Breaking News
Comments
Newest First | Oldest First | Threaded View
selinz
User Rank
Author
re: Next computing target容xascale systems
selinz   11/17/2010 7:47:50 PM
NO RATINGS
I agree. The "changes" that are being made architecturally are probably trivial compared to the changes needed in the software architecture to enable independant thought..

LarryM99
User Rank
Author
re: Next computing target容xascale systems
LarryM99   11/17/2010 2:33:24 PM
NO RATINGS
This generally works much better for a laptop than a PC, since it is much less likely that someone will have changed the configuration. The reality is that most of the time everything is the same, but the OS probably can't make that assumption. Windows 7 actually did a good job of minimizing the time, though. I am painfully reminded of this every time I boot my XP work machine. Larry M.

prabhakar_deosthali
User Rank
Author
re: Next computing target容xascale systems
prabhakar_deosthali   11/17/2010 6:33:13 AM
NO RATINGS
I think the solution for zero-time booting should be very simple. Most of these devices do a lot of initialization , recognition of peripherals and loading their drivers into the working ram and so on. A simple technique would be have a battery backup for this working ram so that when you shut down your laptop/PC or whatever it is , it is just in a snooze mode. You press a button and up starts the computer in an instant. The penalty will be the battery power that will be consumed. Another way could be to cut a CD of the working copy of the OS ( after initial installation and configuration). If this CD is present in your CD drive, it just loads it into the working ram and starts. Any time you have changed the peripherals or done something new, you do a normal boot up from the hard disk.

LarryM99
User Rank
Author
re: Next computing target容xascale systems
LarryM99   11/16/2010 10:51:30 PM
NO RATINGS
According to Kurzweil we are close to the Singularity where we all become last year's model anyway. The fact that we may be getting hardware that is 100x faster than the most sophisticated biological computer but is still so far short of the capability of that computer means that it's a software problem, not a hardware one. More computational power doesn't mean that we know how to either ask it the right questions or program it to get the answers. Larry M.

dirk.bruere
User Rank
Author
re: Next computing target容xascale systems
dirk.bruere   11/16/2010 9:57:11 PM
NO RATINGS
An exascale computer would be approximately 100x the estimated processing power of the Human brain. If the AI problem is cracked after that time the machine will outclass us intellectually to the extent we outclass a mouse.

Robotics Developer
User Rank
Author
re: Next computing target容xascale systems
Robotics Developer   11/16/2010 3:36:02 PM
NO RATINGS
I would think that mankind would benefit in a very significant way if all our machines: laptops, desktops, PDAs, etc. booted in zero or near zero time. Just think how much time is wasted every day waiting for your machine to be ready? How many leave the machine on 24/7 to avoid the delay in power up? How much power could be saved? I do understand the "need for speed"; raw compute power is very helpful for a number of applications modeling complex systems to name one. It was my experience in the ASIC industry that taught me: local / on-chip processing is better. Every time data moved off/on chip, every time external memory or disk needed to be accessed the speed/power costs were significant. Perhaps, we need to change how we perform processing sooner rather than later? Just a thought.

RickMatz
User Rank
Author
re: Next computing target容xascale systems
RickMatz   11/16/2010 2:38:56 PM
NO RATINGS
If more is better, than too much is just enough.

prabhakar_deosthali
User Rank
Author
re: Next computing target容xascale systems
prabhakar_deosthali   11/16/2010 6:05:34 AM
NO RATINGS
I totally agree with selinz that in computing power more is always better. making super computers play chess like games against the world chess champions is just the fun part of it. But seriously thinking, we have hardly scratched the surface when it comes to understanding this gigantic universe. It still takes us a year to reach Mars by a space craft and the universe is millions of light years in size. Todays supercomputers have to increase their performance many million times so that those brilliant minds working on their theories can simulate, model the universe of such a gigantic in size and finally get some clues about what is this universe in which we , the solar system is nothing but a tiny spec.

selinz
User Rank
Author
re: Next computing target容xascale systems
selinz   11/16/2010 1:32:27 AM
NO RATINGS
I understand. However, I recently began reading the Enders game series..Just finished Children of the Mind (see Ender's game series in Wiki if not familiar). I'd never read any of the series and after reading the main 6, it's not hard to imagine the usefulness of more computing horsepower. (audiobooks from the library have become my mainstay during my commute) As we move into 3D input and output, the demands will be taxed. Heck, it still takes my $3K laptop way too long to boot up...

LarryM99
User Rank
Author
re: Next computing target容xascale systems
LarryM99   11/16/2010 12:10:35 AM
NO RATINGS
If we assume that they solve the hardware problems which will enable it to be built and the software problems which will enable it to be programmed, then what exactly are they going to program it to do? Skynet? The meaning of Life, the Universe, and Everything? Or maybe they will race it against a quantum-based PC. Seriously, at some point we need to figure out if this computing paradigm has reached the point of diminishing returns. Larry M.



Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Radio
NEXT UPCOMING BROADCAST

What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Are the design challenges the same as with embedded systems, but with a little developer- and IT-skills added in? What do engineers need to know? Rick Merritt talks with two experts about the tools and best options for designing IoT devices in 2016. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.
Like Us on Facebook
Special Video Section
Linear Technology痴 LT8330 and LT8331, two Low Quiescent ...
The quality and reliability of Mill-Max's two-piece ...
01:34
The quality and reliability of Mill-Max's two-piece ...
LED lighting is an important feature in today痴 and future ...
05:27
The LT8602 has two high voltage buck regulators with an ...
05:18
Silego Technology痴 highly versatile Mixed-signal GreenPAK ...
The quality and reliability of Mill-Max's two-piece ...
01:34
Why the multicopter? It has every thing in it. 58 of ...
Security is important in all parts of the IoT chain, ...
Infineon explains their philosophy and why the multicopter ...
The LTC4282 Hot SwapTM controller allows a board to be ...
This video highlights the Zynqョ UltraScale+ MPSoC, and sho...
Homeowners may soon be able to store the energy generated ...
The LTCョ6363 is a low power, low noise, fully differential ...
See the Virtexョ UltraScale+ FPGA with 32.75G backplane ...
Vincent Ching, applications engineer at Avago Technologies, ...
The LTョ6375 is a unity-gain difference amplifier which ...
The LTCョ4015 is a complete synchronous buck controller/ ...
10:35
The LTCョ2983 measures a wide variety of temperature sensors ...
The LTCョ3886 is a dual PolyPhase DC/DC synchronous ...