Pages:
Author

Topic: [XMR] Monero Speculation - page 95. (Read 3313076 times)

legendary
Activity: 2730
Merit: 1288
July 13, 2020, 08:52:45 AM
I got the perfect title for that chart:  The Last Laugh.  Grin

I wrote down 2102 page of [XMR] Monero Speculation for future quoting. I just wonder how long will take until the quote will happen.

Two more charts. The bottom point of this story is that every trend at some point break.



newbie
Activity: 3
Merit: 0
July 12, 2020, 09:00:45 PM
Well. A groups placement on the map is not that important, as it's just indicating the general area.
Firstly, it's interesting that the US, per head of population (one node per million) actually has less than many European countries.  Germany is 3 per million, Finland is 12 per million - even little Iceland has two nodes - one per 150,000 people.

Concentration in areas or cities is more interesting.  In the USA California on its own has more nodes than most countries (56) - more than the whole of China or Russia - which is perhaps not surprising.

Lets not forget that you can have a node anywhere that servers are rented, I had one in one of the Slavs (forgot which one) because it filled in a gap.

Also, don't forget that those maps does not show a realistic view of actual nodes out there.
For example. None of my nodes is on this map or the map from xmr.to

Yeah, I gathered it was not exactly realistic when I saw 180 nodes in a lake in Germany... Wink

Goals ^^^

Well. A groups placement on the map is not that important, as it's just indicating the general area.

This map doesn't show IP addresses of the nodes, so it can't be verified, or be used as a list of usable remote nodes.
legendary
Activity: 4004
Merit: 1250
Owner at AltQuick.com
July 12, 2020, 06:15:06 PM
Firstly, it's interesting that the US, per head of population (one node per million) actually has less than many European countries.  Germany is 3 per million, Finland is 12 per million - even little Iceland has two nodes - one per 150,000 people.

Concentration in areas or cities is more interesting.  In the USA California on its own has more nodes than most countries (56) - more than the whole of China or Russia - which is perhaps not surprising.

Lets not forget that you can have a node anywhere that servers are rented, I had one in one of the Slavs (forgot which one) because it filled in a gap.

Also, don't forget that those maps does not show a realistic view of actual nodes out there.
For example. None of my nodes is on this map or the map from xmr.to

Yeah, I gathered it was not exactly realistic when I saw 180 nodes in a lake in Germany... Wink

Goals ^^^
legendary
Activity: 2604
Merit: 1748
July 12, 2020, 05:54:18 PM
Firstly, it's interesting that the US, per head of population (one node per million) actually has less than many European countries.  Germany is 3 per million, Finland is 12 per million - even little Iceland has two nodes - one per 150,000 people.

Concentration in areas or cities is more interesting.  In the USA California on its own has more nodes than most countries (56) - more than the whole of China or Russia - which is perhaps not surprising.

Lets not forget that you can have a node anywhere that servers are rented, I had one in one of the Slavs (forgot which one) because it filled in a gap.

Also, don't forget that those maps does not show a realistic view of actual nodes out there.
For example. None of my nodes is on this map or the map from xmr.to

Yeah, I gathered it was not exactly realistic when I saw 180 nodes in a lake in Germany... Wink
newbie
Activity: 3
Merit: 0
July 12, 2020, 01:01:06 PM
Firstly, it's interesting that the US, per head of population (one node per million) actually has less than many European countries.  Germany is 3 per million, Finland is 12 per million - even little Iceland has two nodes - one per 150,000 people.

Concentration in areas or cities is more interesting.  In the USA California on its own has more nodes than most countries (56) - more than the whole of China or Russia - which is perhaps not surprising.

Lets not forget that you can have a node anywhere that servers are rented, I had one in one of the Slavs (forgot which one) because it filled in a gap.

Also, don't forget that those maps does not show a realistic view of actual nodes out there.
For example. None of my nodes is on this map or the map from xmr.to
legendary
Activity: 3836
Merit: 4969
Doomed to see the future and unable to prevent it
July 12, 2020, 11:24:06 AM
Firstly, it's interesting that the US, per head of population (one node per million) actually has less than many European countries.  Germany is 3 per million, Finland is 12 per million - even little Iceland has two nodes - one per 150,000 people.

