The YACoin Superfun Premine Thread

[00400]

I remember when it was IXC and then I0C...
so... maybe a "Y0Coin" with fair and the same rules for Win/Linux from the beginning

I vote on relaunch.
YACoin start was really suspicious. Anyway the idea of coin for CPU miners looks attractive. So I say lets clone it, but make a fair one.

BTW. Y0Coin is maybe even better name than YAC. Like: Yo man, got some change?

[cryptohunter]

I usually wait for thread to reach 15-20 pages before downloading hoping some of zillion geeks here would already found out if there is something fishy.

Yeah the problem with that is that by page 2  99% of the coins have been hoovered up the gpu farms and the developer.

It's same for me, I don't have any special mining equipment, I don't really mine so 2nd or 20th I can't get anything  

lol well i do have 2 pc's with 7950 but it takes forever for me to get it set up and running, since it took me ages to realise i was not meant to be putting that yac command into cgminer, even when i found out it was meant for the windows cmd box i didnt know you had to cd to the directory of the prog before running it.

Anyway was fun learning, and people on this forum are quite cool they are pleased to give a hand to others.

After i found out what to do it was great starting to mine solo, until i was told my next issue was my latency would ensure no blocks being accepted.

House is for sale, moving to a place somewhere nearer a town or a place with decent internet connection. My top speed is 1-2mb line here and ping is pretty bad too.

I got 3 blocks of yacs after a day but then stopped mining when i jumped out of bed today to see all the blocks i must have mined over night, sadly .......not even 1 additional yac. Switched off the miner in disgust.

I did better on cnc, bobs pool was cool.  For now i'm just collecting lots of different coins. I like to have a selection. Even snagged a bitbar chunk or 2. That was cool.

[aikklond]

Also adding the addnode part from the first post on the official thread limited my connections to 8, and I didn't solve any blocks or get above 8 connections until I removed that code (after wasting 8 hours).

how do you remove this code?

[to3m]

Excellent conspiracy theories in this thread.  You guys continue to entertain me.  The reality is that you had to be there for the launch.  I was browsing the forums at 1AM PST and saw the launch of it, downloaded, and profited big time.  All you guys who were hours late and got caught out by the massive zerg rush can continue QQ'ing.

Well some of us WERE there for the launch... and it didn't launch

--Tom

[markm]

another scam, good job we have some smart people who can detect this kind of behaviour. Let's hope this does not make it to an exchange.

Someone should put out a decent cpu only coin asap  because ppl will keep yacoin going because they can mine it at the same time as gpu mined coins.

Look around for gosh sakes, there are still, and have always been, several chains you can mine very very nicely with even just a single-core CPU.

Look at BBQcoin now, it is now in the phase where all the people who mined it with CPUs for months or a year or more are getting paid handsomely, and meanwhile now that BBQcoin is too high difficulty for CPUs their CPUs are still mining any of several other coins that, just like BBQcoin had been for so long, they are glad the big miners are ignoring.

The payoff for a rig with GPUs and CPUs, with the CPUs mining low difficulty chains where CPUs can get lots of coins, has been much much higher for the CPUs than the GPUs. Those CPUs made more money than the GPUs.

The CPUs have turned out to be far more profitable than the GPUs because although in both cases there are only so many coins available per day, most of the chains the GPUs work on divide up those coins among more actual human beings on any given day than the chains the CPUs are working on.

So while you as a GPU miner might be one of a thousand or ten thousand or more miners divvying up the coins of a high difficulty chain, you as a CPU miner can often find chains where you are one of only a hundred or less people divvying up the coins on any particular day. (A lot of CPU miners seem to just pick up some number of coins of each chain that they feel is enough to give them a nice payoff when the chain comes back into the limelight, instead of dedicating cores long term like a person with a huge GPU farm would be likely to do if they bothered to put their GPU rigs' CPUs to work.)

