Author Topic: Masternodes Testing (Round 2)  (Read 154906 times)

georgem

  • Tech Admin
  • ******
  • Posts: 956
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #225 on: February 10, 2015, 02:11:07 pm »
Is it possible that the "high powered" mns keep voting each other (good and bad votes) and filling up all available votes per block, so that all lower mns will not have the slightest change of EVER placing a vote in a block?

If an mn is already elected, does it keep receiving good votes in the blocks, or only possibly negative ones if ever.

Strumpet!

  • Full Member
  • ***
  • Posts: 158
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #226 on: February 10, 2015, 02:59:47 pm »
Is it possible that the "high powered" mns keep voting each other (good and bad votes) and filling up all available votes per block, so that all lower mns will not have the slightest change of EVER placing a vote in a block?

If an mn is already elected, does it keep receiving good votes in the blocks, or only possibly negative ones if ever.

If someone who groks c++ could look at the code and draw a diagram for idiots like me of exactly how it works that would be great.  ;D

InstantX seems to be back in action, I'm getting Masternode confirmations straight away again, albeit not quite so many as last time.

georgem

  • Tech Admin
  • ******
  • Posts: 956
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #227 on: February 10, 2015, 03:11:44 pm »
Is it possible that the "high powered" mns keep voting each other (good and bad votes) and filling up all available votes per block, so that all lower mns will not have the slightest change of EVER placing a vote in a block?

If an mn is already elected, does it keep receiving good votes in the blocks, or only possibly negative ones if ever.

If someone who groks c++ could look at the code and draw a diagram for idiots like me of exactly how it works that would be great.  ;D

InstantX seems to be back in action, I'm getting Masternode confirmations straight away again, albeit not quite so many as last time.

Maybe start with a blockchain byte chart... Dissecting the blockchain byte by byte....
I don't know much about C++, but I like to fire up my trusty HEX Editor once in a while...

I will start a new thread....

MyFarm

  • Sr. Member
  • ****
  • Posts: 259
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #228 on: February 10, 2015, 04:12:44 pm »
I was laying in bed last night and a concern arose.

At present, we are seeing around 150 masternodes be elected per day.  For testing, this means we're looking at another 7-8 days before they are elected and competition begins.  However, once someone puts up a higher value masternode, will it take 10 days for it to become unelected and drop off and get paid the whole time?  And are we looking at 18 days just to test this?

I understand the need for a slower election process, but I worry this is too slow for mainnet as well.

georgem

  • Tech Admin
  • ******
  • Posts: 956
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #229 on: February 10, 2015, 04:17:08 pm »
I was laying in bed last night and a concern arose.

At present, we are seeing around 150 masternodes be elected per day.  For testing, this means we're looking at another 7-8 days before they are elected and competition begins.  However, once someone puts up a higher value masternode, will it take 10 days for it to become unelected and drop off and get paid the whole time?  And are we looking at 18 days just to test this?

I understand the need for a slower election process, but I worry this is too slow for mainnet as well.

I think the goal of this test Round 2 is to find out what could be made better next time.
I agree that the way the voting happens at the moment gives me the impression that we have some kind of bottleneck problem, that even threatens to sabotage the whole competitive game we were supposed to have with the masternodes. What good is competition when mns with more SPR get treated much better than the rest?

MyFarm

  • Sr. Member
  • ****
  • Posts: 259
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #230 on: February 10, 2015, 04:33:55 pm »
I was laying in bed last night and a concern arose.

At present, we are seeing around 150 masternodes be elected per day.  For testing, this means we're looking at another 7-8 days before they are elected and competition begins.  However, once someone puts up a higher value masternode, will it take 10 days for it to become unelected and drop off and get paid the whole time?  And are we looking at 18 days just to test this?

I understand the need for a slower election process, but I worry this is too slow for mainnet as well.

I think the goal of this test Round 2 is to find out what could be made better next time.
I agree that the way the voting happens at the moment gives me the impression that we have some kind of bottleneck problem, that even threatens to sabotage the whole competitive game we were supposed to have with the masternodes. What good is competition when mns with more SPR get treated much better than the rest?

I don't mind them getting treated a little better such as getting elected first like they are now, but they shouldn't get THIS big of a headstart.  A 7+ day payment headstart over lower masternodes is way too much.

