• Steam recently changed the default privacy settings for all users. This may impact tracking. Ensure your profile has the correct settings by following the guide on our forums.

Introducing points system (EXP) and cross platform leaderboards

In my opinion, that would be too much manual work, instead of working on tracker features. And it would get too subjective/personal. How would Exophase separate "spam games" from easy, indie or short games? Every gamer has his own perception of what that would be. And it exists on all platforms, not just Steam.

Edit: And you can still get the top place, by playing games with rare achievements. Easy games are barely worth EXP now compared to how much it used to be in the past.
Oh yeah I definitely agree with the too much manual work point. There are a ton of games that full under this distinction and I imagine it would be hell for those attempting to do it. That said, I feel like there is definitely a big distinction between spam games and shorter/smaller games, but I understand why some cases may potentially be a bit indecisive.
 
Hi, can anyone help me understand how achievements that haven't been earned by anyone are counted? I'm currently 1 of only 2 players tracked on Metal Gear 2 for Retro - I've earned 43/58, which includes every achievement that's been earned by anyone, and I'm showing 427 exp out of a possible 427, since the unearned ones supposedly count for 0 exp. However, I'm also showing only 45% completion, which indicates that unearned achievements do count toward completion %, and heavily so.
 
Hi,
trophies that have not yet been earned by anyone actually count 0 points towards the EXP. However, for the completion percentage, instead of the EXP, the platform-specific point system is used. So for example for Xbox Live it is gamerscore, and for PSN it is trophy points. Thus, for PSN the completion percentage should always be equal to the one that PlayStation shows to you. Hence, to get 100% completion you really need to have all possible achievements. However, getting the largest amount of EXP does not require to have them all, but only those that have already been owned by someone else before.
 
Last edited:
Hello, I have more thant 400 achievements on World of Warcraft (Blizzard), but only 1 EXP. A bug or is this explainable? Thanks !
 
Hello, I have more thant 400 achievements on World of Warcraft (Blizzard), but only 1 EXP. A bug or is this explainable? Thanks !
My best guess is maybe the system is automatically thinking it is a spam game due to the amount of achievements in the game. It will need to be looked at by @x3sphere
 
My ranking on the Steam Leaderboard is approx. 61000 and on the total Leaderboard it is approx. 35000 (I know, not very high ;) ). I would have expected that the total Leaderboard has more players, therefore I should have a lower ranking on the total Leaderboard compared to the Steam Leaderboard. I'm wondering for some time why it shows up like this, can someone explain it to me? :)
 
My ranking on the Steam Leaderboard is approx. 61000 and on the total Leaderboard it is approx. 35000 (I know, not very high ;) ). I would have expected that the total Leaderboard has more players, therefore I should have a lower ranking on the total Leaderboard compared to the Steam Leaderboard. I'm wondering for some time why it shows up like this, can someone explain it to me? :)

The cross-platform leaderboard only includes users that have actually created an account on the site. The Steam leaderboard has a lot more users that were added through other means (usually by creating a gamercard but not registering on the site).
 
Just curious, why is the EXP for Ghost of Tsushima (PS4) so disproportionally high?

edit: now I see it, they added new trophies with the new DLC that no one has gotten yet. That might be the reason. I still think it's kinda weird that I got 18k EXP just from that game even though I don't have any of those super-rare trophies and only earned the more common ones from the main game. Not complaining though, lol
 
Last edited:
Yeah it's due to the DLC. Only 0.01% of owners have earned some of the DLC trophies. It'll go back down as more earn the DLC trophies.

Awhile back we used to calculate the EXP for DLC sets independently but most people didn't want that (since they are worth almost nothing then).
 
I know I'm late to the discussion on this topic, but are we sure using the site's rarity data instead of each platform's rarity data is the best idea? I understand total players/unlocks/rarity data can't be seen for every platform, but doesn't using data from as many players as possible result is better accuracy when calculating these things? Limiting it to just the people who have used this site feels like purposefully... shooting the calculations in the foot from the beginning, if that makes sense.

The RetroAchievements platform in particular feels hampered by this as it's a much smaller community that most other platforms to begin with, and only a small fraction of that community has used this site. I don't know if Retro is one of the platforms that offers this data (I would assume it does, given it's on their site) but if it is I feel like it would be very beneficial to use it. Lots of games have few or no players tracked here.

