Relax wrote:Nanoscale application today is the main development tech of the future. Computing plateaued a decade ago. Communications 4 decades ago with the only difference changed from analog/digital to near 100% digital. Aerodynamics, 4 decades ago. Chemistry, 4 decades ago. Carnot Engines? 3 decades ago. Still waiting on Stirling engine tech...
Not exactly. You are painting with too broad a stroke. I understand your intent, but it is too general a statement.
Innovation, application, software design and architecture design, a few examples, hasn't quite plateaued
or peaked by a long shot. Software design might never, as it simply isn't afforded the opportunity. The constant evolving and changing of Operating Systems and the constant demand for more powerful hardware has never allowed the software industry to catch up. A fledgling programmer cannot become completely efficient in program design because the platform changes faster than his coding can keep up -- or his motivation to attempt.
If it levels out will afford a chance for software to evolve.
There is still very much room for software design to grow. Especially within the new paradigm of system architecture in the fairly new arena of HSA (Heterogeneous System Architecture) which I've currently placed many of my own projects on hold to assist in writing completely new algorithms to help facilitate the growth of this new industry. When insane amounts of money is offered to you for your programming skills, you tend to jump.
Computer design and manufacturing changes with both the availability of different technologies and the needs and demands of users
and software requirements. Current CPUs and GPUs are designed as separate components that don't work together efficiently. HSA solves that problem — SEE:
https://youtu.be/Me0g_FsgU0U and
https://youtu.be/Me0g_FsgU0U — and many others and enlightens the road to future advancements. I'm afraid that the computer industry is limited only by our own imaginations.
Programmers are forced to rely too much on brute force. As we are nearing certain physical and intrinsic limitations, our minds are forced elsewhere to completely untapped regions.
My personal Cray offers computing speeds of 176 teraflops. Of course, it isn't anywhere near the 17.59 petaflops offered by the world's fastest silicon champion the Titan, but that comes with a hefty price tag of $60M of which I cannot afford to indulge my whims. My Cray, bought as soon as it hit the market in 2013 was acquired for less than $600,000 maxed out, representing a $3M dollar value, achieved by innovative computer
design. That was achieved just three years ago. Which has given me a head start and advantage in testing and developing HSA written code. CPU performance, GPU performance and energy efficiency of the related HSA is already astounding and is only the tip of the iceberg. There are many more such areas in computing that has not plateaued because the mountain hasn't even been attempted yet. Yet is sitting there, idle. Awaiting minds that aren't... idle.
The thing to note is that there are several plateaus to be reached in the computing industry. Manufacturing, design, materials, innovation, technological innovation, energy efficiency, software design, languages etc., etc., etc., and even the grouping of these ideologies.
As a plateau is approached it becomes better for the end user because it becomes more appealing to the "better" programmers to invest enormous resources and time developing unprecedented code for machines, systems and architectures that won't become obsolete before releasing, therefore belaying development. HSA has that aim.