[XMR] Monero - A secure, private, untraceable cryptocurrency

[smooth]

For 'thin' clients that will definitely be required for broad and widespread adoptance (mobile phones, etc), I think that the processing required for hashing some POW would be prohibitive.

Why would you mine on a cellphone? thin clients will only get onboard after some blockchain-level online wallet or lightclient is developed...  

This wasn't about actual mining just a tiny amount of PoW to earn the right to submit a transaction and discourage spam without paying a fee. Bitmessage does something similar, and some may be unaware that PoW was invented as an anti-spam measure for email.

However, in rpietila's altcoin thread we talked about a billion cell phones mining at night while they are plugged into a charger. That was pretty interesting.

 

[coins101]

I coded a simple script which parsed this topic, boolberry, bytecoin, ducknote and darkcoin. Figured it would be interesting to see these results.

...

Notice the number of hero and legendary members? Darkcoin has 57 (2.23%) hero and 9 (0.35%) legendary versus Monero with 56 (3.94%) hero and 12 (0.85%) legendary. Percent wise monero has attracted more than twice legendary members and almost twice hero members.

In other words, it seems oldtimers gravitate towards monero/cryptonote and ignore darkcoin.

Wow, very neat! Thanks for scraping and parsing like a boss

lol.

We are all just the guys on the ground that keep the threads bumping along.

[bclcjunkie]

thanks smooth! looks like i'm good to go then.. but yeah the block height keeps increasing in those -4 messages...

anyone else sees the same? can the devs confirm if i should resync or ignore those -4 days messages? sometimes it would continue syncing backwards until i have to close and restart the session again and then it gets synced properly... i've never seen this sort of behavior before.. wallet shows current block height and correct balance..

Ignore the -4 days messages. That means the peer is '-4 days ahead" meaning 4 days behind. i.e. they are stuck. Your node will drop them and continue working. As long as your height is correct (diff command in daemon or bc_height command in wallet) you are fine.

I'm not sure what you mean by syncing backwards unless your height is decreasing instead of increasing. I don't think that is possible. (If you see it, report it.)

[smooth]

The first number in those messages is your height. It is increasing because your node is working properly.

If those connections are outgoing, then once your node establishes a set of stable outgoing connections those messages will stop. If they are incoming you can ignore them or if they really bother you, block the bad nodes in your firewall, but they are harmless.

thanks smooth! looks like i'm good to go then.. but yeah the block height keeps increasing in those -4 messages...

anyone else sees the same? can the devs confirm if i should resync or ignore those -4 days messages? sometimes it would continue syncing backwards until i have to close and restart the session again and then it gets synced properly... i've never seen this sort of behavior before.. wallet shows current block height and correct balance..

Ignore the -4 days messages. That means the peer is '-4 days ahead" meaning 4 days behind. i.e. they are stuck. Your node will drop them and continue working. As long as your height is correct (diff command in daemon or bc_height command in wallet) you are fine.

I'm not sure what you mean by syncing backwards unless your height is decreasing instead of increasing. I don't think that is possible. (If you see it, report it.)

[xulescu]

Just mentioning that building Monero on Ubuntu 12.04 x64 is noob-safe.

install ubuntu, wifi driver, update, upgrade, dist-upgrade, run install_monero.sh, (choose make essentials, compile boost)

Also, it would be a good idea to correct the idea that Monero has a complicated recursive block reward (and further proof of intentional obfuscation). Shameless plug:

Monero has an exponentially decreasing emission rate:

Let a = 2^-20 * 10^-12, then block reward is R_n = M a (1 - a)ⁿ

It's like Bitcoin, but ~2x faster and without the sudden halvings.

[kingscrown]

decent launch im holding for now

[5w00p]

Why would you mine on a cellphone? thin clients will only get onboard after some blockchain-level online wallet or lightclient is developed...  

Where did I say anything about mining in my post, Nekomata?
You misunderstood my post and asked the question and then paraphrased what I wrote about a thin client by using the "word" lightclient.
It's all good, though.

[vokain]

Monero got big and useful much quicker than I expected! As soon as I gave it some time, I now see. Thank you to everyone involved

[digicoin]

Need a XMR shill in Darknet 

[vokain]

Need a XMR shill in Darknet  

At the same time, wouldn't the best shill help it become the Lightnet?  

[qqqq]

Just to give a rough guide to our thinking on fees, we are currently considering 0.01/2 kb. 0.005 didn't work too well given the way dust is handled (it creates a lot of 0.005 change, although in the future we will probably improve dust handling). So most small transactions (<= 2 kb) would have a fee of 0.01, larger ones of around 20 kb would be close to the current 0.1 and continue to provide the same protection against spamming of large transactions.

For reference, at current prices

0.01 XMR is approx. 0.02 USD
0.0001 BTC is approx. 0.05 USD (common fee per kb used with bitcoin, though there is a range in use)

With this setting, we are about 5x cheaper than Bitcoin per kb. However, Bitcoin has the ability to send some zero-fee tx (though not all wallets support this or do it by default) and we don't. Also the 2 kb increment is larger than Bitcoin's 1 (typical) kb increment, which also increases effective fees somewhat, so 5x is an overestimate.

