#Weird gamma / poly-gamma equation.
1 messages ยท Page 3 of 1
We will get cooked
Product of gammas in denominator ๐ญ
See
We are getting cooked by products
We don't know product of gammas
wait
1/gamma(x) has a product representation
i remebered
i forgor that weistras existed
Bro zeta and Bernoulli have ;-;
Aah
Bro but this for every complex number right?
Bro but we are cooked
Bro
How are we even gonna multiply
yeah
We only have two options
Bernoulli on me
And gamma Bose on u
Eta definition?
If we use eta definition?
Do u know product of etas?
eta dirichlet?
It's helpful
If we can derive products of
Etas
Ezily
Can we?
Like eta 2 eta4 .... Eta(2n)
Fr
i mean they have a strong relation
Yes
Like
zeta=eta/(1-2^(1-s(
Something
Like that
Imma check fast
Yea it's correct
Now
How do find product of etas?
Any definition
Or basic series
@hallow geyser
Bro but see
We had used zeta in terms of bernoulli
I don't much definitions for eta
Aah
Do u know
If yes pls tel
L
We begin
@hallow geyser
thats the problem
there shoul be a parenthesis before the Zetsa
We need zeta
yeah
Any other definition
Wait imma check quick
Wait
@hallow geyser
U good at integrals
yeah
We have eta(s)=1/gamma(s) * integral from 0 to infinity x^(s-1)/e^x+1
yse
i will leave you some home work
before you leavke this is the exercisse i want to leave to you
prove this
k
For real
but with integrals
k
Okok
Okok
bye
i dont even need limits
Ok
;-;
can u explain it for a bit i havent read the entire conversation
It involves 3 core function
i wanna see whether i can get ahything from it
Leave it
ok go on
u aint getting any progress if you aint tellin
See this
?
I tried the basic method
Not the op method
Or not my method
The basic method is
P=product of zetas
Take log both sides
logP=sum of log zetas
raise to e power
P=e^lim n tends to infinity sum of log zetas
See this
,w summation from r=1 to infinity ln[zeta(2r)]
So product is simply P=e^(0.599395)
;-;
I cannot find it manually
I failed manually
,w e^(0.599395)
yes?
wha
Did u understand
yes cuz it approaches 1 very fast
Like see only 6 zetas and it is 1.8208
because the bernoulli definition of the zeta function has a factorial in the denominator
Yes
hence of course the denom grows very fast
;-;
But leave the Bernoulli definition
No
What
trust
See what I send you
Is this
1.821...
It is derived from the Bernoulli definition
What I sent u above
But it involves b(n^2+n) and the Barnes G function
well what is it represented in terms of bernoulli definition
๐
?
can u give me one
well given the fact that the sum of all even bernoulli numbers is zeta(2)-1
you can represent it better
hi
read on ur own
takes a while
good luck ๐ธ
!filetype
Please post images (such as PNGs or JPGs) of the question rather than other filetypes such as PDFs which have to be downloaded. Non-image downloads can potentially contain viruses or other security risks.
trying to combine the polygamma definition of the bernoulli numbers we just found into the fact that the sum of bernoulli numbers is zeta(2)-1
now what the fuck is that
lllooooooooool
No it does it
k
Idk why it's not doing
Like see
,w zeta(4)
Sum of all zetas -1 is simply 1-eta(1) lol
Sum of all etas-1 is simply -eta(1)
Do u understand
are you using the dirichlet eta
but thats regularization only
dont think it applies to positive even zetas
No
Sum of all [zetas-1] is 1-eta(1)
And sum of all [etas-1] is simply -eta(1)
Lol
Let me try sum of all gamm-1
Lol
Yo
Wait let me try
,w Sum[ipi/2+ln(digamma(2n-1,1)4n/(2pi)^2n),{n,1,infinity}]
Wolfram Alpha doesn't understand your query!
Perhaps try rephrasing your question?
Click here to refine your query online
this thing i shella dumb
,w summation from r=1 to inf. ln(digamma(2r-1,1)4r/(2ฯ)^(2n)])
,w Sum[i*Divide[pi,2]+ln(40)digamma(40)2n-1(44)1(41)*4Divide[n,Power[(40)2pi(41),2n]](41),{n,1,infinity}]
bruh
Wolfram Alpha doesn't understand your query!
Perhaps try rephrasing your question?
