Lottery Programming Question
nigel
pseudo-code to loop round the ~14M combos of a 6/49 lottery including
restarts, which they're prepared to publish?
Evil Nigel
Colin Fairbrother
nigel
I take that as a 'no' then.
I can see two ways of doing it - either flattening the structure by
looping through the CSNs and calculating each combo from the CSN.
Restarts can be achieved by remembering the last CSN processed.
The other way involves coding the equivalent of For/Next loops manually
(with lots of GoTos) and remembering the last combo processed.
Evil Nigel
Colin Fairbrother
> Colin Fairbrother wrote:
> > On Apr 3, 11:48 pm, nigel <use...@nospam.com> wrote:
>
> >>To save me re-inventing the wheel, does anyone have any code or
> >>pseudo-code to loop round the ~14M combos of a 6/49 lottery including
> >>restarts, which they're prepared to publish?
>
> >>Evil Nigel
>
>
> I take that as a 'no' then.
>
> I can see two ways of doing it - either flattening the structure by
> looping through the CSNs and calculating each combo from the CSN.
> Restarts can be achieved by remembering the last CSN processed.
>
> The other way involves coding the equivalent of For/Next loops manually
> (with lots of GoTos) and remembering the last combo processed.
>
> Evil Nigel
On the contrary the answer is yes. The code provided at the link
enables you to iterate through all the combinations of six integers.
For each iteration after any combination of the integers you can do
some processing and opt out of that loop. Did you notice this line if
using VBA, "Use Continue For instead of GoTo a label to opt out of a
For loop" .
Provide the code showing how far you get in doing what you want to do
and explain where you are having difficulty and a way forward can be
easily given.
Colin Fairbrother
PS An experienced programmer can pick up in a few lines the level of
aptitude in some code. For a novice it is better to be humble and
always explain your abbreviations. Impressive code speaks for itself
and may simply be the implementation of an idea or way of going about
it as was the case with John Rawson in calculating coverage in the
shortest time. Michael Harrington has provided some excellent code in
this area.
nigel
It looks as though you don't understand the question. Perhaps an example
would help.
You kick off a run, saving after every combo tested. When it gets to say
1,13,14,17,25,42 you experience a power cut and your computer dies.
When the power returns and you want to resume your run, where in your
code can you enter or restore the numbers 1,13,14,17,25,42 so that
processing can resume without having to restart from 1,2,3,4,5,6?
Evil Nigel
Colin Fairbrother
>
> - Show quoted text -
Simple - save progressively to a database.
Colin Fairbrother
nigel
Saving to a database (or spreadsheet cells) is simple, restarting is not
so simple.
Evil Nigel
Gerry
Colin has software that will predict power failures and save the work
nanoseconds before the outage. He keeps it hidden with all those
wheels he refuses to publish.
nigel
> Colin has software that will predict power failures and save the work
> nanoseconds before the outage. He keeps it hidden with all those
> wheels he refuses to publish.
One of my number-crunching progs has two phases - a generation phase
which produces raw data and a consolidation phase which lumps it all
together and producs sets of weights.
As the consolidated data grew larger and larger, updating it took longer
and longer until it took all night. Obviously with a run lasting that
length of time it was sensible to build in restarts. Initially I 'saved'
after processing each row of raw data but that pushed the execution time
way beyond acceptable limits. In the end I settled for 'saving' when the
time ended in 00 seconds, a compromise between what is an acceptable
amount of work to redo if the computer fails and what is an acceptable
amount of extra prog execution time in order to achieve it.
Many of my number-crunching progs have 'save's and 'restarts' built in.
It's standard practice in IT for long runs. Certainly virtually every
role-play computer game these days has the facility to save a
game/character at a certain point and restart it later.
Evil Nigel
Colin Fairbrother
Nigel
You waffle on using words that don't impress and others that show how
little you know about coding and never actually give us some nitty
gritty ie code for what appears to be a silly objective. I think of
myself as a relational database programmer where saving data is second
nature - in fact I had to restrict myself from doing so in this field
of interest as I used arrays more and more rather than record sets
etc.
You simply use the mod function wirth a number such as 1,000, 10,000
50,000 or 100,000 etc you pick from a drop down: Where you save to is
a matter of what you are using - in your case I understand you use
Excel so a cell on a worksheet is OK just as is a text file but I
would use a table.
Time for you to show us a bit of code. If you don't know how to save
and retrieve data or skip to a starting point in a loop then you
really should be reading an introduction to Excel VBA programming book
or asking your questions on an Excel newsgroup where they would expect
you to show the snippet where you are stuck,
Colin Fairbrother.
nigel
> Nigel
>
> You waffle on using words that don't impress
I'm not trying to impress - you are.
> and others that show how
> little you know about coding and never actually give us some nitty
> gritty ie code for what appears to be a silly objective.
I'm asking for code - you're the one who isn't supplying.
> I think of
> myself as a relational database programmer where saving data is second
> nature - in fact I had to restrict myself from doing so in this field
> of interest as I used arrays more and more rather than record sets
> etc.
What you think of yourself is not an issue here.
>
> You simply use the mod function wirth a number such as 1,000, 10,000
> 50,000 or 100,000 etc you pick from a drop down: Where you save to is
> a matter of what you are using - in your case I understand you use
> Excel so a cell on a worksheet is OK just as is a text file but I
> would use a table.
