Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.physics > #712865 > unrolled thread

What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

Back to article view | Back to sci.physics

Contents

  What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Keith Stein <keithstein111@gmail.com> - 2018-11-15 17:17 +0000
    Re:What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Libor Striz <poutnik4REMOVEnntp@CAPITALSgmail.com> - 2018-11-15 19:54 +0100
      Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2018-11-15 20:32 +0100
        Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Shon Wible <shwn@shouve.wd> - 2018-11-16 20:22 +0000
        Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Sergio <invalid@invalid.com> - 2018-11-19 17:26 -0600
          Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Lochù Umská <lare@cpi2alwa.cc> - 2018-11-19 23:35 +0000
            Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Sergio <invalid@invalid.com> - 2018-11-19 22:05 -0600
              Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Elden Lathan <an@tnettw.ea> - 2018-11-22 19:25 +0000
          Gravitational waves (was: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<) Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2018-11-20 03:11 +0100
          Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Tom Roberts <tjroberts137@sbcglobal.net> - 2018-11-19 22:23 -0600
          Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< benj <benj.fake@gmail.com> - 2018-11-20 11:46 -0500
            Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2018-11-20 21:26 +0100
            Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Tom Roberts <tjroberts137@sbcglobal.net> - 2018-11-20 15:51 -0600
              Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Paul Edgar <lara@uldewan.lg> - 2018-11-21 18:12 +0000
                Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Sergio <invalid@invalid.com> - 2018-11-21 13:43 -0600
      Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Keith Stein <keithstein111@gmail.com> - 2018-11-15 20:54 +0000
        Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Sergio <invalid@invalid.com> - 2018-11-19 21:57 -0600
          Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2018-11-20 09:16 +0100
            Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< moroney@world.std.spaamtrap.com (Michael Moroney) - 2018-11-20 17:48 +0000
          Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Keith Stein <keithstein111@gmail.com> - 2018-11-20 12:47 +0000
          Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Sergio <invalid@invalid.com> - 2018-11-20 10:01 -0600
        Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Edward Prochak <edprochak@gmail.com> - 2018-11-20 08:22 -0800
          Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< benj <benj.fake@gmail.com> - 2018-11-20 12:10 -0500
            Re: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Sergio <invalid@invalid.com> - 2018-11-21 13:50 -0600

Page 1 of 2  [1] 2  Next page →

#712865 — What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

How to Program a Computer eh!

[toc] | [next] | [standalone]

#712876

Keith Stein <keithstein111@gmail.com> Wrote in message:
> How to Program a Computer eh!

They have programmers for that.
-- 
Libor Striz aka Poutnik ( a pilgrim/wanderer/wayfarer)

"Humour is the only effective weapon against stupidity."
Miloš Forman


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

[toc] | [prev] | [next] | [standalone]

#712884

Libor Striz wrote on 15.11.18 19:54:
> Keith Stein <keithstein111@gmail.com> Wrote in message:
>> How to Program a Computer eh!
> 
> They have programmers for that.

Speaking as a professional software developer, it is seldom efficient to
employ a professional software developer (the term “programmer” as a job
description went out of fashion some 30 years ago, and rightly so) for this
because you have to understand the physics, too.

This is why learning to use computers for solving physics problems is a part
of the undergraduate physics curriculum.  So, for a change, Keith Stein is
(probably accidentally) correct this time.

For example, if I understood Prof. Barish correctly, as gravitational wave
equations cannot be solved algebraically, it will be very important in the
future to find efficient ways for solving them numerically with computers.

-- 
PointedEars

Twitter: @PointedEars2
Please do not cc me. / Bitte keine Kopien per E-Mail.

[toc] | [prev] | [next] | [standalone]

#712980

Thomas 'PointedEars' Lahn wrote:

> Libor Striz wrote on 15.11.18 19:54:
>> Keith Stein <keithstein111@gmail.com> Wrote in message:
>>> How to Program a Computer eh!
>> 
>> They have programmers for that.
> 
> Speaking as a professional software developer, it is seldom efficient to
> employ a professional software developer (the term “programmer” as a job
> description went out of fashion some 30 years ago, and rightly so) for
> this because you have to understand the physics, too.

