Topic: [ANN] Catcoin - 0.9.1.1 - Old thread. Locked. Please use 0.9.2 thread. - page 4. (Read 131044 times)

Ok...  Now that we're rolling towards block 46331, here's what to expect once we get to the new code.

First, a reminder:  Sending to/from Cryptsy should be normal up to our fork.  Hold off on transfers at the fork until we post that things are working - let me risk my CATs before you risk yours.

As everyone knows (painfully knows...), the PID algo that seemed perfect on testnet started to gallop once we forked into it on mainnet.  We've been in this fast/slow block mode for far too long while we dealt with attacks and forks and searched for a difficulty adjust algo that worked better for our relatively small coin in this 'modern' environment.

We worked on tweaks and filters for PID, single algos, and combinations - including DGW2 and 3 and 1- and 3-block Digishield.  The winner by far was the dual random Digishield/PID that's in V9.2.0.  All of the minimum block time kludges have been removed (as well as all of the other attempts to fix things since V8).

This combination moves around enough to show that it's alive, but will never drop the difficulty too low the way our current code does.  Yes, we've attracted coin switch miners in the past when our difficulty droped low enough (read:  too low) and that's given blocks away and then kicked our diff up too high.  As we've said all along, the goal has been to stabilize the difficulty while resisting hoppers and common exploits as much as possible.  Hopefully this is what'll happen in mid-July.

This code runs a 9-12 minute block time during smooth periods, resists hash 'attacks' very well while properly responding to new miners on the network.  It also does the best job of ramping the difficulty back down if an attacker stays on us for many hours then pulls their miners.  Fingers crossed CATs.  If all goes well, this'll finally be the stability we've all wanted since the coin launched at Christmas, 2013.

Andy


http://catcoinwallets.com/images/testnet/24b.jpg


http://catcoinwallets.com/images/testnet/34b.jpg




Ada the mighty huntress is on fork watch...

Well, it didn't take our resident hacker long to start messing with us.  IP 107.170.40.107 is a former network seed node run by hozer.  Way to stay true to your stripes...

ouch; have just had my linux wallet running the whole time.. oh well

The good chain is 5 hours back right now and on block 44177.  There's a catcoin.conf in the Windows distro package - add those two seed nodes to your existing catcoin.conf (delete any others - especially if any start with 170.x.x.x).

These are the network seed nodes.  They're in the code as well - you shouldn't have to add them.  But do clean any nodes out of appdata/roaming/catcoin/catcoin.conf that aren't these:


Andy

Current block, at the time of this post, is 44177, showing 4 hours behind, because the last block was almost 5 hours ago.

The official explorer reflects that too: http://catcoinwallets.com:2750/chain/Catcoin

seems to be stuck 6 hours behind?
i'm up around 44193..

There you go... ADVERTISING YOUR NON OFFICIAL CODE...

Take your crap elsewhere!

https://github.com/CatcoinOfficial/CatcoinRelease/compare/master...tmagik:master

Interesting discussion about bitcoin:

http://cointelegraph.com/news/114657/chinese-mining-pools-call-for-consensus-refuse-switch-to-bitcoin-xt


s/bitcoin/catcoin/

If you read the change history, there are two active catcoin developers that have made substantial changes to the code. They don't agree with each other right now.

It sets a particularly bad precedent to release a hardfork based on block number that doesn't even do the miner-voting that has been demonstrated successfully in Bitcoin-core multiple times.

Again Troy, you are trying to start beating on the dead horse again.  NOBODY of the community accepts your solution of REJECTING LEGIT BLOCKS BECAUSE 3 MINUTES HAS NOT PASSED.  We only had the 30 seconds in there to help stop the galloping, which DID NOT HELP!  The only one bringing up the FUD, is YOU!  Trying to claim that someone else GOT YOUR CODE .. ummm the only way they can get it is if you post it somewhere, AND ADVERTISE IT!

Take your FUD spreading elsewhere. 

The wallet's up.  Thanks again for the heads-up folks.  Andy

Windows wallet:
http://catcoinwallets.com/win/Catcoin_0.9.2.0.zip
SHA256: A3FAC5C54F2E22B2918823140A4D121BA25160DFB3E2AEDB073A14E30A5204B9

i can appreciate the fact you've put effort in to prevent this, and it is always disappointing when a coin gets whored out - however a lower block time will deter people looking for 'blocks ahead of time' etc.. increase competition per block but will devalue the coin as well i suppose..

ive got a coin i plan to release soon - and it is *this* that puts me off each time..