Don't understand how i can get 0.01 fee cause right now i have to pay 0.1 on any transactions.

Error: not enough money to transfer, available only 1.700000000000, transaction
amount 1.800000000000 = 1.700000000000 + 0.100000000000 (fee)

[smooth]

Just to give a rough guide to our thinking on fees, we are currently considering 0.01/2 kb. 0.005 didn't work too well given the way dust is handled (it creates a lot of 0.005 change, although in the future we will probably improve dust handling). So most small transactions (<= 2 kb) would have a fee of 0.01, larger ones of around 20 kb would be close to the current 0.1 and continue to provide the same protection against spamming of large transactions.

For reference, at current prices

0.01 XMR is approx. 0.02 USD
0.0001 BTC is approx. 0.05 USD (common fee per kb used with bitcoin, though there is a range in use)

With this setting, we are about 5x cheaper than Bitcoin per kb. However, Bitcoin has the ability to send some zero-fee tx (though not all wallets support this or do it by default) and we don't. Also the 2 kb increment is larger than Bitcoin's 1 (typical) kb increment, which also increases effective fees somewhat, so 5x is an overestimate.

Don't understand how i can get 0.01 fee cause right now i have to pay 0.1 on any transactions.

Error: not enough money to transfer, available only 1.700000000000, transaction
amount 1.800000000000 = 1.700000000000 + 0.100000000000 (fee)

The above was not describing the current client. I was throwing out some of our preliminary thoughts on the per-KB fee model that isn't released yet.

The current version uses 0.1 fee regardless of transaction size (in bytes).

[rpietila]

Just to give a rough guide to our thinking on fees, we are currently considering 0.01/2 kb. 0.005 didn't work too well given the way dust is handled (it creates a lot of 0.005 change, although in the future we will probably improve dust handling). So most small transactions (<= 2 kb) would have a fee of 0.01, larger ones of around 20 kb would be close to the current 0.1 and continue to provide the same protection against spamming of large transactions.

For reference, at current prices

0.01 XMR is approx. 0.02 USD
0.0001 BTC is approx. 0.05 USD (common fee per kb used with bitcoin, though there is a range in use)

With this setting, we are about 5x cheaper than Bitcoin per kb. However, Bitcoin has the ability to send some zero-fee tx (though not all wallets support this or do it by default) and we don't. Also the 2 kb increment is larger than Bitcoin's 1 (typical) kb increment, which also increases effective fees somewhat, so 5x is an overestimate.

So - once I estimated that the whole blockchain (100GB) was for sale for $41k and deemed that too cheap. Now the price is per kB, and it works out to $1 million, a 25-time increase. This sounds indeed much better and while $1M is still very little, the coin's market cap is also small.

It is telling how concentrated the blockchain usage is towards manual transaction that a 20x increase in fees resulted in only 30% reduction in tx. This means that if the blockchain growth will need to be limited in the future, the fees have to be raised really high to compel off-chain activity.

[Wexlike]

When the 0.1 fee will be fixed ?

When we replace it with per-kb fees.

Just please make the fees high enough that the blockchain is not sold out too cheaply. It is possible to lower them when the price rises. The recent decision to up them by 20x did not lead to price crashing.

I would like to hear your specific suggestions with supporting analysis.

If "you" was singular, I'd like to know many variables first, to start to analyze it:

- number of tx per day
- distribution of the sizes of tx, how much user can affect it
- does mixing factor affect it
- is there any reason to compete for space in 1 block as with Bitcoin
- is there any way the user could/should be able to affect the fee
- how did the recent 20x increase of fee affect tx
- what is the realistic upper limit of blockchain size now, in 1, 2, 3, 4, 5, etc years so that it is still usable
- blockchain prunable to what extent, when?

After these, I proceed to calculate if there is any need to limit the blockchain usage or if the antispam is the only reason for fees. If the limiting is required, then it takes some time to try to develop a "lifespan" for the coin by probably subsidising the fee now so that later Monero is so big that we can face the problem with larger resources. But I won't go into specifics without getting the information first.

ADD: Before even seeing the facts, I would say that there exists the following usage:

- Transfers. These are quite large and not much affected by the fees.
- Payments. These may be small but can be reconfigured such that they happen less often, which would reduce the usage given higher fee.
- Micropayments. These proliferate is the fees are small, and become uneconomic if they are raised. I don't believe there are many micropayment innovations functioning in Monero, and by keeping the fees high there will not be, either. It is a new thing that has not existed before, but seems that Monero blockchain is not the right place for it.
- Spam/Attacks.

If fee is set so high that people are grudging when making a transfer, it is a problem because they sell out of a coin that is too expensive to transact with. If fee is too high for payments, the coin loses economy/services. There were never any micropayments to begin with.

I am not a fan of high transaction fees. After all, the low transaction fee lead to the latest successful attack - and that is a good thing! With higher fees that attack might not be feasible - true, but this just means the attacker would be a future force with more money in the pockets.

That is why i propose: keep the fees low, give attackers a chance to attack. After all it makes the coin so much more secure and our devs handled the last attack so well!

