What can be done to reverse the decline?

[iAmCaffeine]

I am seriously considering leaving Conquer Club if something doesn't change with the dice.

[jmyork82]

can somebody point me to a link where they decided to change the dice away from the random.org dice?

[iAmCaffeine]

can somebody point me to a link where they decided to change the dice away from the random.org dice?

Good luck; it was around lackattack's time I think. Whilst they remain on a limited pool of something like 50,000 rolls, there won't be any realistic results. People have left due to dice many times over and many more will do as well in the future.

[degaston]

can somebody point me to a link where they decided to change the dice away from the random.org dice?

Good luck; it was around lackattack's time I think. Whilst they remain on a limited pool of something like 50,000 rolls, there won't be any realistic results. People have left due to dice many times over and many more will do as well in the future.

This is where I first heard about the possibility:
http://www.conquerclub.com/forum/viewtopic.php?f=6&t=194323&start=75#p4333844

It was eventually confirmed to me that the site was never replacing the dice data file. I never heard any explanation as to why they did this. Maybe they didn't want to pay for the data, or maybe something broke the code and they never bothered to fix it.

[iAmCaffeine]

can somebody point me to a link where they decided to change the dice away from the random.org dice?

Good luck; it was around lackattack's time I think. Whilst they remain on a limited pool of something like 50,000 rolls, there won't be any realistic results. People have left due to dice many times over and many more will do as well in the future.

This is where I first heard about the possibility:
http://www.conquerclub.com/forum/viewtopic.php?f=6&t=194323&start=75#p4333844

It was eventually confirmed to me that the site was never replacing the dice data file. I never heard any explanation as to why they did this. Maybe they didn't want to pay for the data, or maybe something broke the code and they never bothered to fix it.

Thanks for that. I expect there were monetary issues as well as "performance", but the site wouldn't want to announce that. Do note Mets was a moderator at the time of posting that.

[jmyork82]

So let me get this right, they changed the dice from Random.org to a set list of dice options that are chosen by random.org in order to balance the outcomes of 1's and 6's?

I'm just trying to summarize and make sure I'm on the same page as everybody.

[degaston]

can somebody point me to a link where they decided to change the dice away from the random.org dice?

Good luck; it was around lackattack's time I think. Whilst they remain on a limited pool of something like 50,000 rolls, there won't be any realistic results. People have left due to dice many times over and many more will do as well in the future.

This is where I first heard about the possibility:
http://www.conquerclub.com/forum/viewtopic.php?f=6&t=194323&start=75#p4333844

It was eventually confirmed to me that the site was never replacing the dice data file. I never heard any explanation as to why they did this. Maybe they didn't want to pay for the data, or maybe something broke the code and they never bothered to fix it.

Thanks for that. I expect there were monetary issues as well as "performance", but the site wouldn't want to announce that. Do note Mets was a moderator at the time of posting that.

They could get 10,000 dice rolls per day for free. These could be rotated into the data file, or combined with other data (previous data files, or pseudo-random data) to get as many truly random rolls as they want.

[degaston]

So let me get this right, they changed the dice from Random.org to a set list of dice options that are chosen by random.org in order to balance the outcomes of 1's and 6's?

I'm just trying to summarize and make sure I'm on the same page as everybody.

My understanding is:

• When they were using random.org, they would get a new set of 50,000 rolls each day, and re-use that as many times as was needed during that day. (I think they used to use about 1 million rolls per day)
• At some point, they stopped getting new data, and went with a single 50,000 roll data file that was never replaced.
• For truly random data, you would expect each number to show up an equal number of times in the long run, but this file had an unequal distribution of numbers, (fewer 1's, more 2's and 4's) which eventually showed up in everyone's dice stats. - leading to the thread that I linked.
• Rather than fixing the source of the problem, they modified the data file to equalize the number of times each digit occurs. It's a much worse solution than even using a standard pseudo-random number generator.

[Dukasaur]

can somebody point me to a link where they decided to change the dice away from the random.org dice?

Good luck; it was around lackattack's time I think. Whilst they remain on a limited pool of something like 50,000 rolls, there won't be any realistic results. People have left due to dice many times over and many more will do as well in the future.

This is where I first heard about the possibility:
http://www.conquerclub.com/forum/viewtopic.php?f=6&t=194323&start=75#p4333844

It was eventually confirmed to me that the site was never replacing the dice data file. I never heard any explanation as to why they did this. Maybe they didn't want to pay for the data, or maybe something broke the code and they never bothered to fix it.

