BBQKorv
|
|
May 13, 2013, 02:19:46 PM |
|
Could you watch how the Elacoin launch goes and learn from that, this could give you much better chances to succeed in here.
|
|
|
|
toruonu
Member
Offline
Activity: 88
Merit: 10
|
|
May 13, 2013, 02:25:58 PM |
|
Recommendation. Start with a reasonable difficulty so that even a low adoption won't totally block the coin, but that it doesn't fly with 3-5 blocks / sec either. Then make the retargeting for difficulty happen very fast initially with gradual slowdown until it comes to where you think it'll be mature.
I'd recommend starting off with retarget every 10 blocks, then at 100 blocks every 25 blocks, then at 500 blocks every 100 blocks, then at 2000 blocks every 250 blocks etc. This way a spike in hashrate will quickly change difficulty and the coins won't be minted like nuts meaning that a delay of 5 min in launch won't mean a difference of 10k blocks. It will also make the blockchain easier to access as initially most coins that fly with loads and loads of blocks per sec make later adopters lose a lot of time and blocks because of pure blockchain catchup.
Secondly don't reward the first 1k blocks at all. By that time the difficulty estimate should give a more stable rate, then make a gradual buildup until 25k blocks or what ever you planned.
Thirdly. Implement a failsafe. If you target the blocks at every 2.5min. Then if a block isn't found in N * 2.5 min the difficulty halves (you'll have to figure out what a good N is). Rinse and repeat. This way if for whatever reason the diff is driven up by someone in a short timeframe and then he leaves the rest can still continue to mine it and not be abandoned at high difficulty.
If I think of something else I'll post. And make sure you don't fuck up the launch. Try to choose a time that is mostly reasonable in all populated timezones. Probably late evening Asia, day EU, morning US is something that might work so that the maximum amount of people can be online for the launch. Give the time at least 24h in advance and then also launch it at the time. Giving people a countdown with 2h to go ... 1h to go, 30 min to go etc would be useful too.
|
|
|
|
cryptohunter
Legendary
Offline
Activity: 2100
Merit: 1167
MY RED TRUST LEFT BY SCUMBAGS - READ MY SIG
|
|
May 13, 2013, 02:32:20 PM |
|
Recommendation. Start with a reasonable difficulty so that even a low adoption won't totally block the coin, but that it doesn't fly with 3-5 blocks / sec either. Then make the retargeting for difficulty happen very fast initially with gradual slowdown until it comes to where you think it'll be mature.
I'd recommend starting off with retarget every 10 blocks, then at 100 blocks every 25 blocks, then at 500 blocks every 100 blocks, then at 2000 blocks every 250 blocks etc. This way a spike in hashrate will quickly change difficulty and the coins won't be minted like nuts meaning that a delay of 5 min in launch won't mean a difference of 10k blocks. It will also make the blockchain easier to access as initially most coins that fly with loads and loads of blocks per sec make later adopters lose a lot of time and blocks because of pure blockchain catchup.
Secondly don't reward the first 1k blocks at all. By that time the difficulty estimate should give a more stable rate, then make a gradual buildup until 25k blocks or what ever you planned.
Thirdly. Implement a failsafe. If you target the blocks at every 2.5min. Then if a block isn't found in N * 2.5 min the difficulty halves (you'll have to figure out what a good N is). Rinse and repeat. This way if for whatever reason the diff is driven up by someone in a short timeframe and then he leaves the rest can still continue to mine it and not be abandoned at high difficulty.
If I think of something else I'll post. And make sure you don't fuck up the launch. Try to choose a time that is mostly reasonable in all populated timezones. Probably late evening Asia, day EU, morning US is something that might work so that the maximum amount of people can be online for the launch. Give the time at least 24h in advance and then also launch it at the time. Giving people a countdown with 2h to go ... 1h to go, 30 min to go etc would be useful too.
yep, like the sound of all that.
|
|
|
|
cre8r (OP)
Member
Offline
Activity: 100
Merit: 10
|
|
May 13, 2013, 02:36:58 PM |
|
Recommendation. Start with a reasonable difficulty so that even a low adoption won't totally block the coin, but that it doesn't fly with 3-5 blocks / sec either. Then make the retargeting for difficulty happen very fast initially with gradual slowdown until it comes to where you think it'll be mature.
