Groups > comp.lang.javascript > #27772 > unrolled thread

Bignumber conversion and arithmetic on unsigned operand pairs using +,-,* now work in anybase.

Started by	jonas.thornvall@gmail.com
First post	2015-05-20 06:11 -0700
Last post	2015-06-29 04:25 -0700
Articles	20 on this page of 83 — 11 participants

Back to article view | Back to comp.lang.javascript

Page 2 of 5 — ← Prev page 1 [2] 3 4 5 Next page →

#28148

From	"Evertjan." <exxjxw.hannivoort@inter.nl.net>
Date	2015-06-24 09:45 +0200
Message-ID	<XnsA4C36348A26Beejj99@194.109.6.166>
In reply to	#28147

Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote on 24 Jun 2015 in 
comp.lang.javascript:

> Dr J R Stockton wrote:
> 
>> […] Thomas 'PointedEars' Lahn […] posted:
>>> According to the ODE and Merriam-Webster,
>> 
>> I think that you should have explained the difference between the ODE
>> and the OED.
> 
> I do not think so.

Order of das Deutsches Empire.

<https://en.wikipedia.org/wiki/Teutonic_Order>

-- 
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

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

#28149

From	jonas.thornvall@gmail.com
Date	2015-06-24 03:42 -0700
Message-ID	<11f64750-4842-453f-8344-2c906df809b7@googlegroups.com>
In reply to	#28148

Den onsdag 24 juni 2015 kl. 09:45:43 UTC+2 skrev Evertjan.:
> Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote on 24 Jun 2015 in 
> comp.lang.javascript:
> 
> > Dr J R Stockton wrote:
> > 
> >> [...] Thomas 'PointedEars' Lahn [...] posted:
> >>> According to the ODE and Merriam-Webster,
> >> 
> >> I think that you should have explained the difference between the ODE
> >> and the OED.
> > 
> > I do not think so.
> 
> Order of das Deutsches Empire.
> 
> <https://en.wikipedia.org/wiki/Teutonic_Order>
> 
> -- 
> Evertjan.
> The Netherlands.
> (Please change the x'es to dots in my emailaddress)

Talking about order i just noticed how much better the arithmetic look if you actually rollback the digits.

http://jt.node365.se/arithmeitc07.html

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

#28155

From	jonas.thornvall@gmail.com
Date	2015-06-25 01:13 -0700
Message-ID	<e88550fc-02c5-4877-a04b-e23c9cbcaf20@googlegroups.com>
In reply to	#28149

Den onsdag 24 juni 2015 kl. 12:42:46 UTC+2 skrev jonas.t...@gmail.com:
> Den onsdag 24 juni 2015 kl. 09:45:43 UTC+2 skrev Evertjan.:
> > Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote on 24 Jun 2015 in 
> > comp.lang.javascript:
> > 
> > > Dr J R Stockton wrote:
> > > 
> > >> [...] Thomas 'PointedEars' Lahn [...] posted:
> > >>> According to the ODE and Merriam-Webster,
> > >> 
> > >> I think that you should have explained the difference between the ODE
> > >> and the OED.
> > > 
> > > I do not think so.
> > 
> > Order of das Deutsches Empire.
> > 
> > <https://en.wikipedia.org/wiki/Teutonic_Order>
> > 
> > -- 
> > Evertjan.
> > The Netherlands.
> > (Please change the x'es to dots in my emailaddress)
> 
> Talking about order i just noticed how much better the arithmetic look if you actually rollback the digits.
> 
> http://jt.node365.se/arithmeitc07.html

Changed the baseconversion search into using threevalued logic for faster match.
I am far from sure it is the optimal way to do it.

http://jt.node365.se/baseconversion5.html

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

#28152

From	Thomas 'PointedEars' Lahn <PointedEars@web.de>
Date	2015-06-24 19:31 +0200
Message-ID	<1997407.asJGmFjWh6@PointedEars.de>
In reply to	#28148

Evertjan. wrote:

> Thomas 'PointedEars' Lahn […] wrote […]:
>> Dr J R Stockton wrote:
>>> […] Thomas 'PointedEars' Lahn […] posted:
>>>> According to the ODE and Merriam-Webster,
>>> 
>>> I think that you should have explained the difference between the ODE
>>> and the OED.
>> I do not think so.
> 
> Order of das Deutsches Empire.
> 
> <https://en.wikipedia.org/wiki/Teutonic_Order>

Guess again.