[smooth]

Just to give a rough guide to our thinking on fees, we are currently considering 0.01/2 kb. 0.005 didn't work too well given the way dust is handled (it creates a lot of 0.005 change, although in the future we will probably improve dust handling). So most small transactions (<= 2 kb) would have a fee of 0.01, larger ones of around 20 kb would be close to the current 0.1 and continue to provide the same protection against spamming of large transactions.

For reference, at current prices

0.01 XMR is approx. 0.02 USD
0.0001 BTC is approx. 0.05 USD (common fee per kb used with bitcoin, though there is a range in use)

With this setting, we are about 5x cheaper than Bitcoin per kb. However, Bitcoin has the ability to send some zero-fee tx (though not all wallets support this or do it by default) and we don't. Also the 2 kb increment is larger than Bitcoin's 1 (typical) kb increment, which also increases effective fees somewhat, so 5x is an overestimate.

So - once I estimated that the whole blockchain (100GB) was for sale for $41k and deemed that too cheap. Now the price is per kB, and it works out to $1 million, a 25-time increase. This sounds indeed much better and while $1M is still very little, the coin's market cap is also small.

Thank you for explaining this model again. It didn't make a lot of sense to me the first time (though we had other things on our mind) but now it does.

It is telling how concentrated the blockchain usage is towards manual transaction that a 20x increase in fees resulted in only 30% reduction in tx. This means that if the blockchain growth will need to be limited in the future, the fees have to be raised really high to compel off-chain activity.

Keep in mind we were starting from a very low base (as your $41K number indicates). A 2x increase might have resulted in almost no decrease at all. A 20x increase from the current level would probably reduce usage by more than 30% (possibly 100%), and 2x would probably have very significant effect now.

[myagui]

@smooth, @rpietila:
 
I don't think that we can take such statistical conclusions to the recent variations in the number of transactions, without taking into account the concern that users would have had during the brief time of the attack/fork and the couple of days that followed. I'm not saying that the higher transaction fee did not impact on number of transactions, but just that in this particular instance, it was not really the only motivation for less transactions.

In any case, I'm looking forward to per-kb fees, and appreciate the rationale that you've both presented on this matter.
~ Myagui
  

[onetwentyfive]

For 'thin' clients that will definitely be required for broad and widespread adoptance (mobile phones, etc), I think that the processing required for hashing some POW would be prohibitive.

Why would you mine on a cellphone? thin clients will only get onboard after some blockchain-level online wallet or lightclient is developed...  

This wasn't about actual mining just a tiny amount of PoW to earn the right to submit a transaction and discourage spam without paying a fee. Bitmessage does something similar, and some may be unaware that PoW was invented as an anti-spam measure for email.

However, in rpietila's altcoin thread we talked about a billion cell phones mining at night while they are plugged into a charger. That was pretty interesting.

 

https://play.google.com/store/apps/details?id=com.miner.pro&hl=en

Android based miner that just recently added support for cryptonote coins.. On my Galaxy Note 3 (quad core) I've yet to find a share, and set to the highest priority I can literally watch my battery go down! But it works, I tried it with LTC and got around 3kh/s..

[OrientA]

Monero got big and useful much quicker than I expected! As soon as I gave it some time, I now see. Thank you to everyone involved

Do you see it is adopted some where?

[OrientA]

Just to give a rough guide to our thinking on fees, we are currently considering 0.01/2 kb. 0.005 didn't work too well given the way dust is handled (it creates a lot of 0.005 change, although in the future we will probably improve dust handling). So most small transactions (<= 2 kb) would have a fee of 0.01, larger ones of around 20 kb would be close to the current 0.1 and continue to provide the same protection against spamming of large transactions.

For reference, at current prices

0.01 XMR is approx. 0.02 USD
0.0001 BTC is approx. 0.05 USD (common fee per kb used with bitcoin, though there is a range in use)

With this setting, we are about 5x cheaper than Bitcoin per kb. However, Bitcoin has the ability to send some zero-fee tx (though not all wallets support this or do it by default) and we don't. Also the 2 kb increment is larger than Bitcoin's 1 (typical) kb increment, which also increases effective fees somewhat, so 5x is an overestimate.

So - once I estimated that the whole blockchain (100GB) was for sale for $41k and deemed that too cheap. Now the price is per kB, and it works out to $1 million, a 25-time increase. This sounds indeed much better and while $1M is still very little, the coin's market cap is also small.

It is telling how concentrated the blockchain usage is towards manual transaction that a 20x increase in fees resulted in only 30% reduction in tx. This means that if the blockchain growth will need to be limited in the future, the fees have to be raised really high to compel off-chain activity.

For me, all my transactions are to receive the coin from pool and send to exchange. I am paid once per hour, with lots of dust. So it would be better that pools payout once every few hours, that would decrease transaction size dramatically.

[neverminer77]

Monero got big and useful much quicker than I expected! As soon as I gave it some time, I now see. Thank you to everyone involved

Do you see it is adopted some where?

Once it gets a decent GUI, users will flock in. Next stage is a solid ecosystem. Monero is still young but it's getting there dude. Don't you worry about that!