Thanks for that. I expect there were monetary issues as well as "performance", but the site wouldn't want to announce that. Do note Mets was a moderator at the time of posting that.

When Mets posted that, it was long after the fact. The original announcement was made by lackattack on June 21st, 2010. Five-and-a-half years ago.

see: http://www.conquerclub.com/forum/viewtopic.php?f=634&t=120373#p2649912

The site at that time was using more than a million rolls a day. The cost to keep refreshing them at random.org was becoming expensive, but even more importantly, the communication with random.org was causing significant server lag, so lackattack switched to the famous 50K file. It was never a secret. The announcement was posted in public and always has been public knowledge.

However, I don't believe the site is using the 50K file any more. I believe that's old news.

[Dukasaur]

[*]Rather than fixing the source of the problem, they modified the data file to equalize the number of times each digit occurs. It's a much worse solution than even using a standard pseudo-random number generator.[/list]

I don't know why you assume that, instead of some other solution.

[iAmCaffeine]

However, I don't believe the site is using the 50K file any more. I believe that's old news.

I have not witnessed, nor been informed of, anything that would lead me to believe they've improved.

[Dukasaur]

However, I don't believe the site is using the 50K file any more. I believe that's old news.

I have not witnessed, nor been informed of, anything that would lead me to believe they've improved.

http://www.conquerclub.com/forum/viewtopic.php?f=6&t=194323&start=25#p4254756

The original data that degaston collected in August of 2013 no longer holds true.

At that time, almost everyone on the site had average dice stats of 3.51 instead of the expected value of 3.50. That was one of the key pieces of evidence that degaston used to establish that there was a shortage of 1s in the 50K file. http://www.conquerclub.com/forum/viewtopic.php?f=6&t=194323&start=25#p4254756

Later that same month, bigWham bought the site, and in January of 2014 made major structural changes, including new servers, a new method for calculating the scoreboard, etc., etc. Basically a major overhaul of the game engine.

Since then, the number of people with 3.51s has dwindled steadily. Most people who have joined since January of 2014 have 3.50 (if they have a large-enough number of games to be statistically useful).

It's unfortunate that bigWham is so secretive and won't make any announcements, but you can follow the timeline well enough.

• June 2010 lack switches to the 50K file
• August 2013 degaston crunches the numbers and proves the 50K file has a shortage of 1s. Note that it took 3 years of the 50K file to get to this point.
• January 2014 bigWham overhauls the site, and people who join after January of 2014 don't have a shortage of 1s in their dice stats, and even among people who joined before 2014 the sign of the bias starts to dwindle.

It's pretty clear to me that whatever else was done in the Jan. 2014 overhaul, it included an overhaul of the dice. All the data that degaston accumulated was before that date. Since then, there is less and less evidence of bias.

[degaston]

[*]Rather than fixing the source of the problem, they modified the data file to equalize the number of times each digit occurs. It's a much worse solution than even using a standard pseudo-random number generator.[/list]

I don't know why you assume that, instead of some other solution.

I think that's what I was told in a PM that I received after I showed that the dice file was not being replaced in this thread. It took a huge effort before I was believed. The PM is gone now, so I can't say for sure what it said, but I've posted this kind of comment several times, and no one has ever disputed it before. The last thing I see from lack in that thread you referenced is:

I have changed the way the dice work once again, it should now be theoretically impossible to cheat:

This is how the intensity cubes now work:
We have a series of 50,000 true random numbers from random.org
Each time the game engine processes an assault or auto-assault, it select a random spot in the series to read from using a pseudo-random computer function
Each time the game engine generates a random intensity cube, the next number is read in sequence from the series (e.g. in a 3v1 attack 4 numbers are read sequentially)
The series of 50,000 true random numbers from random.org is replaced every hour

So are you able to say how the dice are generated now?

[Dukasaur]

[*]Rather than fixing the source of the problem, they modified the data file to equalize the number of times each digit occurs. It's a much worse solution than even using a standard pseudo-random number generator.[/list]

I don't know why you assume that, instead of some other solution.

