[ANN] Catcoin - Scrypt meow!

[SlimePuppy]

Before anyone thinks 'fork' they might want to read this:
https://forum.feathercoin.com/index.php/topic,2178.0.html

Can we resist this type of complex 51%-with-hax-fest?

[Maverickthenoob]

Before anyone thinks 'fork' they might want to read this:
https://forum.feathercoin.com/index.php/topic,2178.0.html

Can we resist this type of complex 51%-with-hax-fest?

Upon review that was an entirely different scenario. That was a direct and deliberate attack on the coin.  "The board" isn't talking about 51%ing anything, we want the community to reach a consensus on this coin and how to fix it ASAP, so we can begin implementing this fix. The hash rate is in free fall, we must do something relatively quickly or we're soon going to be on the list of dead coins. These are not threats at all, just facts. We're wasting time talking about hozer's fork, because unless someone dropped 200 MHash on it, it will not be official. I'm sure we're not the only coin with a cloned git repo.

SO, will the 1 block 36 average method satisfy members of the community to fix the diff jumps for the short-medium term? If we added a max 20% jump per re-target, would that alleviate more concerns? Is there another solution you'd rather see? What are we going to do immediately to save this coin? You know what my vote is, and most members of the board agree with me, however, we will not put through a change with out the large acceptance of the community. So lets start accepting or suggesting! Lets get this party started.

[etblvu1]

I ran a simulation of what would happen if we continuously adjusted difficulty based on taking the average time it took to solve the last 36 blocks, and used the ratio of (10 minutes) / (average time it took to solve last 36 blocks), to determine what the difficulty should be for the next block. I used real data pulled from catchain.info for when each block was generated. The blue lines indicate the algorithm we are using now. The red indicates the simulation of what is being proposed (unless I misunderstood the formula, or made a mistake in my simulation code).

I am going back over my code to make sure I did not make any mistakes, so I want to make a disclaimer that this is only the initial result. But since there is talk of a fork, I wanted to get  a graph of the simulation out as soon as possible, for discussion.

Etblvu1

[janos666]

It might will be a little difficult to get a real consensus now.

Last time, I was loud about my opinion that the first fork should absolutely not happen the way it just happened in front of my eyes in a matter of a few hours (which is a fast blink in a coins's life).
I knew that any rushed decision will drive away many supporters and that this low retarget time will convert CAT into a GLX kind of yoyo coin.
But regardless of this, I still kept with this coin because I thought the CAT name and the relatively strong community is more important than these things on the long run (I didn't consider that there might won't be a long run from that point, regardless of the strong name and community).

But now I have absolutely no idea. I can't say anything, but: do what you want. I can only hope it won't make things worse.

The only thing I know is that I invested one month worth of mining income into CAT (income, not net profit!), so I won't do a panic sell at 0.0003 BTC/CAT and cut the losses right now because that would mean an actual net loss for that month. I will be one of the last solo miners with ~4.5Mh/s if everything else seem to break down around this coin and still have a little hope.
But this also means that I reached the point where I can't invest anymore into this coin. I can't afford to buy any more CAT in a hope that it will worth at least as much later as I invest and that my purchases might help the coin to survive it's dark moments.
And I can't honestly suggest anything after I feel so stupid about my earlier decisions regarding this coin.

[SlimePuppy]

Upon review that was an entirely different scenario. That was a direct and deliberate attack on the coin.

Apparently so.  Yet the 'guy' is still out there.  This is something else BTC didn't have to deal with when it was less than 1 month old.

 "The board" isn't talking about 51%ing anything, we want the community to reach a consensus on this coin and how to fix it ASAP, so we can begin implementing this fix.

I don't recall anyone suggesting the board is planning a coup.

The hash rate is in free fall, we must do something relatively quickly or we're soon going to be on the list of dead coins. These are not threats at all, just facts.

I've been living with my pad less than an arms-length away with two pages open since just after Christmas:  Coinium and the catcoins.biz/charts.  It appears that the hash rate is recovering again with 12 blocks to go before the next adjustment.  Are we truly in need of another 'emergency' fork?  Not to fight, but is it clear to the 'board' this is the case?

SO, will the 1 block 36 average method satisfy members of the community to fix the diff jumps for the short-medium term?

How the hell do we know?    Has it been looked-at?  Back tested?  Or are we 'doing something, even if it's wrong'?

Has anyone looked at other coins to see what/how they're adjusting?