So basically if your CPUs have not been raking in more money over the long term than your GPUs that is not due to lack of special anti-GPU hashing systems, it is simply due to your failure to put your CPUs to profitable work.

-MarkM-

[WindMaster]

I phucking knew that something was fishy from the begining and made a thread to document the orphan rate. Windows users had 90%+ orphan rate, linux users about 10-15%:

To be clear though, I immediately modified the client after I pulled it from git (8 hours past launch because I slept through it, grrr) to increase the maximum number of outbound connections, all 1560 servers had public IP's so inbound connections were possible, and all 1560 servers also had tweaks to the source so they all connecting with each other to form a large interconnected low-latency mesh within the Yacoin network.  Those factors probably significantly helped with my orphan rate.  Still, I was 8 hours late to the party and a single PC at the right time early on would've mined much more than I did.  But I still think it was probably a fair fight 8 hours in.

[mr_random]

To be fair I made two separate posts querying what the starting difficulty would be in that topic and to my surprise not a single person responded or picked them up.

Everyone forgets the 'scam' claims in the run up to a coin launching and conveniently stop caring, because they hope they are going to get rich from it. As soon as they don't get rich, out come the pitchforks.

[WindMaster]

tl;dr pretty much everyone running the Windows binaries from the onset was screwed.

Possible alternative, everyone that didn't immediately start figuring out how to port the scrypt chacha code and (relatively slowly changing) variable N to the cgminer GPU kernel were actually screwed from the onset regardless whether they were running Windows or Linux?  From experience running a substantial server farm mining YAC yesterday, and looking at the difficulty and block solving rate today, I'm pretty skeptical that CPU mining accounts for most of the apparent hash rate.

[markm]

can we, as a community, create a coin that we all AGREE is fair terms?  like right from the get-go, a solution that works for all of us?

So far the fairest really seem to be the ones that even right now you can rake in lots of coins of using just a CPU or a single GPU.

They are fair because everyone has had years to mine them and thus everyone has had plenty of time to pick up pretty much however many coins they want.

Some of them have even been handed out free on the side to Bitcoin miners by merged mining pools, so even if the miners were too short sighted to care about the coins themselves their pool operator nonetheless gave them their fair share so that come the payoff the miners will not be  looking back saying "it is so unfair, all those years I could have picked these things up for almost nothing and now that everyone appreciates their value they are going to cost me a fortune to buy!"

-MarkM-

[maxsolnc]

I was mining with near 200 Xeon cores (13 servers), they were running Windows. when it was 1M coins total, I had 10K - 1% , it is really good for small cluster. My servers are in USA and I even couldn't open port ! so I don't think that windows clients are 2 times slower.

[markm]

guess i'll better get a linux distro for any future coins before my 7950 arrives

Supposedly Linux isn't great at Scrypt, some have said to use Windows for that.

For example I can get 405 or so MHashes of SHA256 from a 5870, but fiddle with it all I want (using Fedora core 17) I don't seem to be able to get more than 255KHashes of Scrypt. Yet people claim one should get 1/1000 as many Scrypt hashes as SHA256 hashes out of one's GPUs.

-MarkM-

[jimhsu]

Didn't do too badly on the launch 2 days ago with just a few systems, even just Windows (I don't have the fancy Xeon servers or EC2 instances or whatever that you people have). Then again, I only found about the thread and got in at block #200 (difficulty 0.0002) just because I was doing work way past my bedtime and happened to see the thread.  

[markm]

I mined around 400 on win 7 with i5 3570k oc'ed to 4.5Ghz

I was even late to party and running it in a sandbox for a few hours

So windows mining was differently possible

I think the biggest problem was/is network latency.

You can have a SUPER fast computer, but if your net is slow with high latency you will never get those solved blocks accepted since someone with better network latency already got their block accepted before you..

No, the big problem is insanely low initial difficulty.

The latency "problem" is only even noticeable because the blocks are too insanely easy to solve.

