Gridcoin StakeV8 in effect for 4000 blocks: stats and analysis
Hello. As promised, I bring you a few stats from the Gridcoin block-chain. Let me start with the pretty picture:
The picture represents a difficulty plot from recent days. To smooth it a little, 28 blocks are grouped together, average is plotted as a green line and min-max with a pink band. We can see that the value moves randomly around 2, precisely average difficulty since block 1010200 is 1.89312234. The fact that the difficulty is mostly enclosed in 1-3 and does not jump all over the place, shows how stable the chain is. Apart from the two dips around blocks 1010500 and 1012600, where debug log indicated forks, there is nothing more interesting.
Average block spacing was 96 seconds (899.37 bpd) and now is almost the same, 95.4 seconds (906.1 bpd). Average block size is too almost the same, 1.6kB vs 1.4kB.
In the last 4000 blocks, there were 2873 transactions (excluding stake) and 3322 blocks without any user transactions, which is 4.24 transaction per non-empty block on average. Yeah, it is not comparable to performance of bitcoin. Before the transition, there were sightly more transactions, but only a little.
Now for the DPoR stats. From the last 4000 blocks, 42.6% are from investors, and the top CPIDs have 8%, 5% and 2% respectively. Only 48% (1939) blocks have non-zero BOINC reward, this is a decrease from 79% (3160) before v8. Also there were only 14% investor blocks, but the top CPIDs remained the same with 7%, 4% and 2%. This is probably caused by small researchers loosing their magnitude boost, while the top CPIDs are probably more wealthy and therefore did not notice a difference. As you saw in my previous article, we are working on this known problem.
In total, 92866 GRC was paid in research rewards (4000 blocks) and 148076 in interest. The average DPoR payment increased a little from 34.2 GRC to 47.9 GRC, but the average interest payment increased drastically from 6 to 37 GRC. I don't have an explanation to the increase in interest paid.
On the bright side, we have an increase in superblock production. Average time between superblock decreased from 53.4 hours to 35.3 hours. And 96.6% of blocks are from nodes running the latest version.
As a last note, I would like to let you know that a new version of wallet, fixing some issues, will enter a testing stage soon. It is still work in progress, but I can tell that it will probably not be a mandatory one. So stay tuned for updates.
Thank you for the analysis but a question arises:
"I total, 92866 GRC was paid in research rewards (4000 blocks) and 148076 in interest."
Is this a temporary statistic or does this mean that the GRC system that is in place now rewards holded capital more than value generating labor?
You know that if this is happening that it translates once again to the common saying: the rich will get richer and the poor get poorer.
If this is the case, that isn't what I signed up for. I have to deal with that enough in life.
With regards.
I agree. The primary purpose of GRC is to reward research.
PoS interest is causing the danger of mass inflation. DPoR payout is capped at~48k GRC while interest has no cap and with he new protocol creates a far greater percentage of GRC than DPoR. There are many conversations taking place. One in #poll-drafts on slack is discussing removing the reward for PoS altogether along with fixed block rewards and general PoS/DPoR intricacies.
I am in favor of 0%-0.2% interest, depending on how the math works out. I will be expanding on this idea soon... if I find the time.
Unfortunately, many of the current protocols greatly favor investors over crunchers. That is not the system I signed up for either. Keep the conversation going and don't be afraid to speak up. There are no bad ideas.
@bullshark what you just said about why the poor get poorer is out of place and frankly, pretty ignorant. With regards to GRC, I do not think we should build our protocols on such assumptions.
Further:
This is why we target existing Idle Processing Potential. There is no need for mom, dad, and grandma to go purchase a GPU. They already have enough IPP in their house to contribute to BOINC. We need to build protocols which help cover the expenses of directing IPP toward BOINC.
You say you didn't sign up to that system but then you must not have read the Gridcoin home page because it's one of the main selling points. Earn 1.5 % APR.
I do agree that inflation could get out of hand and that there should have been some ratchet where the interest gradually decreases to nothing. Either that or fixed block rewards that reduce over time.
However, In order for it to go to nothing, there would need to be a lot more transactions and the fees would need to be higher otherwise there would be no incentive in keeping the wallet running or to secure the network.
Fixed block reward does not need to reduce over time, cause block per day is constant. It is the "coin*age" interest that is actually increasing: every day, there are more coins and from these coins, even more interest is paid. The "coin*age" interest creates exponential inflation, while BOINC rewards inflate linear, therefore interest is bound to take-over research at some point in the current system.
Yeah, changing to fixed block rewards would be good but it will make it even harder to stake so I thing we need to come up with the best solution for easily rewarding researchers first.
If that is one of the main selling points, we may as well sell up and leave. The research should be the only main selling point. Everything else is extra. Without BOINC, GRC is utterly worthless.
Make an announcement to take away all interest in the next release and see what happens to the value of the coin.
It is exactly the model GNT, SPARC, etc operate on. I don't see why GRC, being the only coin with an established research platform, would suddenly fail like that.
I am not saying we should remove POS, but it should not be the focus in either development or marketing.
Can't argue with that :-)
For the change in interest, I would like to have a poll, please see the task, and the linked thread and add your ideas there.
straight from my response:
Instead, I suggest we slow down a minute and have some discussions in the community hubs before considering an actionable poll. Please join #poll-drafts on slack or any channel really and join the conversation. I have posted an example of what an opinion poll might look like in that channel.