That's called numerical analysis,scientific computation, etc, you moron. 
And it has very little to do with "programmers". One more proof that you 
stupid around here are  completely deranged in science and physics. Idiots.

[toc] | [prev] | [next] | [standalone]

#713426

On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:
> Libor Striz wrote on 15.11.18 19:54:
>> Keith Stein <keithstein111@gmail.com> Wrote in message:
>>> How to Program a Computer eh!
>>
>> They have programmers for that.
> 
> Speaking as a professional software developer, it is seldom efficient to
> employ a professional software developer (the term “programmer” as a job
> description went out of fashion some 30 years ago, and rightly so) for this
> because you have to understand the physics, too.

true

> 
> This is why learning to use computers for solving physics problems is a part
> of the undergraduate physics curriculum.  So, for a change, Keith Stein is
> (probably accidentally) correct this time.

in the old days, you had to do the calculations/equations by hand, now
they are in libraries like Java libraries on your computer, and they are
well tested.

> 
> For example, if I understood Prof. Barish correctly, as gravitational wave
> equations cannot be solved algebraically, it will be very important in the
> future to find efficient ways for solving them numerically with computers.
> 

they use supercomputers for that.  You can get time on one if you try.
https://www.tacc.utexas.edu/systems/stampede2

[toc] | [prev] | [next] | [standalone]

#713429

Sergio wrote:

> On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:

>> For example, if I understood Prof. Barish correctly, as gravitational
>> wave equations cannot be solved algebraically, it will be very
>> important in the future to find efficient ways for solving them
>> numerically with computers.
>
> they use supercomputers for that.  You can get time on one if you try.
> https://www.tacc.utexas.edu/systems/stampede2

That cretin just gave the proof on not knowing what solving, algebra, 
numerically and computers stands for. Tootaly greek to him.

[toc] | [prev] | [next] | [standalone]

#713468

On 11/19/2018 5:35 PM, Lochù Umská wrote:
> Sergio wrote:
> 
>> On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:
> 
>>> For example, if I understood Prof. Barish correctly, as gravitational
>>> wave equations cannot be solved algebraically, it will be very
>>> important in the future to find efficient ways for solving them
>>> numerically with computers.
>>
>> they use supercomputers for that.  You can get time on one if you try.
>> https://www.tacc.utexas.edu/systems/stampede2
> 
> That cretin just gave the proof on not knowing what solving, algebra, 
> numerically and computers stands for. Tootaly greek to him.
> 

na, try the site on the link, see what the requirements are for your
project on their super computer,

you do supply the equations, algebra, and boundary conditions,

with things like collision of two black holes in time, you have to model
it.

far to many data points, and conditions to recalcualte by hand, over and
over,  a simulation may tens of millions of calculations per second over
a two hour period, cant do that by hand.

[toc] | [prev] | [next] | [standalone]

#714025

Sergio wrote:

> On 11/19/2018 5:35 PM, Lochù Umská wrote:
>> Sergio wrote:
>> 
>>> On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:
>> 
>>>> For example, if I understood Prof. Barish correctly, as gravitational
>>>> wave equations cannot be solved algebraically, it will be very
>>>> important in the future to find efficient ways for solving them
>>>> numerically with computers.
>>>
>>> they use supercomputers for that.  You can get time on one if you try.
>>> https://www.tacc.utexas.edu/systems/stampede2
>> 
>> That cretin just gave the proof on not knowing what solving, algebra,
>> numerically and computers stands for. Tootaly greek to him.
>> 
>> 
> na, try the site on the link, see what the requirements are for your
> project on their super computer,
> you do supply the equations, algebra, and boundary conditions,
> with things like collision of two black holes in time, you have to model
> it.

Ahh ney. Nothing, not enough. You're not even touch the newbeginner level. 
There's a whole branch of science in making stuff distributable. So much 
to learn, you guys, completely uneducated, in "computer science".

Not a wonder people are hating homepage designers. They are thinking they 
are scientists.