Concentration in areas or cities is more interesting.  In the USA California on its own has more nodes than most countries (56) - more than the whole of China or Russia - which is perhaps not surprising.

Lets not forget that you can have a node anywhere that servers are rented, I had one in one of the Slavs (forgot which one) because it filled in a gap.
legendary
Activity: 3976
Merit: 1421
Life, Love and Laughter...
July 12, 2020, 09:07:08 AM
he bi yearly forks were a net negative in my humble opinion.  I was glad the consensus was to stop them after the final try at RandomX.  And that we felt the need to call the hard forks "network upgrades" was sort of proof that we knew we were doing something that is anti-decentralization for the tradeoff of defeating ASICS.  As much as I loved taking out bitmain on that first algorithm tweak, it was a bit of a defeat to keep doing that.

We basically switched to the term scheduled network upgrade, because people associated the term hard-fork with a chain split that would create a new coin (and thus holders would be eligible for some kind of dividend). There used to be tons of questions on Reddit whether users would be getting some kind of dividend / new coin.

Wait so are we not doing it biannually anymore? I do like the terminology “network upgrade” too.

I believe idea is to do upgrades when they are ready and that is about twice a year. So nothing really changed. In the past almost all upgrades needed to be hardforks. Last one was not. There is misunderstanding in crypto community that Monero was doing hardforks to change mining algo. That is not true. Monero was doing hardforks to upgrade its protocol with new things, while doing it they also could tweak mining algo a bit. It was just an opportunity and if that opportunity would not exist Monero would have ASIC miners before RandomX.  Next upgrade will be hardfork again.   At the end the diferences between hardfork upgrade and not hardfork upgrade are that when hardfork you need to upgrade your wallet. So that makes whole network with the latest software and that is good. Since last upgrade was not hardfork we can still use v0.15 wallets.


And some long term hopium from reddit



I got the perfect title for that chart:  The Last Laugh.  Grin
legendary
Activity: 2730
Merit: 1288
July 12, 2020, 06:53:50 AM
I believe idea is to do upgrades when they are ready and that is about twice a year. So nothing really changed. In the past almost all upgrades needed to be hardforks. Last one was not. There is misunderstanding in crypto community that Monero was doing hardforks to change mining algo. That is not true. Monero was doing hardforks to upgrade its protocol with new things, while doing it they also could tweak mining algo a bit. It was just an opportunity and if that opportunity would not exist Monero would have ASIC miners before RandomX.  Next upgrade will be hardfork again.   At the end the diferences between hardfork upgrade and not hardfork upgrade are that when hardfork you need to upgrade your wallet. So that makes whole network with the latest software and that is good. Since last upgrade was not hardfork we can still use v0.15 wallets.


I suppose I let me hopefulness get ahead of me.  I see the "network upgrades" as something that fly in the face of decentralization, and thought since we were not (hopefully) going to stop adjusting for ASICS we might stop forking.

I don't know if I want to open the blocksize, and governance cans of worms in the same day though Wink haha.

Part of the next upgrade is CLSAG. It will reduce transaction sizes and transaction times for if I remember right about 20%.  You can not do that without a hard fork. It is simple. Cant be done. We will have 2 chains one with 20% savings and one without savings. Nodes and miners will decide which to follow.  Centralisation in Monero development is called MORE PRIVACY, SMALLER TRANSACTIONS and FASTER TRANSACTIONS.
legendary
Activity: 2604
Merit: 1748
July 12, 2020, 04:48:00 AM

As to the nodes...  This website has been up for a long time:
https://monerohash.com/nodes-distribution.html

I would love to see a graph.  I am not sure what impact RandomX had on node count.
Thanks for this…

I just got lost in looking at node spread and finding the Monero network’s secret submarine centres.

Firstly, it's interesting that the US, per head of population (one node per million) actually has less than many European countries.  Germany is 3 per million, Finland is 12 per million - even little Iceland has two nodes - one per 150,000 people.

Concentration in areas or cities is more interesting.  In the USA California on its own has more nodes than most countries (56) - more than the whole of China or Russia - which is perhaps not surprising.

There are a handful of cities which have huge amounts of nodes listed in one dot, but it's probably just geographically untraceable nodes lumped together.

