Move FTC to factor four diff swing over 7 days
-
Many altcoins get screwed up when their developers try to come up with “the best, the fastest, the most advanced” settings. Most of these folks don’t even know the story behind Geist Geld.
Meanwhile, [url=https://github.com/ghostlander/FeatherCoin/commit/1f7a9e74895eaa723daea351b6e1a2294b864330]a small patch[/url] to fix broken getnetworkhashps.
-
Thanks for the patch Ghostlander. Can you make the changes on the Feathercoin/Feathercoin repo? You can edit it on the webpage. This will commit it to a temp branch with a pull request for me to accept. Currently I have to commit everything on your fork to pull your last commit.
-
Done. Our master branches are out-of-synch a bit since I’ve committed some of your recent changes myself.
-
this GME code only change the calculation of the hashrate approximation, this is [b]not[/b] involve in retarget calculation. This represent the hashrate average number for 30 blocks in the stats page of FTC, this is used to determine what you will get in the client when requesting the network hashrate in its console.
FTC was 120 and Ghostlander just make a patch to set it to 30 like the stat page (see the change link 5 post ago)
-
[quote] Well Gamecoin (GME) is a lesson in how not to do diff adjustments. [/quote]
There is no other coin in our position. We now have significant Hash rate to make attacks expensive.
We have been a consistent value, for a significant time, on the exchange, so even scammers have built up an investment in FTC.
The software changes have been successful, and have worked to stabilise the coin. Future changes look technically feasible and few side effects; look good to further increase network security. + Its still a community driven, open source coin…
Can I say thanks for that. Well Done.
-
I have pulled the commit from Ghostlander for better default values in the getnetworkhashps function.
https://github.com/FeatherCoin/FeatherCoin/pull/20
I spoke to Ghostlander last week about his branch for 0.6.4.4, there was an issue that he was ironing out. One question which he brought up was how to implement the .25 damping. I have not spent any time on this yet but have been very busy all the same on behalf of our project. If anyone has code they would like to put forward for this then please do and we can incorporate it.
-
when you get the nActualTimespan calculated (in an if for this patch like line 930) you just add 3 full target and then divide it back by 4 so if you get (in hour) 4.25 it is (4.25 +(3*5.25))/4 = 5
[quote]nActualTimespan += 3 * nTargetTimespanCurrent;
nActualTimespan = nActualTimespan/4;[/quote]you can have a nSample in it also depending on how you calculate nActualTimespan (as I have done in the spread sheet as 8 126 sample to construct the 504/126: so (nActualTimespan = (8 sample +(24* nTargetTimespanCurrent ))/32 )
-
[url=https://github.com/ghostlander/FeatherCoin/commit/69ec264e6c2405d4a2048267c2c3221990c25368]Update for 0.6.4.4 beta 2[/url]
Implements a new retarget strategy at block #87948 supporting combined average windows of 126 and 504 blocks with .25 damping. Tested on the livenet to report statistics in 504/2016 mode, seems fine. Ready for testnet.
-
edited: deleted, my mistake the code is ok the new naming just tricked me
-
I’d rather the testing phase irons out any kinks and nicks as to prevent another hard fork a few months from now when we realize the current rushed update wasn’t properly tested.
-
[quote name=“erk” post=“28576” timestamp=“1379643357”]
[quote author=mnstrcck link=topic=3447.msg28574#msg28574 date=1379641688]
I’d rather the testing phase irons out any kinks and nicks as to prevent another hard fork a few months from now when we realize the current rushed update wasn’t properly tested.
[/quote] Nice theory, now the reality, what testing phase?
[/quote]Ghostlander and groll both have been running calculations and simulations on getting the math right. The data is in this thread. Not sure what you’re getting at, maybe you can take the tongue out of your cheek and clarify?
-
[quote name=“erk” post=“28579” timestamp=“1379645562”]
When you test a coin modification you use the coins testnet. To my knowledge that has not been done, and I would imagine there would have been a call for participants.
[/quote]It’s up next re: Ghostlander’s last reply.
-
Speak of the testnet and it shall appear.
-
[quote] Here is a graph of our network hash rate and difficulty for the last 2 weeks. We’ve had about 1.7GH/s of loyal miners and 9GH/s (12GH/s peak) of coin hoppers. 15% vs. 85% [/quote]
That’s a great graph of the Feathercoin Hash Rate and Difficulty (Ghostlander), can we have it live on the forum or, every 2 weeks? How’s the change to 0.6.4.4 effected the Hash Rate, since September?
-
I actually contacted the chap who runs multipool to get the code for the graphs we have been using for illustration purposes. I intend to get it up on the stats page.
-
[quote name=“wrapper0feather” post=“31818” timestamp=“1382194677”]
[quote] Here is a graph of our network hash rate and difficulty for the last 2 weeks. We’ve had about 1.7GH/s of loyal miners and 9GH/s (12GH/s peak) of coin hoppers. 15% vs. 85% [/quote]That’s a great graph of the Feathercoin Hash Rate and Difficulty (Ghostlander), can we have it live on the forum or, every 2 weeks? How’s the change to 0.6.4.4 effected the Hash Rate, since September?
[/quote]Multipool uses HighCharts.com to generate these graphs. I guess we can, too. Their service is free for personal or non-profit use.
