[X] My Assistant

So is it possible that this Isekai season would last for entire year until June, and skip a whole season ?

possible but not likely. Boss has already started working on the new release. And usually he finishes what he starts first, before taking on a next improvement. If HV would take up so much time, there would not be any time to tweak the search engine further, for instance. So I personally don't expect the delay to be that long.

I was planning to attend the next Isekai season

I am still planning to attend the next Isekai season.

Should I kill that plan and pick the current season's progress on?

I'd wait. Work on next season has been started. But we just don't know when work will be finished.

Saturday I was wondering, "When will I get another crystalized phazon"

Then Monday raised its hand, lol

The galleries are down for some reason. Here's an directly link to HV if anyone needs it.
https://hentaiverse.org/

The Market still has an issue with rewarding bot-like behavior. You can run into somebody who's willing to overbid by 1% up to some limit, then drop down and repeat it again, endlessly all day long. Or make eternal 1c adjustments to have a more desirable position in the stack, since only the leading price is limited to 1% adjustments. This behavior is easy to codify as an algorithm, and with the way the market works, it does increase profits.

I repeat my suggestion to try limiting the number of price adjustments per day per item, so people aren't forced to meatbot to compete. And as before, I note that there is no RiddleMaster on price adjustments to block bots from "playing the game a different way".

The Market still has an issue with rewarding bot-like behavior. You can run into somebody who's willing to overbid by 1% up to some limit, then drop down and repeat it again, endlessly all day long. Or make eternal 1c adjustments to have a more desirable position in the stack, since only the leading price is limited to 1% adjustments. This behavior is easy to codify as an algorithm, and with the way the market works, it does increase profits.

I repeat my suggestion to try limiting the number of price adjustments per day per item, so people aren't forced to meatbot to compete. And as before, I note that there is no RiddleMaster on price adjustments to block bots from "playing the game a different way".

I can see that one could write a bot to do what you describe. But how should anyone be able to profit from it? Being 2nd best seller and selling faster than others? Still will be a reasonable price that you get, and the best seller will still be first. Overbid 1% repeatedly? You'll end up buying at the seller's price - not the best deal, you could have waited. Or are there items where the spread is very high?

Do you have examples of this happening? That would help the case for change for sure.

I've been doing this with someone on Infusions of Storms for the past several days.

(IMG:[files.catbox.moe] https://files.catbox.moe/bzbno0.png)

When the buy price goes up to 211c, they drop down to 190. If you stay at 211, you're now a chump buying at 10% more than you need to, because you could drop down to 192. But if you do, then they go to 194, and the one-upping continues back to 211.

When you have two resellers competing, this strategy allows the "ever-present" party to either be in the lead, or force the other party to give up most of the potential profits to prevent the cycle from starting again.

Similarly on the sell side, they will walk you down to 228 by 1 credit increments, then jump back up to 241 under the next large sell order. So an "ever-present party" will either be in the lead, or force the other seller to accept a much lower profit in order to not have to sit there and noodle prices all day long.

With two ever-present parties with an equal update rate, it breaks down to a 50/50 shot of being in the lead, assuming neither one is willing to remain at a disadvantageous price. If one has a substantially higher update rate, they stay in the lead.


Also, in a case like Unicorn Horns here, whoever has the higher update rate on the second-highest price will get to buy when a large dump eventually occurs.

(IMG:[files.catbox.moe] https://files.catbox.moe/er0gns.png)

This post has been edited by Necromusume: Feb 1 2023, 22:59

When you have two resellers competing, this strategy allows the "ever-present" party to either be in the lead, or force the other party to give up most of the potential profits to prevent the cycle from starting again.

With two ever-present parties with an equal update rate, it breaks down to a 50/50 shot of being in the lead, assuming neither one is willing to remain at a disadvantageous price. If one has a substantially higher update rate, they stay in the lead.

If you have two people meatbotting and both choose to give up the price point the other is unwilling to pay, then yeah, you'll get that. Either don't give up that price point, or let them have it while keeping your second place bid. Just because they dropped their bid doesn't mean they dropped off the planet, they are probably still lurking and still willing to place bids up to that same price again.

Limiting the number of price adjustments you can do won't really fix anything, you just get a contest of who runs out of adjustments first - in fact, the party who had the highest price to begin with will probably lose, since they probably spent some of their quota already. Which would probably be even more frustrating.

(And before people suggest it, I'm not making any fundamental changes to how the market works. It's far too much time and effort.)

What about this: Constrain the valid market bids to be members of the geometric series 10*(1.01^n), so the valid prices differ by 1% all the way up & down the stack. There would be move up/down buttons so users don't have to compute it, and entering a price would give you the nearest valid bid. That would get rid of the 1c bid warring that we're currently doing for position within the stack. Would make it faster and more efficient for all parties to find the stable prices where the noodling can stop.

What about this: Constrain the valid market bids to be members of the geometric series 10*(1.01^n), so the valid prices differ by 1% all the way up & down the stack. There would be move up/down buttons so users don't have to compute it, and entering a price would give you the nearest valid bid. That would get rid of the 1c bid warring that we're currently doing for position within the stack. Would make it faster and more efficient for all parties to find the stable prices where the noodling can stop.

Your example was a ~200C item, which would only be constrained by +/- 2C at that level, so even leaving out the fact of it being very awkward UI-wise, it wouldn't make much of a difference. I don't think there is a problem with today's system except if, like in your case, people drop their bid after they have "won".