Kassel in Germany has a dot in the middle of a lake with 180 nodes, Paris has one in the middle of the Seine with 126 nodes, London has a 64 node dot in the Thames and Toronto has two dots, one of 83 nodes is in the middle of the harbour.  

Maybe all those boating accident stories have more than a grain of truth to them Wink
legendary
Activity: 2268
Merit: 1141
July 12, 2020, 03:01:05 AM
he bi yearly forks were a net negative in my humble opinion.  I was glad the consensus was to stop them after the final try at RandomX.  And that we felt the need to call the hard forks "network upgrades" was sort of proof that we knew we were doing something that is anti-decentralization for the tradeoff of defeating ASICS.  As much as I loved taking out bitmain on that first algorithm tweak, it was a bit of a defeat to keep doing that.

We basically switched to the term scheduled network upgrade, because people associated the term hard-fork with a chain split that would create a new coin (and thus holders would be eligible for some kind of dividend). There used to be tons of questions on Reddit whether users would be getting some kind of dividend / new coin.

Wait so are we not doing it biannually anymore? I do like the terminology “network upgrade” too.

The idea is to merely perform a scheduled protocol upgrade when it is needed (i.e. there are significant consensus changes). I guess, in practice, this implies forking every nine to twelve months. The six months forks were, in my opinion, too much of a strain on the ecosystem, even if they don't include a PoW change.
legendary
Activity: 3766
Merit: 5146
Note the unconventional cAPITALIZATION!
July 11, 2020, 10:04:51 PM
Good stuff cAPSLOCK, some food for thought.

Adaptive blocksize has not really been tested with massive volumes for long periods, so I guess it has seemed chain bloat is not issue not worth worrying about.  That doesn't mean, however no one has been thinking about it at the dev level.  Yes, it's a more complicated tech, so block size being larger sort of makes sense.

TBH personally I don't know much about how many nodes there are, but the blockchain is 60-odd Gigs, it doesn't seem 'that' large after 6 years.  But then if you look at orders of magnitude increases in transactions, I don't know - has anyone modeled growth v blocksize at different tx volume levels?

At what level is it a possible problem - and are nodes already pitifully small in number?

Ref: current activity /price, it does look different. On Polo XMR is trading five times what BSV or BCH are, so it's refreshing. OK, I know that isn't the case everywhere of course - but it's a good sign.  I did actually go there to play with some alts and I sold a few XMR, expecting to buy back in the (recently) 'inevitable' fall back when BTC makes a move - but honeypony seems to be doing it's own thing. 

I sort of hope I don't get a chance to get these back - but let's see if Monero has decided it's time to properly breakout.  It has to happen sooner or later.

As to the nodes...  This website has been up for a long time:
https://monerohash.com/nodes-distribution.html

I would love to see a graph.  I am not sure what impact RandomX had on node count.
legendary
Activity: 2604
Merit: 1748
July 11, 2020, 04:31:56 PM
Good stuff cAPSLOCK, some food for thought.

Adaptive blocksize has not really been tested with massive volumes for long periods, so I guess it has seemed chain bloat is not issue not worth worrying about.  That doesn't mean, however no one has been thinking about it at the dev level.  Yes, it's a more complicated tech, so block size being larger sort of makes sense.

TBH personally I don't know much about how many nodes there are, but the blockchain is 60-odd Gigs, it doesn't seem 'that' large after 6 years.  But then if you look at orders of magnitude increases in transactions, I don't know - has anyone modeled growth v blocksize at different tx volume levels?

At what level is it a possible problem - and are nodes already pitifully small in number?

Ref: current activity /price, it does look different. On Polo XMR is trading five times what BSV or BCH are, so it's refreshing. OK, I know that isn't the case everywhere of course - but it's a good sign.  I did actually go there to play with some alts and I sold a few XMR, expecting to buy back in the (recently) 'inevitable' fall back when BTC makes a move - but honeypony seems to be doing it's own thing. 