[toc] | [prev] | [next] | [standalone]

#713453 — Gravitational waves (was: What Every Physicist Needs to Know..............<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<)

Sergio wrote:

> On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:
>> This is why learning to use computers for solving physics problems is a part
>> of the undergraduate physics curriculum.  So, for a change, Keith Stein is
>> (probably accidentally) correct this time.
> 
> in the old days, you had to do the calculations/equations by hand, now
> they are in libraries like Java libraries on your computer, and they are
> well tested.

Virtually *nobody* is using Java for this.  Languages of choice for
numerical scientific computation are Python (with NumPy), C/C++, and
Fortran.  Languages of choice for algebraic computation are Wolfram Language
(in Wolfram Mathematica and Wolfram Alpha), MathWorks MATLAB/FreeMat, and
Waterloo Maple.

>> For example, if I understood Prof. Barish correctly, as gravitational wave
>> equations cannot be solved algebraically, it will be very important in the
>> future to find efficient ways for solving them numerically with computers.
> 
> they use supercomputers for that.  You can get time on one if you try.
> https://www.tacc.utexas.edu/systems/stampede2

Do you even know what a differential equation is?

You need to read more carefully.  I did NOT say that the gravitational wave
equations cannot be solved, but that they cannot be solved *algebraically*.
It is precisely supercomputers with which they can be solved *numerically*,
but according to Dr. Barish, *efficiency* of the algorithms is lacking, as
you indicate yourself, computing time is limited/expensive.  That is, in the
best case, one should not need a supercomputer to solve them.

You are still in violation of RFC 5536 and the AUP of your provider.

F’up2 sci.phyiscs.relativity

-- 
PointedEars

Twitter: @PointedEars2
Please do not cc me. / Bitte keine Kopien per E-Mail.

[toc] | [prev] | [next] | [standalone]

#713469

On 11/19/18 5:26 PM, Sergio wrote:
> On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:
>> For example, if I understood Prof. Barish correctly, as gravitational wave
>> equations cannot be solved algebraically, it will be very important in the
>> future to find efficient ways for solving them numerically with computers.

Yes.

> they use supercomputers for that. 

Competition for supercomputer time is fierce, and LIGO could probably 
use just about all available time on all available supercomputers. But 
solving such equations is tailor-made for GPUs and their SIMD 
architecture. Indeed, some supercomputers have nodes with integrated 
GPUs for just such types of problems (e.g. Titan at ORNL). But clusters 
with GPU-rich nodes are a significantly cheaper approach, and one that 
most university groups can afford.

Note: it is challenging to use GPUs effectively and efficiently, and I 
suspect that is what Prof. Barish is thinking of.

	I am not a member of the LIGO collaboration, and don't
	know how they are actually doing computations....

Tom Roberts

[toc] | [prev] | [next] | [standalone]

#713506

On 11/19/2018 6:26 PM, Sergio wrote:
> On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:
>> Libor Striz wrote on 15.11.18 19:54:
>>> Keith Stein <keithstein111@gmail.com> Wrote in message:
>>>> How to Program a Computer eh!
>>>
>>> They have programmers for that.
>>
>> Speaking as a professional software developer, it is seldom efficient to
>> employ a professional software developer (the term “programmer” as a job
>> description went out of fashion some 30 years ago, and rightly so) for this
>> because you have to understand the physics, too.
> 
> true
> 
>>
>> This is why learning to use computers for solving physics problems is a part
>> of the undergraduate physics curriculum.  So, for a change, Keith Stein is
>> (probably accidentally) correct this time.
> 
> in the old days, you had to do the calculations/equations by hand, now
> they are in libraries like Java libraries on your computer, and they are
> well tested.
> 
>>
>> For example, if I understood Prof. Barish correctly, as gravitational wave
>> equations cannot be solved algebraically, it will be very important in the
>> future to find efficient ways for solving them numerically with computers.
>>
> 
> they use supercomputers for that.  You can get time on one if you try.
> https://www.tacc.utexas.edu/systems/stampede2
> 
Who needs time on a "supercomputer" you have on right in your laptop. 
It's only a couple years behind current supercomputer levels . Just wait 
a short time and there you are.

