So when the change was made from KGW to DGW3 (around block 124000) we start seeing clever solve REALLY low difficulty blocks.. because as you point out, DGW3 makes very extreme changes to the difficulty. Here is where I kind of disagree and am hoping others can step in and shed some light. In my opinion, the algorithm should have nothing to do with profitability. The market will determine that. Notice how the MAX difficulty of the blocks clever solves is tightly correlated to the bittrex NLG price? That is no coincidence. The super-low difficulty blocks are just a bonus. The miners are the ones who should be worried about profitability, not the algorithm.. Assuming the new algorithm allows dedicated miners to get their fair share of the rewards (right now clever is robbing them, so they go elsewhere) I would expect miners to come back to NLG. In a fair market (no massive jump pools influences), demand drives price, price drives mining hashrate. IE: If there is demand for the NLG, the price goes up and this will attract new miners to NLG.. thus the hashrate goes up as those miners compete for the coins to sell. I'll try to create another chart soon through december 1.. maybe this time with clever solved blocks as blue, everyone else's solved blocks as orange. I expect to see 2 things. Orange on average will be higher than blue.. and bioMike's efforts over the past month will drastically reduce the difference between the highs and the lows. If we then add a secondary axis charting a block gap moving average, we should see the strong correlation you describe between a tighter difficulty range and lower block gaps. Sorry for the long-winded post. I think I've gotten a little carried away  --Mark Markanth, great post, mate. I just wanted to clarify something in terms of the bolded sentances above. With regards to clever, the algo change is specifically about profit. We need to limit the period of mining profitability and we need to do it faster than we currently do. It is a profits game for them, so we need to take that away from them. I like your graphs, but I'm not sure if you're looking at them correctly. The correlation between difficulty and price might be getting skewed, if I'm reading your post correctly. Correct me if I'm wrong(it's 4am and the baby is fussy). The reason you're seeing a direct correlation is because clever is causing it. Clever isn't a traditional miner that stores their coins and spends a little here and there. They instadump their coins for BTC/LTC. I'd suspect you'd see a trailing correlation if you took a closer look. Clever mines a super low series of blocks, and then the price drops shortly after. Price isn't driving difficulty in our case right now, but rather difficulty driving price. Clever mines easy blocks, clever sells easy blocks. If it wasn't for the dedicated miners propping up the nethash, clever would probably own the whole chain and the price would be much lower than it is now. You are 100% correct though that in a free market environment, price would drive difficulty via more dedicated miners. We just aren't in a free market right now. I would suspect you are correct with the orange/blue observation. The bottom of that graph would be very blue. If you have the time, please chart it. Anyone else could as well. You just need two data sets, one for clever's blocks, and one for everyone else. -Fuse |
|
|
|
Question about effects of hashrate on blockgaps. So we currently have about 5 GigaHash of dedicated mining of NLG.. and theoretically clever can jump in when the difficulty is low with up to 320 GigaHash to mine some quick blocks although last communication from Terk said they were only dedicating 1/8 of their hash power . Based on above, on a test net, if the dedicated steady hashrate is 5MegaHash can we see the same effects that clever has on the chain by jumping in on low blocks with 320MegaHash? I understand the difficulty would be totally different but the behavior of the algorithm would be the same no? Hoping someone can educate me --Mark 5GH? I've always seen it between 8 and 12GH pre-DGW3 and post-DGW3 . Hardcoreminers reports as such, as well. Current "Getmininginfo": "networkhashps" : 10754201215 or 10.7GH. Am I missing something? I know that the numbers are not 100% true, but for it to be 50% of getmininginfo actuals? Educate me as well, please I think he was referring to the portion of the network that isn't clever. -Fuse |
|
|
|
Question about effects of hashrate on blockgaps. So we currently have about 5 GigaHash of dedicated mining of NLG.. and theoretically clever can jump in when the difficulty is low with up to 320 GigaHash to mine some quick blocks although last communication from Terk said they were only dedicating 1/8 of their hash power . Based on above, on a test net, if the dedicated steady hashrate is 5MegaHash can we see the same effects that clever has on the chain by jumping in on low blocks with 320MegaHash? I understand the difficulty would be totally different but the behavior of the algorithm would be the same no? Hoping someone can educate me --Mark The problem with the current algo is that it overshoots the "sweet spot" on profitability vs difficulty. Regardless of how much hashrate clever throws at NLG, if that sweet spot is reached faster, and it isn't overshot to the point of our current difficulty swings, we're good. The problem with DGW3 is that it takes longer to retarget to the sweet spot, and then you have 5-10 blocks that are too low in difficulty, so the algorithm overshoots the sweet spot by leaps and bounds. It leaves us in the 1000 difficulty range and we struggle to find the next block. So it takes forever and a day, and then the difficulty plummets way below the sweet spot, rinse/repeat. Hashrate aside, the problem we need to solve is making the difficulty reach that sweet spot without overshooting or taking to long, and then keeping it there. I know I'm beating a dead horse, but our DIGI tests have done that. However, to truly test it's effectiveness, we need a profitability algo like clever uses so we can see the jump in and out. Or a really smart community member that could provide some advanced math on profitability so we could guess at in and out points. -Fuse |
|
|
|
|
Damnit, /GJ... making me learn GO. lol
Let me see if I can figure this out.