-- 
PointedEars
FAQ: <http://PointedEars.de/faq> | SVN: <http://PointedEars.de/wsvn/>
Twitter: @PointedEars2 | ES Matrix: <http://PointedEars.de/es-matrix>
Please do not cc me. / Bitte keine Kopien per E-Mail.

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

#28153 — OT: English dictionaries (was: Bignumber conversion and arithmetic on unsigned operand pairs using +,-,* now work in anybase.)

From	"Christoph M. Becker" <cmbecker69@arcor.de>
Date	2015-06-24 19:54 +0200
Subject	OT: English dictionaries (was: Bignumber conversion and arithmetic on unsigned operand pairs using +,-,* now work in anybase.)
Message-ID	<mmeqs6$21d$1@solani.org>
In reply to	#28152

Thomas 'PointedEars' Lahn wrote:

> Evertjan. wrote:
> 
>> Thomas 'PointedEars' Lahn […] wrote […]:
>>> Dr J R Stockton wrote:
>>>> […] Thomas 'PointedEars' Lahn […] posted:
>>>>> According to the ODE and Merriam-Webster,
>>>>
>>>> I think that you should have explained the difference between the ODE
>>>> and the OED.
>>> I do not think so.
>>
>> Order of das Deutsches Empire.
>>
>> <https://en.wikipedia.org/wiki/Teutonic_Order>
> 
> Guess again.

Spoiler: Oxford Dictionary of English[1].  By the way, the Wikipedia
article explains the difference between the ODE and the OED.

[1] <https://en.wikipedia.org/wiki/Oxford_Dictionary_of_English>

-- 
Christoph M. Becker

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

#28156

From	jonas.thornvall@gmail.com
Date	2015-06-25 01:57 -0700
Message-ID	<d58dbc53-80d4-4404-a1c4-8d3617c4d8ad@googlegroups.com>
In reply to	#27780

Den onsdag 20 maj 2015 kl. 17:58:52 UTC+2 skrev Evertjan.:
> jonas.thornvall@gmail.com wrote on 20 mei 2015 in comp.lang.javascript:
> 
> > And i though Janevert said i could not do it, no overflow.
> 
> Impolite to misquote my name.
> 
> I never said you cannot do something,
> only that you should not do some things.
> 
> I never discussed overflow with you.
> 
> I only critisize your methods 
> and understanding of mathmatics and scripting.
> 
> -- 
> Evertjan.
> The Netherlands.
> (Please change the x'es to dots in my emailaddress)

Was this the correct way for three valued? I try to change basconversion for faster match.

function arrCompareTwo(A, B)
{
   var AA = A.length;
   var BB = B.length;
   if(AA > BB) return false;
   if(AA < BB) return true;
   // AA = BB, compare indexes from biggest to smallest.
   for(i = AA - 1; i >= 0; i -- )
   {
      if(A[i] < B[i]) return true;
      if(A[i] > B[i]) return false;
  }
   return null;
}

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

#28157

From	"Evertjan." <exxjxw.hannivoort@inter.nl.net>
Date	2015-06-25 12:44 +0200
Message-ID	<XnsA4C481AA8D7A8eejj99@194.109.6.166>
In reply to	#28156

jonas.thornvall@gmail.com wrote on 25 Jun 2015 in comp.lang.javascript:

> Was this the correct way 

"the correct way", what nonsense!

> for three valued?

"three valued" ??

> I try to change basconversion for faster match. 

>    if(AA > BB) return false;
>    if(AA < BB) return true;

>       if(A[i] < B[i]) return true;
>       if(A[i] > B[i]) return false;
>    return null;

Customary, and imho much better, is to return +1 , 0 , -1

-- 
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

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

#28159

From	jonas.thornvall@gmail.com
Date	2015-06-25 04:09 -0700
Message-ID	<ceedf973-66db-4800-b9fa-f1582d2cf79b@googlegroups.com>
In reply to	#28157

Den torsdag 25 juni 2015 kl. 12:44:57 UTC+2 skrev Evertjan.:
> jonas.thornvall@gmail.com wrote on 25 Jun 2015 in comp.lang.javascript:
> 
> > Was this the correct way 
> 
> "the correct way", what nonsense!
> 
> > for three valued?
> 
> "three valued" ??
> 
> > I try to change basconversion for faster match. 
> 
> >    if(AA > BB) return false;
> >    if(AA < BB) return true;
> 
> >       if(A[i] < B[i]) return true;
> >       if(A[i] > B[i]) return false;
> >    return null;
> 
> Customary, and imho much better, is to return +1 , 0 , -1
> 
> -- 
> Evertjan.
> The Netherlands.
> (Please change the x'es to dots in my emailaddress)

Well i thouught it was three valued logic.