I know it is early, therefore the poll "task" is marked as "preparation", to make it more clear, I now added a "draft" mark too. On the other hand, I think there is enough info in #106 already. And it is now over a a year old.
@bullshark this is a response to you as well.
The thread is over a year old and the majority of it is general questions and discussion, which is great, however most of the discussion before May is restated afterward, so it's a very long and repetitive thread with weeks at a time of dead-air.
Also, the only real proposal in the thread was posted by dc7dgrid's only 11 days ago. The discussion in general talks about a lot of potential options, politics, philosophy, and intent, but only that one post 11 days ago explores potential paths and their consequences.
I think we should give time for more posts like dc7dgrid and wait for a discussion to take place around detailed protocol proposals. The alternative is to rush into something that is generally agreed upon by a loosely structured discussion between a few people.
If we encourage serious and detailed proposals, and if we create several opinion oriented polls before an actionable poll, more people might join the discussion.
What is your idea of a timeline for addressing the PoS incentive protocol? When do you want to create the actionable poll?
The opinion poll is a very good idea. I tried one, but failed to make it clean enough. Please have a look: https://github.com/Erkan-Yilmaz/Gridcoin-tasks/issues/151
An we already had a poll about fixed block rewards, but is was not stated there explicitly (PoSv3 = fixed rewards + some other things), see: https://github.com/Erkan-Yilmaz/Gridcoin-tasks/issues/43#issuecomment-325420116
and if you can, please continue the discussion there
As you know, this discussion has been going on for over a year on GitHub. See GitHub Issue #106. How long before a poll gets made for this topic or do we wait another year?
Hi jringo, maybe you can help clear up some of my confusion.
I've been under the impression that we try to create 48k DPOR coins per day and coin creation through DPOR is metered via the Magnitude Unit. I've also been told there is a total of 115,000 mag.
Therefore, to hit our target of 48k coins per day, we'd need to see an average magnitude unit of 0.417 (48,000 divided by 115,000).
I'm aware the mag unit moves up and down depending on recent DPOR coin creation, but I've been around since March and I've never seen the Mag Unit above 0.3.
It seems we're only creating ~50% of our target. Why?
Also, what's the point of a moving mag unit? If we haved a fixed magnitude unit, wouldn't we create the target amount of DPOR coins (on a long-term basis)?
Mag unit also (wrongly) depends on the total magnitude in superblock. I think the rest of the function is not right as well.
I'm not sure. I think @ravonn or @tomasbrod would be able to give you a much better answer than I would = )
I have to say that i find the ratio also strange. I think its to much in favour of interest. I do not know if this was always this way. Its the first time i come across numbers for this - but to my understanding it is 6 times more interest after the update. This means the BOINC contribution is valued less. Which feels wrong from my point of view.
Yes, it is strange. But remember, that the AMOUNTS paid were not changed. StakeV8 only indirectly affects, WHEN you will be paid the interest or BOINC reward. If you average it over long time, it should be the same (assuming no transactions happen, which destroy interest, etc).
The formula for interest is and was
Tx.Value*Tx.Age*InterestRate, which is also called Coin*Age interest. (assuming correct units).You have to have a lot of capital to purchase mining hardware so the "rich will get richer" applies there as well. The poor get poorer because they spend their money unwisely and don't invest a portion of income.
Are you suggesting the reward for investors is removed?
"You have to have a lot of capital to purchase mining hardware so the "rich will get richer" applies there as well. " Well that rewards labor and effort so that is a plus in my book, my worry is that in the future, running BOINC wont be the primary means to generate this currency.
BOINC is the only reason GRC has value. If the miners leave, the investors will see their investment plummet to zero.
So if we were running on DPoS like the STEEM blockchain and the only rewards were for research do you think the coin would be valuable? Or would people just sell everything they earn?
Sure - the real value that can drive GRC in the long run is allowing project admins to buy, or encourage, research on their project. This is why competitors like GNT, SPARC etc have such a relatively high value despite having no functional product.
I would love to see Gridcoin ported to Graphene.
I think @cm-steem has mentioned in the past. Projects do not want to pay GRC to the volunteers because of tax/legal concerns.
Maybe a third-party donation site could work. A one where people can donate any crypto to a project and it automatically gets converted to GRC and rained on the researchers.
I too, would like to see either Gridcoin or a new BOINC coin on Graphene. The more I read about PoS, it seems to be much less secure than PoW of DPoS.
That is a fair concern. The projects could just rain the GRC though, which would probably not classify as paying the crunchers? I am guessing here, and the law would probably be different depending on where in the world this were to happen...
Thank you for the detailed statistics!
Thanks for putting together this useful information. It's good to know that the blockchain is stable and that there are more updates coming soon :-)
Is isn't. At all. Everything is being sugarcoated. The NN is currently eating POR rewards and the POR rewards are underpaid as the mag unit is broken, which several people highlighted in June.
Thanks for good stats.
I mined 15 GRC with iMac mostly and now the rate of mining is 2 GRC per day
Hib@tomasbrod thanks alot for this worthy information! I already had the impression that everything is running quite ok currently and this is stressing this as well. Thanks for your hard work Tomas !
I would assume this is due to most interest payment having been tacked onto POR rewards pre V8. Now, as miners stake much less frequently, they no longer generate as many small interest payments which pushed up the average.