-Fuse
|
|
|
|
This sounds great, but if we're going to do a new algo change, I think it's best to have proof.
With proof I mean: maths. Input/output of the algorithm calculation. Throwing of 320GH/s (the amount of hashing power that cleverminig currently has).
This is not something that can be tested in a test network, unless you happen to have 320GH/s at your disposal..
When the sim is released, it will be very easy to add another algorithm such as digishield.
First and foremost, everything I bring up is with the utmost respect. When I lose faith in the team, the community will know it because I will flat out say it. I am primarily posting this because I feel lost in trying to help this coin, and I want guidance. That being said, the problem with your math is that clever will never throw 320GH at NLG. If they were going to increase their mining it would be because NLG mining is profitable. If NLG mining is profitable, we'll have more miners, and we'll have a higher difficulty. Mining is all about proportions. Clever will never increase their proportion if it's not profitable. If they did, our proportion would probably increase as well. But you forget something, too. Clever won't stick around if they are cut off the chain with an algo change. If you limit the number of blocks clever can mine before it pushes profitability out of reach, which DGW3 doesn't do, you shut clever down. 1000GH won't mean anything if you only get 3 blocks before it shuts down your profitability. Also, throwing 320GH at NLG is basically saying that every other coin, including LTC are inferior to NLG at that point, and I'm pretty sure we'll have a lot more on our plate than clever at that point. Think terahashes. But at that point, you could implement a standard LTC algo and call it a day. Anyone with enough hashpower can disrupt a blockchain. That's a fundamental flaw with mining. 24Kilo and I almost burnt each others houses down in a heated argument over this. Hashrate is king. No algo change is going to change that. The only way you could change that is implementing code that said that only X amount of blocks can be mined from a certain address in a certain amount of time, which is back to the centralized mining debate. The other approach is POS, and that's a crap-shoot at best. I've never seen a POS coin that didn't have forking issues, or even worse. So yes, 320GH is a big scary monster that can trample our town like Godzilla on a Sunday stroll through the park. However, that 320GH is governed by profit, and an algo change like DIGI would minimize that profit in a single git pull. There is at least a year of data that can be mined from various coins that made algo changes. There's testnet data that can be put together. There's enough information out there that could be used to make an educated decision. My question is what is it that the dev team wants? If you want a 320GH testnet, then tell the community you want it, and we'll work on it. If that's what it's going to take to convince the team of a proposed change, we need to know that. We're flying blind here, mate. We need input as to what direction to go to better support you and your team. -Fuse |
|
|
|
to all the guldencoin lovers have a good weekend! CONGRATZ Fuse!!  i was forget that Congratulations on your second daughter Fuse! On the mining front it looks like clevermining is now under 50% at these prices. Also buyers will come back once the simulator and algorithm change takes place, it looks as though the market is starting to feel this is only going to take place next year. I also cannot with a good conscience vote for /GeertJohan until the simulator has been completed so hopefully that happens this month. I thought investeerder was a bit of a asshole but looks like he will be right and all of us wrong.  I don't think it's a big deal that we have to wait until next year for the simulator and algorithm change if the price stays as is, it's just a major issue if it goes over 500. No... I think your first impression was probably right. The simulator and the algorithm change aren't mutually exclusive. So even if /GJ can't finish the simulator before next year, we could still push forward with an algorithm change. Like I mentioned previously, my team has tested DIGI on a testnet and it performs very well. We made a slight change to accommodate the longer block times, but it performs much better than the current DGW3 implementation. I could put up a git pull request right now and we could move forward. But that's not up to me. We need the dev team to back a change. The simulator will only take us so far, and from what /GJ said, it seems like it's hard programmed to only do traditional and DGW3 algos in the first release. IMO, we already know what that looks like. Other algorithms should be the focus... not what we already know doesn't work. This is why my team went ahead and decided to do independent testing. We need a solution to the problem at hand. If the dev team wants to make a change to DIGI, I'm sure all of this clever nonsense would be behind us and we could move on with bigger and better things. If they don't want DIGI, we need a clear indication of what expectations are, and my team will continue to test solutions. We can lead a horse to water... -Fuse |
