Стоим на пляже, смотрим как Мзымта впадает в Чёрное море, зацениваем радугу заодно.
Идея двери. Рама сварная, в неё пррсто доски вставлены вертикально. Минусы: щели, слышимость пердежа. Возможно решаемо.
This media is not supported in your browser
VIEW IN TELEGRAM
Стадион Фишт в standby mode
This media is not supported in your browser
VIEW IN TELEGRAM
Охрана трампа отпинывает айфоны.
Why humanity realized the dead end of parallel buses so late?
Today every normal link is point-to-point differential signaling pair or wires and communication is SERIAL and self-clocking. Such a pair works great at several GHz on "usual" cheap PCM boards without gold-plated contacts. When you trying to implement 64-wire parallel interface (plus clock signal), you cannot make physical board that works after ~500 MHz (and cheap enough). You get timing skew at "normal" PCB designs at several hundreds of MHz, because you cannot make those 64 lines perfectly similar to expect they transmit the "pulses" with similar speed, so when you receive your 64-bit word on receiving end you can know that all those bits are related to the same word. Making 64 lines with the same characteristics are hard. That is what killed PCI bus (also the fact that it is a BUS, not point-to-point, so you block others while you use the bus) and maybe all other parallel buses.
So my question is why it took so much time for radio-electronics-related humanity to get an idea that serial point-to-point differential signalling pair is much more versatile in many aspects? Why the world's engineering took so much time making different kinds of parallel interfaces? Couldn't it have been calculated that a pair of wires would operate at frequencies unavailable to a synchronized pile of wires?
Today every normal link is point-to-point differential signaling pair or wires and communication is SERIAL and self-clocking. Such a pair works great at several GHz on "usual" cheap PCM boards without gold-plated contacts. When you trying to implement 64-wire parallel interface (plus clock signal), you cannot make physical board that works after ~500 MHz (and cheap enough). You get timing skew at "normal" PCB designs at several hundreds of MHz, because you cannot make those 64 lines perfectly similar to expect they transmit the "pulses" with similar speed, so when you receive your 64-bit word on receiving end you can know that all those bits are related to the same word. Making 64 lines with the same characteristics are hard. That is what killed PCI bus (also the fact that it is a BUS, not point-to-point, so you block others while you use the bus) and maybe all other parallel buses.
So my question is why it took so much time for radio-electronics-related humanity to get an idea that serial point-to-point differential signalling pair is much more versatile in many aspects? Why the world's engineering took so much time making different kinds of parallel interfaces? Couldn't it have been calculated that a pair of wires would operate at frequencies unavailable to a synchronized pile of wires?