Lasers, robotics, is that only the beginning, or not?

[GTOM]

At DARPA Robotics Challenge while robots could solve a numbver of things, they were still slow and clumsy compared to humans, and required some amount of remote control, when they had to face with complex tasks.
Of course, it isn't good to send humans into radiation and hazards like that, but what are your opinions, robots present capabilities only the beginning, or that is what we could achieve with decades of continuous development? As far as i know processors arent that far from their physical limit, and a robot has limited space for a brain, especially, if it has to be shielded from radiation, EMP and similar hazards.

About lasers, they are still week compared to kinetics, they are happy to take out the most fragile targets. (I found the coilgun and EMP tests more impressive.)
What do you think, can we achieve a new breakthrough on this field, what can be the limits of their development?

 

[mfb]

but what are your opinions, robots present capabilities only the beginning, or that is what we could achieve with decades of continuous development?

My opinion: yes to both.

As far as i know processors arent that far from their physical limit, and a robot has limited space for a brain,

Human brains are within the physical limits, so there is certainly space for improvement.

especially, if it has to be shielded from radiation, EMP and similar hazards.

Good electronic components can survive more than a factor of 1000 more radiation than humans. EMP can be an issue, but usually not for emergencies.

About lasers, they are still week compared to kinetics, they are happy to take out the most fragile targets. (I found the coilgun and EMP tests more impressive.)
What do you think, can we achieve a new breakthrough on this field, what can be the limits of their development?

Accelerating something massive is just way more convenient if you want to destroy things, but I guess lasers will get some applications.

 

[GTOM]

How far present day processors from their physical limits, can they made several magnitudes faster or more compact without making them extremely sensitive?

Lasers sure have many useful civil and military applications already. (Although certain SF fans like to see death rays instead of oversized defence systems.)

Atomic rockets claimed that free-electron lasers have a maximum theoretical efficiency of 65% however i hasnt found another source. Can this only applies when they generate microwaves, or shorter wavelengths also?

 

[mfb]

How far present day processors from their physical limits, can they made several magnitudes faster or more compact without making them extremely sensitive?

That's the billion-dollar question...
With current technologies, there is certainly some limit - probably within 20 years if the current improvements continue. New technologies could change this completely.
Just to throw in a number: If you could somehow perform a single operation with 100 atoms in 1 nanosecond, 1kg of this material would be able to perform ~10³² operations per second. That is more than 10 orders of magnitude above the best supercomputers. Is such a thing possible? I don't know.

Atomic rockets claimed that free-electron lasers have a maximum theoretical efficiency of 65% however i hasnt found another source. Can this only applies when they generate microwaves, or shorter wavelengths also?

Efficiency for what? Conversion of electric energy to light? Looks a bit high, but I cannot rule it out.
Free-electron lasers for microwaves looks odd. Most of those lasers are used in the infrared to x-ray range.