Click here to refine your query online
its sitalinning my text
,w Limit[ln(digamma(2n-1,1)/(2*pi)^2n),n->infinity]
do what
,w summation from r=1 to infinity ln[zeta(2r)]
bro thats not it trust
yes im tryna do the math just leave me alone pls
Okok
IT DOES NOT DIVERGE
U sure?
Ik
DID YOU READ THE FLIPPING LINK I SENT U
just use norlund sum
I am saying why u playing with Bernoulli
i am 1000000000% sure it does not diverge using norlund sum
i thought i explained it pretty clearly
do you not listen
.
,ln(0.1)
wtf does that fucking do
,w ln(0.001)
Understand
listen bro that doesnt do shit
We have -ves bro
See
But it is not saying bout sum of log(B2n)
logically
Understand
if the sum of B2n is convergent
And stop shouting
so is the sum of log B2n
no
Bernoulli numbers are very less
if you read the link
Like 1/30
,w ln(1/30)
I read the link till the point he reached the sum
then why should i get sum log b2n right
bro has no bloody clue
THATS MY POINT
yes he is
No
its in the bracket
Id think
why would it be otherwise
See
Bernoulli numbers are no functions
We represent their number is sub script
In*
Now idk
.
Wolfram Alpha doesn't understand your query!
Perhaps try rephrasing your question?
Click here to refine your query online
Lol
i give up
@hallow geyser would really appreciate it if you could read our convo and provide us with some info
tysmmmm
See if
U differentiate the function x/(e^x-1) n times and put lim x tends to 0
U get the nth Bernoulli
Number
tf
ok
That is lim x tends to 0 nth derivative fo x/(e^x-1)
Is nth Bernoulli number
U can find 2,3
Using this
why dont u demonstrate
Pro tip
As sum of all bernoulli is
Zeta(2)-1
Sum of lim x tends to 0 all derivatives of x/(e^x-1) is also zeta(2)-1
why
can u write it down
Later
cant identify your jumbled up mess
Going
hello
seem you reached a dead end
@halcyon hearth you might wanna see this
it seem that there is no discovered closed form for the product
By elementary estimates, the constant lies in the open interval (Pi/6, exp(3/4)). - Bernd C. Kellner, May 18 2024
my take on this is that one may get a closed form of the product using inequalities
Decimal expansion of zeta(2)zeta(4)...zeta(2k)...
If u(k) denotes the number of Abelian groups with group order k , then Product_{k>=1} zeta(2*k) = Sum_{k>=1} u(k)/k^2.
This constant C is connected with the product of values of the Dedekind eta function on the upper imaginary axis. The product runs over the primes, where i is the imaginary unit: 1/C = Product_{prime p} (p^(1/12) * eta(i * log(p) / Pi)). - Bernd C. Kellner, May 18 2024
this is the best i found
and i sugest that if you are not interested in spending too muchh time in this to drop it and move on
@left sigil
i will be going now sincei have work to get done
bye guys
bye
Bye
Bruh
bruh
have you done the proof?
do it
very good excerssice
wut
Now can we discuss
wut
Can u tell me zeta definition
Which has operation range for every rational number
1
les than 1 of bigger?
Bigger
Is this valid for every rational number ?
I don't think this is general definition right?
Sure?
yeah
Ok
,w summation from r=2 to 5 1/[(r)^(phi)-1]
@halcyon hearth what are u dropping
yes
you dont need me anyways
im only 15
ill learn more about different stuff then ill take this on (and probably fail
)
Everyone fails
But those who accept that they failed
Become better
@halcyon hearth
wise aproach, although don't get ahead of yourself to early
@halcyon hearth it's better to have a more positive mindset
try to focus on what you can learn and how you can grow from the challenges you propose to yourself.
Every step you take provides you with new perspectives and knowledge.
and, don't put yourself down because of your age.
tbf my dream career isnt even maths
not my best thing
its sorta just a side hobby i do
Okok
Bro
Chill
U are just 15
U have years to find your dream career
We just discuss here casually
@halcyon hearth
@hallow geyser
sup
Yo
Wanna do something
Let me make myself clear
Here for a personal project imma here
U tell the topic
k
We ball
hmmm
look
Yea
this might be a very interesting exercice
Ok
so i think for a series representation here
for all non negatie integers]
Ok
So leave I cannot get it to gamma
See I was thinking to use the Bose integral definition

hmmm
No see
I got
(ฯ^3)/3-integral from 0 to 2ฯ zeta'(1+e^it).e^it.t.dt
Is the eta zeta definition for complex ?