Again I understand the data isn't available for every platform and there's no "perfect" way of calculating this stuff either way, but I feel like using it when it is available has to be better than not. It's like we're currently taking a poll with 100 people instead of 10,000 people, even though they're available. Even if the calculations have already been refined to help account for smaller player counts I feel like we may as well include as much data as possible to get the most complete picture.
 
I know I'm late to the discussion on this topic, but are we sure using the site's rarity data instead of each platform's rarity data is the best idea? I understand total players/unlocks/rarity data can't be seen for every platform, but doesn't using data from as many players as possible result is better accuracy when calculating these things? Limiting it to just the people who have used this site feels like purposefully... shooting the calculations in the foot from the beginning, if that makes sense.

The RetroAchievements platform in particular feels hampered by this as it's a much smaller community that most other platforms to begin with, and only a small fraction of that community has used this site. I don't know if Retro is one of the platforms that offers this data (I would assume it does, given it's on their site) but if it is I feel like it would be very beneficial to use it. Lots of games have few or no players tracked here.

Again I understand the data isn't available for every platform and there's no "perfect" way of calculating this stuff either way, but I feel like using it when it is available has to be better than not. It's like we're currently taking a poll with 100 people instead of 10,000 people, even though they're available. Even if the calculations have already been refined to help account for smaller player counts I feel like we may as well include as much data as possible to get the most complete picture.

I have thought about doing it like that for smaller platforms due to the limited amount of data we have on those. I feel like we have a big enough sample size for platforms like PSN, Xbox and Steam though.

I think the main problem with using platform data is only a small % of the player base are seriously into achievement hunting. That probably isn't true for Retro Achievements - but talking in regards to the bigger platforms like PSN. So any achievement that is remotely difficult to earn will almost always be ultra rare. While technically accurate, it doesn't give a good gauge of how difficult the achievement is to attain - since on here, the user base represents more skilled players. I believe SteamHunters started off with using the rarity calculations from Steam but have since moved to using their own data for this reason mostly. Although they also provide an option to see points calculated either way, and a leaderboard specific to each points implementation, so perhaps we could do something like that too.

But yes, using the platform data would probably be an improvement for services like Retro Achievements. And possibly Google Play, Origin, etc... basically any service that we don't have enough users tracked.
 
Last edited:
I have thought about doing it like that for smaller platforms due to the limited amount of data we have on those. I feel like we have a big enough sample size for platforms like PSN, Xbox and Steam though.

I think the main problem with using platform data is only a small % of the player base are seriously into achievement hunting. That probably isn't true for Retro Achievements - but talking in regards to the bigger platforms like PSN. So any achievement that is remotely difficult to earn will almost always be ultra rare. While technically accurate, it doesn't give a good gauge of how difficult the achievement is to attain - since on here, the user base represents more skilled players. I believe SteamHunters started off with using the rarity calculations from Steam but have since moved to using their own data for this reason mostly. Although they also provide an option to see points calculated either way, and a leaderboard specific to each points implementation, so perhaps we could do something like that too.

But yes, using the platform data would probably be an improvement for services like Retro Achievements. And possibly Google Play, Origin, etc... basically any service that we don't have enough users tracked.

I guess taking that sort of focused approach makes sense. It's a shame the achievement hunting community is split across the various different sites though, I don't suppose you could pull the user data from other communities like SteamHunters, True Achievements etc.? That could offer a more "complete" but still achievement-hunting focused sample size. I imagine the other communities may not want to share that data though.

In any case, yes, I would say RetroAchievements is quite unique among the other platforms as it's basically a prerequisite to be into achievements/achievement hunting to earn any of those achievements in the first place, since achievements aren't set up by default like they are on Xbox, Steam, etc. You have to be interested enough to go out of your way to set up an account on their site and download the correct software, get the specific game roms that match their database, turn achievements on in RetroArch and login etc. You essentially avoid the issue you've described with the other platforms right from the beginning, so there's really no reason not to include data from the full userbase.
 
But yes, using the platform data would probably be an improvement for services like Retro Achievements. And possibly Google Play, Origin, etc... basically any service that we don't have enough users tracked.