If the number of miners who were able to get up and running within the first few pages of the thread had been too few to even have solved the first block in that tiny amount of time, then by the time there were enough miners aboard to have a decent chance of getting the first block solved the blocks would have come faster and faster and faster until within just an hour or two after launch there would have been enough miners to have blocks being solved at about the target speed and even then chances are high that within another few pages they would have been solved faster than the target speed, but hopefully by then a retarget would be coming soon enough to keep adjusting back to the target speed fast enough to keep latency from ever being a noticeable problem.

-MarkM-

[fisheater]

if the init diff is high, then many small computing people won't be able to get anything. It will need the mining pools being set up at the beginning.

In any case, those with big rigs will always have big advantage, no matter what we do about it.

[tacotime]

tl;dr pretty much everyone running the Windows binaries from the onset was screwed.

Possible alternative, everyone that didn't immediately start figuring out how to port the scrypt chacha code and (relatively slowly changing) variable N to the cgminer GPU kernel were actually screwed from the onset regardless whether they were running Windows or Linux?  From experience running a substantial server farm mining YAC yesterday, and looking at the difficulty and block solving rate today, I'm pretty skeptical that CPU mining accounts for most of the apparent hash rate.

This is pretty likely.  If I can pull 400 kh/s on a 2700k, my guess is that a 7970 could pull several mh/s at the current difficulty/N.

[tacotime]

Ahh, It's always the same....  A small bunch calling premine and crying about being late.

In 2 hours time there will be 10 new altcoins launched and new folks to whinge about those.

So. Fair play, but this ain't going to spoil YacCoin.

I'm not complaining, I have lots of them.

[markm]

if the init diff is high, then many small computing people won't be able to get anything. It will need the mining pools being set up at the beginning.

In any case, those with big rigs will always have big advantage, no matter what we do about it.

There does seem to be need for a good clear step by step manual walking coin developers through the changes that p2pool will need, so that as they decide each thing they are changing to make their coin they can make the corresponding change to a pull for p2pool.

So far no one seems to really know what some of the strange hashes or coded strings in p2pool are let alone how to generate them or where to obtain their values for your new coin.

But yeah it makes sense that if thousands of machines will be mining, the chance of any individual one of them solving a block in the target time has to be one in the number of machines, thus that if several thousand cores will be mining it should take any individual core thousands of times the target time, on average, to find a block.

-MarkM-

[robotrebellion]

YAC's failure to launch at the announced launch time was a breach of trust. This business with the binaries makes matters worse.

At this point I'm comfortable lumping it in with all the other pre-mined coins.

[bitdwarf]

Possible alternative, everyone that didn't immediately start figuring out how to port the scrypt chacha code and (relatively slowly changing) variable N to the cgminer GPU kernel were actually screwed from the onset regardless whether they were running Windows or Linux?

I joined at 9 minutes with 40 blocks mined, and my cheap single i5 2500k Linux rig held for quite a while with ~10% of all coins (I kept checking it, cos it was the first coin I joined since the start and wanted to see how the minting rate was). It ended making 20k YACs throughout the day. So basically anyone who pointed a common Linux desktop to the network would have got a substantial share of the early coins.

Definitely there weren't big guys around with huge rigs from the start, maybe because nobody was expecting a CPU coin. It took a little while for people to react and set up their farms.

[TheSwede75]

How to spot a SCAM COIN:

Launched at non-peak forum hours with no, little or faulty launch time.
Convoluted design for the purpose of discouraging public mining in early blocks.
Quickly dropping block reward to reward 'prepared' miners (aka. inner circle)
Very low starting difficulty that quickly increase (also to reward inner circle and pre/instaminers)
Ultra fast time to market. aka. Appears on trading solutions within a few days. To enable early miners fast and profitable exit.
LOTS of forum PR (spamming altcoin forums)

Feel free to add to this, comment, criticize etc. I would love to build a common definition that people can learn from.