I sort of hope I don't get a chance to get these back - but let's see if Monero has decided it's time to properly breakout.  It has to happen sooner or later.
legendary
Activity: 3766
Merit: 5146
Note the unconventional cAPITALIZATION!
legendary
Activity: 3766
Merit: 5146
Note the unconventional cAPITALIZATION!
July 11, 2020, 11:42:52 AM
I believe idea is to do upgrades when they are ready and that is about twice a year. So nothing really changed. In the past almost all upgrades needed to be hardforks. Last one was not. There is misunderstanding in crypto community that Monero was doing hardforks to change mining algo. That is not true. Monero was doing hardforks to upgrade its protocol with new things, while doing it they also could tweak mining algo a bit. It was just an opportunity and if that opportunity would not exist Monero would have ASIC miners before RandomX.  Next upgrade will be hardfork again.   At the end the diferences between hardfork upgrade and not hardfork upgrade are that when hardfork you need to upgrade your wallet. So that makes whole network with the latest software and that is good. Since last upgrade was not hardfork we can still use v0.15 wallets.


I suppose I let me hopefulness get ahead of me.  I see the "network upgrades" as something that fly in the face of decentralization, and thought since we were not (hopefully) going to stop adjusting for ASICS we might stop forking.

I don't know if I want to open the blocksize, and governance cans of worms in the same day though Wink haha.
legendary
Activity: 3836
Merit: 4969
Doomed to see the future and unable to prevent it
July 11, 2020, 11:41:17 AM
he bi yearly forks were a net negative in my humble opinion.  I was glad the consensus was to stop them after the final try at RandomX.  And that we felt the need to call the hard forks "network upgrades" was sort of proof that we knew we were doing something that is anti-decentralization for the tradeoff of defeating ASICS.  As much as I loved taking out bitmain on that first algorithm tweak, it was a bit of a defeat to keep doing that.

We basically switched to the term scheduled network upgrade, because people associated the term hard-fork with a chain split that would create a new coin (and thus holders would be eligible for some kind of dividend). There used to be tons of questions on Reddit whether users would be getting some kind of dividend / new coin.

I only argued for the term "scheduled Update" for years (way before monero reddit) but its nice to see it finally adopted no matter what term they used.
legendary
Activity: 3766
Merit: 5146
Note the unconventional cAPITALIZATION!
July 11, 2020, 11:36:40 AM
^ Great post cAPSLOCK, thanks.

You can now pay Lightning invoices directly from https://xmr.to/
https://old.reddit.com/r/Monero/comments/ho5ibv/xmrto_added_lightning_network_payments/

Well damn this is cool Cool

I forgot, is there any work in progress for a “LN” type of network for Monero?

And cAPS, what makes you think running nodes will become more centralized than what Bitcoin already is? Or is that not what you’re saying?

Well I take a conservative stance in the blocksize debate.  I think full node centralization is a VERY STRONG net negative.  I think the end game in that scenario is you have to have a network pipe, storage space and computation resources that are really only currently owned by Governments and Google/Amazon size corporations.  Once any blockchain gets to this point it will become possible for the nodes to collude to reverse transactions etc.  And we have YEARS of proof it will happen.  Even with BTC, the most decentralized blockchain in existence, we came damn close with Binance just a little while back.  I am not sure if they could have done it...
Code:
[3:43:01 AM] Vitalik Buterin: ok can you guys stop trading
[3:43:05 AM] Tristan D’Agosta: Okay

Currently BSV is the existential leader in this camp design wise.  Not only do they cheer on node centralization, but they actually also imply if not right out say that they think censorship is a feature, not a bug.  I think BSV is a steaming pile of shit for other obvious reasons but the experiment of an unbounded block limit is at least valid.

But if we centralize the nodes, then we open an attack vector for censorship, and frankly the whole damn thing loses all it's value at that point.  No need to burn all this electricity, and continue to use a distributed database.

Well, Monero has an unbounded blocksize, theoretically.  And at the moment it is no big deal because the traffic on our chain is so small it does not really matter.  But if usage keeps going up (and I think it will) then eventually it will get harder to run (and sync!) a full node.  Monero ALREADY has way higher overhead in space as well as computational requirements.  When is the last time you synced a node from the bottom to the top?

But Monero, unlike BSV, is actually a cypherpunk project and the developers and community value decentralization.  Even the people buying drugs off the dark web do, whether they really know it or not.  And I think the hybrid blocksize technique we are using is VERY interesting... but it leaves the door open for a route to possible centralization.

