primer-
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
July 08, 2014, 11:13:06 AM |
|
You can hear the crickets here throughout the day. Dev, release the fucking PoS fix(increase diff by x10) and let the community decide where to go from there!!
|
|
|
|
glop
|
|
July 08, 2014, 11:27:00 AM |
|
Another idea that would be better long term wise:
I could take novacoin's source code, and add the Dcrypt and PoB aspects to it. That is not a small job since there are a lot of differences. It will probably not be a 1 week project.
It is better for the long run of Slimcoin as all the commands and bugs that were in peercoin are added/fixed.
The new client will be made backwards compatible with this blockchain, so you will not lose any coins at all.
Great!!! Very good idea!! At least if you want to keep PoS.. if you choose not to keep PoS fork from bitcoin Why not Blackcoin instead of novacoin, Blackcoin has a more active dev team on the source code? PS: Please split up dcrypt and PoB into seperate classes and files for that matter, and made them accessable by methods and functions I do not know for novacoin, but blackcoin POS works nicely it could be an inspiration.
|
|
|
|
slimcoin (OP)
|
|
July 08, 2014, 11:34:22 AM |
|
Yes, but I looked into forking and have established that adding the PoB again will not be easy. Although I am not implementing it from scratch like before, I am afraid I may miss some little thing which could throw the entire client off again.
|
-Much Donate BTC-1D5pnma7E1CP6cquHujycVy79EyXJ3eY
|
|
|
glop
|
|
July 08, 2014, 12:00:28 PM Last edit: July 08, 2014, 12:13:09 PM by glop |
|
For me in the POS process 2 major fix to increase the viability of slim (as other POS coin) : Performance: POS should not take any power from computer or a very low part. Multiplicity: POS should not increase the number of transactions. After calculation of the interest, the base + interest should be in a new unique transaction and the old one to be deleted. I think if you address that part It could helps a lot for the precedent part. If you manage to fix these two issues you will be really close to success (at least technically)
|
|
|
|
slimcoin (OP)
|
|
July 08, 2014, 12:12:21 PM |
|
I have decided not to fork since it is said a LOT easier than done.
I still need some information as to the PoS using CPU. Someone said that blackcoin's worked, I could take a look. What would really help would be some information when it happens. Is it durring the creation of the stake, or afterwards? This is not a bug that I programmed, it probably was around when I forked Slimcoin from Peercoin. If someone knows how it was fixed, please tell.
|
-Much Donate BTC-1D5pnma7E1CP6cquHujycVy79EyXJ3eY
|
|
|
glop
|
|
July 08, 2014, 12:17:39 PM Last edit: July 08, 2014, 12:29:32 PM by glop |
|
I have decided not to fork since it is said a LOT easier than done.
I still need some information as to the PoS using CPU. Someone said that blackcoin's worked, I could take a look. What would really help would be some information when it happens. Is it durring the creation of the stake, or afterwards? This is not a bug that I programmed, it probably was around when I forked Slimcoin from Peercoin. If someone knows how it was fixed, please tell.
In blackcoin the POS process is not implemented the same way. In some other coin I know, the POS problems exist as well an I do not think they are fixed yet. As I known first, the performance problem come from the multiplicity of the transaction into the wallet. More you have transaction more it takes CPU for POS...... The second is the POS calculation of interest multiply the quantity of transaction into the wallet because the processes separate in two part the old transaction creating two new transactions and so on every loop (increasing by 2^n the number of transaction by the time). Also if you connect to the precedent point ...... I had this problem on another coin: I have let my wallet open for POS for weeks and by the time I had 40000 transactions and my CPU has jump to the ceiling at 100%
|
|
|
|
slimcoin (OP)
|
|
July 08, 2014, 12:28:52 PM |
|
So there is no fix for this yet. The second is the POS calculation of interest multiply the quantity of transaction into the wallet because the processes separate in two part the old transaction creating two new transactions and so on every loop (increasing by 2^n the number of transaction by the time). Also if you connect to the precedent point ...... I do not understand what you mean by this.
|
-Much Donate BTC-1D5pnma7E1CP6cquHujycVy79EyXJ3eY
|
|
|
glop
|
|
July 08, 2014, 12:35:07 PM |
|
So there is no fix for this yet. The second is the POS calculation of interest multiply the quantity of transaction into the wallet because the processes separate in two part the old transaction creating two new transactions and so on every loop (increasing by 2^n the number of transaction by the time). Also if you connect to the precedent point ...... I do not understand what you mean by this. I have not looked to the SLIM code but in POS code generally the process to calculates interest is: if the coin is old enough then take the transaction and calculates the interest then take the total of the transaction + interest and then create two new transaction with the half of the total (creating two new transaction) then delete the old transaction. This is the general process for POS, I think it is the same for SLIM, isn't it ? So if SLIM uses the same schema, the number of transaction into wallet have a tendency to increase by 2^n and the performance dropping in parallel.
|
|
|
|
slimcoin (OP)
|
|
July 08, 2014, 12:40:02 PM |
|
Yes, that is it. But in that process, when does that happen?
Does it only happen when the coins are old enough, which is after the 1 week wait time, or when the coins can create a valid PoS block.
|
-Much Donate BTC-1D5pnma7E1CP6cquHujycVy79EyXJ3eY
|
|
|
glop
|
|
July 08, 2014, 12:42:14 PM |
|
Yes, that is it. But in that process, when does that happen?
