00:00:11.301,00:00:14.301
dev77: +1
00:00:29.603,00:00:32.603
Ferric: wooohoo
00:00:34.288,00:00:37.288
Leo Marquand: 8pm here that's better
00:00:35.791,00:00:38.791
Nye Liu: hey ferric :)
00:01:02.591,00:01:05.591
Ferric: yoyo
00:02:00.216,00:02:03.216
Zantetsu: Priority fees fluctuate so much per epoch that it's hard to see a significant difference atm
00:03:06.941,00:03:09.941
Max Resnick: likely will have longer term effects because a lot of the point of 96 was to make it so that side contracts were less desirable
00:03:14.970,00:03:17.970
Michael Hubbard: there's definitely an increase this epoch vs the previous in total block fees but yeah it will take several epochs to establish a median but then we're seeing like 100% increase and decrease again some epochs
00:04:19.091,00:04:22.091
David B: FD giving high skip rates to us
00:04:23.782,00:04:26.782
Bryan Cole: Agreed. Way too soon to see the effect. Max - are you expected to see a downward trend in priority fees since side-deals will no longer be as effective, e,g, priority fee marketplace competition will actually work?
00:04:41.855,00:04:44.855
Adi Bungarzan: At some point will be required for sfdp validators to switch to FD on mainnet?
00:05:15.043,00:05:18.043
David B: I’m running a 9374f
00:05:33.783,00:05:36.783
Max Resnick: Should be overall priority fees will increase in the long term relative to the counterfactual where 96 did not go into effect because without 96 validators prefer tips from side channels e.g. JITO tips because they are twice as effective.
00:05:44.511,00:05:47.511
Leo Marquand: Can you repeat the name of the event in NYC 2025 ? Did not have the time to look at it
00:05:45.235,00:05:48.235
Adi Bungarzan: Cool. Thanks
00:06:05.236,00:06:08.236
Leo Marquand: ok thanks !
00:06:38.243,00:06:41.243
Tim Garcia: <https://x.com/pkxro/status/1869422824924021100>
00:07:00.400,00:07:03.400
Bryan Cole: Interesting Max. I was sort of thinking that users were increasing priority fees because they were running at a 50% disadvantage (versus jito). e.g. if they wanted to compete with jito they would have to pay 10, versus 5 for jito. but now they can both pay 5.
00:07:17.877,00:07:20.877
Ben Hawkins: the developer focused event will be invite only. Any validators can get an invite.
00:07:32.586,00:07:35.586
Solomon Ponomarev: We also have some sponsorship options for workshop rooms if anyone wants to run a validator-focused day of programming!
00:07:54.908,00:07:57.908
Dan Albert: tldr of pre-announcement (formal announcement and website launch coming soon next week):
- deep dive technical/developer event: May 19-20
- product/general conf: May 22-23
00:09:01.376,00:09:04.376
Max Resnick: I think all of these things will play out over the longer term. Substitution between priority and other fee channels takes time because you actually have to change code. Similarly the way validators prioritize jito tips vs in protocol priority fees is also coded in and takes time to change as the market evolves
00:09:35.704,00:09:38.704
Bryan Cole: Brilliant, thanks Max.
00:09:40.575,00:09:43.575
Max Resnick: e.g. jito 70/30 jito tip priority fee split is something we might see change but they actually have to change it in the code
00:10:02.914,00:10:05.914
Italo Casas: great workshop, we should probably find a way to save the content in text format too
00:10:29.782,00:10:32.782
Tim Garcia: <https://github.com/solana-foundation/solana-improvement-documents/pull/123>
00:10:42.154,00:10:45.154
Zantetsu: Interestingly those SIMDs aren't accepted proposals
00:10:48.266,00:10:51.266
Tim Garcia: Intermediate vote credits: <https://github.com/solana-foundation/solana-improvement-documents/pull/217>
00:10:49.307,00:10:52.307
Zantetsu: They are still pull requests
00:11:08.975,00:11:11.975
Tim Garcia: SIMD 228: <https://github.com/solana-foundation/solana-improvement-documents/pull/228>
00:11:37.175,00:11:40.175
Tim Garcia: Yeah, there is a bit of a process exploration here about when governance should happen. Before or after acceptance.
00:12:07.172,00:12:10.172
Max Resnick: That is mostly a process thing Zantetsu typically proposals aren't merged in until after acceptance particularly for economic things although this is something we should explore, its a bit harder to checkout the PRs than if they are in the main repo
00:12:37.983,00:12:40.983
Bryan Cole: Makes sense. That's a pretty important choice. If something isnt going to make it through governance then should it go through the SIMD process?