I think that's what I was told in a PM that I received after I showed that the dice file was not being replaced in this thread. It took a huge effort before I was believed. The PM is gone now, so I can't say for sure what it said, but I've posted this kind of comment several times, and no one has ever disputed it before. The last thing I see from lack in that thread you referenced is:

I have changed the way the dice work once again, it should now be theoretically impossible to cheat:

This is how the intensity cubes now work:
We have a series of 50,000 true random numbers from random.org
Each time the game engine processes an assault or auto-assault, it select a random spot in the series to read from using a pseudo-random computer function
Each time the game engine generates a random intensity cube, the next number is read in sequence from the series (e.g. in a 3v1 attack 4 numbers are read sequentially)
The series of 50,000 true random numbers from random.org is replaced every hour

So are you able to say how the dice are generated now?

No, I have no idea. But I do know that the data you collected in 2013 does not resemble stats I've looked at more recently. With full respect to the work you did and the rigour of your methods, I believe you were right in 2013 but something changed in January 2014.

I can't say what they do nowadays, but I'm pretty confident it's no longer the same thing. And if they went and changed it, I don't know why you would assume that they changed it in a bad way instead of changing it in a good way.

Granted, you may have received a PM about it, but without seeing it I can't comment on whether you might be misinterpreting what it said.

[degaston]

No, I have no idea. But I do know that the data you collected in 2013 does not resemble stats I've looked at more recently. With full respect to the work you did and the rigour of your methods, I believe you were right in 2013 but something changed in January 2014.

I can't say what they do nowadays, but I'm pretty confident it's no longer the same thing. And if they went and changed it, I don't know why you would assume that they changed it in a bad way instead of changing it in a good way.

Granted, you may have received a PM about it, but without seeing it I can't comment on whether you might be misinterpreting what it said.

I understand that something changed, but if you don't know what it was, how can you be confident that it was not just the data file?

Even after I had provided plenty of evidence that the dice were biased, the position of the team member who was following the thread was:

You can trust that I have taken notice, and I hope it's evident that I'm not ignoring your results. But this result is still not enough to prove that the dice are non-uniform when considered sitewide, since there could be some hidden selection bias in your sample (which is small compared to the site as a whole). More importantly, since this effect is relatively small, if you want this to be a site priority then you need to collect the data suggesting that the dice are not fair.

So at that time, it seemed like they were not aware that the dice file was not being updated, and even if it wasn't, as long as it affected everyone equally, it wasn't going to be a high priority.

I understand that there could be financial and response-time related issues that would make going back to frequent Random.org updates impractical. I outlined a few methods that would allow the dice to be generated quickly and for free, but never received any response to those. In any case, changing the dice algorithm would probably be a moderately complicated project that would have needed some testing before it was released, but I see no mention of it on the Beta Tester thread.

Also, fixing something that many people had been complaining about for years would be kind of big news. If they had gone back to truly random dice, I would think that they would have wanted people to know about it, and would have announced or posted it somewhere. And finally, human nature being what it is, the easiest fix to the problem that had been exposed would have been to simply replace the data file with one that had been artificially generated to have each number appear the same number of times.

If you tell me that they've completely changed the generation method, then I'll accept it, but it doesn't seem like you really know.

[Beast Of Burson]

I am seriously considering leaving Conquer Club if something doesn't change with the dice.

Re: dice sucks
Postby iAmCaffeine on Fri Jul 10, 2015 3:57 am

In real life you're playing one game at a time with the same set of dice with what, 2-5 people? On here there are thousands of active games at the same time using dice from a limited pool, there are going to be all kinds of ridiculous results that we don't expect. A couple of weeks ago I lost eight large escalating games where I had over 75% chance to take the sweep each time and it was shit, but it's not a bug. Furthermore, it's a human's nature to remember the bad times more than the good, and we also see more bad results than good because we will obviously attack 18v2 (which I've lost before) but we're not gonna try 5v20.

Really? You post this when somebody else cries about the dice?

Can you say hypocrite?

[nippersean]

[*]Rather than fixing the source of the problem, they modified the data file to equalize the number of times each digit occurs. It's a much worse solution than even using a standard pseudo-random number generator.[/list]

I don't know why you assume that, instead of some other solution.

I think that's what I was told in a PM that I received after I showed that the dice file was not being replaced in this thread. It took a huge effort before I was believed. The PM is gone now, so I can't say for sure what it said, but I've posted this kind of comment several times, and no one has ever disputed it before. The last thing I see from lack in that thread you referenced is:

I have changed the way the dice work once again, it should now be theoretically impossible to cheat:

This is how the intensity cubes now work:
We have a series of 50,000 true random numbers from random.org
Each time the game engine processes an assault or auto-assault, it select a random spot in the series to read from using a pseudo-random computer function
Each time the game engine generates a random intensity cube, the next number is read in sequence from the series (e.g. in a 3v1 attack 4 numbers are read sequentially)
The series of 50,000 true random numbers from random.org is replaced every hour

So are you able to say how the dice are generated now?

