monoxide
|
|
January 31, 2014, 07:44:48 PM |
|
very interesting. i think this can go far.
|
|
|
|
s4w3d0ff (OP)
Sr. Member
Offline
Activity: 322
Merit: 250
Spray and Pray
|
|
January 31, 2014, 07:49:41 PM |
|
Try to check manualy the value fmod(1,0.00024414) ? Maybe function fmod returns 0 if the true reminder is less than 0.00001. Sorry for multiply edits Ok, I made program fmod with following code: #include <stdio.h> #include <math.h>
int main () { float a, c; a = 1; c = 0.00024414; printf("Remainder of %f / %f is %lf\n", a, c, fmod(a,c)); return(0); }
which returned : Remainder of 1.000000 / 0.000244 is 0.000003
Then realized that in my coins code I was using "double" instead of "float" to define the difficulty. So I proceeded to change my fmod code to: #include <stdio.h> #include <math.h>
int main () { double a, c; a = 1; c = 0.00024414; printf("Remainder of %f / %f is %lf\n", a, c, fmod(a,c)); return(0); }
Which still gave me: Remainder of 1.000000 / 0.000244 is 0.000003
Still don't understand why my coins code thinks that 0.000003 doesn't fall between 1 and 0.0000001...
|
|
|
|
tabjohn
Member
Offline
Activity: 149
Merit: 10
TABJOHN
|
|
January 31, 2014, 07:59:15 PM |
|
Try manualy check all values of: diff = GetDifficulty(); fmod(nHeight, diff); and condition fmod(nHeight, diff) > 0.0000001. Then you should be able to find what is wrong... Probably diff = GetDifficulty() is 0.
|
|
|
|
twistedtrick
|
|
January 31, 2014, 08:00:49 PM |
|
Interested idea, going to keep my eyes here.
|
|
|
|
s4w3d0ff (OP)
Sr. Member
Offline
Activity: 322
Merit: 250
Spray and Pray
|
|
January 31, 2014, 08:14:00 PM Last edit: January 31, 2014, 10:05:06 PM by s4w3d0ff |
|
Probably diff = GetDifficulty() is 0.
I don't think that is true because at block 4 ( fmod(4, diff) ) the code decided to make an anomaly and the diff never changed between blocks 1 - 10 (because of such a low hash rate). If diff = 0 then none of the blocks should of generated an anomaly. I'll keep working on it... I just don't want this to happen when I change the code back to "< 1 && > 0.1" it could be a disaster.
|
|
|
|
tabjohn
Member
Offline
Activity: 149
Merit: 10
TABJOHN
|
|
January 31, 2014, 08:26:03 PM |
|
Yeah, you right. So probably the condition itself has some problem with comparision double/float such small numbers
|
|
|
|
Errror
Jr. Member
Offline
Activity: 47
Merit: 25
ggwp
|
|
January 31, 2014, 08:40:08 PM |
|
people in crypto are really creative...
|
⚫ ⚫ ⚫ ico.TrueGame.io ⚫ ⚫ ⚫ BLOCKCHAIN BASED GAMES OF CHANCE ▄ ▄▄ ▄▄▄ PRE-SALE 5th March ▄▄▄ ▄▄ ▄[/cente
|
|
|
bryhardt
Member
Offline
Activity: 89
Merit: 10
|
|
January 31, 2014, 08:57:44 PM |
|
This is great and got my gears turning a bit. But in the end you could calculate the odds of getting the remainder you are looking for and just using a random number generator (not as cool and wouldn't be as fun)? I hope to be available when you are ready to put this out, this is good stuff.
|
|
|
|
s4w3d0ff (OP)
Sr. Member
Offline
Activity: 322
Merit: 250
Spray and Pray
|
|
January 31, 2014, 11:16:53 PM |
|
Probably diff = GetDifficulty() is 0.