00:12:39.202,00:12:42.202
Ben Hawkins: The thinking around signaling governance before SIMD acceptance is similiar to the way we always look to Jump and Anza giving approval before merging.
00:15:01.935,00:15:04.935
Ben Hawkins: On economic changes we are looking for validator approval along with Azna and Jump. Once all have signaled approval then we would merge.
00:15:58.863,00:16:01.863
Ben Hawkins: None of these are binding and SIMD's can be withdrawn after the fact but I think it's valuable to make sure all pertinent parties think its reasonable to move forward towards making the change.
00:16:00.963,00:16:03.963
Max Resnick: Very in favor of 123, these block rewards distributions are already very common out of protocol. The proposal allows a way to verify this for delegators. This is nice for validators in that it may simplify their acounting and it protects delegators and it makes the data more public.
00:16:30.637,00:16:33.637
Zantetsu: Definitely, it's important to give stake pools more tools to squeeze validators harder :)
00:17:02.763,00:17:05.763
Eddie Nuta: :)
00:17:23.173,00:17:26.173
Alexey Kolpin: +1
00:17:48.296,00:17:51.296
Zantetsu: All joking aside it's inevitable so we might as well make the mechanism clean
00:19:07.204,00:19:10.204
Tim Garcia: SIMD 218 Intermediate vote credits: <https://github.com/solana-foundation/solana-improvement-documents/pull/217>
00:19:47.274,00:19:50.274
Zantetsu: I'm all for 218. My mods are out there and I welcome efforts to kill them :)
00:19:52.180,00:19:55.180
Bryan Cole: this is the anti-backfill simd
00:20:04.651,00:20:07.651
Michael Hubbard: or auto-backfill depending on how you look at it
00:20:28.454,00:20:31.454
Ferric: good way to get your code merged in Zan, open source and get us all to run it anyway ;)
00:24:49.784,00:24:52.784
Tim Garcia: SIMD 228 Market-Based Emissions <https://github.com/solana-foundation/solana-improvement-documents/pull/228>
00:25:45.329,00:25:48.329
Zantetsu: The issue for me is that "overpaying" and "security" are both not very well defined terms. The justification for the SIMD doesn't provide much concrete data that could be evidence for these assertions.
00:26:22.521,00:26:25.521
Michael Hubbard: not sure comparison is helpful, i'd rather try identify what security budget is necessry from first principles
00:27:17.662,00:27:20.662
Chainflow: I tend to agree with what Zan is saying, however feel like this SIMD is an improvement over what we have today. So if seen as an iterative step in the right direction, I'm generally in support of it.
00:27:53.827,00:27:56.827
Uri Klarman: Michael, isn't the first principles highest need is to maintain 67% of SOL staked, so if majority vote on something, they are over 33%?
00:27:58.083,00:28:01.083
Zantetsu: The ETF would pass staking returns on to end users anyway so why would staking be negative pressure on ETF usage?
00:28:15.036,00:28:18.036
Zano: high emissions made sense as a way to bootstrap the network; this type of proposal seems directionally correct
00:28:38.556,00:28:41.556
Michael Hubbard: Uri, define "vote on something", consensus threshold is 67%, so if 50% is staked then consensus threshold is 33%. but why is 33% the critical threshold?
00:28:46.571,00:28:49.571
Michael Hubbard: smart emissions make sense
00:28:48.403,00:28:51.403
Zantetsu: The evidence for what is "needed" needs to be really strong though. Not just conjecture.
00:29:39.915,00:29:42.915
Zantetsu: Michael all SOL chooses whether it is happy with the current security or not; if it is, it doesn't need to stake, the existing stake proxies its security concerns. If it is not happy with current security, then it stakes.
00:30:07.103,00:30:10.103
Dan Albert: staked SOL ETF going live in canada btw: <https://www.newsfilecorp.com/release/240405>
00:30:10.983,00:30:13.983
BM | ADRA: SOL is being used in DeFi through LSTs
00:30:12.717,00:30:15.717
Michael Hubbard: yes I agree but the incentives should align to some threshold of stake percent
00:31:13.666,00:31:16.666
Uri Klarman: Michael - I think Tushar covered it in the lightspeed podcast, but I don't remember it well enough to make it properly, but it removes an attack vector - smtn like it prevents having 66% ability to vote on an old fork
00:31:42.392,00:31:45.392
Michael Hubbard: yeah that assumes the economic and market ability to aquire 33% of stake
00:31:47.065,00:31:50.065
Bryan Cole: so.... people staking SOL at 10.55% undermines defi because they dont want to move SOL into DEFI because they need to make at least 10.55% or else why use defi? (unless there is some other non -apr reason.