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

#28162

From	John Harris <niam@jghnorth.org.uk.invalid>
Date	2015-06-25 15:08 +0100
Message-ID	<pp2ooadg2emnqn9t1vdug9f3jdfp7s9e4v@4ax.com>
In reply to	#28159

On Thu, 25 Jun 2015 04:09:34 -0700 (PDT), jonas.thornvall@gmail.com
wrote:

>Den torsdag 25 juni 2015 kl. 12:44:57 UTC+2 skrev Evertjan.:
>> jonas.thornvall@gmail.com wrote on 25 Jun 2015 in comp.lang.javascript:
>> 
>> > Was this the correct way 
>> 
>> "the correct way", what nonsense!
>> 
>> > for three valued?
>> 
>> "three valued" ??
>> 
>> > I try to change basconversion for faster match. 
>> 
>> >    if(AA > BB) return false;
>> >    if(AA < BB) return true;
>> 
>> >       if(A[i] < B[i]) return true;
>> >       if(A[i] > B[i]) return false;
>> >    return null;
>> 
>> Customary, and imho much better, is to return +1 , 0 , -1
>
>Well i thouught it was three valued logic.

That's not what comparison does. It shouldn't return false just
because one number is bigger than the other.  

It's LT, LE, EQ, GE, and GT that return true or false.
  
  John

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

#28164

From	jonas.thornvall@gmail.com
Date	2015-06-25 09:30 -0700
Message-ID	<85fcbff1-f4bb-407c-a79e-329faf59958e@googlegroups.com>
In reply to	#28157

Den torsdag 25 juni 2015 kl. 12:44:57 UTC+2 skrev Evertjan.:
> jonas.thornvall@gmail.com wrote on 25 Jun 2015 in comp.lang.javascript:
> 
> > Was this the correct way 
> 
> "the correct way", what nonsense!
> 
> > for three valued?
> 
> "three valued" ??
> 
> > I try to change basconversion for faster match. 
> 
> >    if(AA > BB) return false;
> >    if(AA < BB) return true;
> 
> >       if(A[i] < B[i]) return true;
> >       if(A[i] > B[i]) return false;
> >    return null;
> 
> Customary, and imho much better, is to return +1 , 0 , -1
> 
> -- 
> Evertjan.
> The Netherlands.
> (Please change the x'es to dots in my emailaddress)

I managed to get the algorithm to work faster using 3 values, but to be honest i am not quite sure why it work. And the bignumb functionality broke down.

http://jt.node365.se/baseconversion5.html

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

#28172

From	Scott Sauyet <scott.sauyet@gmail.com>
Date	2015-06-25 13:12 -0700
Message-ID	<ea5903a4-1654-45c5-a40c-dae22330d54d@googlegroups.com>
In reply to	#28164

jonas.thornvall@gmail.com wrote:
> I managed to get the algorithm to work faster using 3 values, but to be 
> honest i am not quite sure why it work. 

So no, you did not get an algorithm working.

You cast a spell.

  -- Scott

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

#28174

From	"Evertjan." <exxjxw.hannivoort@inter.nl.net>
Date	2015-06-25 22:21 +0200
Message-ID	<XnsA4C4E37EE4E09eejj99@194.109.133.246>
In reply to	#28172

Scott Sauyet <scott.sauyet@gmail.com> wrote on 25 Jun 2015 in 
comp.lang.javascript:

> jonas.thornvall@gmail.com wrote:
>> I managed to get the algorithm to work faster using 3 values, but to be 
>> honest i am not quite sure why it work. 
> 
> So no, you did not get an algorithm working.

Is an algorithm that does not "work" an algorithm?

> You cast a spell.

-- 
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

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

#28176

From	Scott Sauyet <scott.sauyet@gmail.com>
Date	2015-06-25 13:28 -0700
Message-ID	<98b92e03-533c-4e86-94b3-831547a710f0@googlegroups.com>
In reply to	#28174

Evertjan. wrote:
> Scott Sauyet wrote
>> jonas.thornvall@gmail.com wrote:
>>> I managed to get the algorithm to work faster using 3 values, but to be 
>>> honest i am not quite sure why it work. 
>> 
>> So no, you did not get an algorithm working.
>>
>> You cast a spell.
> 
> Is an algorithm that does not "work" an algorithm?

I don't know.  But I do know that if you don't understand it, it's not
programming.  It's incantations.