I don't think that is true because at block 4 ( fmod(4, diff) ) the code decided to make an anomaly and the diff never changed between blocks 1 - 10 (because of such a low hash rate). If diff = 0 then none of the blocks should of generated an anomaly. I'll keep working on it... I just don't want this to happen when I change the code back to "< 1 && > 0.1" it could be a disaster. tabjohn you were partially right! The first couple of blocks the difficulty must be near zero (or it is zero) but when you "getmininginfo" it says 0.00024414 (damn lairs!). I adjusted the refresh time to 80 secs (so every 2 blocks instead of 3) and my third block generated an anomaly ! So we are all good! Thank you so much for helping me out! The coin is back on schedule! Just some finishing touches and the coin should be released by Feb 1st at noon!Something else I realized about my formula tho is that if the difficulty ever goes higher than the current block number then the remainder is going to always be the block number! Example: 50 % 100 = 50; 50 % 345.342 = 50! So to answer everyone's questions on is lower hash rate better than higher hash rate? The answer is sometimes... Because if you pump to much hashing power into this coin you would raise the difficulty too high thus making it literally impossible to generate an anomaly! This is something I did not expect to happen but I like it. It forces the miners to be smart about there hash rates instead of "brute forcing" the network.
|
|
|
|
alphateam
|
|
January 31, 2014, 11:36:23 PM |
|
So 1st february 8pm for GMT
|
|
|
|
Hazard
Legendary
Offline
Activity: 980
Merit: 1000
|
|
January 31, 2014, 11:39:44 PM |
|
Literally the worst idea I've ever seen pitched on this forum.
|
|
|
|
tabjohn
Member
Offline
Activity: 149
Merit: 10
TABJOHN
|
|
January 31, 2014, 11:46:01 PM |
|
You right, but I think that diff will never be higher than the block number. The higher diff, the lower probability of hitting anomaly. At diff < 1 the anomaly occurs almost every time. The key thing will be also diff retarget. It is scheduled for 3 blocks but with some restrictions? Say max +15% and -50% or something like that?
|
|
|
|
s4w3d0ff (OP)
Sr. Member
Offline
Activity: 322
Merit: 250
Spray and Pray
|
|
January 31, 2014, 11:56:21 PM |
|
You right, but I think that diff will never be higher than the block number. The higher diff, the lower probability of hitting anomaly. At diff < 1 the anomaly occurs almost every time. The key thing will be also diff retarget. It is scheduled for 3 blocks but with some restrictions? Say max +15% and -50% or something like that?
I changed the re-target to every 2 blocks... but I might change it to every block before the release.
|
|
|
|
atleticofa
|
|
February 01, 2014, 12:08:27 AM |
|
I'm in!!!
|
|
|
|
drakoin
|
|
February 01, 2014, 12:55:10 AM |
|
First idea I have is to give the agents (wallets) a history. Deep trust only comes with long history. Depending on the trust level, an agent can or cannot do certain things. History is something that can be made to be expensive to fake. Look at this forum here. Newbies can do a lot, but not everything. Hard to fake 1000 accounts with high activity level. What do you think?
|
no sign of a signature
|
|
|
rnasynthetase
Newbie
Offline
Activity: 16
Merit: 0
|
|
February 01, 2014, 01:32:37 AM |
|
This coin is going to smash 42 coin on most expensive coin ever we should try to get this to a exchange sooner then later
|
|
|
|
s4w3d0ff (OP)
Sr. Member
Offline
Activity: 322
Merit: 250
Spray and Pray
|
|
February 01, 2014, 04:26:05 AM |
|
|
|
|
|
Nolo
|
|
February 01, 2014, 04:50:18 AM |
|
If someone would compile win binaries I'd jump on this one.
|
Charlie Kelly: I'm pleading the 5th. The Attorney: I would advise you do that. Charlie Kelly: I'll take that advice under cooperation, alright? Now, let's say you and I go toe-to-toe on bird law and see who comes out the victor? The Attorney: You know, I don't think I'm going to do anything close to that and I can clearly see you know nothing about the law. 19GpqFsNGP8jS941YYZZjmCSrHwvX3QjiC
|
|
|
anon4250158
|
|
February 01, 2014, 08:37:32 AM |
|
Please wait for windows binaries to be available before launching. I would like to get in on this.
|
|
|
|
|
|