I've already explained that saving is not a problem.
> Time for you to show us a bit of code.
I'm asking for help. Since none is forthcoming I'll bite the bullet and
work it out for myself.
> If you don't know how to save
> and retrieve data or skip to a starting point in a loop then you
> really should be reading an introduction to Excel VBA programming book
> or asking your questions on an Excel newsgroup where they would expect
> you to show the snippet where you are stuck,
You really don't understand, do you! Excel is irrelevant.
Evil Nigel
Colin Fairbrother
We have seen no evidence that you can work out anything! Never a code
snippet, never a report just fucking waffle, waffle waffle.
Colin
nigel
>
>
> We have seen no evidence that you can work out anything! Never a code
> snippet, never a report just fucking waffle, waffle waffle.
>
> Colin
I've sussed it - you're using argument inversion.
I ask for (pseudo)code which you don't have and don't know how to work
out so you ask to see mine.
If you look in the archives you'll see quite a lot of (pseudo)code from
me, so you state I've never published anything and just waffle. (I can't
claim I was the first but I certainly published the six nested loops for
lottery combos many years back in this ng).
Just yesterday I explained how my 'randomness' test for lottery combos
'worked', although I didn't publish the code explicitly because I don't
use Excel and it would have been unhelpful for people without access to
my spreadsheet and its built-in functions, who would have been unable to
ascertain what the functions do.
Evil Nigel
sm...@gmx.de
Nigel,
I had the same problem like you,
to continue or to start a program loop at a certain step.
For your example: 1,13,14,17,25,42
I programmed it in this way:
x = 0
// use x as a switch
For a = 1 To 44
For b = a + 1 To 45
For c = b + 1 To 46
For d = c + 1 To 47
For e = d + 1 To 48
For f = e + 1 To 49
Print a; b; c; d; e; f
If x = 0 Then
a = 1
b = 13
c = 14
d = 17
e = 25
f = 41
x = 1
EndIf
// here insert your code
Next f
Next e
Next d
Next c
Next b
Next a
Also you can insert six input commands for a to f
to start with other numbers of your choice
Manfred
nigel
Hi Manfred,
Many thanks, that's very simple and elegant.
I know of one or two languages where that wouldn't work because they
don't allow the outer a-f to be altered inside their for/next loops but
it will definitely work in my spreadsheet of choice and should work in
Excel too.
Thanks again,
Evil Nigel
Colin Fairbrother
Nigel
Your original question was about resuming processing after a power
failure. The only difference between the code Manfred supplied and the
standard code here
http://groups.google.com/group/lottogroup/web/probability-primer-or-refresher
for enumerating the combinations without integer repetition is that
instead of a counter Manfred has used a boolean variable and you would
see this used extensively using arrays in Michael Harrington's code
which I referred you to. So you thank Manfred for something you state
you posted years ago and make out this solves your problem of
resumption after a power failure. Don't you realize this bullshit
gives away your pretentiousness? Where is the boolean variable value
stored and how does this help if it has to be recalculated to know the
position in the outer loop?
As I pointed out before if you have a long process and in the case of
a power failure or simply running out of space in an Access database
(2GB) and don't want to start from the beginning again then you need
to save your position in the outer loop for which you need a counter
number to retrieve from wherever you have stored it after say checking
a check box on your form or worksheet
To say that I or anyone else wouldn't understand your code is a cop
out. Generally, you've been left alone because you've come across as a
pretentious would be programmer who time and again has given away your
ignorance by contradictory statements and we've felt a little sorry
for you..
So, what are you doing? I can see how Stig Holmquist aka Bertil Junior
aka Bertil Senior was sniffing around as you are attributing magnitude
to the identifiers and then working out Standard Deviations. This
nonsense has well and truly been debunked by me. For those reading an
introduction to Standard Deviation can be read here: -
http://lottoposter.com/forum_posts.asp?TID=224
For debunking of attributing magnitude to the integers see here: -
http://lottoposter.com/forum_posts.asp?TID=90
Interestingly, you can have a Pick 6, Pool 7 Lotto game as a sub-game
from a Pick 6, Pool 49 game. Simply let -
1 be also 8, 15, 22, 29, 36 or 43
2 be also 9, 16, 23, 30, 37 or 44
3 be also 10, 17, 24, 31, 38 or 45
4 be also 11, 18, 25, 32, 39 or 46
5 be also 12, 19, 26, 33, 40 or 47
6 be also 13, 20, 27, 34, 41 or 48
7 be also 14, 21, 28, 35, 42 or 49
How do your Standard Deviations work for this example?
Colin Fairbrother
Colin Fairbrother
wrote:
> On Apr 8, 8:15 pm, nigel <use...@nospam.com> wrote:
>
>
>
>
>
> > Colin Fairbrother wrote:
>
> > > We have seen no evidence that you can work out anything! Never a code
> > > snippet, never a report just fucking waffle, waffle waffle.
>
> > > Colin
>
> > I've sussed it - you're using argument inversion.
>
> > I ask for (pseudo)code which you don't have and don't know how to work
> > out so you ask to see mine.
>
> > If you look in the archives you'll see quite a lot of (pseudo)code from
> > me, so you state I've never published anything and just waffle. (I can't
> > claim I was the first but I certainly published the six nested loops for
> > lottery combos many years back in this ng).