So what happens when the storage space, bandwidth, and computational resources required to validate the XMR blockchain start to become prohibitive?

We need MORE fully validating nodes, not less.  Especially for Monero.

What do you folks think?  I know we have a fair share of big blockers and other apostates insofar as BTC maximalist doctrine goes among our numbers...  I would love to hear arguments that my reasoning is wrong.  And maybe I am just arrogant, but so far I have not been converted. 

I should listen to Articmine some more... I think he has some fairly strong opinions in this regard, and I don't think he is just a big-blocker nutcase. Wink



P.S. - On a side note it sure is nice to see little XMR marching up a little... but is it gonna break out of it's everlasting rut?  It has tended to only do this while BTC is rangebound itself.  Once BTC moves violently in EITHER direction XMR tends to deflate rather quickly.  I would LOVE to see that trend ended.  Something DOES feel a little different this time...  What is up with the alts?
legendary
Activity: 2730
Merit: 1288
July 11, 2020, 07:19:05 AM
he bi yearly forks were a net negative in my humble opinion.  I was glad the consensus was to stop them after the final try at RandomX.  And that we felt the need to call the hard forks "network upgrades" was sort of proof that we knew we were doing something that is anti-decentralization for the tradeoff of defeating ASICS.  As much as I loved taking out bitmain on that first algorithm tweak, it was a bit of a defeat to keep doing that.

We basically switched to the term scheduled network upgrade, because people associated the term hard-fork with a chain split that would create a new coin (and thus holders would be eligible for some kind of dividend). There used to be tons of questions on Reddit whether users would be getting some kind of dividend / new coin.

Wait so are we not doing it biannually anymore? I do like the terminology “network upgrade” too.

I believe idea is to do upgrades when they are ready and that is about twice a year. So nothing really changed. In the past almost all upgrades needed to be hardforks. Last one was not. There is misunderstanding in crypto community that Monero was doing hardforks to change mining algo. That is not true. Monero was doing hardforks to upgrade its protocol with new things, while doing it they also could tweak mining algo a bit. It was just an opportunity and if that opportunity would not exist Monero would have ASIC miners before RandomX.  Next upgrade will be hardfork again.   At the end the diferences between hardfork upgrade and not hardfork upgrade are that when hardfork you need to upgrade your wallet. So that makes whole network with the latest software and that is good. Since last upgrade was not hardfork we can still use v0.15 wallets.


And some long term hopium from reddit

hero member
Activity: 1873
Merit: 840
Keep what's important, and know who's your friend
July 11, 2020, 07:05:56 AM
he bi yearly forks were a net negative in my humble opinion.  I was glad the consensus was to stop them after the final try at RandomX.  And that we felt the need to call the hard forks "network upgrades" was sort of proof that we knew we were doing something that is anti-decentralization for the tradeoff of defeating ASICS.  As much as I loved taking out bitmain on that first algorithm tweak, it was a bit of a defeat to keep doing that.

We basically switched to the term scheduled network upgrade, because people associated the term hard-fork with a chain split that would create a new coin (and thus holders would be eligible for some kind of dividend). There used to be tons of questions on Reddit whether users would be getting some kind of dividend / new coin.

Wait so are we not doing it biannually anymore? I do like the terminology “network upgrade” too.
legendary
Activity: 2268
Merit: 1141
July 11, 2020, 05:09:06 AM
he bi yearly forks were a net negative in my humble opinion.  I was glad the consensus was to stop them after the final try at RandomX.  And that we felt the need to call the hard forks "network upgrades" was sort of proof that we knew we were doing something that is anti-decentralization for the tradeoff of defeating ASICS.  As much as I loved taking out bitmain on that first algorithm tweak, it was a bit of a defeat to keep doing that.

We basically switched to the term scheduled network upgrade, because people associated the term hard-fork with a chain split that would create a new coin (and thus holders would be eligible for some kind of dividend). There used to be tons of questions on Reddit whether users would be getting some kind of dividend / new coin.
legendary
Activity: 2702
Merit: 2053
Free spirit
July 11, 2020, 04:54:53 AM

Continuing the nice outlook today   Smiley


Pages:
Jump to: