From: Jan Panteltje on 14 Dec 2009 12:48 On a sunny day (Mon, 14 Dec 2009 09:27:13 -0800) it happened Joerg <invalid(a)invalid.invalid> wrote in <7onarkF3osi2cU1(a)mid.individual.net>: >My Hameg is fully "transistorizated" :-) Was Hameg a sync scope? I have seen them, but did not like them. I designed and build my second scope in the end of the sixties, it had RTL logic, a delayed timebase, full trigger, but of course the bandwidth was limited. All solid state of course, Think I gave it away, just before I started travelling the world... I do remember it used a 710 comparator for the TV trigger... TV video output transistors for the deflection, nice 10 turn pot for the timebase delay. You needed dual timebase to get a good look at the vertical TV interval, especially when doing things with video recorders. The whole thing was on one 10ox160 mm Eurocard... Also build a transitor TV one such eurocard in those days...
From: Joel Koltner on 14 Dec 2009 21:37 "Joerg" <invalid(a)invalid.invalid> wrote in message news:7oo8gfF3qa1tjU1(a)mid.individual.net... > It's like with laptops, when do you really, really need the horsepower of > that dual-core? You're seeing CPUs go to multiple cores because the ability to just keep cranking the clock rate has largely petered out. Hence you might as well ask, "When do you really need the horsepower of a 1GHz CPU?" :-) I'd say that any simulation you run today that takes more than a few seconds can benefit significantly from multiple cores -- if they can be made inexpensively enough, of course. I like John's idea that eventually we might have, e.g., a 1000 core CPU that would be fast enough to let you tweak component parameters and end up with a complete SPICE simulation including monte carlo analysis done in effectively real-time. ---Joel
From: Joerg on 14 Dec 2009 21:48 Joel Koltner wrote: > "Joerg" <invalid(a)invalid.invalid> wrote in message > news:7oo8gfF3qa1tjU1(a)mid.individual.net... >> It's like with laptops, when do you really, really need the horsepower >> of that dual-core? > > You're seeing CPUs go to multiple cores because the ability to just keep > cranking the clock rate has largely petered out. Hence you might as > well ask, "When do you really need the horsepower of a 1GHz CPU?" :-) > That's actually a good question. It has taken over a decade until laptop designers have re-learned what the older guys who designed the Compaq Contura series knew all along and did in the early 90's: Clock speed reduction when horsepower is not needed. That way they squeezed 6h of runtime out of a simple NiCd pack on my Contura 410. Only with the recent advent of netbooks like the Samsung NC-10 has that ever been topped with a regular stock battery. So I promptly bought that netbook. > I'd say that any simulation you run today that takes more than a few > seconds can benefit significantly from multiple cores -- if they can be > made inexpensively enough, of course. I like John's idea that > eventually we might have, e.g., a 1000 core CPU that would be fast > enough to let you tweak component parameters and end up with a complete > SPICE simulation including monte carlo analysis done in effectively > real-time. > But that will reduce the quality of coming generations of engineers because they assume they can let the machine do the thinking, just like with automatic transmissions in cars :-) -- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
From: Joel Koltner on 14 Dec 2009 22:07 "Joerg" <invalid(a)invalid.invalid> wrote in message news:7oobnuF3qe5ngU1(a)mid.individual.net... > But that will reduce the quality of coming generations of engineers because > they assume they can let the machine do the thinking, just like with > automatic transmissions in cars :-) I think that most of the damage in electronics has already been done, Joerg. We're at a point where the Internet provides incredible opportunities to study and learn even very esoteric bits of engineering that would have been nigh impossible (or at least incredibly time-consuming) even 20 years ago, yet only a tiny fraction of those studying to become EEs take advantage of it as a means to improve the quality of their work beyond what is strictly required of them in a classroom or workplace setting. Hence the primary remaining concern is that those aspiring EEs in China or India or Botswana aren't yet quite as dumbed-down yet as the ones in the US today and hence there'll be fewer and fewer EE jobs here. That being said, at least when I was in grad school, the Indian and Asian students I knew really weren't that different from the white kids from the U.S. -- most of them were more interested in just getting the sheepskin than in becoming top designers (and apparently some of the female Indian students were planning to use their EE degrees as a means to being able to go back home, work for a couple of years, get married, and drop out of the industry entirely -- the degree was primarily a means to gain access to more desirable guys... whereas for the rather small number of U.S. woman in the program, that didn't seem to be the case...). Hey, don't cars with automatics generally get a mile or two per gallon better mileage than manuals? ;-) ---Joel
From: JosephKK on 15 Dec 2009 01:02
On Sun, 13 Dec 2009 10:45:45 -0600, "Tim Williams" <tmoranwms(a)charter.net> wrote: >"JosephKK" <quiettechblue(a)yahoo.com> wrote in message >news:vr5ai5tjh0b59t76e4362bufh7ufgp0n97(a)4ax.com... >> The last time i saw a synchronizer type 'scope was over 40 years ago, it >> was a neighbors prize possession and 10+ years old even then. > >http://webpages.charter.net/dawill/tmoranwms/Elec_Scope1_lg.jpg > >Well, it says "TRIG LEVEL" on the outer knob, but it has this peculiar >"STABILITY" on the inner knob. > >I never did completely understand the circuit in this thing (I have the kit >instructions, so I have all the drawings), it's drawn so as to make a >minimum of sense. > >Tim I usually do pretty well figuring out Heathkit stuff. Could you send me a pdf copy? |