Real_McCoy <McCoy@somewhere.co.tw> wrote:> Do you still sue your slide-rule? >No it went completely bankrupt the last time I sued it and now just mopes in the back of the drawer. fred.
Software vs. Hardware. The Harsh Truth
Started by ●October 22, 2005
Reply by ●October 23, 20052005-10-23
Reply by ●October 23, 20052005-10-23
Bevan Weiss wrote:> [...] Algorithms exist whether in hardware or > software, it's just so much easier to make changes in software that > people often don't put as much thought into them. No one would just > throw in an inverter or a multiplier to try and solve some hardware > problem they were having, instead they'd spend many hours working out > the best way, simply because implementing any change is much harder. > Software is just typing three or four characters.Those parts of the system that you can hit with a hammer (not advised) are called Hardware; those program instructions that you can only curse at are called Software. - Unknown :-) Carlos --
Reply by ●October 23, 20052005-10-23
Jerry Avins <jya@ieee.org> writes:> Real_McCoy wrote: >> "Radium" <glucegen1@excite.com> wrote in message >> news:1130017709.670750.190780@o13g2000cwo.googlegroups.com... >> >>>Find out why >>> >>>1. http://www.rebelscience.org/Cosas/Reliability.htm#WhyBad >>> >>>2. http://www.rebelscience.org/Cosas/Reliability.htm#Good >>> >> This is no longer a black and white topic. In the near future all >> hardware >> will be derived from software. eg algorithms will be simulated in software >> and go straight to hardware via FPGA compilers and such like. They already >> exist in a limited capacity for LabVIEW. Yuu draw your virtual instrument >> block diagram, press a button and its hardware! >> The days of fiddling with transistors are over. > > Over for whom? Who will design the FPGAs?Excellent point, Jerry. Kinda like stating we no longer need to learn arithmetic since we've got calculators. -- % Randy Yates % "So now it's getting late, %% Fuquay-Varina, NC % and those who hesitate %%% 919-577-9882 % got no one..." %%%% <yates@ieee.org> % 'Waterfall', *Face The Music*, ELO http://home.earthlink.net/~yatescr
Reply by ●October 23, 20052005-10-23
Real_McCoy wrote:> "Jerry Avins" <jya@ieee.org> wrote in message > news:GpmdnZd0qJSNCMbeRVn-iQ@rcn.net... > >>Real_McCoy wrote:>>>The days of fiddling with transistors are over. >> >>Over for whom? Who will design the FPGAs? Progress centralizes and >>categorizes. The days when every family rendered tallow to made its own >>candles are long gone, but prudent households still keep candles handy. >> >> > > Yes I meant for the majority of applicatiosn guys - of course we still need > engineers to design the lower level stuff in the first place. I look at it > this way though - look at the history of tools (ordinary tools). Most > joiners use automatic screwdrivers nowadays and power saws and drills. 30 > years ago or more it was all done by hand for the most part. > Once you design the basic tools, they in turn are used to design more > sophisticated ones and the systems bootsraps itself. Its still good to have > an ordinary screwdriver at hand though! Do you still sue your slide-rule?There was a time when a machinist had to use hand tools well enough to build a passable lathe (demonstrated by proficiency with a few sub-tasks) from castings in order to graduate from apprenticeship. It may once have made sense, but that time was long past when the requirement was dropped. Knowing how things are done by hand may help the tool user, just as familiarity with assembly language may help a HLL programmer. I don't dispute your claim, but yes: one of my slide rules is on my desk as I write. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●October 24, 20052005-10-24
"Randy Yates" <yates@ieee.org> wrote in message news:irvnn08c.fsf@ieee.org...> Jerry Avins <jya@ieee.org> writes: > > > Real_McCoy wrote: > >> "Radium" <glucegen1@excite.com> wrote in message > >> news:1130017709.670750.190780@o13g2000cwo.googlegroups.com... > >> > >>>Find out why > >>> > >>>1. http://www.rebelscience.org/Cosas/Reliability.htm#WhyBad > >>> > >>>2. http://www.rebelscience.org/Cosas/Reliability.htm#Good > >>> > >> This is no longer a black and white topic. In the near future all > >> hardware > >> will be derived from software. eg algorithms will be simulated insoftware> >> and go straight to hardware via FPGA compilers and such like. Theyalready> >> exist in a limited capacity for LabVIEW. Yuu draw your virtualinstrument> >> block diagram, press a button and its hardware! > >> The days of fiddling with transistors are over. > > > > Over for whom? Who will design the FPGAs? > > Excellent point, Jerry. Kinda like stating we no longer need to learn > arithmetic since we've got calculators. > -- > % Randy Yates % "So now it's getting late, > %% Fuquay-Varina, NC % and those who hesitate > %%% 919-577-9882 % got no one..." > %%%% <yates@ieee.org> % 'Waterfall', *Face The Music*, ELO > http://home.earthlink.net/~yatescrThat's all true but I thought the whole point in rapid prototyping was just that - that it's supposed to be rapid! Some guy sitting down programming in assembler is going to take 10 times longer though the end result maybe far more efficient.I have often pondered the evolution of ordinary tools - from stone-age man onwards and how if you had the knowledge you would re-build our technological society from nothing! Suppose all technology was lost for some reason and you had to do everything from scratch - what would be the sequence to get to modern machine tools (assuming the knowledge was still there).You would start from stone and work your way up! McC McC