  -- Scott

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

#28177

From	jonas.thornvall@gmail.com
Date	2015-06-25 15:56 -0700
Message-ID	<fd9f39b6-3507-40f5-acd2-9ca3b8d6814f@googlegroups.com>
In reply to	#28176

Den torsdag 25 juni 2015 kl. 22:28:22 UTC+2 skrev Scott Sauyet:
> Evertjan. wrote:
> > Scott Sauyet wrote
> >> jonas.thornvall@gmail.com wrote:
> >>> I managed to get the algorithm to work faster using 3 values, but to be 
> >>> honest i am not quite sure why it work. 
> >> 
> >> So no, you did not get an algorithm working.
> >>
> >> You cast a spell.
> > 
> > Is an algorithm that does not "work" an algorithm?
> 
> I don't know.  But I do know that if you don't understand it, it's not
> programming.  It's incantations.
> 
>   -- Scott

Well in this case you were right ;) it is really not working, but i got the  ideas to get it working.

1. If compare in **looptwo** returns null it is a multiple that will reduce the number to zero.
2. This was harder, there is a case that i can't fetch without modification of existing code. I have to fetch the case where the number is a multiple of the base and **loop** in return null, direcly after rollup Powers. There will be if(loop!=null) do all the code ELSE there is a 1 and rest is zeros. 

I probably should make a flowchart tomorrow and start fresh. But the above is enough to get the code working that i can see clearly.

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

#28191

From	jonas.thornvall@gmail.com
Date	2015-06-27 19:33 -0700
Message-ID	<373265b1-370e-4983-9aac-ff3fe192b8a9@googlegroups.com>
In reply to	#28176

Den torsdag 25 juni 2015 kl. 22:28:22 UTC+2 skrev Scott Sauyet:
> Evertjan. wrote:
> > Scott Sauyet wrote
> >> jonas.thornvall@gmail.com wrote:
> >>> I managed to get the algorithm to work faster using 3 values, but to be 
> >>> honest i am not quite sure why it work. 
> >> 
> >> So no, you did not get an algorithm working.
> >>
> >> You cast a spell.
> > 
> > Is an algorithm that does not "work" an algorithm?
> 
> I don't know.  But I do know that if you don't understand it, it's not
> programming.  It's incantations.
> 
>   -- Scott

No more need for incantations.

Well sometimes the solutions is closer than one thought. Now the baseconversion do break the GCD loop as soon the value found, no need to squeeze it any further. It is alot faster doing conversion into big bases.

Further improvements Scott, i can take suggestions.

It really work now as i wanted from beginning.
http://jt.node365.se/baseconversion5.html

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

#28192

From	jonas.thornvall@gmail.com
Date	2015-06-27 19:45 -0700
Message-ID	<981c05fd-9012-4350-bf7f-b8e6a75a6149@googlegroups.com>
In reply to	#28191

Den söndag 28 juni 2015 kl. 04:33:52 UTC+2 skrev jonas.t...@gmail.com:
> Den torsdag 25 juni 2015 kl. 22:28:22 UTC+2 skrev Scott Sauyet:
> > Evertjan. wrote:
> > > Scott Sauyet wrote
> > >> jonas.thornvall@gmail.com wrote:
> > >>> I managed to get the algorithm to work faster using 3 values, but to be 
> > >>> honest i am not quite sure why it work. 
> > >> 
> > >> So no, you did not get an algorithm working.
> > >>
> > >> You cast a spell.
> > > 
> > > Is an algorithm that does not "work" an algorithm?
> > 
> > I don't know.  But I do know that if you don't understand it, it's not
> > programming.  It's incantations.
> > 
> >   -- Scott
> 
> No more need for incantations.
> 
> Well sometimes the solutions is closer than one thought. Now the baseconversion do break the GCD loop as soon the value found, no need to squeeze it any further. It is alot faster doing conversion into big bases.
> 
> Further improvements Scott, i can take suggestions.
> 
> It really work now as i wanted from beginning.
> http://jt.node365.se/baseconversion5.html

It would be nice if the algorithm also did break when the positive remainder after multiplication less than the base multiple, but i am not it is worth the work?

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

#28193

From	"Evertjan." <exxjxw.hannivoort@inter.nl.net>
Date	2015-06-28 09:17 +0200
Message-ID	<XnsA4C75E7BE4F61eejj99@194.109.6.166>
In reply to	#28192

jonas.thornvall@gmail.com wrote on 28 Jun 2015 in comp.lang.javascript:

> It would be nice if the algorithm also did break

A broken algorithm being no more than an incantation.