00:32:08.158,00:32:11.158
Bryan Cole: (just using current staking APR 10.55% is place holder)
00:32:11.569,00:32:14.569
Ionut Scirlet: I fully agree with Vishal and Tushar. All are real reasons due to rules in many jurisdictions and dilution in capital markets, it’s pure math.
00:33:20.720,00:33:23.720
Ben Hawkins: 600m * .045 = 27m
00:34:05.139,00:34:08.139
Matthew Cammell: I'm not an expert in any of these area's but I think trying to play market is a tough thing to do, I'm not entirely convinced the inflation schedule is broken as the market knows has this information and Solana is doing great and continues to be better and faster. (sorry if this is a ill informed take)
00:34:28.682,00:34:31.682
Michael Hubbard: Thanks Ben, I'd written down the last number after applying tax. 600m * 0.045 * 0.5 * 0.5 = ~$1bn annual sell pressure
00:34:41.288,00:34:44.288
BM | ADRA: Bryan, people who want to use DeFi stake their assets through LSTs and utilize them across multiple DeFi protocols
00:35:37.433,00:35:40.433
BM | ADRA: So, I wouldn’t agree that high yield on staking removes people’s interest in using DeFi with SOL
00:36:21.491,00:36:24.491
Max Resnick: issuance creates expected cap losses which cannot be offset by ordinary income gains from validator rewards
00:36:25.562,00:36:28.562
Max Resnick: which is very odd
00:42:19.527,00:42:22.527
Tushar Jain: for context, Ethereum's staking participation rate is ~27.5%
00:43:11.156,00:43:14.156
Евгений Быстров: Is security about % SOL staked or about distibution of stake also?
00:44:11.981,00:44:14.981
Zantetsu: That's disingenuous honeslt.y
00:44:24.037,00:44:27.037
Zantetsu: It's been 3+ years since staking participation was anywhere outside of a fairly narrow range.
00:44:55.564,00:44:58.564
Zantetsu: It may have been as high as 90% or as low as 20% back when it was still effectively in development maybe ..
00:45:00.901,00:45:03.901
Eddie Nuta: Ethereum APY is also a reason for that 27.5%
00:45:21.714,00:45:24.714
Chainflow: Past performance is no guarantee of future results ;)
00:45:29.449,00:45:32.449
James Stewart: I think thats the intent though to free up capital for defi
00:46:10.693,00:46:13.693
Chainflow: But the stakers would have to be comfortable using defi
00:47:14.480,00:47:17.480
Zantetsu: The problem is that delegators only represent the interest in higher inflation rates since delegators are the ones benefitting from inflation.
00:47:33.381,00:47:36.381
Zantetsu: Non-delegated SOL doesn't even have a validator to represent their interests in any vote.
00:48:18.451,00:48:21.451
Ben Hawkins: Futarchy fixes that issue Zan.
00:48:20.815,00:48:23.815
Max Resnick: my sense is that delegators care about the overall returns on Sol not just the yield
00:48:41.382,00:48:44.382
Michael Hubbard: Max, in theory yes but short term vs long term thinking is hard sometimes
00:48:53.985,00:48:56.985
Zantetsu: @Ben so we should be proposing some kind of futarchy based voting process here instead of validator governance?
00:49:03.787,00:49:06.787
Max Resnick: yeah sticker shock is maybee something that matters on a behavioral level
00:49:05.204,00:49:08.204
James Stewart: So whats the delegators incentive to lower staking yields?
00:49:10.986,00:49:13.986
João Gustavo: There is a big difference between inflation and hyperinflation, this comparison makes no sense.
00:49:21.117,00:49:24.117
Ben Hawkins: I don't think it's ready for prime time and not proven out but I do think it fixes issues that you point out.
00:51:37.959,00:51:40.959
Tim Garcia: 744
00:52:26.199,00:52:29.199
Nye Liu: I would eventually like to plug the snapshot projecdrt
00:52:27.349,00:52:30.349
David B: Tim mentioned that HW could be an issue with FD skip rate but I have Amd epyc 9374f, 3x crucial t700’s, 512gb ram and when running agave I was in the top 200 to 500 rank on test net with little to no skip rate and was 916 in SFDP queue and almost immediately when switching to FD seen helliish skip rates and a lot of transaction count not advancing which ended up causing me to loose like 30,000 test sol and now almost to 1400 in SFDP queue
00:52:27.936,00:52:30.936
Nye Liu: :(
Tim Garcia: All right, welcome everyone to the Salana Foundation validator discussion February 13, 202 a lot of things on the agenda today. so first of all, validator updates. we'll talk briefly about Accelerate NYC, which is a both developer and sort of product focused conference in New York. It's happening in May. I hope to see a lot of you there. some educational workshop ideas and then the majority of the meetup I think we'll be talking about these SIMD 123, 218 and 228 for a potential governance vote. Michael is going to kick it off. I think and others are here to talk through 228 specifically. So let's get into it. first of all, validator updates.
