Would this trading system work for D4? It combines free trade with BoA

With D2 having non-BOA free trade and D3 having BoA very restrictive trade, the D4 team seems to recognize these polar opposite camps and is trying to find something for both. They don’t want to alienate one of these camps if possible.

Probably holes but following is the gist. (D3 item names are used throughout even though we’re talking D4).

Items dropped in the game would be BoA. The game’s keeping track of how many times a specific item has dropped game-wide. The first day after launch there might be 10,000 Furnaces in circulation game-wide. Something else like a common blue dagger might have qty 2,000,000 by comparison.

In addition to real in-game drop quantities they will have the coded drop rates of every item in their tables or code. With furnace it may be 1-in-8,000 drop rate or something and blue dagger might be 1-in-40 or something.

This would be two sources of item drop information - the real in-game drop quantities and the drop tables themselves.

To enable free trade with BoA items, couldn’t they just use a tiering system with a trade window? 2 Players would place their BoA items into the window - the game would have a tiering system that would require an item in the same tier (same approximate rarity interpolated from both dropped quantity and drop rate) to be traded to get an item of that same rarity. You have to give something good to get something good.

So for example, if I wanted a 1-in-8000 Furnace, I would need to already have my own 1-in-8000 drop rate item and it could be something like an In-Geom just as example, that I don’t want. So because In-Geom is in the same high tier as Furnace for sake of argument, 2 players could trade them. Once they’re done trading in the window, the BoA changes ownership but BoA is still in effect - just different owners.

Wouldn’t this deter botting and RMT? I couldn’t just hop onto a site and pay $10 to get a Furnace. I’d have to already have a similarly rare item to trade for it.

A drawback would be if a top tier item was unpopular because they didn’t spec it right or it’s relatively crap. Then there’s incentive to go on a website and pay $10 to get someone to give you their popular good 1-in-8,000 item for your unpopular crap 1-in-8,000 item.

So why couldn’t their balance team “rate” or change item tiers based on the in-game usefulness or power of the item? Thus even if 2 items have the same rarity, they may be adjusted and placed in different tiers.

At this point, they would have 3 metrics - the real in-game drop qty, the drop rate, and some rating from the balance-team that could adjust it’s tier.

The whole gist is just to look at using BoA as a start, use a trade window that keeps BoA in effect just changing owners, and uses a tiering or rating that is a combination of many factors to ensure that items of roughly the same rarity are needed in order to get an item of roughly the same rarity.

Very specific small example would be if they had 50 items in the game, 10 items in 5 major tiers. Tiers ranked from 5 lowest to 1 highest. You get a drop that’s in tier 1 you can trade for another in tier 1 but not for any other item in any other tier. If something in tier 1 is unpopular or underperforms, balance team might drop it to tier 2. They would most likely also want to slightly adjust drop rates if they are changing tiers as well - lower-tier items tending to drop more frequently. In practice, in D4, they might have 100’s of tiers.

I like free trade. But I also don’t like all the RMT nonsense that it brings. This seems like a middle ground that also allows for community-based non-RMT trading websites - list the item, what you want for it, and so on.

1 Like

Problem is, as you also mention, item rarity doesnt necessarily translate to the quality of the item. Often not at all.
And as such, it would not deter RMT much. Nor botting at all.

As for rating all the items based on how good they really were. That would be an extreme task. Might be somewhat doable for uniques, with their fixed stats. Though still unlikely to lead to a very good rating - I mean, if the devs could rate items that well, then they could also balance them well.
But for rares and legendaries, with “billions” of different permutations? Not realistic.

Nah. Stick with BOA. It ain’t broke, it don’t need fixing.

1 Like

With D3 and primals vs ancients that seems pretty easy to automatically rate. Since an ancient that rolls perfectly is basically a primal (minus the GG reroll potential). I think they could come up with a system that rates items based on stat totals, stat priorities, and all that.

There’s already kind of a tiering with items just based on them being magic, rare, legendary, primal, and so on.

Anyway - I do agree they’d need to come up with an algorithm that rates items based on some kind of criteria - but even if it was a “rough” rating - this kind of BoA free-trade mix might go a long way to bridging the gap between D2 and D3 camps I think. They’d need to spend time coding a rating system that might not be all that hard.

1 Like

I still maintain that if they want to have trading the best way to do it is to have self-found and trading be separate modes, then design trading entirely around people who want to trade.

RMT is going to happen one way or the other once you have a decent trading system. The amount of restrictions you’d need to place to stop it would be so harsh you might as well just not do trading.

It would actually be quite a bit of a task to nail down a system that could enforce only being able to trade an item for an item of equivalent value. Especially one that can handle the ever changing market prices, and one that could not be gamed by people looking to exploit the system.

and a system based largely on the colour rarity of the item is barely doing to do anything.