Mr. Spread

  • Core Developer
  • ******
  • Posts: 105
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #231 on: February 10, 2015, 05:26:10 pm »
EDIT: btw, the miningprivkey=xxxxxxxxxxxxxxxxxxxxxxxx doesn't work in spreadcoin.conf.
Not on my PCes anyway... I might be missing something...
Are you sure you set up everything correctly? datadir is SpreadTest2Coin and config name is spreadcoin.conf.

Mr. Spread

  • Core Developer
  • ******
  • Posts: 105
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #232 on: February 10, 2015, 05:30:17 pm »
I was laying in bed last night and a concern arose.

At present, we are seeing around 150 masternodes be elected per day.  For testing, this means we're looking at another 7-8 days before they are elected and competition begins.  However, once someone puts up a higher value masternode, will it take 10 days for it to become unelected and drop off and get paid the whole time?  And are we looking at 18 days just to test this?

I understand the need for a slower election process, but I worry this is too slow for mainnet as well.
Electing from zero to 1440 is slow because there is a long queue of masternodes to be elected. When there will be around elected 1440 masternodes election/deelection will happen faster.

once someone puts up a higher value masternode, will it take 10 days for it to become unelected
What made you think it will take 10 days? It will happen in 1-2 hours.

def15

  • Jr. Member
  • **
  • Posts: 62
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #233 on: February 10, 2015, 06:14:21 pm »
I was laying in bed last night and a concern arose.

At present, we are seeing around 150 masternodes be elected per day.

Electing from zero to 1440 is slow because there is a long queue of masternodes to be elected. When there will be around elected 1440 masternodes election/deelection will happen faster.

Does that mean 150 masternodes coin join/leave the network per day?

georgem

  • Tech Admin
  • ******
  • Posts: 956
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #234 on: February 10, 2015, 06:17:38 pm »

Does that mean 150 masternodes coin join/leave the network per day?

Yes, I ask myself the same question.

What is the maximum possible votes we can have per day and per block?

jjjordan

  • Newbie
  • *
  • Posts: 48
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #235 on: February 10, 2015, 06:59:58 pm »
EDIT: btw, the miningprivkey=xxxxxxxxxxxxxxxxxxxxxxxx doesn't work in spreadcoin.conf.
Not on my PCes anyway... I might be missing something...
Are you sure you set up everything correctly? datadir is SpreadTest2Coin and config name is spreadcoin.conf.
Thanks, that worked. I thought spreadcoin.conf should go where wallet.dat goes in SpreadTest2Coin/testnet3/
But obviously it has to be all alone by himself (blockchain, wallet etc are in the ../testnet3/)

georgem

  • Tech Admin
  • ******
  • Posts: 956
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #236 on: February 10, 2015, 07:02:46 pm »
EDIT: btw, the miningprivkey=xxxxxxxxxxxxxxxxxxxxxxxx doesn't work in spreadcoin.conf.
Not on my PCes anyway... I might be missing something...
Are you sure you set up everything correctly? datadir is SpreadTest2Coin and config name is spreadcoin.conf.
Thanks, that worked. I thought spreadcoin.conf should go where wallet.dat goes in SpreadTest2Coin/testnet3/
But obviously it has to be all alone by himself (blockchain, wallet etc are in the ../testnet3/)

You are right, the conf file has to be in the folder above, not the folder with the wallet.dat

Mr. Spread

  • Core Developer
  • ******
  • Posts: 105
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #237 on: February 10, 2015, 07:18:03 pm »
Found some strange issue on this test version, maybe a bug, when using "getinfo":

Code: [Select]
girino@vps1:~/git/spreadcoin/src$ ./spreadcoind getinfo
error: {"code":-1,"message":"map::at"}

other commands seem to work correctly.

(using linux, ubuntu 14.04, compiled with GCC 4.8.2)

Same here, I think GilAlexander reported it too.
I wasn't able to reproduce it but it should be fixed in latest master.

MyFarm

  • Sr. Member
  • ****
  • Posts: 259
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #238 on: February 10, 2015, 09:17:34 pm »
1600+ testnet masternodes.

Congratulations to Mr. Spread and the Spreadcoin community.  Truly amazing work!

georgem

  • Tech Admin
  • ******
  • Posts: 956
    • View Profile
Re: Masternodes Testing (Round 2)
« Reply #239 on: February 10, 2015, 11:17:42 pm »
395 / 1665

I am slowly starting to get rid of my 151er mns and turning them into 300er mns.

All my <=151 mns have yet to be elected, ...

so I guess I will wait some more, but will slowly start to fuse them together one by one...