I can tell you why you can't stop the 'spam'. You are thinking in cult doctrine. If it was real spam, and I was selling Viagra for example - you could easily ban keywords and urls. Instead, stop being a sheep, think like a cult leader. Recoginse that this 'spam' is just some bullshit that you tell to the sheep.

When you do that, solution will present itself. Observe. 'spam' -> 'FUK talks bad things about Monero on our IRC' (Don't say that out loud obviously, that will get you excommunicated) Solution? Get off-the-shelf sentiment analyser, detect anyone who 'talks bad things about Monero' and ban them.

I'm not an expert and read some books and did a little gpu mining on Monero

BUT could you point me to a cloud operator where I can lease some hardware mining equipment? No pools please.

Try #monero, but cloud mining is extremely unlikely to be profitable.

Especially GPU mining for monero.

yes thanks but want to switch to ryzen

have another question about nounces

I suppose that the nounces are used in monero like in bitcoin.

the generation of nounces is this always incremental?

Whatever the miner feels like.

Incremental is just easy/fast.

are the different schemes configurable?

Sure. With the monerod miner, just replace the ++ with whatever function you want.

There's not much point though, since you can't tell by looking at the starting nonce whether you'll find a hash below target without actually doing the work.

true, is there some parallellism involved and / of configurable?

If you're on about fingerprinting the mining software, the monerod miner starts to increment from a random value, so that should not be fingerprintable since only the end nonce gets published.

Hashing can be parallel, yes. Nonce generation is pointless to make parallel, it's just trivially fast and threaidng it would slow things down a lot.

Maybe "generation" is a misnonmer, it implies substantial work.

yes indeed meaning hash generation

Ah, then definitely.

but because it is random it is pure brute force

Mining is brute force due to the properties of a good hash function.

so in order to be profitable you need to go above a certain threshold with your mining hardware hash generation ability.

Yes, but since this doesn't seem to be about coding, see #monero for more.

Unless you have questions about the code :)

not for the moment, though I'm massevly interested in the RandomX mining code and such.

thanks a lot for your time, I will come back with my questions on the code (c/c++ I have seen) if the gods permit it ;-)

Sure.

Nice documentation by the way on the code around RandomX.

so,. because monero's gonna rocket to the moon and we'll inch ever closer to ossification, should we try and get some rough consensus going about that tx_extra?

im liking the "allow tx_extra in coinbase tx" options. leaves wiggle room, but ensures user-tx uniformity

so only miners can stuff something into tx_extra?

yeah

i mean, it seems to have a rough consensus. I'm just curious if we're at the stage where we throw it really out there (i.e., reddit) to see what further sentiment might exist

Why does the Saviour of NASA take a group achievement award and present it as a proof of individual glory? twitter.com/hyc_symas/status/1203709575226183683

master is on xmrchain (v0.17.0.0-dcba757dd)

moneromooo how is the password applied when opening wallet? I mean how many passwords/second can be tested?

i.e. I want to know the safe password length

cn/0 is bottleneck there

so cn/0 hashes per seconds

I think I've found the code but there's also kdf_rounds parameter, so more than 1 cn/0 hash is calculated?

I think by default only 1

Defaults to 1.

If I set kdf-rounds from command line when creating the wallet, do I need to set it every time I open it?

I think so.

That's interesting. If cn/0 is used, then those original ASICs can be used to crack weak passwords

correct. Still fairly slow I would wager

I did some math: given 1 joule/kh spent (2x the efficiency of Antminer X3), price of energy $0.03/kWh, $1 is enough to check 120 billion passwords. $1M is enough to check 120e+15 passwords. 10 characters password [a-zA-Z0-9] should be enough

people use passwords that short? 0_o

I use 1 character passwords, on the assumption no cracker will think I'm that dumb and start at 2.

If the password is random and uses large enough character set, it can be short, but 10 characters is reasonable minimum

just spent 30 minutes and memorize whole secret spend key (only 64 characters)

sech1: the peeps behind hashcat looked at it some years back - and concluded it was pretty difficult

* just spend 30 minutes and memorize whole secret spend key (only 64 characters)

No way

I spent a few days to memorize Pi to 30 digits

Maybe I'm just bad at remembering random numbers...

Did you go around in circles better after that ?

.merges

was cryptonote project moved to monero? it seems to be dead since 2015, is there an equivalent project today?

Monero is based off cryptonote.

The two have diverged over the years.

There's a number of forks of more recent CN code AFAIK.

i see

is there any guide on forkin monero just like the cryptonote one? it seems pretty self explanatory, I was reading monero source and it's very big with lots of extra stuff that cryptonote doesn't have

No.

.merge+ 7088 7238 7310 7326 7358 7459 7460 7615 7620 7621 7622 7623 7631 7632 7636 7637 7638

Added

v0.17.0.0-dcba757dd uptime 0d 6h 42m 17s

on xmrchain

any concerns with adding monero-project/monero #7387 ? we tested it on mac and linux and seems to work

bruh

.merges

.soon

“I thought, ‘I’m going to pump it and dump it,’ because I was interested and taking the ideas and implementing them in bitcoin. The bitcoin code base was far more interesting to me than monero, and I thought, ‘I’m not going to work on this codebase, it’s terrible,'” he recalls - fluffypony in an interview about Monero

If it works, why not.

functional tests will require 2 new python dependencies is the downside

but cmake will skip the test if they are missing like with requests

Common ones ? If it's stuff that doesn't come in OS packages, it looks like a bad idea.

psutil monotonic

not sure how "common" they are

mingw-w64-x86_64-python-psutil & mingw-w64-x86_64-python-monotonic in MSYS2.

iDunk: could you test if 7387 breaks functional tests on windows? I never got tests running on Windows

Looking at it it's a bit weird it cals pi rather than, say, CN, as a "let's use up CPU" function :)

IIRC, functional_tests fail on other stuff on Windows, but I'll try.

I've got those two on fedora. Phew.

moneromooo: I think you can close 7272 as it was replaced by 7308 if I understand this correctly

Yes, I think it did. I'll close.

.merge+ 7520 7538 7542 7549

Added