>
> > Just yesterday I explained how my 'randomness' test for lottery combos
> > 'worked', although I didn't publish the code explicitly because I don't
> > use Excel and it would have been unhelpful for people without access to
> > my spreadsheet and its built-in functions, who would have been unable to
> > ascertain what the functions do.
>
> > Evil Nigel
>
> Nigel
>
> Your original question was about resuming processing after a power
> failure. The only difference between the code Manfred supplied and the
> for enumerating the combinations without integer repetition is that
> instead of a counter Manfred has used a boolean variable and you would
> see this used extensively using arrays in Michael Harrington's code
> which I referred you to. So you thank Manfred for something you state
> you posted years ago and make out this solves your problem of
> resumption after a power failure. Don't you realize this bullshit
> gives away your pretentiousness? Where is the boolean variable value
> stored and how does this help if it has to be recalculated to know the
> position in the outer loop?
>
> As I pointed out before if you have a long process and in the case of
> a power failure or simply running out of space in an Access database
> (2GB) and don't want to start from the beginning again then you need
> to save your position in the outer loop for which you need a counter
> number to retrieve from wherever you have stored it after say checking
> a check box on your form or worksheet
>
> To say that I or anyone else wouldn't understand your code is a cop
> out. Generally, you've been left alone because you've come across as a
> pretentious would be programmer who time and again has given away your
> ignorance by contradictory statements and we've felt a little sorry
> for you..
>
> So, what are you doing? I can see how Stig Holmquist aka Bertil Junior
> aka Bertil Senior was sniffing around as you are attributing magnitude
> to the identifiers and then working out Standard Deviations. This
> nonsense has well and truly been debunked by me. For those reading an
> For debunking of attributing magnitude to the integers see here: -http://lottoposter.com/forum_posts.asp?TID=90
>
> Interestingly, you can have a Pick 6, Pool 7 Lotto game as a sub-game
> from a Pick 6, Pool 49 game. Simply let -
> 1 be also 8, 15, 22, 29, 36 or 43
> 2 be also 9, 16, 23, 30, 37 or 44
> 3 be also 10, 17, 24, 31, 38 or 45
> 4 be also 11, 18, 25, 32, 39 or 46
> 5 be also 12, 19, 26, 33, 40 or 47
> 6 be also 13, 20, 27, 34, 41 or 48
> 7 be also 14, 21, 28, 35, 42 or 49
>
> How do your Standard Deviations work for this example?
>
>
> - Show quoted text -
Correction:
The Pick 6 from Pool 7 game as a sub-game from a Pick 6 from Pool 49
game wouldn't work because we want Combinations without repetition. We
could have 7 sets of 7 balls each set a different colour but a way to
exclude the same colour being picked twice or more would be needed.
Colin Fairbrother
nigel
>
> Your original question was about resuming processing after a power
> failure. The only difference between the code Manfred supplied and the
> standard code here
> http://groups.google.com/group/lottogroup/web/probability-primer-or-refresher
> for enumerating the combinations without integer repetition is that
> instead of a counter Manfred has used a boolean variable and you would
> see this used extensively using arrays in Michael Harrington's code
> which I referred you to. So you thank Manfred for something you state
> you posted years ago and make out this solves your problem of
> resumption after a power failure. Don't you realize this bullshit
> gives away your pretentiousness? Where is the boolean variable value
> stored and how does this help if it has to be recalculated to know the
> position in the outer loop?
You really haven't got a handle on this, have you!
> So, what are you doing? I can see how Stig Holmquist aka Bertil Junior
> aka Bertil Senior was sniffing around as you are attributing magnitude
> to the identifiers and then working out Standard Deviations. This
> nonsense has well and truly been debunked by me.
Don't tell me - you've spent twenty years as a psychologist and you
completely understand why some sequences look more random to the human
eye than others.
Bertil read and understood my initial post - that there was no
predictive value in what I was looking at, something else you don't seem
to be able to grasp.
Yes, my work was nonsense - I pointed out some of the flaws myself, and
in doing so I hope others who are interested in what I was attempting
will therefore have less mistakes to make.
Evil Nigel
Colin Fairbrother
What you are on about is -
1 Attributing magnitude to the identifiers for testing purposes.
2 Thinking for the conventional numerical order in which the integers
are presented that some combinations are more likely to occur than
others.
You should realize that including Numerical Order there are 49! =
49x48x47x46x45x44x43x42x41x40x39x38x37x36x35x34x33x32x31x30x29x28x27x26x25x24x23x22x21x20x19x18x17x16x15x14x13x12x11x10x9x8x7x6x5x4x3x2x1
orders or permutations of 49 integers each of which is just as valid
as Numerical Order.
Basically this is just a variation on the Sums nonsense whichj has
been totally debunked by me. See -
http://lottoposter.com/forum_posts.asp?TID=143
and also
http://lottoposter.com/forum_posts.asp?TID=142
The upshot is proportionality is maintained between the groupings for
all possibilities and that of a reasonable sample and no advantage
gained.
Colin Fairbrother
Gerry
> Nigel
>
> What you are on about is -
> 1 Attributing magnitude to the identifiers for testing purposes.
> 2 Thinking for the conventional numerical order in which the integers
> are presented that some combinations are more likely to occur than
> others.