|
|
|
I also gave a little birthday present  have fun my friend I thought there was a second person . Thank you very much Strataghyst! It's greatly appreciated! -Fuse |
|
|
|
Well, welcome to this lovely world then Hope! Here is your first Guldencoin wallet: GaHopexjuAgEkqn7BzREsvNrPK8s2a4Vxf
I've put in some coins for you. Maybe later you remember you got your first Guldens from "Uncle LTEX"
I've sent the private key to daddy... Thank you for this very gracious gift. I'm not sure if there was another amount sent by another person, but if so, thanks to the other possible person. I be sure to let Hope know that Uncle LTEX bought her her first car when NLG hits $1. After 3 days in the hospital, everyone is now home and getting rest. I'll be home until the 15th, so I'll be pretty active here in my downtime. If anyone needs me, shoot me a PM. Cheers, Fuse |
|
|
|
|
I just want to give a very gracious thanks to all of you who gave well wishes. This is our second daughter. Our first, who is now 4yo, is named Liberty. This baby, born 14:54 pst, was 6lbs 7 oz and 18". Her name is Hope. Mom and the baby are happy and healthy, and dad is exhausted from trying to sleep in these very uncomfortable chairs lol.
Additionally, thanks to everyone again for your understanding with the pool issues today. A special thanks to those that messaged me here and through other sources to let me know that things weren't right.
I'll be home tomorrow, and I'll drop back in and see how things are going.
-Fuse
|
|
|
|
|
It looks like blocks are processing properly again. Initial thoughts are that there was an outage with my hosting provider, but I will need to get to a computer to really dig into the logs.
Again, I am truly sorry for not being on top of this sooner. It always seems like thing go boom when you can't get to them lol
-Fuse
|
|
|
|
|
First let me apologize for not getting to this issue sooner. I am currently sitting in a hospital room with my wife and newborn child. I dont have a computer with me so I was a bit out of the loop.
I have restarted the services on the pool but I'm working from my phone so I cant test functionality. Can some test the pool snd let me know. If needed, I can track down a computer and get this fixed in the nect couple hours.
-Fuse
|
|
|
|
I just wanted to post this in case anyone hadn't seen it yet.  With more hashpower on the network dedicated to an actual pool, the longest block time is 50% of what it was previously, and the difficulty max has been much less. So there are two points to be noted here. First, more miners do help the network. We knew this all along, and it's a fundamental principle of the blockchain- more miners=more stable chain. Secondly, it proves the effectiveness of multipools against our current DGW3 implementation. The swings in difficulty are too great, and they are causing clever to be able to capitalize on too many low difficulty blocks before shooting the difficulty to the moon. Again, something we all knew. I bring this up because my team has been testing DIGI over the last couple weeks on a testnet. Initial results are very promising. Retargeting is occurring quickly and efficiently. We are seeing very good retargeting when large hashes come and go. We've also found out some other key dynamics that are working against us in some cases. I will let 24Kilo comment on those. However, at this time I am standing behind my initial assertion that DIGI is going to fix this issue. When we get closer to releasing data, we'll provide the charts and darts for community review. -Fuse |
|
|
|
Really, I do not think adding this latest merchant (mentioned 3 posts above) is helpful to the Guldencoin cause and future.
Or maybe I am wrong and have been expecting too much of Guldencoin. Anyway, I still hope NLG will become a serious digital currency used in the Dutch economy.
Guldencoin being associated with merchants that have sh.. in their brandname, will not do any good for NLG being taken seriously.
