>>3435041>You were claiming the A7r3 sensor had to have better metalsNo that's just history revision. The context was you claimed we had no idea how advanced they were, which is where I pointed out you were wrong, we do have a ballpark clue in the materials used.
And yes, you do indeed need superior metals that provide better characteristics the smaller you make your transistors. So would the TV cables in your house IF you insisted on making them of 7nm materials.
>That IS the point because it costs money, space, and power to add the circuitry to bridge different bus widths.And yet that is exactly what our modern sensors are build to do, they are build to send out various signal types depending on what the customer needs.
Another gotcha moment that blew your mind.
>Phy?I see you don't even understand the topic at hand.
I'll explain to you one more time, the sensors are built to talk to many types of bus signals that links the sensor to the processor/buffer.
I know for a fact that you have never looked at a sensor whitepaper, otherwise you would have known this.
>375 MHz for a 256-bit bus.You don't even know if the entire sensor pipeline is 256-bit wide. That's just a number you conjured up.
Regardless, the wider you make the sensor, the smaller transistor size you need.
An no, Rolling Shutter can indeed be migitated by increasing the speed of the sensor, the faster the sensor, the less rolling shutter.
That's why I consider it one of the bottlenecks that can be solved my increased throughput.
>On a DSLR most of the chipsSo 99% throttled down like you said. Not much distinction here.
Potaeto Potaato.
>Throttled down.Not by 99% like you claimed. Just partially reduced. Mirrorless image sensor isn't a CPU.
>CPUs had nap modes and shut off functional unitsThis is actually irrelevant, because they still need to retain similar battery life when they record video for example.
Regardless of how much they throttle down in idle, neither of us knows how much.