[toc] | [prev] | [next] | [standalone]

#713558

benj wrote:
> On 11/19/2018 6:26 PM, Sergio wrote:
>> On 11/15/2018 1:32 PM, Thomas 'PointedEars' Lahn wrote:
>>> For example, if I understood Prof. Barish correctly, as gravitational wave
>>> equations cannot be solved algebraically, it will be very important in the
>>> future to find efficient ways for solving them numerically with computers.
>>
>> they use supercomputers for that.  You can get time on one if you try.
>> https://www.tacc.utexas.edu/systems/stampede2
>
> Who needs time on a "supercomputer" you have on right in your laptop. 
> It's only a couple years behind current supercomputer levels . Just wait 
> a short time and there you arc.

It is hilarious how, out of utter ignorance, you think that you have a clue.

F’up2 sci.physics.relativity

-- 
PointedEars

Twitter: @PointedEars2
Please do not cc me. / Bitte keine Kopien per E-Mail.

[toc] | [prev] | [next] | [standalone]

#713590

On 11/20/18 10:46 AM, benj wrote:
> Who needs time on a "supercomputer" you have on right in your laptop. 
> It's only a couple years behind current supercomputer levels . Just wait 
> a short time and there you are.

How silly and ignorant.

The supercomputer cori.nersc.gov has more than 700,000 cores and over a 
million gigabytes of RAM, with access to > 200 petabytes of disk. It 
requires > 10 megawatts of power. Each of those numbers is > 100,000 
times a typical laptop. No mount of "waiting" will get a laptop to 
approach any of those values.

     Yes, I have run calculations on cori, and other supercomputers.

Tom Roberts

[toc] | [prev] | [next] | [standalone]

#713767

Tom Roberts wrote:

> On 11/20/18 10:46 AM, benj wrote:
>> Who needs time on a "supercomputer" you have on right in your laptop.
>> It's only a couple years behind current supercomputer levels . Just
>> wait a short time and there you are.
> 
> How silly and ignorant.
> 
> The supercomputer cori.nersc.gov has more than 700,000 cores and over a
> million gigabytes of RAM, with access to > 200 petabytes of disk. It
> requires > 10 megawatts of power. Each of those numbers is > 100,000
> times a typical laptop. No mount of "waiting" will get a laptop to
> approach any of those values.
>      Yes, I have run calculations on cori, and other supercomputers. 
> Tom Roberts

Total delusion. Surprised to watch that. Without proper coding granulation 
to the multicore your "super"computer is useless. It will still run over 
single core or a few. Moreover, excessive stupid implemented granulation 
makes the tasks running even slower than along a single core.

[toc] | [prev] | [next] | [standalone]

#713790

On 11/21/2018 12:12 PM, Paul Edgar wrote:
> Tom Roberts wrote:
> 
>> On 11/20/18 10:46 AM, benj wrote:
>>> Who needs time on a "supercomputer" you have on right in your laptop.
>>> It's only a couple years behind current supercomputer levels . Just
>>> wait a short time and there you are.
>>
>> How silly and ignorant.
>>
>> The supercomputer cori.nersc.gov has more than 700,000 cores and over a
>> million gigabytes of RAM, with access to > 200 petabytes of disk. It
>> requires > 10 megawatts of power. Each of those numbers is > 100,000
>> times a typical laptop. No mount of "waiting" will get a laptop to
>> approach any of those values.
>>      Yes, I have run calculations on cori, and other supercomputers. 
>> Tom Roberts
> 
> Total delusion. Surprised to watch that. Without proper coding granulation 
> to the multicore your "super"computer is useless. It will still run over 
> single core or a few. Moreover, excessive stupid implemented granulation 
> makes the tasks running even slower than along a single core.
> 

 yep, good for modeling thousands of  interconnected points

massively parallel machines


http://www.ncsa.illinois.edu/news/story/alya_code_scaled_to_100000_cores_on_blue_waters_supercomputer

https://www.gc.cuny.edu/Page-Elements/Academics-Research-Centers-Initiatives/Doctoral-Programs/Computer-Science/Course-Listings/Spring-2016/Massively-Parallel-Programming


an INTRO here, not bad;
https://computing.llnl.gov/tutorials/parallel_comp/

[toc] | [prev] | [next] | [standalone]

#712892

On 15/11/2018 18:54, Libor Striz wrote:
> Keith Stein <keithstein111@gmail.com> Wrote in message:
>> How to Program a Computer eh!
> 
> They have programmers for that.