Tim Garcia: Agave 2111 and Fire Dancer V305 2000111 is recommended for mainet. if you have not upgraded, please do so. Especially RPC operators. we are not yet at the cutoff for feature activations to fork off 20 versions. but it is approaching. So again, if you're an RBC operator, please update quickly before feature activations fork you off. SIMD96 has been saw a lot of talk about that on X. So, 96 is out there. Full priority fees are now going to validators. very curious to see any data on that. Is there any preliminary feedback or thoughts from that activation? Looking in the comments here. Done yet? All right.
Tim Garcia: yeah, hope to see data soon. I'm sure Dan or someone will have data out there. Ucoming feature gates. So, SIMD 133 get epoch stakes, which should be important for validator governance tooling is currently eighth in the test net queue. it's slated for 2.1 activation. So, just need to get through the queue of activations on test net and then on mainet to get activated. probably in 2 to 3 months timeline I'm guessing there's about six or seven future activations in the mainet queue once we get through those we'll have to go through the ones in testn net typically we do about two activations a week so that's where I'm getting the two or three month timeline and then sd 207 raise block limits to 50 million cus
Tim Garcia: I know a lot of people have been interested in this one as well. it's 11th in the queue on testn net. so we should also have this for 2.1 in probably three or more month timeline there. Any questions on that for mainet? yeah, just some chatter about 96, but sounds like no comments. Moving ahead, moving on testn net. so we have more than 80% of stake on fire dancer on testn net. So thank you to all the testn net sftp operators that switched over. really important for testing. fire dancer is also on mainet. We want to make sure we test that client well and sort of go through some feature activation cycles there as well.
Tim Garcia: so this is not the long-term state of test net, but for the next week or so, fire dancer will be the super majority. if you haven't switched over yet, please do so. Fire dancer 305 20011. And, the plan is after we do some testing there on testn net, we will switch to agave v2.2. so just a heads up, we'll be going back to either agave or fire dancer within a week or so. Any questions on testn net? Yeah, some comments about fire dancer skip rates. I've talked to the fire dancer team. I know they're working on sort of a changes to optimize a little bit for lower-end hardware.
Tim Garcia: there's been a couple comments in Discord, but essentially there's a bit of a discrepancy between the hardware that's sort of best suited for Agave versus the hardware that might be better suited for Fire Dancer. And, rightfully, most operators on test net have hardware that's better suited for Agave, which is a maybe not very high core count, but higher base clock CPUs. the thought is that fire dancer might work better on a base clock maybe not being as important but number of cores being more important. So we'll see how that shakes out but sort of TLDDR they're planning to make some improvements there. there's a question on whether or not SFTP will ever require operators to switch to fire dancer on mainet.
Tim Garcia: I don't anticipate we'll ever say that you have to run this version or that version on mainet. we'll still increase min versions for both clients, but I don't think you must run this client. That's the whole point of having client diversity is to have operators just following the comments. yeah, there's a question about the event in NYC is called Accelerate. there's not a lot online about it.
Solomon Ponomarev: the week of that event will be publicly announced next week.
Tim Garcia: Excuse but yeah, I'll talk about it in just a sec. H great. Yeah, Saul, who just spoke, he's running it or not running it, but sort of putting together the event, helping to figure out programming and such. So, if you have questions, I'll put his tweet up here in just a sec.
Tim Garcia: yeah, so Accelerate 2025 is happening between May 19th and should be 23rd. here is Saul's post from before on it. So I'll put that in the chat. and Solomon is the one who is again putting together the agenda. So please reach out to him if you got thoughts on technical content that would be good for that event. yeah and sort of my plea to you all is I hope to see a lot of validators there. it would be great to have kind of validator focused events especially around technical conferences like this. super excited to see a lot of Yep. Chat is really buzzing today. Hard to keep up.
Tim Garcia: Yeah. Any questions about Accelerate? All right. Educational workshop.
Bryan Cole: Sorry Tim,…
Tim Garcia: Yeah.
Bryan Cole: real quick question. Ben said the developer focus event will be invite only. is that the entire event or is there a specific just developer side piece that's invite only