No, I have no idea. But I do know that the data you collected in 2013 does not resemble stats I've looked at more recently. With full respect to the work you did and the rigour of your methods, I believe you were right in 2013 but something changed in January 2014.

I can't say what they do nowadays, but I'm pretty confident it's no longer the same thing. And if they went and changed it, I don't know why you would assume that they changed it in a bad way instead of changing it in a good way.

Granted, you may have received a PM about it, but without seeing it I can't comment on whether you might be misinterpreting what it said.

Duk - is there any way you can find out exactly what the current method of dice generation is and if there are any plans for change?

To put an end to the conjecture in this tread.

[owenshooter]

Duk - is there any way you can find out exactly what the current method of dice generation is and if there are any plans for change?

To put an end to the conjecture in this tread.

we already know what it is... they are not going to change it... it is not conjecture, for those of us that are active in the forums/cc and have been for years. once Random.org was abandoned, there was a significant change in the dice... the black jesus has spoken...-Jésus noir

[iAmCaffeine]

I am seriously considering leaving Conquer Club if something doesn't change with the dice.

Re: dice sucks
Postby iAmCaffeine on Fri Jul 10, 2015 3:57 am

In real life you're playing one game at a time with the same set of dice with what, 2-5 people? On here there are thousands of active games at the same time using dice from a limited pool, there are going to be all kinds of ridiculous results that we don't expect. A couple of weeks ago I lost eight large escalating games where I had over 75% chance to take the sweep each time and it was shit, but it's not a bug. Furthermore, it's a human's nature to remember the bad times more than the good, and we also see more bad results than good because we will obviously attack 18v2 (which I've lost before) but we're not gonna try 5v20.

Really? You post this when somebody else cries about the dice?

Can you say hypocrite?

I can. Can you also note the time difference between those two posts?

Also, that post simply offered some explanation, nothing more.

[Condestável]

I wonder what were the CC psychologists who came up with this "humans remember the bad times more than the good" anyway.

I remember and I cling to the best moments, both in life and in this faux-random game.

Good moments here are an absolutely rare crap. Epic landslides here are the fucking norm, not the exception.

[degaston]

I wonder what were the CC psychologists who came up with this "humans remember the bad times more than the good" anyway.

I remember and I cling to the best moments, both in life and in this faux-random game.

Good moments here are an absolutely rare crap. Epic landslides here are the fucking norm, not the exception.

If that were true, then your overall battle stats for 3v2 and 3v1 would not all be at 0%.

[Beast Of Burson]

Not that my opinion matters:

What about a survey from new players after about a month or two on here to find out what THEIR take on the site is?...
Might give you guys a little more perspective as to what to change/upgrade to make people want to stay instead of sprinting away as fast as they can.
From the horses mouth so to speak.

[riskllama]

Not that my opinion matters:

What about a survey from new players after about a month or two on here to find out what THEIR take on the site is?...
Might give you guys a little more perspective as to what to change/upgrade to make people want to stay instead of sprinting away as fast as they can.
From the horses mouth so to speak.

not a bad idea, BoB...
say, why do you have a canadian bald eagle in front of the stars & stripes??? you are aware you guys are importing them from us now, right?

[degaston]

Not that my opinion matters:

What about a survey from new players after about a month or two on here to find out what THEIR take on the site is?...
Might give you guys a little more perspective as to what to change/upgrade to make people want to stay instead of sprinting away as fast as they can.
From the horses mouth so to speak.

Even better might be to contact people who haven't been around for a while and ask them why they left. I left the site for close to a year, and was never contacted to find out why. Of course, this would require that the powers that be have some interest in opinions other than their own.

[degaston]

say, why do you have a canadian bald eagle in front of the stars & stripes??? you are aware you guys are importing them from us now, right?

Citation?

On August 9, 2007, the bald eagle was removed from the federal list of threatened and endangered species. After nearly disappearing from most of the United States decades ago, the bald eagle is now flourishing across the nation and no longer needs the protection of the Endangered Species Act.