>
> You should realize that including Numerical Order there are 49! =
> 49x48x47x46x45x44x43x42x41x40x39x38x37x36x35x34x33x32x31x30x29x28x27x26x25x24x23x22x21x20x19x18x17x16x15x14x13x12x11x10x9x8x7x6x5x4x3x2x1
Woo-Hoo
Colin wants everyone to know he can count backwards from 49
Jack Ricci
concerning evil nigel:
> You waffle on using words that don't impress and others that show how
> little you know about coding and never actually give us some nitty
> gritty ie code for what appears to be a silly objective.
...Colin...Anyone ever teach you any Public Relation skills? To be a guru,
mentor, teacher, celebrity, expert or critic in any field, ...you must be
personable, or no one will pay much attention to you except for all the
wrong reasons. It's best you watch some hardcore videos on constructive
communication and being human first...Your " hitman mentality " way of
getting to the top does not impress a whole lot of the entire world.
Nigel?...Of all people to pick on...Leave the guy alone, Colin. But I'm
certain Nigel will see the humour in it :)
LottoHackJack
------------------------------------------------------------------------------------------------------------------------------------------------------------
"Colin Fairbrother" <colin.fa...@gmail.com> wrote in message
news:4ce4050c-d17f-4f76...@n34g2000yqb.googlegroups.com...
> On Apr 8, 2:06 am, nigel <use...@nospam.com> wrote:
>> Gerry wrote:
>> > Colin has software that will predict power failures and save the work
>> > nanoseconds before the outage. He keeps it hidden with all those
>> > wheels he refuses to publish.
>>
>> One of my number-crunching progs has two phases - a generation phase
>> which produces raw data and a consolidation phase which lumps it all
>> together and producs sets of weights.
>>
>> As the consolidated data grew larger and larger, updating it took longer
>> and longer until it took all night. Obviously with a run lasting that
>> length of time it was sensible to build in restarts. Initially I 'saved'
>> after processing each row of raw data but that pushed the execution time
>> way beyond acceptable limits. In the end I settled for 'saving' when the
>> time ended in 00 seconds, a compromise between what is an acceptable
>> amount of work to redo if the computer fails and what is an acceptable
>> amount of extra prog execution time in order to achieve it.
>>
>> Many of my number-crunching progs have 'save's and 'restarts' built in.
>> It's standard practice in IT for long runs. Certainly virtually every
>> role-play computer game these days has the facility to save a
>> game/character at a certain point and restart it later.
>>
>> Evil Nigel
>
> Nigel
>
-------------------------------------------------------------------------------------------------------------------------------------------------
Colin Fairbrother
Using your computer calculator %SystemRoot%\System32\calc.exe you find
that it is so large that it reverts to scientific notation. The
highest you will get with System 32 using your computer calculator and
expressed decimally is 29! which is 8841761993739701954543616000000.
So, 49! divided by 29! is 68796453067268024353471856640000.
Multiply the two large numbers together and you have 49!
Colin Fairbrother
http://LottoPoster.com
Colleen Goodsister
That should scare the shit out of those dumbass quick pickers
who don't know jack squat but somehow manage to scoop all the
fkn jackpots.
John Griffin
I did this a long time ago, as part of a program someone wanted
to use to select a random sample of combinations for some reason.
I posted a VB6 form file which includes a function which will
return all combinations of any format, not just 6/49, one per
function call. It doesn't have any for/next stuff and there's no
'go to' in it. You could use that kind of loop to generate all
the 'bonus balls' for each combination. Using 6/49 as an
example, you just use something like combs=getnextcomb(newcomb()
as long, outof as long, take as long), where outof=49, take=6,
and the array newcomb(take) returns the next combination. The
function returns a serial number or 0 after 44,45,46,47,48,49.
I'll find the function it among the junk on my disk and post it
if anyone really cares.
Making a restart capability would be easy enough. If I remember
right, I used some static variables in the function and one of
them (the one that says to continue instead of initializing and
starting at the first combination) would need to be handled
differently.
--
"You may have to give up a piece of your pie so that someone
else can have more." --Obama, explicitly endorsing Karl Marx's
"Ideal state of communism: From each according to his ability,
to each according to his need."
Colin Fairbrother
Rec.Gambling.Lottery..
The worms will squirm their dance of death.
Long live Truth, Logic and Free Thought.
Death to the evil False Accusers and Ignoramuses - (sorry, that should
be Slow Death).
Colin Fairbrother
When the palefaces unite on the basis of skin colour, just like the
blacks and hispanics, there will be another whitey/gringo USA
president.
nigel
> nigel <use...@nospam.com> wrote:
>
>
>>To save me re-inventing the wheel, does anyone have any code
>>or pseudo-code to loop round the ~14M combos of a 6/49 lottery
>>including restarts, which they're prepared to publish?
>>
>>Evil Nigel
>>
>
>
> I did this a long time ago, as part of a program someone wanted
> to use to select a random sample of combinations for some reason.
>
> I posted a VB6 form file which includes a function which will
> return all combinations of any format, not just 6/49, one per
> function call. It doesn't have any for/next stuff and there's no
> 'go to' in it. You could use that kind of loop to generate all
> the 'bonus balls' for each combination. Using 6/49 as an
> example, you just use something like combs=getnextcomb(newcomb()
> as long, outof as long, take as long), where outof=49, take=6,
> and the array newcomb(take) returns the next combination. The
> function returns a serial number or 0 after 44,45,46,47,48,49.