I'd disagree. When there is full-fledged adoption of the coin, there will be far worse people using it than someone who sends a joke gift. Adoption is adoption is adoption. What's to say this store's owner doesn't take the coins they make from any sales and goes down to the pub and buys a beer with the coin. If a closed-loop system is the end goal of NLG, as it should be for any crypto, adding another vendor, regardless of how obscure it is, just helps to make that loop that much easier to close. Hell, I was almost tempted to buy a gift for clever when I saw this. -Fuse |
|
|
|
As someone who has been on the receiving end of many a CS-gas canister training sessions in the military, it's not a fun experience. It's no joke if you want to stop a group of people quick. Be careful with these on a windy day too, as the gas will carry for a long distance. It almost hangs in the air sometimes. Even on the outskirts of the cloud, people will feel the effects.
If I could get these shipped legally to the US, I would be all over it for SHTF.
-Fuse
hey, that is true! these are no toys that you wanna prank your friends with! use only if necessary ! oh, CS is illegal in USA ? If these are the "pull the pin, and throw the canister" type of CS, I've never seen them for sale anywhere here. If it's legal in the US, I might pick some up. Cost with shipping to US? Escrow? Pictures of actual product? -Fuse |
|
|
|
|
As someone who has been on the receiving end of many a CS-gas canister training sessions in the military, it's not a fun experience. It's no joke if you want to stop a group of people quick. Be careful with these on a windy day too, as the gas will carry for a long distance. It almost hangs in the air sometimes. Even on the outskirts of the cloud, people will feel the effects.
If I could get these shipped legally to the US, I would be all over it for SHTF.
-Fuse
|
|
|
|
NLG on cryptsy opens the door to other multipools that use the cryptsy APIs. Let's focus on multipool mitigation before we throw the tuna into shark infested waters. -Fuse |
|
|
|
Best time to buy is now or in 2 months time just before the algo change takes place. Plenty other projects coming to keep more people coming back for more while we down our sorrows on the blockchain.
Algo change take another 2 month? Why so long? mean we have to wait 3 month for iOS to apple... maybe delay tactic from team for another announcment? That's just a wild guess of investeerder. The only thing we can say for sure is that the simulator is almost finished. ...and my team is already testing algorithm changes. -Fuse |
|
|
|
Can't we block him ourselves with the low orbit ion cannon targeting his ip and guldencoin rpc port. That's a no-go area... don't go there... that is a direct assault on the site and could get you in a lot of mess in many countries (even in the Netherlands). Even among the majority of IT people DDOS is NEVER an option. I have to agree with BioMike on this one. As much as I hate what clever is doing to this coin, going down the DDOS road is not something I would ever consider. -Fuse |
|
|
|
|
I have great respect for the dev team and how they have built up this coin. I often tell people, and I've said in this thread as well, that they are the only team I've dealt so far with that does not over-promise or under-deliver.
That being said, when the simulator is done, I really hope the dev team will be ready to take community recommendations and move on an algo. I've pushed from the beginning to move to DIGI. Terk reaffirmed my thoughts with his comments in that PM... obviously he thinks DIGI will knock his pool out of the chain. I was swayed by one of my team members that BTM was the way to go, but we found out shortly there-after that BTM was borked, and it failed miserably. Custom algos are great and all, but I know that DIGI works, and my team has had some good initial findings with our testnet.
However, what worries me is the fact that clever is still mining NLG, even when it can't possibly be profitable. The storage/Bittrex address shows that. So why continue to mine, other than to hurt the coin on an insta-dump later on? My main concern is that regardless of the effectiveness of whatever algo we chose, Terk will use his enormous pool of hashrate to continue to rape NLG for the hell of it. I'm starting to think regardless of what he mines, he can absorb a considerable amount of profit loss to "play" with coins. NLG is just a bug under his shoe.
One thing we could do that would cut clever out of the picture is ask Bittrex to suspend trading. If that's where the coins are in fact ending up, maybe just pull the rug out from under him. Or ask the Bittrex owner to remove the API calls for NLG. Without a way to gauge exact profitability, it would be a solid argument to his miners that he is in fact using their hashes to ruin a coin, rather than making him money.
And yes, until we hear from Terk, it is all just speculation. But I'd seriously doubt he'd say anything when he knows he's the reason for 3 months worth of community angst. How could he seriously explain why he's doing what he's doing... "oh, I just didn't know"... yeah, ok.
-Fuse
|
|
|
|
|
Show Posts