Numbers?
yes
I think its for s>0 right
It's for complex numbers
Lol
Okok
Wait idea
Do u know the zeta functional equation
Zeta(s)=2^(ฯ)^s-1 sin(ns/2) zeta(1-s)gamma(1-s)
Ya
Ook
Now what if we use
s=-e^(itheta)
yeah
I mean we can get zeta(1+e^it) in terms of zeta(-e^it)
i think we can't
no
If u apply it it becomes zeta(1+e^-it)
kings rule cz I am in india currently
It's said like it
let me xplain
f(b+a-x)=f(x) rule
Do u know?
no
nah
Ok
you know that that e^ix its the complex that lie down that circumference?
Yes
Ya ok
What
then you know complex analysis theorem?
crap g
don't u think there is another way too
k nows its good
Yea
be prapared\
gamma was a closed smooth curve in the complex plane
Ok
then we can aply the following theorem
Ok
Yes ok i see
sowe gota find the res of the zeta(1+z)/z
Fr
i see
aaaaaaa this thing
And now
We apply this propert
So our integral remains i
So we get
2I=int from 0 to 2ฯ zeta(e^it)+zeta(e^-it)
Fr
true
Now do u know
cosx=(e^ix+e^-ix)/2
We reaching the point with this one fr ๐ฅ๐ฅ๐ฃ๏ธ
yeah
Now imma just thinking this only
Should I use this
Or I should simply express zeta(1+e^it) and zeta(1+e^-it)
As seriss
Infinite series*
wait
no i was going to see if the series expansion was valid to all negative integres
Fr
We gonna ball with this one
Now I use the general definition
For real
See
General definition
Valid for
Real>1
We have
Re>1
For all values in the interval
No wait
I feel imma wrong
I feel need to break
so you cant use it since in 3pi/4 it goes to negative 1
No prob
We gonna break
At each ฯ/2 intevral
Let me compute the sum first
Na this not working bro
It getting complicated for intervals
Aah
@hallow geyser proceed with the complex analysis
we just gota fin the Residues of the function
,w residue zeta(1+z)/z
Lol
Fire AF
We have residue as gamma
yes i think
Lol
ah but then
For real
so its 2pigamma
Lol goated fr
Ok
fr ๐ฃ๏ธ ๐ฅ
Lol bro a exotic sum
your turn now
k
Your turn
hmmmmmmmm
Ok u type imma get water
k
there are two thing i wanted to discuss
Ok
one related with actuall math
Ok
and the other one is more filosofical related
i only go time like for halve of any of the topics :/
What?
its getting mad late in my country
Ok
you remember goodel shit?
so there are matha statements that can't be proven true or false
Yes fr
I feel cz they are both true and false w.r.t to the observation or reference
Like 1+1/2+1/3+1/4+.... Seems finite from a pre-uni one's prospect
But is infinite from uni one's prospect
fr
then, imagine that there are probles that can't be proven using math
Yea ok
and there is no way of prooving that are unsolvable
Fr
is;t like a shiti carboard game
but noo one
fr
So it's neither false nor true simultaneously
bro
Once a philosopher said
hmm?
facts
imagine saying you proved a problem using numerical proof for some fintie number a
Yes ok
For real
that one can say 2.0000000000000000000000000000000000
Imma say it again
For real
Let me give an example
Assume our computational limit is
10^8
k
,w zeta(242)
Now
One with the limit of 10^8 will says it's 1
At that time
But as computational limit increases one says it's not 1
It's irrational
true
and imagine that the other way around
Fr
lets say 1.123425675734215658696875 its irrational
For real
but then you compute a little and see that its just
We can say its irrational cz we don't have the way to express in terms of ratio
Who knows when we advance
1.123425675734215658696875777000000000000000000000
Yes for real
like its wierd to be studying this and then realizing that
there is no way of proving
So even if their digits add and become something like
yeah
true
facts
The only reason why we cannot prove Riemann hypothesis numerically
Cz u cannot compute infinity numerically
the only reliable proofs are that of logic ones, that fall from theorems
true
shiti engenier axioms would say that pi=e=3
then pi+e=6
For real
Lmao ๐คฃ
I once heard
That
hmm?
From someone
That we are a simple civilization
Because of how simple we take things
huh
He wrote that
The small assumptions and neglects we make
If we start considering it
We may change
Like see
He said
That
If for once
researchers
Lol
Don't focus on problem solving
But focus how accurately they are solving it
We can be better
bro had beef withy phisicist!!!
He said the small assumptions we make
Make a big difference in reality
He said if we can narrow down the difference
man
We can predict nature
that guy