>
> I'll find the function it among the junk on my disk and post it
> if anyone really cares.
Yes please. Manfred's solution is fine, but it would be interesting to
compare alternatives to see how quickly they run.
>
> Making a restart capability would be easy enough. If I remember
> right, I used some static variables in the function and one of
> them (the one that says to continue instead of initializing and
> starting at the first combination) would need to be handled
> differently.
I'm using VBA/Excel inter alia, so it's easy (but slow!) to swap stuff
between VBA and spreadsheet cells with the occasional save.
Evil Nigel
Gerry
> When the palefaces unite on the basis of skin colour, just like the
> blacks and hispanics, there will be another whitey/gringo USA
> president.
Ah yes, even the neutral Aussies can be race baiters.
A free market capitalist who doesn't fuck over his allies will be
"change to believe in."
Skin color(colour) is optional.
nigel
> Skin color(colour) is optional.
I think women with transparent skin would make interesting porn models.
There would be no need for the man to pull out for the cumshot.
Evil Nigel
Gerry
Kowa Bonga !!
Don't hold back Nigel. What else is on your mind ?
Jack Ricci
"nigel" <use...@nospam.com> wrote in message
news:4cmdnTLPu6OpRlzW...@brightview.co.uk...
Transparent clothes would be all that is needed for me to enjoy. If you find
any, please forward their pics to ri...@mnsi.net
Thanks.
Jack
Gerry
> Rec.Gambling.Lottery..
>
> The worms will squirm their dance of death.
fuckin' toadie suckup
nigel
> Don't hold back Nigel. What else is on your mind ?
I'm a man :)
Sex sex sex sex sex money.
Sex sex sex sex sex cars.
Sex sex sex sex sex food.
Sex sex sex sex sex and more sex.
Evil Nigel
Gerry
You should be very happy considering how much the lottery
keeps screwing you :)
John Griffin
This function gets all of the 13983816 combinations for 6/49 in
about 20 seconds on my 2.8ghz gadget, running it in the VB6
environment. From the one time I tried to use VBA, my guess is it
would take about 20 minutes.
The first time getcomb is used, it sets global variable
'Started' to true, sizes the array that returns the combinations,
and sets the current combination to 1,2,3,4,5,5 for 6/49,
1,2,3,4,4 for 5/52, etc.
To find a combination, it adds 1 to the value in the last
position or the position it just backed into. If that exceeds
the maximum value of 49 in the 6th spot, 48 in the 5th, etc., it
just backs up one position and repeats that test. When it backs
up to position zero, there are no more combinations. When it
stores a new value in the last position, it exits with a good
combination. When it stores a new value in an earlier position,
it stores the same value in the next one and moves forward.
Example: after 2,7,45,47,48,49, it has 2,7,46,46,x,x, then
2,7,46,47,47,x, then 2,7,46,47,48,48, and finally
2,7,46,47,48,49.
I noticed someone talking about calculating factorials. Don't
remember who...if that had something to do with a problem
calculating the number of combinations, the function "NoOfCombs'
below might be helpful. It doesn't do even one multiplication
and it will handle any practical lottery format, like choose 111
out of 1999, for example.
===========================
dim Started as boolean
dim Serial as long
Function getcomb(Combo() As Long, From As Long, Take As Long) As
Long
'uses module-level data: Started (yes/no) & Serial number
Dim NextNumber As Long
Dim i As Long
Dim where As Long
If Started <> True Then
Started = True
Serial = 0
ReDim Combo(Take)
For i = 1 To Take - 1
Combo(i) = i
Next i
Combo(Take) = Take - 1
End If
where = Take
Do
NextNumber = Combo(where) + 1
If NextNumber > From - where + Take Then
where = where - 1
Else
Combo(where) = NextNumber
If where = Take Then
Serial = Serial + 1
getcomb = Serial
Exit Function
Else
where = where + 1
Combo(where) = NextNumber
End If
End If
If where = 0 Then
getcomb = 0
Exit Function
End If
Loop
End Function
===========================
This function calculates number of combinations using natural
logs, using the usual rules:
First, log(N!)=sum of log(x), x=1 to N
and log(N!/(N-n)!) is sum log(x), x=N-n+1 to N
Second, log(a/b)=log(a)-log(b).
By definition, e^(log(a)-log(b))=a/b (The exp function)
Adds .5 to fix loss of precision
=================
Function NoOfCombs(pick As Long, outof As Long) As Double
'calculate number of combinations for pick/outof
Dim i As Long
Dim logs1 As Double
Dim logs2 As Double
logs2 = 0
logs1 = 0
For i = 1 To pick
logs1 = logs1 + Log(i)
Next i
For i = outof - pick + 1 To outof
logs2 = logs2 + Log(i)
Next i
NoOfCombs = Int(Exp(logs2 - logs1) + 0.5)
End Function
nigel
<snip>
Cheers,
Evil Nigel
Gerry
Best tv show ever !!
Well, one of them...
Colin Fairbrother
>
> - Show quoted text -
John Griffin
Interesting code but a bit slow. Using just standard code as at
http://groups.google.com/group/lottogroup/web/probability-primer-or-refresher
all the combinations can be enumerated into an array in just 4
seconds on my 2.7 GHz laptop using VBA.(not 20 minutes as you guess
and quite a bit less than your 20 seconds). If I used VB 2008 it wouid
be significantly less.
