Mel Gross
Offline
• • •
Upload & Sell: Off
|
 p.94 #2 · Canon 5D Mark II master thread | |
Mike1 wrote:
Jerry, you're missing the point. Look at CPU technology & MHz wars of the early 2000s, for a while, CPUs seemed to be going up in MHz, the highest MHz ceiling for ANY production CPU is 3.8 MHz, which was done by Intel's Pentium 4 back in 2004. The physical limitations of the CPU core only allows up to 3.8 MHz without damaging the CPU by causing instability to the system or overheating. The voltage was too high, power consumption went up to a whopping 124W idle and 226W under a full load while most of the other CPUs from AMD consumes between 100-162W (idle & under full load). For those that do not know about computers, anything that consumes 100W=enough to burn an egg after a minute or so. Afterwords MHz hype was abandoned, Intel decided to concentrate on lower voltage and hyperthreading eventually evolves into to dual core. Systems became cooler, power consumption was lower, and MHz was dropped to between 1.6-2.4 MHz again. The same goes for the physical limits of the sensor. You can't defy physics, sure you can make it bigger with software, but it won't be any better than what's already physically possible. It's like having an optical zoom vs. a digital zoom in a P&S. Another comparison is like having a 4 or 6 cylinder car, sure you can double the horsepower to make it more powerful, and faster, even with the most sophisticated computer system installed into it, you won't ever be able to make it fast enough to compete against F1 cars or the more powerful dedicated drag racers. Again, it comes back down to physics....Show more →
Again, what you're saying isn't true. You are making assumptions that have little basis in reality.
Just as there is no problem with increasing cpu speeds, as technology proceeds, though it did become more difficult, there is little impeding a raise in sensor density. That doesn't mean that all of these technologies have no limits, of course they do.
But those limits are way off.
I would rather believe Canon, when they state that it's possible to achieve 65 MP FF sensors, and that 50 MP sensors are certainly possible, then listen to people who know little about it claim they they think it's not possible.
Using the old car analogy isn't useful. The two areas have nothing to do with each other. You simply can't compare mechanical systems that must deal with weight, mass, inertia, physical strength, etc, with purely electronic systems which have to deal with none of those issues, and have completely different issues of their own, and completely different methods of dealing with them.
What you are missing about cpus as well, is that todays Intel cpus in a number of cases, can be clocked much higher than the old, supposedly higher clocked Prescott chips, and that's on air, no liquid cooling, where they can be clocked even higher. In a year, we will see officially clocked chips at the level of the last Prescotts, 3.8 GHz. After that, clocks will continue to rise as we get to 22nm. What happens after that, no one know now. But something else will come along.
This is what we are seeing with camera sensors. It was thought, several years ago, that 18 MP for a FF sensor was the max that could be done. We now know that at least 30 MP can be done, because it IS being done.
We haven't seen the last of this by a long shot. Get used to it.
|