[janos666]

I ran a simulation of what would happen if we continuously adjusted difficulty based on taking the average time it took to solve the last 36 blocks, and used the ratio of (10 minutes) / (average time it took to solve last 36 blocks), to determine what the difficulty should be for the next block. I used real data pulled from catchain.info for when each block was generated. The blue lines indicate the algorithm we are using now. The red indicates the simulation of what is being proposed (unless I misunderstood the formula, or made a mistake in my simulation code).

I am going back over my code to make sure I did not make any mistakes, so I want to make a disclaimer that this is only the initial result. But since there is talk of a fork, I wanted to get  a graph of the simulation out as soon as possible, for discussion.

Etblvu1

The biggest error factor is that you didn't take it into account that hash power is a factor of the real-time profitability.
Different diff target algorithm attracts miners differently.

[etblvu1]

The biggest error factor is that you didn't take it into account that hash power is a factor of the real-time profitability.
Different diff target algorithm attracts miners differently.

Thank you for your feedback. Yes, I agree and am aware that if the profitability is different, people would react differently, and that would have affected the hash power, and therefore the timing of when the blocks were found. However, my point was not to do a simulation in a vacuum, but apply the formula to actual historical data, so the "shape" of what comes out from the formula is more obvious than something that sounds nice and fuzzy in the purely theoretical realm. I am convinced that basing the difficulty on how long it took to mine the last 36 blocks, would result in even less stability than we have now. If the graph is skewed because it is based on what the algorithm would created based on how people reacted to the existing algorithm, just imagine how people would have reacted to this one.

Also, I want to reiterate, I still need to verify the code. Since a proposed fork was imminent, I wanted to do a simulation and post a graph as soon as possible, so people are not deciding whether or not to support this fork purely based on an abstract description, but also consider what a simulation shows the resulting difficulty might look like.

Etblvu1

[janos666]

Absolutely just an idea out of thin air! But what about something like this?
Retarget can happen after every block.
The minimum relative change is 5%, the maximum is 10% and the diff is always truncated (or rounded) to be an integer (to smooth out the small fluctuations even further).
On top of that, there is check after every 10 blocks and a forced retarget with 25% limit if the diff would have changed by >=10% >=5 times during the last 10 /1 <=10% ratargets.
After that, there is a retarget with 50% limit after every 30 blocks if the diff would have changed >=25% >=2 times during the last 3 /10 <=25% checks.
And as a last safety belt, there is a check after every hours and a forced, limitless retarget, based on the average of the last 5 blocks if a >=25% would have been necessary during the last  1 /30 retarget.

However, I must point it out again that smoother and less limited diff recalc is always a weapon in the buyers hands.

A perfectly smooth retarget means that buyers are free to lower the prices after each and every transactions as long as they wish (literally without limits until they hit 1 mBTC and get bored of investing more) because the coin supply is virtually limitless: lower price -> less miners -> lower diff = same amount of constant coin supply (for cheaper and cheaper). The net hasrate is dropping and dropping... Well, actually the same thing happens what we can see right now.

But may be the smoother retarget allows easier resurrections after investors also realize that they want some profit at the end, so prices must start to go up at some point, or everybody looses.

[mashac]

"Too Many Cooks Spoil the Broth" - is what happening here. Even if I hated this coin initially, this is really sad to see.

[mavis369]

I have been loyal to Cats, but I am worried now, I have 762.71029709 CAT , does anyone want to buy them for 0.00050000 BTC. I do not mine, so I am losing money everyday, like all of us I know, but? I bought most of mine when they first came out and for 0.0010000 BTC each.

[janos666]

I have been loyal to Cats, but I am worried now, I have 762.71029709 CAT , does anyone want to buy them for 0.00050000 BTC. I do not mine, so I am losing money everyday, like all of us I know, but? I bought most of mine when they first came out and for 0.0010000 BTC each.

Hah. Well, I have several thousands. Do YOU want to buy from ME at 0.0007? LOL.
Hold on and see what happens! You can't really loose a lot more if the coin dies. But you might have a chance to make a profit after all. Even if it takes months.

[kisa2005]

whenever talking about "profitability" please lets distinguish between various concepts. Apologies I am not tech, but trying to clarify the economics. Please be specific which of the following you mean when stating that the coin is unprofitable to mine:

1.  ... as it doesn't cover energy cost over difficulty cycle at current price

2. ... as it doesn't cover energy cost and hardware amortisation over over difficulty cycle at current price

3. ... as it doesn't cover energy cost and hardware amortisation over during high difficulty at current price

4. ... as it is at current price less profitable then mining some other coins during high difficulty

5. ... as it is at current price less profitable then mining some other coins over difficulty cycle

Which of those constitutes a real problem for loyal miners?
Which of those states is a problem primarily for professional hoppers?