Call EnumCombWoR(49,6)
Start Time: 4/13/2010 10:42:47 AM
Finish Time: 4/13/2010 10:42:51 AM
Time taken: 4 secs
Count of combinations in array: 13983816
Value of element 13983816 in array: 44 45 46 47 48 49
I am very intrigued how your code is usefull in recovering from a
power failure static or no static variable. To the best of my
knowledge, as I pointed out, the value of the point in the processing
has to be saved in some way to hard disk or some other accessable
storage. Amazing how Nigel accepts this now as if he knew it all his
life but it does raise the question why did he pose his original
question.
Colin Fairbrother
Gerry
The eyes glaze over reading about all this VBA,VB6,GHZ stuff
but it appears you have more time to practice pissing farther
than anybody else.
Or maybe use the time savings to stroke your dick.
Colin Fairbrother
discussion and useful articles where all are welcome to contribute
with the exception of the like of the previous poster with his
absolute rubbish posts.
http://groups.google.com/group/lottogroup
Colin Fairbrother
Gerry
All lottogroups are rubbish except RGL.
Give yourself a cheap thrill and ban me from all of them. Good luck with
banning me from RGL.
gARY
wrote:
So, why are you hanging here Colin?
Just asking,
a
Gerry
Colin responded to my post but he thinks if he deletes my comment
and doesn't mention my name he isn't really responding to my post.
nigel
All time top posters (Google stats):
1st Colin Fairbrother 7 posts
2nd Draco Merest 1 post
Evil Nigel
Gerry
guffaw
John Griffin
>
> Interesting code but a bit slow. Using just standard code as
> at
> http://groups.google.com/group/lottogroup/web/probability-prime
> r-or-refresher all the combinations can be enumerated into an
> array in just 4 seconds on my 2.7 GHz laptop using VBA.(not
> 20 minutes as you guess and quite a bit less than your 20
> seconds). If I used VB 2008 it wouid be significantly less.
Actually when that trivial special case code you pointed out and
my general solution are run in the same environment (Compiled by
VB6 instead of running in the debug environment), the timings are
around 1.1s for the special case thing versus about 1.3s for
mine, modified to store the combinations in an array.
How long does it take EnumCombWoR to do a 5/59 thing? <snicker>
How about that 4/24 or the 10/80 keno we have in Washington?
<snicker><snicker> How about whatever other format you might
encounter? <snicker> <snicker> <snicker>
> Call EnumCombWoR(49,6)
No one writing a serious lottery program will use that routine if
they know there's a general solution. Okay, maybe...if they want
to add new code for every format... yuk!
> Start Time: 4/13/2010 10:42:47 AM
> Finish Time: 4/13/2010 10:42:51 AM
> Time taken: 4 secs
> Count of combinations in array: 13983816
> Value of element 13983816 in array: 44 45 46 47 48 49
>
> I am very intrigued how your code is usefull in recovering
> from a power failure static or no static variable. To the best
> of my knowledge, as I pointed out, the value of the point in
> the processing has to be saved in some way to hard disk or
> some other accessable storage. Amazing how Nigel accepts this
> now as if he knew it all his life but it does raise the
> question why did he pose his original question.
Code at this level is not involved in system recovery. What I
meant when I said it would be easy to restart that process is
that it could be modified so that the main program calling it--
the one that would be involved in providing for and implementing
such a recovery--could restore its data and continue with the
next combination after the last checkpoint. Just get a UPS and
don't worry about it.
My guess is that the reason Nigel asked his question was that he
wanted an answer better than the beginning programmer's code you
pointed out. Anyway, he did know about the restart issue.
Everyone does. The checkpoint concept was known even before the
first IBM 7094 program I ever wrote. If you don't have a UPS,
have the main program write checkpoints to a flash drive, which
is both useful and accessible in that process. If you do it
right, you don't need to restart at the exact point where your
power failed, just at the last checkpoint.
John Griffin
Colin Fairbrother
Thank you for your courteous reply.
I don't know whether you followed the posts in this group in your very
long absence but an interesting one came up in 2009 started by me here
http://groups.google.com/group/rec.gambling.lottery/browse_thread/thread/396c92e7bf9a0242/f08dab533949b15f
and continued here
http://groups.google.com/group/rec.gambling.lottery/browse_thread/thread/0c60afe7b167d2d2#
The final code given by Michael Harrington through the mentoring of
John Rawson according to your way of looking at things would be
classified as beginner's code. For the code I supplied at
http://groups.google.com/group/lottogroup/web/probability-primer-or-refresher
it is not meant to be anything other than beginning code.
Interestingly, I've looked at various other approaches and found no
reason to change including yours. I would be interested to know
whether John Rawson in his CoverMaster program, which is commonly used
by current contributors to this group, uses code which is "snicker"
proof.
Why guess the reason Nigel wanted the code to resume when it is given?
Nigel was enthralling us with his quest for finding the most likely or
unlikely combination or most "nonrandom" for the next UK Lotto draw
and this involved days of calculating. On one occasion after many
hours he complained that it had all been to no avail after a blackout
and he had to start again. Doesn't come across to me as someone
familiar with using the mod function to determine the interval as I
suggested to record the point reached in say the outer loop in his
spreadsheet so that he could resume his processing from that point.