Thanks again iju76 and barrysty1e.  This is on me.  I built the Windows wallet.  I tested each of the previous six builds for this code release and they worked perfectly on my personal database and on a fresh machine with no database.  I did not, however, fully test the final build before zipping it - I just made sure it ran and connected to the network and got the latest blocks.

I've got the new Windows wallet built and am watching it sync with the network.  Once this is done, I'll get a copy on my personal machine and test against my existing CAT wallet.  If all goes well, I'll push this to catcoinwallets.com and let you know it's ready to download.

Thanks - and again - please accept my apology.
Andy

The problem is you have no way to enforce that everyone runs the 'official' code, unless you change the copyright and try to DMCA anyone who runs hacked code. As far as I can tell, someone else leased a ton of hash and brought up a new pool that ran my code, and probably made a reasonable profit doing so.

What's going to happen is someone's going to write hacked code (or even just a hacked *miner*) that will mess with the timestamps and then be able to effectively control the difficulty adjustment.

I advise you to go *experiment* with a miner that always sets 44 second timestamps on testnet and see what happens.

I have no problem with people that admit it, and that this blockchain was started out by instaminers who profited handsomely. But miners can't profit if nobody buys the coin, and nobody is going to buy the coin if they know that instaminers are the only ones profiting. So there's got to be more going on than just a new code release that is different instaminers trying to game the difficulty adjustment.

We've got what looks to me like the same old altcoin pump-and-dump cycle going on here. Someone is going to make a profit instamining by exploiting the new difficulty adjustment. I know it's going to happen, and what I would like is for some discussion about how adjusting the difficulty every block requires re-examining some fundamental assumptions that are made in the Satoshi codebase.

Like the assumption that timestamps don't matter.

I'd argue that any adjustment that happens every block requires paying very close attention to timestamps, and that it probably requires a minimum block time, as well as that the network reject blocks more than say 10 minutes in the future.

The problem is that what I get is some FUD from people who stand to make money instamining instead of rational discussion. Maybe I am expecting to much.

Would you all prefer I just said nothing, and silently exploited the code like everyone else? Or would you rather I take the time to develop and publish the code? Or maybe we could possibly admit that the stock PID algorithm with a 3 minute minimum instead of a 30 second one might be a reasonable compromise that instaminers, small miners, and people that like cats can all live with?

Sadly barrysty1e, around block 43xxx, is the last attempt that hozer tried to derail catcoin using HIS rogue "catbox" code, and (most likely leased) a TON of hash rate.  I have added checkpoints that should keep it from actually being derailed.

This new release should help stop the galloping of the difficulty at low hash rates, and we should finally start smoothing out and get regular blocks.  As stated in the past many times, if EVERYONE runs the "official" code, and keeps steady hashing on the coin, then when switch-pools do pop on and off, it won't affect the difficulty swing as hard, and it will absorb the impact.

Blak

Troy - Blak's already outlined why your suggestion is off the mark.  Feel free to bring exploit code to -dev if you think you have a leg to stand on.  Details talk; BS/FUD walks.

New Folks:  Disregard Troy/hozer/tmagik/technomagik/catcoiner.  Trade with extreme caution.  Get deliverables in advance.

thats bit harsh, we're all instaminers..
id refer you to the blocks at the start of the chain that saw > 1000mh (especially in early 2014!)

This looks like a nice present for the insta-miners:

https://github.com/CatcoinOfficial/CatcoinRelease/commit/0d5a0ec09bbbf1c0b2ad3941c751419b86942c7a#diff-7ec3c68a81efff79b6ca22ac1f1eabbaR1392

I detailed why it's a problem awhile back. If you really all want to run this, then can the instaminer(s) who have exploit code for the above please let me know where to transfer catstat.info so we can make this transition as smooth at possible so it does not interrupt your plans?

PM me and I'll probably get back to you in a couple of weeks when I check back how the transition went.

Good luck, and enjoy your new (or maybe the original) insta-miner overlords!

*blinks*

did that just happen?
thanks guys.