Why I’m Against the GMX v2.2 + v2.3 Network Fee Proposal
I’ve taken a close look at the fee smoothing proposal and want to share my thoughts. The main idea is to increase fees when the network isn’t congested to subsidize costs when it is. Sounds reasonable at first, but as a trader who operates during low-traffic periods, this actually makes things worse for me.
Why This Hurts Traders Like Me:
I mostly trade when the network is not congested, meaning I’ll be paying higher fees without benefiting from the subsidies during peak times.
Effectively, I’ll be subsidizing traders who operate during high congestion, even though I intentionally avoid those times to save on fees.
This is basically a penalty for “smart” trading—why should I pay extra just because I choose to trade at optimal times?
On the forum, woofbark pointed out that only 5% of transactions experience high fees, so the congestion issue is likely overblown. On the other hand, Jonezee argued that fee spikes do cause problems, but I’m not convinced this is the right way to smooth them out.
What Could Be Done Instead?
Implement a truly dynamic fee system that adjusts in real-time based on network demand rather than preemptively raising fees.
Consider cross-chain solutions or other ways to reduce network congestion instead of shifting costs onto traders.
Give users a choice—let those who want fee stability pay an optional surcharge rather than enforcing it on everyone.
Final Thoughts
I’m not against improving the fee structure, but this proposal punishes the wrong users. Instead of smoothing fees, it makes them less predictable and raises costs for most traders.
its possible that those other solutions are further ahead in development, while you are right that all base users are now charged more upfront with this implementation, the result is higher performance during congestion, given the reversibility of the proposed solution, its worth considering that the entirety of the platform achieves greater outcomes for all parties, thats the intent,
this only raise the gas fee to 50 cents, for you, is this not manageable? for 100-1000 usd positioning, i suppose no
time boost kind of deflects as it caters to the bidder, between the bidder for priority and congestion, we need to insure a baseline before gasless txs.
once pectra adds more blobs this implementation will look trivial long term
(~4 months?)
therefore i am in support still, while deeply appreciating the dialogue on this matter
I see your point, but this proposal hasn’t been implemented yet—we’re still discussing its potential impact (or am i wrong?). You’re talking about improved performance, but there’s no guarantee that this mechanism will actually fix the issue.
50 cents might not seem like much, but that’s not really the point—the issue is that this artificially increases fees, putting traders who intentionally avoid congestion at a disadvantage.
Also, if Pectra is expected to solve this problem in about four months, wouldn’t it make more sense to wait for that rather than pushing through a controversial fee increase now?
totally but the lead dev is recommending this, quite a bit of weight for me, apologies, but traders just got a significant decrease in fee charges that i believe covers this
network congestion will still be an issue after pectra, onchain trading continues to surge but you are right the next 6 months the rest of the market that makes up our trade environment will scale, time boost and pectra will help as well as v2.2 v2.3 developments
so you and woofbark are not okay with experimenting with this given the reversibility? and recent fee decrease
I get your point, and I respect that the lead dev is backing this, but I still think there are some issues to consider.
Yes, trading fees were reduced, but that doesn’t change the fact that network fees are increasing for those who trade during low-traffic periods. Should one really compensate for the other?
As for the proposal being reversible, sure—but why introduce a potentially harmful change when in just a few months it might not even be necessary? If congestion is still an issue after Pectra, let’s see how it plays out first before making changes that shift costs onto different traders.
Also, just to clarify—I’m not here to push my opinion onto anyone, I’m just sharing my thoughts since this topic was brought up. I respect different perspectives, but from my point of view, this proposal could negatively impact traders like me.
I seldom comment on threads but based on community feedback i do understand the pov from both sided that for a perp dex, we cant have high fees and disallow traders to participate during high network congestions, while on the otherhand we cant seem to force traders whom actively only trade in low volume to pay additional cost.
Maybe Team could simulate the estimated cost to cover the trading fees and see if we could draw from the treasury to subsidy fees during high congestion/re-adjust fee proportion to allocate maybe 0.5-1% fees to be used for subsidies to maybe allow free trades/subsidized trades instead.
I do understand this is not a practical solution and end of day this is a issue for all onchain perps to fix, practical solution be L3 where only gmx runs on the chain, txn are then bundled and posted on the eth chain or arb to still ensure security.
That is why avax has the subnetting ideas etc, we got to accept the fact that gas prices be affected by other factors not just gmx but other dapps on the same chain and might make trading impossible for retail. So that is just a suggestion.
we may widen our section of the pie, and widen out those spikes, creating more demand, kind of front running time boost and pectra in a way, with timeboost bidding on gas, on the table anyways, these costs may be realized anyways
based on the current txn costs, during times of non-congestion the increase in txn cost would be ~$0.16
i understand that this is also a discussion in principle, that users who exclusively do not trade during times of congestion would not want to subsidise costs for users who do trade during those times
our thinking is that if we can reduce the costs during times of congestion, this leads to a better experience for all, so more users do not have to be as concerned that they would not have sufficient funds to pay for transaction costs during times of volatility to withdraw collateral or close a position
this would help to make the protocol more user friendly for all users, leading to more volume, which leads to higher APRs, which leads to higher liquidity, lower price impact, in turn benefiting all users
having said these, this is just a proposal for the community to vote on and i hope that my position as lead dev does not overly sway people into this solution
the snapshot proposal has been created, but if the quorum of 50k votes is not reached then the proposal will not be implemented
we will continue to work on alternative solutions regardless of the result of the vote, appreciate everyone’s input on this as we shape the solutions