There is some code that I would like to see which gives a comparable
speed to that of John Rawson's in his CoverMaster program for finding
the next best plays by coverage in say a 6/49 Lotto game whether it be
for the first, ninth or one hundredth line. Perhaps you could spare a
second or two to show how to go about it which is better than my
currently slow approach especially for the initial 50 lines or so.
Colin Fairbrother
duncan smith
[snip]
>
> You really don't understand, do you! Excel is irrelevant.
>
> Evil Nigel
>
It certainly is.
#############################python code##########################
def iter_combs(n, k, start=None):
if k > n:
raise StopIteration
if start is None:
current = range(k)
else:
current = list(start)
while current[0] != n - k:
yield list(current)
for index in range(-1, -k - 1, -1):
if current[index] != n + index:
current[index] += 1
for other_index in range(index + 1, 0):
current[other_index] = current[index] + other_index
- index
break
yield current
###########################end code###########################
>>> combs = iter_combs(49, 6)
>>> i = 0
>>> for comb in combs:
if i == 1000000:
break
i += 1
>>> comb
[0, 8, 13, 25, 30, 37]
>>> combs.next()
[0, 8, 13, 25, 30, 38]
# (re)start from a particular combination
>>> combs = iter_combs(49, 6, [0, 8, 13, 25, 30, 37])
>>> combs.next()
[0, 8, 13, 25, 30, 37]
>>> combs.next()
[0, 8, 13, 25, 30, 38]
# exhaust generator
>>> for comb in combs:
pass
>>> comb
[43, 44, 45, 46, 47, 48]
>>> combs = iter_combs(49, 6)
>>> combs.next()
[0, 1, 2, 3, 4, 5]
>>>
The only issue is that my code generates combinations of integers (in
lex order) in the range 0 to n-1. It would need adjusting to generate
from the range 1 to n. The easiest (but not most efficient) way would
be to change the final line to:
yield [i+1 for i in current]
and to add the following line at the start of the function so that the
'start' argument would take a combination in the range 1 to n:
start = [i-1 for i in start]
Duncan
Lee Frank
"John Griffin" <thathi...@yahooie.com> wrote in message
news:Xns9D5A850A37871th...@130.133.4.11...
> I hate to say this, but...ROTMFFLMMFAO!
rolling on the motherfucking floor
laughing your motherfucking ass
off?
i can see why you'd hate to say it.
i'd certainly practice denial too if i
were to be fucking my mom on the
floor. don't you think she at least
deserves a bed for your shenanigans?
gARY
> "John Griffin" <thathillbi...@yahooie.com> wrote in message
>
> news:Xns9D5A850A37871th...@130.133.4.11...
>
> > I hate to say this, but...ROTMFFLMMFAO!
>
> rolling on the motherfucking floor
> laughing your motherfucking ass
> off?
Question mark here? If you ask a question and then follow like thus...
> i can see why you'd hate to say it.
> i'd certainly practice denial too if i
> were to be fucking my mom on the
> floor. don't you think she at least
> deserves a bed for your shenanigans?
...you have already made up your mind the answer.
It seems, oneself knows everything so why waste screen ink talking to
oneself?
Good luck,
a
nigel
My amateurish attempts at fiction occasionally include rhetorical
questions. Different proofreaders seem to have differing opinions on
whether theose rhetorical questions should be punctuated by a question mark.
I don't know what's technically correct but personally I have a mild
preference for omitting the question mark so that the reader can infer
it's not meant to be read as a serious question, but others have
different opinions.
If the question you objected to is taken to be rhetorical despite the
question mark then what follows isn't necessarily illogical.
Evil Nigel
Gerry
Sure am glad you cleared that up.
nigel
> Sure am glad you cleared that up.
I thought I made things muddy rather than clear, assuming what you wrote
couldn't be sarcastic unless you're an atypical American :)
Evil Nigel
Gerry
> Gerry wrote:
>
>> Sure am glad you cleared that up.
>
> I thought I made things muddy rather than clear,
Yes you did
> assuming what you wrote
> couldn't be sarcastic unless you're an atypical American :)
I'm an American being forced to live in the United States of Obama.
>
> Evil Nigel
>
gARY
>
> - Show quoted text -
However, the 'said' reply could be seen as being aggressive and
undermining.
Likened to (and in order of) Nick UK and Colin F.
Take ya pick,
a
Lee Frank
"gARY" <ga...@justservices.com> wrote in message
news:79a94456-e61a-4037...@q23g2000yqd.googlegroups.com...
Good luck,
a
``````````````````````````````````````````````````
luck's got nothing to do with it.
nigel
>>
>>>>i can see why you'd hate to say it.
>>>>i'd certainly practice denial too if i
>>>>were to be fucking my mom on the
>>>>floor. don't you think she at least
>>>>deserves a bed for your shenanigans?
>>
>
> However, the 'said' reply could be seen as being aggressive and
> undermining.
>
> Likened to (and in order of) Nick UK and Colin F.
>
> Take ya pick,
> a
TBH, I thought the intent was humorous.
I did some Google searching on the issue of rhetorical questions and
question marks. The formal grammar sites seemed to be pretty universally
in favour of mandatory question marks. About half the sites I found
reckoned that rhetorical questions could be punctuated by question
marks, exclamation marks or full stops, but the majority of those seemed
to be personal opinions or to refer to Wikipedia for justification.