Does it only happen when the coins are old enough, which is after the 1 week wait time, or when the coins can create a valid PoS block.
I think when the wallet detect a valid POS block it launch this process .... The problem is not the quantity here it is the separation of the old transaction into two new transactions. It works on paper but not in real time. It was I think originally conceived to improve flow on the blockchain with a small amount of open wallet.....
|
|
|
|
slimcoin (OP)
|
|
July 08, 2014, 12:55:27 PM |
|
If this is part of the PoS and cannot be changed/has not been yet fixed, I could make it so that PoS does not start automatically, but require an option like -start-pos, or something.
|
-Much Donate BTC-1D5pnma7E1CP6cquHujycVy79EyXJ3eY
|
|
|
glop
|
|
July 08, 2014, 01:18:44 PM |
|
If this is part of the PoS and cannot be changed/has not been yet fixed, I could make it so that PoS does not start automatically, but require an option like -start-pos, or something.
A fix could also be to not split the transaction after calculation, something like: Verify if the coin is old enough ( 1 Week ? ) then calculates the interest total= coins + interest then create the new transaction with total and then remove the old coins. Because the quantity of transaction is not only linked to POS but also POW can generates lot of dust coins, it could be also interesting to have a tool to concatenate coins automatically. For example POS could be used as a regulator like, if the coins value > high_limit then split coin else if coins value < low_limit then if (next coin exist) concatenate coin with the next coin
|
|
|
|
primer-
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
July 08, 2014, 01:53:39 PM |
|
...or just make it scarce enough! Higher difficulty + PoS after 3 months would fix it... Would also give us 3 more months to think about a new fix...
|
|
|
|
glop
|
|
July 08, 2014, 01:55:13 PM |
|
...or just make it scarce enough! Higher difficulty + PoS after 3 months would fix it... Would also give us 3 more months to think about a new fix...
It does not fix the problem of quantity of transaction into wallets..... I mean dust coins
|
|
|
|
primer-
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
July 08, 2014, 01:56:48 PM |
|
...or just make it scarce enough! Higher difficulty + PoS after 3 months would fix it... Would also give us 3 more months to think about a new fix...
It does not fix the problem of quantity of transaction into wallets..... I mean dust coins CPU itilization problem is triggered when there are hundreds of small PoS transactions. That wont happen if you increase PoS diff.
|
|
|
|
glop
|
|
July 08, 2014, 02:01:51 PM |
|
...or just make it scarce enough! Higher difficulty + PoS after 3 months would fix it... Would also give us 3 more months to think about a new fix...
It does not fix the problem of quantity of transaction into wallets..... I mean dust coins CPU itilization problem is triggered when there are hundreds of small PoS transactions. That wont happen if you increase PoS diff. I think it is not only small POS transaction but also POW dust received from pools. I have that kind of dust into my wallet, I am sure you have as well (may be bigger dust ). IE: my wallet actually uses CPU and I have no coins into except burned coin and lot of POW dust from pool.........
|
|
|
|
primer-
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
July 08, 2014, 02:05:07 PM |
|
...or just make it scarce enough! Higher difficulty + PoS after 3 months would fix it... Would also give us 3 more months to think about a new fix...
It does not fix the problem of quantity of transaction into wallets..... I mean dust coins CPU itilization problem is triggered when there are hundreds of small PoS transactions. That wont happen if you increase PoS diff. I think it is not only small POS transaction but also POW dust received from pools. I have that kind of dust into my wallet, I am sure you have as well (may be bigger dust ). My pool dust is not causing any cpu spikes whatsoever. I only had spikes when my wallet was minting PoS blocks. The second i encrypted the wallet cpu spikes dissapeared. I fixed the problem by moving all coins to a new wallet.
|
|
|
|
glop
|
|
July 08, 2014, 02:10:17 PM |
|
...or just make it scarce enough! Higher difficulty + PoS after 3 months would fix it... Would also give us 3 more months to think about a new fix...
It does not fix the problem of quantity of transaction into wallets..... I mean dust coins CPU itilization problem is triggered when there are hundreds of small PoS transactions. That wont happen if you increase PoS diff. I think it is not only small POS transaction but also POW dust received from pools. I have that kind of dust into my wallet, I am sure you have as well (may be bigger dust ). My pool dust is not causing any cpu spikes whatsoever. I only had spikes when my wallet was minting PoS blocks. The second i encrypted the wallet cpu spikes dissapeared. I fixed the problem by moving all coins to a new wallet. I think the CPU load is dependent of the quantity of transaction into the wallet whatever it comes (POS POB or POW) and it is generated by the POS process.
|
|
|
|
primer-
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
July 08, 2014, 02:18:46 PM |
|
I think the CPU load is dependent of the quantity of transaction into the wallet whatever it comes (POS POB or POW) and it is generated by the POS process.
Lets leave PoS at 1 year - 1% (or 10% if you insist), as it was before block 15100. Problem solved! DO IT NOW DEVELOPER OR THIS SHIT COIN IS GOING NOWHERE!!
|
|
|
|
primer-
Legendary
Offline
Activity: 1092
Merit: 1000
|
|
July 08, 2014, 02:24:21 PM |
|
When this is done grab a pen and paper and do some math! Difficulty/reward estimation/prediction!
Both for PoW and PoB! Its BROKEN as it is. Add x10 current difficulty and rewards are almost non existant!!
|
|
|
|
|