What about this: Constrain the valid market bids to be members of the geometric series 10*(1.01^n), so the valid prices differ by 1% all the way up & down the stack. There would be move up/down buttons so users don't have to compute it, and entering a price would give you the nearest valid bid. That would get rid of the 1c bid warring that we're currently doing for position within the stack. Would make it faster and more efficient for all parties to find the stable prices where the noodling can stop.

I have another example where this is beneficial. Say I place a few highest bids for items I want to buy before going to sleep. Maybe someone outbids me by the minimum amount and I move down to second in the queue. However, with no bid increment for non-winning bids, I'll quickly move down from 2nd in the queue to 3rd, 4th, or even 10th, despite none of the queue-jumpers being willing to outbid me if I was still winning. Maybe I wake up and see one person ahead of me by 1c and get annoyed. With a minimum bid increment for all bids (not just the highest buy and lowest sell), I would stay 2nd in queue if only one person outbids my leading bid.

Now I can avoid this altogether by placing my highest bids further ahead of the next highest bid, to the point where if one person more impatient than me outbids me, the patient folks still won't outbid me. I don't like the queue jumping, but the current system is certainly usable with these sorts of workarounds, and very much appreciated.

In the infusion of storms case, if you don't drop your bid when you're winning, you're only a chump if someone sells a huge amount at a time. Then the second place bidder "wins" by filling their order at a significantly lower price. The rest of the time you'll still win because the second place bidder gets zero, and zero at 210 each is the same as zero at 190 each, and they ensure you don't get a clear shot at winning at any lower price.

In the unicorn horn case, I certainly dislike the queue jumping for second, but one could fulfill the small overbid to protect their highest buy bid. Still frustrating when you don't catch it in time and end up in second by 1c, and it can feel a bit arbitrary that you can sometimes outbid your rival by 1c and sometimes you can't.

tl;dr - I favor a minimum bid increment for all bids, in order to protect your place in the queue even after you get outbid. There are workarounds, so the market is plenty usable without this feature, but I think this would make the market even better.

This post has been edited by mathl33t: Feb 5 2023, 23:43

Your example was a ~200C item, which would only be constrained by +/- 2C at that level, so even leaving out the fact of it being very awkward UI-wise, it wouldn't make much of a difference.

I didn't want to spam too many examples. In the 500c price range it's already 1/5 the possible prices, and we also have 1.6k, 3.2k, 8k, 20k, 40k, and up. It feels like there's a problem when whenever you look at your sell order in the 20k range for Repurposed Actuators, someone has moved their sell wall under you by exactly 1c. If you're both selling 100 units, and somebody buys up to 100 units, the last to update wins the sale, and sacrifices effectively nothing. Even if you move down by 200c, they will never hesitate to move by only one credit. You have to check all your prices every 5 minutes 24 hours a day. I'm looking for ways to deny (meat)bots the permanent advantage.

It will take a lot less time to find a stable price if there are between 1/5th and 1/200th the options, and if they have to actually reflect on giving up at least 1% of the spread, instead of .005%. If they can price jump impulsively while losing nothing it will never stabilize.

For people just selling or buying, it may not be a big deal, but for reselling, the difference between 2% and 3% returns after fees compounded every few days is significant.

(1 + .02)^N N=10 1.22x N=20 1.49x
(1 + .03)^N N=10 1.34x N=20 1.81x

I can see the up/down arrows being arranged vertically right next to the price list, and visibly moving your bid up and down as you click them, to be finalized when you click the big Commit button at the bottom, with Revert right next to it. I don't think it would be unintuitive.


Really, the point I have been trying to get across is in that compound interest formula. Why would you want to "give up" your "winning price"? It's not just for the hell of it. *If* you can sustain being in the lead anyway, due to having a high enough update rate, the difference in returns per buy-sell cycle *multiplied across a sufficient number of compounding intervals* makes for a *huge increase in profits*. *That* is why it's a risk that people may run bots.

It's even worth mapping out when your primary opponent goes to bed, so you can move down while they're asleep, and move up just before they come back online.

This post has been edited by Necromusume: Feb 6 2023, 01:07

Well, I'll consider adding bid increments, since it's probably an improvement by itself, but I'm not sure fitting any more buttons in is really doable UI-wise since they have to be a certain size with a certain padding to be usable on all devices, and it's pretty cramped as it is.

End of Days. (IMG:[invalid] style_emoticons/default/wink.gif)

only a few days left in the fun-with-monster-names event. 40m prize pool, and 5k hath - every entry, a monster name that is at least a little bit funny, will get a share! There aren't that many entries yet, this is easy income.

1,224,273,530 / 1,269,226,211

Next: 44,952,681

It will take a lot of time

Could there at least be a change that the minimum overbid stays 1% even if there are items on the sell side within 1% of the current highest buy offer? People use this multiple times a day to sneak upwards by as little as 1c when they aren't willing to overbid by a full 1%, which is contrary to the intent of the system.



If there are items within 1%, you get those at the actual price, but you don't get to use one item at +1c to sneak your 5,000 unit order upwards by +1c like the bad old days.

Hey quick question, do you have to be above the equipment level to use it?

AKA be lvl 300 to be able to use a lvl 300 longsword ?

Thanks

Hey quick question, do you have to be above the equipment level to use it?

AKA be lvl 300 to be able to use a lvl 300 longsword ?

Thanks

Yes.

Ok thanks

Is this good enough for lv300?
Magnificent Amber Power Leggings of Protection