1 Like

They can do that with AI. It will have to kick in effect later in Season, but on theory the AI “could know” the value of items. Such AI could also be used for balancing of items. Once you have it you’d basically would solve two great things in aRPGs: trading and balancing.

No, it wouldn’t. Like, at all.

Black Desert Online has no 1 on 1 trading. The only way you can “trade” itens is by a centralized market system where you can only exchange itens for ingame currency (a AH of sorts). The pricing is auto regulated based on suply and demand, so you can’t really set up your price also. You can place pre orders for itens that aren’t available on the market, but that’s it. If a second person puts a pre order together with you, there’s no guarantee that you’ll get the item. With this draconian system surrounding trade, people still do black market RMT in Black Desert Online. So absolutely nothing you can think of will prevent RMT if there’s any form of trade in the game.

1 Like

Not true. Vendor trading is still trading. You can’t RMT with the in-game vendor.

We can agree that D3’s BoA system has virtually eliminated RMT 3rd party sites though right?

Botting in D3 is just paragon botting - it’s not done to put items on a website for RMT or to farm gold. With BoA, the RMT aspect of D3 completely dried up.

2 Likes

Hence why I view it as such a successful answer. Trading as it is in D3 suits me just fine. The absence of constant spamming of chat, because we all know they always respect the trade channel only… just…

None of that is a problem with this game and I love it.

1 Like

You can still find plenty of websites selling stuff. You can’t buy really specific stuff, but you can buy them nonetheless. So no, we can’t agree on that.

edit: What I think we can agree on is that D3’s system is possibly the best you can do to reduce RMT. In the end, RMT nowadays in D3 boils down to you paying someone to play for you.

Creating the AI would also be a lot of work, assuming you want one that would actually be half way decent.

If items have a global maximum drop rate, this encourages bots to loot harder and retain or destroy items so the total availability decreases. Let’s say, for example, The Furnace example of global drop of 8000. 1000 bots get 400 of them. That leaves 7600 “in the wild” with maybe 10 actually in use by the bots. Assuming half aren’t salvaged, this means 200 are just gone… forever.

It creates a means to artificially “corner” the market, in a sense.

Of course, if salvaging merely restores the item back to global drop rate, so there is always X amount, then that’s another thing.

It will, however it will save a lot of work in the future (once implemented). If the plans are D4 to be played for more than 3-4 years, I’d go for it.

All Skelos has to do is travel back to his future timeline where ones are already written, take it, and come back here and profit.

As you also mentioned previously, having AI for balancing should be so much better than whatever they do now.

What’s weird is that AI in anti-botting software is a multi-million dollar industry for eCommerce and web sites. All kinds of big companies use anti-bot software that has AI. Some of the products are pretty successful with a high accuracy rate of not bothering humans in the process.

It’s just that nobody’s really applied it to do anti-botting in gaming yet, at least none that I’ve noticed, but it clearly seems doable.

The latest AI in general just seems completely underutilized in the gaming industry.

For what you’ve explained you think the AI should be able to do, the time needed to get the initial investment back would be massive.

An AI on that scale could easily reach the scope of being on par with developing Diablo 4 itself. Developing AI is something that takes a lot of effort if you want it to do anything beyond extremely basic tasks, especially if machine learning is involved.

That is unless you can just time travel and nab us an AI from the future =P

This is limited in scope though and something like this is often done with a small team. It’s not a quantity type of project like D4 is - gigabytes of graphics, artists, modelers, world design, skill design, animations, level design, item design, character design, story, music, main quest scripting, side quest scriping, all that.

What Skelos is talking about could be done by a small team of experts. It’s a well contained project. Only took a few guys to write core Unix operating system. And Windows NT core etc. That kind of thing.

Also saw that AI beat a pro DOTA 2 team and Bill Gates tweeted about it as being significant. Kind of eye opening.

Anyway, there also may already be some rather decent AI packages for adapting to this problem. Hard to say. But for sure there are AI packages you can get for anti-botting in the ecommerce space. (If Blizz got around to getting more serious about bots.)

I wonder if the DOTA 2 AI cheated like the Starcraft 2 AI. Where they gave it full map view.
In any case those projects definitely also are on a scale that dwarves Diablo4s development.

An AI on the scale that Skelos has mentioned before in other threads would be capable of handling both balancing and trade basically in real time.

Keep in mind the AI AlphaStar that plays Starcraft 2 is developed by a company that does nothing but develop these AI, and that AI literally does nothing but play Starcraft 2.

Starcraft 2 is a game that already has its own AI opponents, but in order to do it using machine learning took an entire dev team by itself. This is based off tech they’ve been working on for 10 years.

Bill Gates tweeted about the DOTA AI being significant because it IS significant. Both in terms of the scale of the project and what they’re actually achieving.