05:33:31 what the best way for a non dev IT person to contribute? basically just run a full node? 05:34:30 if i wanted to pursue a dev skill path, where should I start if all I know is server hardware, windows and linux? 05:51:38 #monero-community-dev:monero.social 05:51:38 This channel is for core development 05:51:47 Hacking, c++ etc 05:52:50 #monero-dev:monero.social is for core dev, hacking etc. 05:52:50 #monero-community-dev:monero.social is for lower level stuff 05:53:39 ok sorry 06:04:02 > <@jeffro256:monero.social> > Another question, with ristretto points. Why are prime order groups harder to implement algorithms for? 06:04:02 > 06:04:02 > They both have different challenges. When the number of possible elliptic curve points is not prime (e.g. Curve25519), you have to constantly worry about sanitizing curve points that are not in the correct subgroup, hence the existence of Ristretto. When you do not use prime subgroups (e.g. secp256k1 in BTC) you have to worry more about Pohlig–Hellman algorithm attacks 06:04:02 But why order h*l instead of l/q attractive? 06:08:27 * But why is order h*l 14:48:40 Well the order of the subgroup we use is actually l, not h*l. h*l is the total number of points on E(Fq). Also, we cannot use q as the order since q is the size of the underlying integer field. 14:49:13 AKA the base generator for curve25519 resides in a subgroup with prime order l 14:50:04 jeffro256[m]: Ive read some time ago that ristretto expands the subgroup to use h*l 14:50:14 but I don't know the correctness of that 14:57:46 I don't think so... If anything, some libraries even use scalars mod l instead of X, Y over mod q 14:58:31 They don't allow users to convert direct from an X value to a point (besides the base point), they have to use scalarmult against G 14:59:05 I would ask dangerousfreedom... They know better 15:00:01 > They don't allow users to convert direct from an X value to a point 15:00:10 That doesn't apply when using DH in X25519 16:01:20 "I don't think so... If anything,..." <- yea. I think sodium does 16:01:44 (they have wrote the zig implementation of ed25519 in its standard library, which I explored some time ago) 18:18:04 Hello. I have got a question about Monero Daemon RPC.... (full message at ) 18:36:39 yes, next_seed_hash is just for reference 21:05:47 .merges 21:05:47 -xmr-pr- 8720 8721 8736 8737 8738 21:11:09 if someone runs a stagenet or testnet node please update to release-v0.18, would be great if we had some testers before we put out the release 21:11:24 (or also mainnet) 21:12:52 the dandelion++ code had changes, there's now a ringct cache and the monero side randomx code had a rewrite, so testing mining and sending transactions would be great 21:17:33 all of the recent testnet spam during p2pool testing was broadcast from 2 nodes / wallet-rpc running release (both mainnet nodes currently have ~2MB/s traffic also and stable). The more testers the better! 21:18:46 my nodes have also been running release-v0.18 and so far everything good 22:07:15 "but I don't know the correctness..." <- It is not expanding to use h*l points. It is remapping to l to avoid the overhead of multiplying by the cofactor, cleaning the scalar or other things. 22:36:50 selsta: someone on Reddit has a problem with this https://github.com/monero-project/monero/issues/7708 22:37:04 https://www.reddit.com/r/Monero/comments/10y6i7x/untransferable_xmr 22:38:01 monerobull[m]: they have to add disable_noise 22:38:38 `--proxy tor,127.0.0.1:9050,disable_noise` 22:38:40 I'll tell them 22:39:52 What's the cause here (for dummies)? 22:40:02 And how does that flag fix it 22:40:58 Ah i see 22:41:09 White-noise feature 22:41:10 the daemon constantly sends noise and in-between the noise it would send the transaction, in this case the transaction is simply to large and the node would switch to a different node before it's fully transmitted 22:41:11 yep 22:41:19 too* 22:41:32 selsta: Woah, something about monero i didn't know 22:42:12 Thanks! 23:39:41 .merge+ 8739 23:39:41 Added