So since we're on the topic of RetroAchievements and calculating points and such, it's worth pointing out that there are technically two separate versions of each achievement. They have "softcore" and "hardcore" versions of each. When playing in softcore mode, save states, slow-motion and rewind features in the emulators are enabled for use, making the achievements in general considerably easier to earn. To make up for this difference, if you unlock an achievement in hardcore mode you also earn the same achievement in softcore, making the hardcore achievement in effect worth double the points. This site seems to only track each achievement once, which I assume means it's either only tracking the softcore unlocks, the hardcore unlocks, or whichever comes first. Regardless it means the difficulty/rarity isn't being calculated properly, right? A softcore player would be indistinguishable from a hardcore player, or if only hardcore unlocks are tracked then softcore-only players aren't being tracked at all.

Is it possible to track the softcore and hardcore unlocks separately? If it is, that would at least solve the difficulty/rarity issue. The problem is it would double the achievement count in the progress bar and someone playing only on softcore could only ever get 50% even if they completed the game. Unless of course you split the softcore and hardcore achievements into separate sets, I guess. The way they do it over there is by having two progress bars on top of each other - blue for softcore and gold for hardcore, so you can still fill the bar up with blue if you only play softcore. The progress percentage goes up to 100% for softcore and 200% for hardcore. I don't know if something similar could be done here though, it's a bit complicated.

To perhaps further complicate this issue they've recently announced they're changing the way softcore is going to work in the near future. While softcore achievements will still be worth the same points as before, they will no longer be counted by their leaderboards in an effort to make the leaderboards more competitive and fair. Since softcore isn't going to be considered competitive going forward, they're also no longer going to moderate softcore play so cheating will be allowed. I suppose the best way to deal with that would be to make the softcore achievements worth 0 EXP here, so that like on their site, softcore achievements just won't contribute to rankings. Although there may actually end up being less cheating on the softcore side since it no longer contributes to their leaderboards anyway, so perhaps there will end up being less cheating rather than more, who knows.

Here's the announcement for softcore changes if you want to read it.

Phew, sorry for the wall of text but that required some explaining! What do you think? Are you able to track softcore and hardcore unlocks separately at all or are we stuck being unable to distinguish them?
 
They unfairly get rid of leaderboards for the casual softcore players and yet you're still complaining even on this website? lol
 
Last edited:
They unfairly get rid of leaderboards for the casual softcore players and yet you're still complaining even on this website? lol
Huh? I'm not complaining about anything, I just realised exophase only tracks achievements from RA once when there are technically two versions of each achievement. Then I saw the announcement about the softcore changes so I pointed that out too, since it's relevant.

I actually pointed out that exophase might not be tracking softcore achievements at all and I think it should if possible.
 
That announcement on RetroAchievement just isn't fair for the majority of softcore players and the years they spent on that website just to get screwed over at the end.
 
Last edited:
It's just not fair for the honest softcore players who play because they just don't have the time to invest. I understand there are cheaters out there but this really isn't fair for those who play fair to just be strip out of leaderboard on every website, y'know? that's just my 2 cent.
It's a tricky subject and this isn't really the place to discuss whether the changes they're making are right, as that has nothing to do with this site.

What could be an issue on this site is the fact that they're planning on making softcore mode a sort of "wild west" option where players will be explicitly allowed to cheat and don't need to follow sections of RA's User Code of Conduct which restrict how achievements can be earned (and by extension, the Unified Achievement Hunting Rules which this website follows). If that ends up being the case, exophase might have to exclude softcore achievements from contributing to the leaderboards just to stay inline with those rules.

I agree that cheaters ruining the RA leaderboards for the honest softcore players is unfair and I sympathise, but how the RA staff deal with that issue is ultimately up to them and should be discussed over there rather than here.
 
So since we're on the topic of RetroAchievements and calculating points and such, it's worth pointing out that there are technically two separate versions of each achievement. They have "softcore" and "hardcore" versions of each. When playing in softcore mode, save states, slow-motion and rewind features in the emulators are enabled for use, making the achievements in general considerably easier to earn. To make up for this difference, if you unlock an achievement in hardcore mode you also earn the same achievement in softcore, making the hardcore achievement in effect worth double the points. This site seems to only track each achievement once, which I assume means it's either only tracking the softcore unlocks, the hardcore unlocks, or whichever comes first. Regardless it means the difficulty/rarity isn't being calculated properly, right? A softcore player would be indistinguishable from a hardcore player, or if only hardcore unlocks are tracked then softcore-only players aren't being tracked at all.

