Jump to content

Featured Comment

Posted

Well, besides stake and provablyfair.me my bets results don't match. And there is a lot of rumours that the 3rd part provablyfair.me it's own by stake anyway. So, why other sites cant return the same results? And if there is one can someone provide? 
 

Posted (edited)

Well, If I’m completely rinsed in an unbelievable way. I want to check the fairness, but if I check it on Stake, what’s the point?
In order to have peace of mind I need to verify these results on my side or a 3rd party website that is not rumoured to be owned by stake
Am I wrong?
 

Independent reproduction is impossible. Not even on a simple game as dice
Lack of reproducibility means trust is still required
Defeats transparency 

Edited by CulitoFuerte
Posted

So, the conclusion here is what? in the end we cannot verify the fairness on other sites besides stake or site related to stake?
Is it just me or what, I mean what am I missing here? Why should I trust the provably fair if I cannot check the results
 

Posted (edited)

sure  stats are abit too green coz i uploaded a really good run multiple times before i knew how the homepage worked, i put little comments in the first pic. Originals at 99.5% at the time i took these, a month back or so.
image.thumb.png.c59faf11a5979e0d89785b95b3760f68.pngimage.thumb.png.b77c8509224e3dc356b47bd6d459386b.png
image.thumb.png.c7b856f2d98906cb1eba0319216c3303.png
image.thumb.png.49c3ba011d5016fa01610fc0f862b041.png

Edited by imalittlepony
Posted
5 hours ago, imalittlepony said:

sure  stats are abit too green coz i uploaded a really good run multiple times before i knew how the homepage worked, i put little comments in the first pic. Originals at 99.5% at the time i took these, a month back or so.
image.thumb.png.c59faf11a5979e0d89785b95b3760f68.pngimage.thumb.png.b77c8509224e3dc356b47bd6d459386b.png
image.thumb.png.c7b856f2d98906cb1eba0319216c3303.png
image.thumb.png.49c3ba011d5016fa01610fc0f862b041.png

Trying to figure out what this is. But Is safe to guess it's not what i m looking for

I'm looking for a third party that has Stake’s algorithm to verify the results after the server seed is revealed. So far, none of the sites I’ve checked can reproduce the results using the server seed, client seed, and nonce.

I only get matching results on Stake itself or on provablyfair.me (which is most likely run by Stake, because if the algorithm cannot be made public, why would provablyfair.me have it?)

Because at least for me makes 0 sense to check the results on their side, it's pointless. It's like asking a scammer if they scammed you

So I don’t understand what is fair or transparent about “provably fair.” How can others trust it?

  • Moderator
Posted
41 minutes ago, CulitoFuerte said:

So not as legit as we thought? Anyone?
I feel scammed. The losses are too aggressive and too well timed with the bet size to be just random. Can we verify the results independently or not?

You can even check it doing the math by hand, not using any external site or app, but it will take several hours to verify a single bet ......

You can find in the web several external sites that provide the basic functions for doing the verification, because the random number generator is based in universal cryptographic functions that were not developed by Stake at all.

Dice bets are the easiest to verify. Just search sites that offer the functions HMAC_SHA256 and SHA256.

Most ppl dont get the right results because usually they try to verify using the hashed server seed instead of using the unhashed server seed.

 

 

Posted
7 minutes ago, maverick528 said:

You can even check it doing the math by hand, not using any external site or app, but it will take several hours to verify a single bet ......

You can find in the web several external sites that provide the basic functions for doing the verification, because the random number generator is based in universal cryptographic functions that were not developed by Stake at all.

Dice bets are the easiest to verify. Just search sites that offer the functions HMAC_SHA256 and SHA256.

Most ppl dont get the right results because usually they try to verify using the hashed server seed instead of using the unhashed server seed.

 

 

Exactly what I said in my previous comment. I understand that a game like BJ is more complex, but at least for a simple game like dice I should be able to reproduce the same results. So far, I could not find a single one. Can you provide a example?

Regarding this part:

“because the random number generator is based on universal cryptographic functions that were not developed by Stake at all”

That is correct, but you are missing something important. There is still an algorithm developed by Stake that defines how the output of those cryptographic functions is interpreted into a game result

 

  • Moderator
Posted

So here I will show you an example for all the non-believers.