[janos666]

whenever talking about "profitability" please lets distinguish between various concepts. Apologies I am not tech, but trying to clarify the economics. Please be specific which of the following you mean when stating that the coin is unprofitable to mine:

1.  ... as it doesn't cover energy cost over difficulty cycle at current price

2. ... as it doesn't cover energy cost and hardware amortisation over over difficulty cycle at current price

3. ... as it doesn't cover energy cost and hardware amortisation over during high difficulty at current price

4. ... as it is at current price less profitable then mining some other coins during high difficulty

5. ... as it is at current price less profitable then mining some other coins over difficulty cycle

Which of those constitutes a real problem for loyal miners?
Which of those states is a problem primarily for professional hoppers?

Either.

It seems like a Scrypt ASIC is not just real, but it's also very close to get mass produced and actually put into work in massive numbers. So, GPU miners (probably still well over 90% of the total scrypt hash speed right now) need to squeeze out as much profit from GPU mining as possible. They can't afford to think in months anymore, they need to constantly switch to the coin which gives them the most BTC in the next X minutes because they can never know when an ASIC farm might completely take them out of business.

I haven't reached my RTI point yet, mostly because I invested into CAT. So, I am an actual example that it's very risky to plan further than a single day. (I said RISKY, not the I completely lost my hope in CAT!)

On the other hand, you might make a lot more profit if you are still willing to take your chances and think in months (instead of joining a multipool, or always mine the latest coin which is the best profitable for a day and then dies tomorow).

[Evilish]

Accepting Catcoin as a payment method in my new Steam and Minecraft shop. 
https://bitcointalk.org/index.php?topic=411976.0

[skillface]

Honestly I'm a little concerned about how this new fork is being handled. This idea of 'expect another release and hardfork next Caturday' doesn't exactly inspire confidence, at least for me. I have a feeling that others share my sentiment and may even give up mining CAT entirely, leaving for another coin less likely to fork every other week. And some people will just scoff at this simply because the dev isn't anywhere to be seen.

I was pretty hopeful for the last fork, and it did help somewhat, but I'm not quite as certain about this one. There needs to be a proper consensus, and not by 'voting with your hash'. That implies that smaller miners like myself have less say than the larger miners. And the whole point of CAT is to appeal to the smaller miners and traders as well as the large ones. Even if I were to 'vote' for the new fork, my 300KH/s isn't going to influence that much now is it?

Still, something needs to be done about the difficulty oscillation. I'm just not entirely sure that re-targeting every block is the correct solution, even if it takes the average of the last 36 blocks.

I feel that the vast majority of the Catcoin community aren't even aware of this test. And given that the new fork is meant to take effect in a short period of time..

[kisa2005]

"Too Many Cooks Spoil the Broth" - is what happening here. Even if I hated this coin initially, this is really sad to see.

your point taken. however, remember how many predictions that the coin would never make it to an exchange, that it would die a few days after hitting cryptcy, that the community is falling apart - all of them proved to be wrong so far.
Those who dedicate themselves to CAT will get over differences, and the community has enough skillful and patient people. Whoever is impatient and wants to dump CATs - please do it, no worries, as it would be better outcome if coins come into caring hands at low price...

[janos666]

I wonder how other altcoins, for example ANC survived these stages and got resurrected later: http://www.cryptocoincharts.info/v2/pair/anc/btc/cryptsy/1-year
I am not here a long ago, so I couldn't observe things like that.
This was my template when I bought more CAT instead of selling all. It took months, and this whole scrypt game might won't continue that long from now, but think about it...

[skillface]

In other news it's looking like BTC might be about to crash again somewhat, something happened to Huobi in China and supposedly there's a lot of panic selling going on. Now may be a good time to buy CAT, perhaps.

[PocketRocketsCasino]

Just a reminder, https://pocketrocketscasino.com/ allows you to play poker including Open Face Chinese and Blackjack for Catcoin.

I was asked by the community to add this currency option but it seems Catcoin interest is dying? Not seen much activity on the subreddit either.

[etblvu1]

Just a reminder, https://pocketrocketscasino.com/ allows you to play poker including Open Face Chinese and Blackjack for Catcoin.

I was asked by the community to add this currency option but it seems Catcoin interest is dying? Not seen much activity on the subreddit either.

I just now set up an account and funded it with 100 Cats tried to fund it with 100 CATs, but the site says only BTC deposits are available now. I am happy to support any site that adds Catcoin support, so help me transfer the CATs in. Thank you for that.

Etblvu1