(There is a subject on which I could be considered something of an
expert. There is a very large Wikipedia article on the subject, but it's
riddled with mistakes. Consequently I take anything on Wikipedia with a
huge dose of salt.)
On that evidence it looks as though I was wrong, doesn't it! :)
Evil Nigel
Gerry
I'm so confused.
What's Wikipedia ?
nigel
>
> luck's got nothing to do with it.
To be perfectly frank, I bet you get a lot of teasing about your name.
Evil Nigel
Lee Frank
"nigel" <use...@nospam.com> wrote in message
news:SqCdnb-fSeiN_FbW...@brightview.co.uk...
it means "honest shelter".
why would someone tease
about something that tugs
at the very strings of their
inherent security, safety and
survival addictions?
Lee Frank
"Gerry" <tul...@earthlink.net> wrote in message
news:4e2dnYBZV5eHlFbW...@earthlink.com...
what planet are you living on?
Gerry
Nigel knew I was joking but clearly you're too fucking stupid to
figure it out.
Stick around dumbass, you will be fun to fuck with.
Gerry
He bunks with Barney Frank
Lee Frank
i'm here. give it your best shot.
Lee Frank
is that the best you got?
pathetic.
nigel
Is your nickname 'Perfect'?
(If you think my sense of 'humour' is convoluted you should see some of
my lottery systems!)
Evil Nigel
Gerry
It's better to let Barney give you his best 'shots'
Lee Frank
if this is what you consider "fucking with me"
you are about the lamest flamer since john griffin.
Gerry
Barney does all the "fucking with you"
> you are about the lamest flamer since john griffin.
Who's john griffin? Do you suck his dick too ?
Gerry
"that all you got"
Lee Frank
so when are you gonna be able
to start fucking with me as you
said? you can see i'm sticking
around for it as you suggested.
Lee Frank
you appear to have some type of puerile,
psychological fear of fucking. my how
you suffer.
>> you are about the lamest flamer since john griffin.
>
> Who's john griffin? Do you suck his dick too ?
see?
gARY
>
> - Show quoted text -
OK, I try again.......
The question was not the problem.
The forcing of the answer (talking to onesself) was the issue after
your own question mark.
Anyways, we move on......
a
gARY
> "Gerry" <tula...@earthlink.net> wrote in message
>
> news:t5adne7WCLjHFVbW...@earthlink.com...
>
>
>
>
>
> > On 4/18/2010 3:20 PM, Lee Frank wrote:
>
> >>news:Hp-dnZNOfenJ51bW...@earthlink.com...
> >>> On 4/18/2010 1:47 PM, Lee Frank wrote:
>
>
> - Show quoted text -
So, an old hat here, to remember JG.
Any idea's peoples?
I guess, Nick UK or Colin.
As of course I would,
a
gARY
> "gARY" <g...@justservices.com> wrote in message
You are wrong there pal.
Or maybe u r right ...
You just found ME!
Good er?
a
Nick UK
>
> So, an old hat here, to remember JG.
>
> Any idea's peoples?
>
> I guess, Nick UK or Colin.
>
> As of course I would,
> a
Being the idiot you are - of course you would! And as per usual, you are
wrong!
I didn't have a clue what ROTMFFLMMFAO meant until the poster 'Lee Frank'
said..
> rolling on the motherfucking floor
> laughing your motherfucking ass
> off?
>
> i can see why you'd hate to say it.
> i'd certainly practice denial too if i
> were to be fucking my mom on the
> floor. don't you think she at least
> deserves a bed for your shenanigans?
1) Brits seldom use the term 'Motherfucking' - it is an Americanism/idiom.
2) The poster 'Lee Frank' types every word in lower case letters: something
I never do.
But of course, being the idiot you are, you wouldn't recognise the glaring
disparities between my posts and his.
Obviously, 'Lee Frank' is a old regular here ( he checked out about two
years ago) and is unable to resist a return to Hotel California.
gARY
> gARY-the-gRASS wrote..
>
> > So, an old hat here, to remember JG.
>
> > Any idea's peoples?
>
> > I guess, Nick UK or Colin.
>
> > As of course I would,
> > a
>
>
> Being the idiot you are - of course you would! And as per usual, you are
> wrong!
I have warned you before, attitude here! Yet again.
> I didn't have a clue what ROTMFFLMMFAO meant until the poster 'Lee Frank'
> said..
>
> > rolling on the motherfucking floor
> > laughing your motherfucking ass
> > off?
>
> > i can see why you'd hate to say it.
> > i'd certainly practice denial too if i
> > were to be fucking my mom on the
> > floor. don't you think she at least
> > deserves a bed for your shenanigans?
>
> 1) Brits seldom use the term 'Motherfucking' - it is an Americanism/idiom.
Frank never wrote ROTMFFLMMFAO, did he!
> 2) The poster 'Lee Frank' types every word in lower case letters: something
> I never do.
I never write that way also, but as an alias I would (and have done)!
> But of course, being the idiot you are, you wouldn't recognise the glaring
> disparities between my posts and his.
See 2 above ^
> Obviously, 'Lee Frank' is a old regular here ( he checked out about two
> years ago) and is unable to resist a return to Hotel California.
We are yet to find out.
a