As I already said, the random number generator in Stake uses functions that were developed for cryptography, those functions were not developed by Stake, and if you are curious you have plenty of information about them on a lot of websites.

I will verify my bet casino:430610983161, its a DICE bet.

image.png.b28acd4a4292294e2105541920df02a9.png

The result is the number 0.06

If we look at the provably fair section of the bet, we get the following numbers =

image.png.aabe48599813cd60b81d831fcd94ffb7.png

The 1st thing to verify is if Stake was honest when it only showed me the hashed server seed. Now that I have the unhashed server seed, I must check that if I actually hash the original server seed, I get the previously shown hashed server seed.

I will use for example the website https://tools.keycdn.com/sha256-online-generator (It is not affiliated with Stake in any way. And you can also use other cryptographic websites).

If you input the unhashed server seed ba3dcff9b17972d2884a7bc8ba8f66aea83ae3c1b5e75d12b5ed0f7714c0b123 , you get as output the correct hashed output c444076cbaf0a9b908cf8c47e011e59ef9f6d22ef6d96d5dfa668d783c5b1de1 .

I must say here a very fine detail only for the ppl that have technical knowledge. The hash function SHA256 theoretically allows that 2 different strings output the same hashed value, this is called a collision of the hash function. So if the hash function wasnt completely safe, Stake could have 2 different unhashed server seeds and cheat on you. BUT if you search for "SHA256 collision" you are going to find that NOBODY in the world so far has been able to generate even 1 collision, even using the most powerful supercomputers of the world. So if NOBODY could generate just 1 collision in the world, its clearly impossible for Stake to generate collisions in all the thousands of seeds it generates every day for all its users. You are safe from this possibility.

For the actual verification of the bet you need a website that calculates the HMAC algorithm.

I will use the website https://www.freeformatter.com/hmac-generator.html , but of course you can use other cryptographic site with this same tool.

You must select Digest algorithm = SHA256

In the place where it says "copy paste the string here" you must paste your client seed plus the character : plus the nonce plus another character : plus the number zero. In my particular example I must paste 3:101051:0

In the field that says "Secret key" I must paste the unhashed server seed, in my example it is ba3dcff9b17972d2884a7bc8ba8f66aea83ae3c1b5e75d12b5ed0f7714c0b123

The result I get is this=

image.thumb.png.73fd85727173c22facee26b4eed5cec8.png

Stake only uses the first 4 hexadecimal numbers from this string (so only look at the first 8 characters and forget the rest).

Those 4 hexadecimal numbers are converted to decimal system, so 00=00 , 2b=43, d7=215 and 53=83 . I do this conversions mentally since Ive been doing it for years, but you can use a calculator or a conversion table or a website, for example  https://www.binaryhexconverter.com/hex-to-decimal-converter 

Now you must divide the first number by 256 , the 2nd number by (256*256), the 3rd number by (256*256*256) and the fourth number by (256*256*256*256). The results you will get are 0.0 , 0.000656127930, 0.000012814999 and 0.000000019325 (you must use 12 significant digits at the right of the comma).

Now you add the four numbers and the result you get is 0.000668962253

Then you multiply it by the number 10001, and you get 6.690291494830

Now you must divide it by 100 and you get 0.06690291494830

You must keep only 2 digits to the right of the comma and discard the rest so the final result is 0.06

As you can see, the result of the verification is 0.06 and the original result of the bet is also 0.06 so the verification is OK, and I didnt use the verification tool of Stake or any website affiliated with Stake.

If you are still paranoid about using external websites or tools , you can do the whole verification using pencil and paper and simple maths , but its a veeeeeeeeryyyyyyyy long and tedious process, in this youtube vid you can see how to do the SHA256 by hand, and the HMAC consist of doing TWO SHA256 hashes plus some extra things......

https://www.youtube.com/watch?v=orIgy2MjqrA

 

  • Moderator
Posted

There are several other games that are almost the same as Dice in the verification maths.

For example Roulette is exactly the same as Dice until you get to the multiplication by 10001, this is changed to multiplication by 37 (in Dice there are 10001 possible outcomes and in roulette 37 different outcomes), and there is no division, you just use the result of the multiplication without the fractional part.

In other games its a bit more complex ......