> when the positive remainder after multiplication 

Multiplications have no remainders, 
neither positive nor negative.

> less than the base multiple, 

What is that?

A "base multiple" being an incatative mantrum?

> but i am not it is worth the work? 

You are not.

-- 
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

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

#28194

From	jonas.thornvall@gmail.com
Date	2015-06-28 02:10 -0700
Message-ID	<dddb8bc5-c8e7-4430-9aad-d6d347d8190d@googlegroups.com>
In reply to	#28193

Den söndag 28 juni 2015 kl. 09:17:24 UTC+2 skrev Evertjan.:
> jonas.thornvall@gmail.com wrote on 28 Jun 2015 in comp.lang.javascript:
> 
> > It would be nice if the algorithm also did break
> 
> A broken algorithm being no more than an incantation.
> 
> > when the positive remainder after multiplication 
> 
> Multiplications have no remainders, 
> neither positive nor negative.

Correct but after multiplication done you can take multresult - number you get a remainder if that remainder is positive and less than the power(basemutliple) you currently work with, then the digit you used in multiplication is correct.

> > less than the base multiple, 
> 
> What is that?

It is a power of the base, it is an instance multiple of base x.

> A "base multiple" being an incatative mantrum?

> > but i am not it is worth the work? 
> 
> You are not.

Whatevert
> 
> -- 
> Evertjan.
> The Netherlands.
> (Please change the x'es to dots in my emailaddress)

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

#28195

From	jonas.thornvall@gmail.com
Date	2015-06-28 02:21 -0700
Message-ID	<63aca0aa-6e84-48e7-9183-076068e8a04b@googlegroups.com>
In reply to	#28194

Den söndag 28 juni 2015 kl. 11:10:19 UTC+2 skrev jonas.t...@gmail.com:
> Den söndag 28 juni 2015 kl. 09:17:24 UTC+2 skrev Evertjan.:
> > jonas.thornvall@gmail.com wrote on 28 Jun 2015 in comp.lang.javascript:
> > 
> > > It would be nice if the algorithm also did break
> > 
> > A broken algorithm being no more than an incantation.
> > 
> > > when the positive remainder after multiplication 
> > 
> > Multiplications have no remainders, 
> > neither positive nor negative.
> 
> Correct but after multiplication done you can take multresult - number you get a remainder if that remainder is positive and less than the power(basemutliple) you currently work with, then the digit you used in multiplication is correct.
> 
> > > less than the base multiple, 
> > 
> > What is that?
> 
> It is a power of the base, it is an instance multiple of base x.
> 
> > A "base multiple" being an incatative mantrum?
> 
> > > but i am not it is worth the work? 
> > 
> > You are not.
> 
> Whatevert
> > 
> > -- 
> > Evertjan.
> > The Netherlands.
> > (Please change the x'es to dots in my emailaddress)

It would nearly half the runtime of the algorithm in average case, but i still not sure it is worthwhile doing.

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

#28196

From	jonas.thornvall@gmail.com
Date	2015-06-28 02:24 -0700
Message-ID	<054f05a4-7c6a-4722-8ad4-66240d05b8fd@googlegroups.com>
In reply to	#28193

Den söndag 28 juni 2015 kl. 09:17:24 UTC+2 skrev Evertjan.:
> jonas.thornvall@gmail.com wrote on 28 Jun 2015 in comp.lang.javascript:
> 
> > It would be nice if the algorithm also did break
> 
> A broken algorithm being no more than an incantation.
Well it isn't broken anymore maybe you should flush the cache of your browser.
> > when the positive remainder after multiplication 
> 
> Multiplications have no remainders, 
> neither positive nor negative.
> 
> > less than the base multiple, 
> 
> What is that?
> 
> A "base multiple" being an incatative mantrum?
> 
> > but i am not it is worth the work? 
> 
> You are not.
> 
> -- 
> Evertjan.
> The Netherlands.
> (Please change the x'es to dots in my emailaddress)

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

Page 2 of 5 — ← Prev page 1 [2] 3 4 5 Next page →

csiph-web

Bignumber conversion and arithmetic on unsigned operand pairs using +,-,* now work in anybase.

Contents

#28148

#28149

#28155

#28152

#28153 — OT: English dictionaries (was: Bignumber conversion and arithmetic on unsigned operand pairs using +,-,* now work in anybase.)

#28156

#28157

#28159

#28162

#28164

#28172

#28174

#28176

#28177

#28191

#28192

#28193

#28194

#28195

#28196