But what you are completely missing Mr.Striz
is just how very much easier physics is, when
you know how to program a computer. You need
to learn to think like a computer, Mr.Striz.
> 
HOW WOULD A COMPUTER SOLVE THE PROBLEM eh!

[toc] | [prev] | [next] | [standalone]

#713467

On 11/15/2018 2:54 PM, Keith Stein wrote:
> On 15/11/2018 18:54, Libor Striz wrote:
>> Keith Stein <keithstein111@gmail.com> Wrote in message:
>>> How to Program a Computer eh!
>>
>> They have programmers for that.
> 
> But what you are completely missing Mr.Striz
> is just how very much easier physics is, when
> you know how to program a computer. You need
> to learn to think like a computer, Mr.Striz.

na,
computers think in binary, 1's and 0's and they have an accumulator,
that is about it, everything else is support.

   64 bit ? means 64 bit wide accumulator


>>
> HOW WOULD A COMPUTER SOLVE THE PROBLEM eh!

[toc] | [prev] | [next] | [standalone]

#713480

Sergio wrote on 20.11.2018 04:57:
>    64 bit ? means 64 bit wide accumulator
No, primarily it means a 64 bit wide *address* *bus*.

Shut up as you do not know what you are talking about.

And stop amok-crossposting.

F’up2 poster

-- 
PointedEars

Twitter: @PointedEars2
Please do not cc me. / Bitte keine Kopien per E-Mail.

[toc] | [prev] | [next] | [standalone]

#713535

On 20 Nov 2018 09:16, Thomas 'PointedEars' Lahn wrote:
>Sergio wrote on 20.11.2018 04:57:
>>    64 bit ? means 64 bit wide accumulator
>No, primarily it means a 64 bit wide *address* *bus*.

Only sort of.

In the old days when the natural data unit size and the address bus size
were often different, "xx bits" meant the natural data unit (accumulator,
if relevant) size.  The address bus width was often different.
DEC in its earlier days had 12, 18, 16, 36 and 32 bit systems before its
64 bit Alpha. The 16 bit PDP-11 had 16, 18 or 22 bits of address space,
depending on the model.

These days, both the natural unit and (theoretical) address bus size
are the same, and the same instructions could manipulate both.
Physical and virtual memory blur things more.

64 bit processors treat address pointers as 64 bit values BUT:

These days, *no* "64 bit" processor can address a full 64 bits of memory,
physical or virtual. Depending on processor, some address lines will 
either be nonfunctional or cause a nonexistent memory fault if set.
A full 64 bits of memory would be over 18 exabytes of memory.
The latest x86-64 can theoretically address 56 bits worth of virtual 
memory, at least internally.

>>Shut up as you do not know what you are talking about.

>And stop amok-crossposting.

>Fup2 poster

[toc] | [prev] | [next] | [standalone]

#713487

 >> HOW WOULD A COMPUTER SOLVE THE PROBLEM eh!
 >

Re: What Every Physicist Needs to Know...........Re: What Every 
Physicist Needs to Know..........Re: What Every Physicist Needs to 
Know............Re: What Every Physicist Needs to Know...........

>>> Keith Stein Wrote:
>>>> How to Program a Computer eh!

>> Libor Striz wrote: >>> They have programmers for that.

 > Keith Stein wrote:
>> But what you are completely missing Mr.Striz
>> is just how very much easier physics is, when
>> you know how to program a computer. You need
>> to learn to think like a computer, Mr.Striz.

   Sergio wrote:> na,
> computers think in binary,

Good point Sergio, we wouldn't want Mr.Striz
to think in binary eh!

> 1's and 0's and they have an accumulator,
> that is about it, everything else is support.

Well it is very clear you know all there is to know,
about computers Sergio, but do you know how to program one?


>     64 bit ? means 64 bit wide accumulator

Too much information Sergio. Don't tell it all.
Keep a bit of mystery eh!

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | sci.physics

csiph-web