I'd recommend starting off with retarget every 10 blocks, then at 100 blocks every 25 blocks, then at 500 blocks every 100 blocks, then at 2000 blocks every 250 blocks etc. This way a spike in hashrate will quickly change difficulty and the coins won't be minted like nuts meaning that a delay of 5 min in launch won't mean a difference of 10k blocks. It will also make the blockchain easier to access as initially most coins that fly with loads and loads of blocks per sec make later adopters lose a lot of time and blocks because of pure blockchain catchup.
Secondly don't reward the first 1k blocks at all. By that time the difficulty estimate should give a more stable rate, then make a gradual buildup until 25k blocks or what ever you planned.
Thirdly. Implement a failsafe. If you target the blocks at every 2.5min. Then if a block isn't found in N * 2.5 min the difficulty halves (you'll have to figure out what a good N is). Rinse and repeat. This way if for whatever reason the diff is driven up by someone in a short timeframe and then he leaves the rest can still continue to mine it and not be abandoned at high difficulty.
If I think of something else I'll post. And make sure you don't fuck up the launch. Try to choose a time that is mostly reasonable in all populated timezones. Probably late evening Asia, day EU, morning US is something that might work so that the maximum amount of people can be online for the launch. Give the time at least 24h in advance and then also launch it at the time. Giving people a countdown with 2h to go ... 1h to go, 30 min to go etc would be useful too.
Thanks for your advices! I'll do my best to follow them.
|
|
|
|
tytanick
Legendary
Offline
Activity: 2660
Merit: 1096
Simplemining.net Admin
|
|
May 13, 2013, 03:31:47 PM |
|
i have error compiling your sources: /bin/sh ../share/genbuild.sh obj/build.h make: *** No rule to make target `obj/scrypt.o', needed by `onecoind'. Stop. make: *** Waiting for unfinished jobs....
|
|
|
|
nocoin
|
|
May 13, 2013, 04:16:53 PM |
|
i have error compiling your sources: /bin/sh ../share/genbuild.sh obj/build.h make: *** No rule to make target `obj/scrypt.o', needed by `onecoind'. Stop. make: *** Waiting for unfinished jobs.... And what kind of compiling instructions might it be? Just compiled, like for any other coin: git clone git://github.com/cre8r/onecoin.git cd onecoin/ qmake make
Or, for headless onecoind git clone git://github.com/cre8r/onecoin.git cd onecoin/src make -f makefile.unix
Of course, all dependencies must be installed.
|
|
|
|
tytanick
Legendary
Offline
Activity: 2660
Merit: 1096
Simplemining.net Admin
|
|
May 13, 2013, 05:35:28 PM |
|
yep, i am compiling by: git clone git://github.com/cre8r/onecoin.git cd onecoin/src make -f makefile.unix and still getting error "no rules for obj/scrypt.o" -D_FORTIFY_SOURCE=2 -MMD -MF obj/rpcnet.d -o obj/rpcnet.o rpcnet.cpp g++ -c -O2 -pthread -Wall -Wextra -Wformat -Wformat-security -Wno-unused-parameter -g -DUSE_IPV6 -DBOOST_SPIRIT_THREADSAFE -I/home/otc/onecoin/src -I/home/otc/onecoin/src/obj -DUSE_UPNP=0 -DSCRYPT_KECCAK512 -DSCRYPT_CHACHA -DSCRYPT_CHOOSE_COMPILETIME -DHAVE_BUILD_INFO -fno-stack-protector -fstack-protector-all -Wstack-protector -D_FORTIFY_SOURCE=2 -MMD -MF obj/rpcrawtransaction.d -o obj/rpcrawtransaction.o rpcrawtransaction.cpp g++ -c -O2 -pthread -Wall -Wextra -Wformat -Wformat-security -Wno-unused-parameter -g -DUSE_IPV6 -DBOOST_SPIRIT_THREADSAFE -I/home/otc/onecoin/src -I/home/otc/onecoin/src/obj -DUSE_UPNP=0 -DSCRYPT_KECCAK512 -DSCRYPT_CHACHA -DSCRYPT_CHOOSE_COMPILETIME -DHAVE_BUILD_INFO -fno-stack-protector -fstack-protector-all -Wstack-protector -D_FORTIFY_SOURCE=2 -MMD -MF obj/script.d -o obj/script.o script.cpp make: *** Brak reguł do zrobienia obiektu `obj/scrypt.o', wymaganego przez `onecoind'. Stop.
