Topic: [NSFW] Official Sexcoin Thread - New MANDATORY upgrade: KGW Timewarp attack FIX - page 97. (Read 307641 times)

Nope just a DDos on me.. I'm kinda flattered they'd take the time to bother...

we can discuss the best percentages approach , but if the hash increase lets say 200% from one block to the next you know we're under a flash attack. At that point the diff should increase and not wait for the 60 block retarget. The same way it should retarget down if hash comes down 20% from one block to the next

There's a couple of things that bother me about how the retarget is calculated. One thing is that it only looks at a time frame and # of blocks. So essentially it runs all the calculations from two blocks, the first and the last. I've been working on a routine that inspects all the solves in pairs or triplets, runs a standard deviation on them, and then throws out data thats way out there. After that it would add a small percentage of the 'off the chart' data, to accomodate the flash. This should give a smooth retarget at almost any retarget interval you pick. In flash cases, we would get a difficulty that was a little higher than 'optimal', and during 'normal' times, there would be no difference.

Would someone please explain to me how a poison pill works?

Correction, Lava was against the faster block time, and it looks like he was right..
From the stats it looks like half the miners are hashing along nicely, but it's reject city for others..

My reject rate is 90%, but I'm on a really really lousy connection with inbuilt lag (VSAT)

It seems to me that a poison pill would make the changes in difficulty larger not smaller.  There are 2 obvious ways to smooth out difficulty shorten adjustment period or make only fractional adjustments.  I think 30 or 60 block adjustments will work fine.  Anything > 5 min and < 1 hour seems pretty solid.  Fractional adjustments would also work if you did say a geometric mean of the previous 2 targets with a double weight on the new target.  This can mess with the time a bit.  You probably could even use the same concept as pools use for PPLNS on difficulty adjustments.

Can you elaborate on that ?

add a poison pill for sudden hash increases

Ok guys lets give it to you straight...

The retargeting is wrong... we've gotten together discussed it on IRC and are organising to take care of it.

We have said that we were going to issue two hardforks from the start which is fortunate, it means that we can fix it quickly, and we'll do an official announcement explaining what exactly went wrong.. Unfortunately we are moving our planned next mandatory client release forward to next week end so you'll have to upgrade again.. we are looking to make it so you'll have two weeks to upgrade.

The basic problem is we should be popping blocks every 60 seconds with a 30 block retarget. Instead we are on 30 second blocks with a 60 block difficulty retarget.. We are all mining at double rewards.

As embarrassing as it is, we've always played it straight, believing that honesty is the best policy, so we admit that we screwed up.

And we've decided that the thirty second blocks give us a competitive edge, the 60 second block retarget still gives us half hourly retargets so.. We are keeping them..

We will reduce the the rewards for blocks by half, and change the block halving to suit..

We know how incompetant this makes us look, but we are willing to wear the blame, and the damage is that miners will be getting more coins for about the next three weeks, and the price will probably drop due to the higher amount of coins hitting the exchanges.. Hopefully it will rebound at the next hardfork.

We know that this is a pain in the ass, but this should be the last mandatory update we need to issue, and the community knew we were doing another one so hopefully it will proceed smoothly.
 

Looks like I can't do high school math anymore.

*Blushing horrifically*

Ok we're going to go with 60 block retarget for now.

The good news is I hit all the right spots.

From my phone
I'm not worried about hash rate for the network. Just waiting for next retarget

Back to RL....

and we are through.. still hashing away... give that man a cigar  

EDIT Nope, we dropped about 4MH off the network... Un-upgraded Solo Miner ?
Edit Dropped 16MH... down to 88MH..

11 blocks to go... I'll watch the switchover... if anything goes wrong, I'll panic for you

Things that suck about RL....

I have obligations today that won't let me watch the switchover.
I am very confident that it will go smoothly.

IF anything should go wrong, I'll be on later this afternoon to react to it.

Wow! A lot has happened in the last 24 hours. Taking the coin off CoinChoose, for now at least, was a good call. The hash-rate graphs pretty much speak for themselves. I think being on Cryptsy is certainly a good thing. I was quickly losing confidence in PhenixEx.com’s willingness to communicate with their clients.  I understand being in beta but the communication coming out of that site is crap. Anyway, I like were this coin is headed. This is great!

CrunchCrunchCrunch.... 140 to go.

well it seems we've mostly got strong hands holding SXC, they just ain't selling cheap. The network is at 76 M/H and thats not flash mining thats new miners coming in.. I'm thinking of paying for a press release by http://bitcoinprbuzz.com/ in a few days, and in 250 blocks we will hit the hardfork, have our "standard" desync fix, and our difficulty retarget will be at every 30 Blocks..

It's all happening...

OK, I've had a bit more of a think about this, each miner contributes a share of 1,000 SXC into a 'syndicate' (ie I contribute 5,000 SXC I have five shares, some-one else contributes 347,000 SXC they have 347 shares) - Affiliate sites refer customers to the SXC (Syndicate) exchange where the customer buys a number of SXC ( 'n'SXC )

The customer buys services with the 'n'SXC - at the end of the billing cycle (daily/weekly/monthly) the affiliate sells the SXC back to the SXC Exchange for a discounted price ('n'SXC - x%)  The remaining amount is paid out to the share holders in the syndicate as a dividend

Naturally if an Affiliate offers more SXC back to the exchange than the exchange has cash on hand there could be problems, however, a get around could be a token given by the affiliate when referring the customer to the SXC Exchange .

Like I said originally, it's a little bit of a closed system as we're only dealing with one digital coin - naturally, a syndicate share holder can 'buy' more shares by contributing more shares at 1,000 SXC pershare & can leave by withdrawing their share at anytime.

I wouldn't want to go lower than 1 share = 1,000 SXC - Miner's who frequent this discussion (we know who we are could be invited with a token like the Zombie / Lavapit invite or something similar.

Anyways, thanks for reading this.

THANK YOU for your comment on "autodumping and automining" earlier !

your insight is spot on i think and something needs to be done..

unlike some other coins i see nothing whatsoever sketchy about your coin.
you had before and you will continue to have my support

best of luck with your goals and congrats on the cryptsy addition too
i hoped it would go on there but was told ages ago it's highly unlikely more than once lol
i guess someone changed their mind for some reason..

Hmm cryptsy seems to be trading about the same as phenixex... interesting to see where it goes from here..

You might want to review your trades, if you are selling SXC.

Normally I would snap up the CoinEX coins quietly, but I am unable to move my funds out of PhenixEX, it's been about a week I've been waiting to withdraw my BTC. I'd say that as we are most of the volume of PhenixEX, Cryptsy is moving in to kill off their volume, as well as having to recognise SXC as a serious altcoin any worthy of addition to their LTC market. Of course moving BTC to LTC then SXC is a bit of a PITA, so the volume on the exchanges that have already had the foresight to list us against BTC may be mitigated somewhat..