Quote from: psychocoin on July 05, 2014, 10:48:01 AM

So, how's the patch coming along?

Well, I was hoping to get some more feedback/criticism/comments on the gist outline, but this doesn't seem to be happening.

I've been preparing a proposal release. I have had a build of it running/mining on the real chain for a couple of days now, and it seems stable. It has the following "feature branches" included so far:

TwoTargets

The full patch for the difficulty time warp fix, still enabled only on testnet, and without the Median change.

GetChains

A trivial port of Domob's getchains patch from HUC, used for monitoring soft fork lengths. This is a great tool for monitoring network fork behavior, and checking for double spend attempts. HUC has seen some epic soft forks and this has become an indispensable tool on MOTO as well, to monitor for attempts to use the time warp.

MoreWarp

Allows time to be sped up or slowed down by up to three orders of magnitude more than the reference client

MapSeeker

Two variants of Minim1ner's map filter(s), tuned for human mining.

TxReplay

Allows you to quickly launch a replay of a block from the context menu of a transaction. This makes it easy to replay blocks you've mined.

TestnetBox

The bitcoin-testnet-box setup modified for convenient moto testing

LtcUpdates

Gitian openssl (etc.) updates merged from upstream

I'll push these branches to my github today for more people to try. I think whether the others step in with feedback on the patch or not we should go forward with launching the change as a community, and make improvements later. IMO, it is important to at least address the attack vector with "something" ASAP, even if that something is not the ideal. Worst case we end up having to hard fork after all, but I see this as an unlikely outcome.

I also have Minim1ner's full bot cleanly merged in, tweaked/optimized, and using the original wallet's build files so it builds cleanly/easily on windos and *nix. I'm assuming OSX will build with the other pullreq that is hanging out there, but I can't test it as all of my mac hw runs linux. However, these patches won't be released until after some fix for the warp is adopted by the network.

@WilliamLie, if I pullreq these patches (with TwoTargets enabled for live net) do you think that you'd merge them? I think the soft fork for the fix will likely go much more smoothly if we're not forced into "forking against" the reference client.

In the future, after the warp attack issue shakes out, I'd like to work on improving the game physics, addressing the question(s) of what happens when we hit the ltc coin caps, and adding some more interesting elements to the competitive aspects.