does not seem that unreasonable to insist that you keep your monies in the country that lead to your wealth?
Generally, billionaires have already moved their assets to tax-havens. Because it is the sane thing to do. And the minute a bill for anti-movement is scheduled for a vote, the rest will be moved before the bill comes into effect.
Instead of taxing leavers, one should provide tax cuts for returners. Their investment locally means jobs will be created locally and taxes will subsequently be paid locally anyway. Win-win for everyone.
It might be legal in YOUR jurisdiction, but at least in the jurisdiction I'm in, it is not - AFAIK - legal to neither underwrite or overwrite costs on the sole purpose of avoiding tax or grooming the pig.
Although I've never worked at a FAANG, but everything except the internal competition to be seen as "the guy" is actually true for any big, multinational company. The alignment, the big leverage of a mistake, the huge code bases that only "the guy who left half a year ago" knows about and the cultural differences (apart from the challenges to work with both Chinese, European and US timezones at once) is all the same.
That said, it gives lots of experience. As does working in startups as well as mature SMEs. Having done all three, I can't say either is better. They're different, but some are better at a certain point of life and not as good in other points of life.
I do think it's a good experience to try either way since the challenges are so different from that of startups. These large multinationals look for different skills some of which you can't really develop at startups.
Now we're at 6-6.2 GHz in turbo mode. Can we go faster?Likely, but I don't see 12 GHz within forseable future. Instead, internal parallelism within a core have increased significantly. The bottleneck is cache reload and flushing from/to RAM. Highly optimized programs /may/ fit in cache (assuming they don't get swapped out due to task switching) but most don't since they use large frameworks and/or are interpreted or runs on Windows that since dawn of time always take 2-3% CPU.
But there's a lot of programs that are unable to utilize multiple cores, so even if there's 20 cores available, 19 are unused and the last one is limited.
Lower voltage allows for somewhat faster clock speeds, allowing faster execution. Microcode based instruction sets (x64) benefit the most from internal parallelism due to the ability to begin an instruction before the previous has completed, while ARM CPUs generally split the entire instructions into independent parts that all are executed simultaneously, but there's a limit on how much that can be parallelized given a certain RAM bus width. Increasing RAM bus width, expecially external RAM bus width is likely where the biggest gain is.
reply