NA

[Litesire]

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.

[1714895399]

1714895399

[1714895399]

1714895399

[1714895399]

1714895399

[1714895399]

1714895399

[1714895399]

1714895399

[Dutchyyy]

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.

[Sharkzz1]

any of you guys have problems with bittrex? want to buy quite some more for my collection right now before it bursts to 1$ 

[ny2cafuse]

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

[/GeertJohan]

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

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.

[ny2cafuse]

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

[/GeertJohan]

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

Hey,

You're making good point here. I understand that we need change, but I simply don't want to make the same mistakes with a new algo.
Just reading the code and performing simple tests simple isn't enough..

The sim is very close to being finished. If you want to help, please go ahead and start converting digishield from C to Go for the simulator. It's not very hard to do for anyone with code experience.
The algorithm must comply to a Go "interface" that is defined like this: https://gist.github.com/GeertJohan/46c30b7d8124476ef100

It would help A LOT if the digi is converted to Go so maybe we can have this at launch time in the simulator, and then we CAN throw 320GH/s at it without even having 320GH/s

[ny2cafuse]

Damnit, /GJ... making me learn GO. lol

Let me see if I can figure this out.

-Fuse

[/GeertJohan]

Damnit, /GJ... making me learn GO. lol

Let me see if I can figure this out.

-Fuse

Awesome! Let me know if you need any help.

[markanth]

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

[BioMike]

Damnit, /GJ... making me learn GO. lol

Let me see if I can figure this out.

-Fuse

If it is funds you need to hire a dev to do this... count me in.

[ny2cafuse]

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

[BioMike]

NOMP pool software claims clever uses their software (I guess with their switching algo's).

[Halofire]

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

[ny2cafuse]

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

[Halofire]

Ah ok, Thanks Fuse.

[markanth]

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

From my point of view.. yes and no.. see graph below.. The main graph is the difficulty of every block clevermining has solved between August 17th (when clever started with NLG) and October 31st. (old graph).  x-axis: block number, y-axis: difficulty.   The small graph in the upper right corner is the bittrex NLG price for the same time period.  



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

          

[BioMike]

Markanth, thank you for your insight. I like the graphs. You can already see my effords in here, also the period where I needed to stop for some time. My point was always to raise the lows.

[veertje]

The super-low difficulty blocks are just a bonus.  The miners are the ones who should be worried about profitability, not the algorithm..
          

See how Clever finds way more blocks since the algo change then before, compared with even higher pricelevels in the past. Those low dif blocks are a bonus indeed for Clever, but in general they get more blocks also, because big miners have left, because of lower profits, due to the GH/s swings of Clever. And that is caused by the algorithm that Clever can abuse for their profit, that big miners have left. Now Biomike compensates, but with a good algo big miners will come back, because Clever can't abuse it anymore in their favor. This algo atm is helping Clever and that is no good in two ways: Big swings in diff and big dedicated miners left because of that.

With KGW there where swings as well, but that had most to do with pricechanges and profitability. Maybe GH/s swings also by Clever already then. but big dedicated miners where still in though at that time. With a new good algo they will come back.

[bram_vnl]

https://twitter.com/ZackNormandin/status/541738586144780288
https://twitter.com/MTC_Marinecoin/status/541799335542136832