|
|
|
|
|
com911
|
|
May 13, 2013, 05:47:50 PM |
|
CPU-only again? So what is the difference between YAC and Onecoin, apart of block reward?
|
|
|
|
garumaru
|
|
May 13, 2013, 05:48:51 PM |
|
Sounds to me this is is like Bitbar for CPU mining.
|
BTC - 19zLbgtvGUttFiT9LXY8XAw1gGRiwXQap4 DGC - DM2qB66qAetszRtQACX683Vd7Bgd6sUsRB
|
|
|
|
xxrforone
Member
Offline
Activity: 115
Merit: 10
NoHoooo
|
|
May 13, 2013, 07:24:32 PM |
|
stupid
|
|
|
|
alex_fun
|
|
May 13, 2013, 07:30:07 PM |
|
lol are you mining his semi finished git?
|
|
|
|
cre8r (OP)
Member
Offline
Activity: 100
Merit: 10
|
|
May 14, 2013, 10:06:03 AM |
|
i have error compiling your sources: /bin/sh ../share/genbuild.sh obj/build.h make: *** No rule to make target `obj/scrypt.o', needed by `onecoind'. Stop. make: *** Waiting for unfinished jobs.... My bad, didn't clean up makefile. Should be fixed now.
|
|
|
|
cre8r (OP)
Member
Offline
Activity: 100
Merit: 10
|
|
May 14, 2013, 06:03:02 PM |
|
I've just made some improvements: Difficulty retargets each block (PPCoin's implementation). Blocks 0-999 will get no reward. Blocks 1000-1499 will receive 10 nano-OTC, 1500-1999 - 20 nano-OTC and so on, the subsidy increasing by 10 nano-OTC every 500 blocks, until block 25500, which will receive the full reward of 500 nano-OTC. The total money amount remains the same - a bit less than one coin.
|
|
|
|
cryptohunter
Legendary
Offline
Activity: 2100
Merit: 1167
MY RED TRUST LEFT BY SCUMBAGS - READ MY SIG
|
|
May 14, 2013, 06:45:56 PM |
|
I've just made some improvements: Difficulty retargets each block (PPCoin's implementation). Blocks 0-999 will get no reward. Blocks 1000-1499 will receive 10 nano-OTC, 1500-1999 - 20 nano-OTC and so on, the subsidy increasing by 10 nano-OTC every 500 blocks, until block 25500, which will receive the full reward of 500 nano-OTC. The total money amount remains the same - a bit less than one coin.
sounds good too, but could we not have gpu not cpu? or a mix? some days it favours cpu others gpu? is it possible?
|
|
|
|
cre8r (OP)
Member
Offline
Activity: 100
Merit: 10
|
|
May 15, 2013, 08:20:12 AM |
|
sounds good too, but could we not have gpu not cpu? or a mix? some days it favours cpu others gpu? is it possible?
I doubt this is possible.
|
|
|
|
Professor James Moriarty
aka TheTortoise
Sr. Member
Offline
Activity: 434
Merit: 250
|
|
May 15, 2013, 08:23:41 AM |
|
why no gpu tough? I mean I get the no fpga and no asic deal since they are not everywhere yet but everyone has a gpu right? If you can't afford a nice gpu than you probably cant afford a nice cpu too
|
|
|
|
nopowerbills
Newbie
Offline
Activity: 12
Merit: 0
|
|
May 15, 2013, 08:56:44 AM |
|
yeah i was going to ask, do you have a launch date,
please no GMT 6:30PM lol
|
|
|
|
wangsw
Newbie
Offline
Activity: 29
Merit: 0
|
|
May 15, 2013, 12:01:28 PM |
|
As of scrypt-jane,i advise Nfactor starts with more than 4,or FPGA will be available like yac did.
|
|
|
|
|