Reply by ●October 24, 20052005-10-24
"kyle" <kyleblay@blerk.org> wrote in message news:JpWdnSmFIYVspsbeRVn-hw@giganews.com...> >Find out why >> >>1. http://www.rebelscience.org/Cosas/Reliability.htm#WhyBad >> >>2. http://www.rebelscience.org/Cosas/Reliability.htm#Good >> >> > > HW: Pentium floating point bug. > Resolution: Buy a new chip > > SW: Security hole in IE > Resolution: Patch > > Software does suffer from the problems mentioned, of course that hasn't > stopped the entire world being completely revolutionized by software!If the world was completely revolutionized by software what does said software run on ? And what level are you taking software to ? Firmware ? HDL (hardware description languages) ? rtl ? netlist ?> None of those failures are surprising, and all of them are the subject of > on-going study by people determined to minimize or reduce their effects. > > Compare today's development env/languages/run-time systems to those of the > 60's or even 80's... Things are improving that are allowing us to scale up > our software systems and make them more reliable.But are they on the whole more reliable ? Alex
Reply by ●October 24, 20052005-10-24
>Suppose all technology was lost for >some reason and you had to do everything from scratch - what would bethe>sequence to get to modern machine tools (assuming the knowledge wasstill>there).You would start from stone and work your way up!Not to be too pedantic but technology IS the knowledge :) I think you mean if we lost all our tools and other physical manifestations of our technologies ;-) This message was sent using the Comp.DSP web interface on www.DSPRelated.com
Reply by ●October 24, 20052005-10-24
Real_McCoy wrote: (snip regarding hardware, software, and FPGAs)> That's all true but I thought the whole point in rapid prototyping was just > that - that it's supposed to be rapid! Some guy sitting down programming in > assembler is going to take 10 times longer though the end result maybe far > more efficient.I have often pondered the evolution of ordinary tools - from > stone-age man onwards and how if you had the knowledge you would re-build > our technological society from nothing! Suppose all technology was lost for > some reason and you had to do everything from scratch - what would be the > sequence to get to modern machine tools (assuming the knowledge was still > there).You would start from stone and work your way up!As I understand it, there is now a strong drive toward using FPGAs over what would have been ASICs in the past. The mask costs are now so high, often over a million dollars, that it doesn't make sense. FPGAs, like DRAM, are a very regular array which makes them relatively easy to design. The required optimization makes it somewhat harder, though. (Especially if they need to be big and fast.) -- glen
Reply by ●October 24, 20052005-10-24
Real_McCoy wrote: ...> ... I have often pondered the evolution of ordinary tools - from > stone-age man onwards and how if you had the knowledge you would re-build > our technological society from nothing! Suppose all technology was lost for > some reason and you had to do everything from scratch - what would be the > sequence to get to modern machine tools (assuming the knowledge was still > there).You would start from stone and work your way up!Technology _is_ knowledge. The day I read about the invention of the point contact transistor (front page N.Y.Times with description inside) I went home and made one using germanium from a WW II surplus diode. Suppose you were in say, England in 1700 and given all the resources you needed to build workable ship-to-ship and ship-to-shore radio. How long would it take you? I suspect that among the hardest part would be finding literate sailors and training them to be Morse code operators. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●October 24, 20052005-10-24
Technology _is_ knowledge. - only half true. Saying is much easier than the doing! (as a Taiwanese engineer once told me).There are lots of academics out there than know teh theory backwards but could not build anything.Knowing the theory of how a rocket balances is relatively easy - getting it to work is quite another matter! Naebad