 

 

Posted
On 12/15/2025 at 6:09 AM, CulitoFuerte said:

Trying to figure out what this is. But Is safe to guess it's not what i m looking for

I'm looking for a third party that has Stake’s algorithm to verify the results after the server seed is revealed. So far, none of the sites I’ve checked can reproduce the results using the server seed, client seed, and nonce.

I only get matching results on Stake itself or on provablyfair.me (which is most likely run by Stake, because if the algorithm cannot be made public, why would provablyfair.me have it?)

Because at least for me makes 0 sense to check the results on their side, it's pointless. It's like asking a scammer if they scammed you

So I don’t understand what is fair or transparent about “provably fair.” How can others trust it?

U can trust it with massive amount of bets + analyzer, here is more clear cut for u 
image.thumb.png.c6cdcd94e9387e98f140741f342eb2be.png

  • 1 month later...
Posted
On 12/17/2025 at 2:42 PM, maverick528 said:

So here I will show you an example for all the non-believers.

For the actual verification of the bet you need a website that calculates the HMAC algorithm.

I will use the website https://www.freeformatter.com/hmac-generator.html , but of course you can use other cryptographic site with this same tool.

You must select Digest algorithm = SHA256

In the place where it says "copy paste the string here" you must paste your client seed plus the character : plus the nonce plus another character : plus the number zero. In my particular example I must paste 3:101051:0

In the field that says "Secret key" I must paste the unhashed server seed, in my example it is ba3dcff9b17972d2884a7bc8ba8f66aea83ae3c1b5e75d12b5ed0f7714c0b123

The result I get is this=

image.thumb.png.73fd85727173c22facee26b4eed5cec8.png

Stake only uses the first 4 hexadecimal numbers from this string (so only look at the first 8 characters and forget the rest).

Those 4 hexadecimal numbers are converted to decimal system, so 00=00 , 2b=43, d7=215 and 53=83 . I do this conversions mentally since Ive been doing it for years, but you can use a calculator or a conversion table or a website, for example  https://www.binaryhexconverter.com/hex-to-decimal-converter 

Now you must divide the first number by 256 , the 2nd number by (256*256), the 3rd number by (256*256*256) and the fourth number by (256*256*256*256). The results you will get are 0.0 , 0.000656127930, 0.000012814999 and 0.000000019325 (you must use 12 significant digits at the right of the comma).

Now you add the four numbers and the result you get is 0.000668962253

Then you multiply it by the number 10001, and you get 6.690291494830

Now you must divide it by 100 and you get 0.06690291494830

You must keep only 2 digits to the right of the comma and discard the rest so the final result is 0.06

As you can see, the result of the verification is 0.06 and the original result of the bet is also 0.06 so the verification is OK, and I didnt use the verification tool of Stake or any website affiliated with Stake.

 

I agree that SHA256 collision resistance stops one specific kind of cheating: swapping the server seed after its hash has been published. That’s correct.

But that’s not why there is a lack of trust.

All your maths and verification talk about a revealed unhashed server seed. Your explanation shows that once the unhashed server seed is revealed, past results can be reproduced and verified. That proves the results are consistent and cannot be changed after the fact, but it doesn’t show when the outcomes were actually generated.

It doesn’t prove that outcomes were precomputed before betting started. It doesn’t prove nonces 1 through ∞ existed ahead of time. It doesn’t prove results weren’t generated on demand, one bet at a time. It doesn’t prove the server seed wasn’t used only when bets actually happened. And it doesn’t prove that the infinite future sequence existed before the seed was revealed.

There’s no way for this verification to tell the difference between:

a system where all outcomes were fixed in advance (all results generated when the seed was created, before the hashed seed is revealed), and

a system where each result is calculated live per bet and infinite future results are generated once the seed is revealed.

Both systems produce the exact same verification output.

So again, SHA256 stops seed swapping. It doesn’t stop real-time outcome generation. Provably fair shows results weren’t changed after the fact, but it doesn’t prove future results existed ahead of time or that outcomes couldn’t be influenced at the moment each bet was placed.

While the math confirms consistency after the reveal, it doesn’t remove the need to trust the system while you’re placing each bet. The verification only shows results weren’t changed afterward; it doesn’t prove they were set in advance.

 

Posted

I don't think so, 

Even if you verify seed. In the in casino's you'll always loose because of human psychology to win more. 

 

That's why they say that house always win. 

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

Privacy Policy Terms of Use