Is it possible to track the softcore and hardcore unlocks separately? If it is, that would at least solve the difficulty/rarity issue. The problem is it would double the achievement count in the progress bar and someone playing only on softcore could only ever get 50% even if they completed the game. Unless of course you split the softcore and hardcore achievements into separate sets, I guess. The way they do it over there is by having two progress bars on top of each other - blue for softcore and gold for hardcore, so you can still fill the bar up with blue if you only play softcore. The progress percentage goes up to 100% for softcore and 200% for hardcore. I don't know if something similar could be done here though, it's a bit complicated.

To perhaps further complicate this issue they've recently announced they're changing the way softcore is going to work in the near future. While softcore achievements will still be worth the same points as before, they will no longer be counted by their leaderboards in an effort to make the leaderboards more competitive and fair. Since softcore isn't going to be considered competitive going forward, they're also no longer going to moderate softcore play so cheating will be allowed. I suppose the best way to deal with that would be to make the softcore achievements worth 0 EXP here, so that like on their site, softcore achievements just won't contribute to rankings. Although there may actually end up being less cheating on the softcore side since it no longer contributes to their leaderboards anyway, so perhaps there will end up being less cheating rather than more, who knows.

Here's the announcement for softcore changes if you want to read it.

Phew, sorry for the wall of text but that required some explaining! What do you think? Are you able to track softcore and hardcore unlocks separately at all or are we stuck being unable to distinguish them?
Yes, it's possible to distinguish hardcore unlocks in the RA API. Maybe we could leave it up to a user option as to whether they want softcore or hardcore achievements to be counted on their profile. And we can have two leaderboards - one for softcore and hardcore, similar to how the PSN leaderboard can be viewed by platform.

If we use the rarity stats from RA, then softcore achievements should be worth less EXP anyway. I don't think we necessarily have to make them worth 0, they will presumably be devalued enough compared to the hardcore achievements.

Open to suggestions... unless I'm missing something, seems like implementing it this way would make it fair to those that use either mode.
 
Yes, it's possible to distinguish hardcore unlocks in the RA API. Maybe we could leave it up to a user option as to whether they want softcore or hardcore achievements to be counted on their profile. And we can have two leaderboards - one for softcore and hardcore, similar to how the PSN leaderboard can be viewed by platform.

If we use the rarity stats from RA, then softcore achievements should be worth less EXP anyway. I don't think we necessarily have to make them worth 0, they will presumably be devalued enough compared to the hardcore achievements.

Open to suggestions... unless I'm missing something, seems like implementing it this way would make it fair to those that use either mode.
That would be better than not distinguishing them at all, however only being able to choose softcore or hardcore tracking would mean hardcore players would lose out on the points from the softcore achievements. It's not a case of earning one or the other; they stack so if you earn the hardcore achievement you also earn the softcore one too. It also wouldn't account for players who aren't strictly in one camp. You can switch between softcore and hardcore mode on a game-by-game or even achievement-by-achievement basis, if you want to. This is why I suggested tracking both softcore and hardcore achievements if possible, it's just a case of figuring out the best way to handle that.

But again that would be great if softcore was remaining the way it currently is, but since they're going to make softcore mode an anything-goes-cheating-allowed mode in the future, it becomes a question of whether the softcore achievements should be able to contribute to any leaderboards at all. Over on RA the people running the site have decided they shouldn't. People have suggested separate hardcore and softcore leaderboards but since they won't be moderating the softcore side any longer and cheating will be allowed, they don't think such leaderboards should be made. If cheating is explicitly allowed it pretty much defeats the point of a leaderboard to begin with, right? Not to mention it breaks the Unified Achievement Hunting Rules.

For now I think figuring out how to have this site track both softcore and hardcore achievements and reflect that on our profile pages is the first step. As you said, using RA's rarity stats to calculate the EXP should make the hardcore achievements worth more than the softcore ones so that's good.

If/when they make the changes to softcore mode on RA, we can figure out if softcore achievements should stop contributing points to the leaderboards here or not. Until then they should contribute to leaderboards as they currently do on RA.
 
Back
Top