Raider.IO Development Updates - Raids and Mythic Plus

[Deleted]

Awesome, that's your opinion. Other people find that they are rewarding enough to be doing.

Yeah whatever... they are not rewarding enough at all

[Alvito]

Unfortunately Blizzard doesn't provide us with the gear someone was wearing on a particular run. We can only take our best guess and go with what the character is wearing when they are logged out at the time of the mythic + scan. We understand that this won't always be accurate, as lots of people change specs and gear sets for various reasons. Down the road we want to give people the ability to log in and edit this information should they choose to do so. Aspyr (the programmer) can provide a little more insight into how all of this works though, when he's online tomorrow

Just pointing out the error, it does seem odd to include a guess at gear though. Why the extra step if you know it is so easily off the mark?

[Ulsoga]

Just pointing out the error, it does seem odd to include a guess at gear though. Why the extra step if you know it is so easily off the mark?

Hmm, well I guess you could view it as a "use it or lose it" analogy. We figured it's better to have these scans that represent a particular snapshot in time, than to just not have them at all. I think (and I may be wrong!) that more often than not the gear and spec are going to be correct for most people. That's not to say that people like yourself are a rarity though. When we get around to adding accounts to the site, these are the sort of things we want to allow users to tweak. It's an interesting problem to try and tackle and there is likely more than one solution. Aspyr probably already has some thoughts on the topic floating around in his head. Feedback on how to better in improve it is welcome though, as always

[Reloe]

Yeah whatever... they are not rewarding enough at all

oh yea wasn't worth at all to do that arcway+19 within 30minutes to get those 10mil AP, I would've totally gotten that by doing worldquest... right.

But to get back to the actual point. I like what Emancptr suggested, just needs some thinking work to figure out the best way to handle the score, because I think a M+18 3chest should still be higher ranked than a what he called "Mythic+19 -1chest" as 3chesting something is way harder than beating the same dungeon on one level higher with 120% of the normal timer.
Like 3chesting a +18 should give about the same score as 1chesting a +20 key, because both of those runs result in a +21 key, but the generally idea to make the score based on the completion time instead of just subtracting like 0,01 for the 2nd best run like wowprogress does is definelty something i'd like to see.

[Aspir]

Just pointing out the error, it does seem odd to include a guess at gear though. Why the extra step if you know it is so easily off the mark?

I did want to clarify that the gear *is not* a "guess" in the sense that we are making things up. The uncertainty that Ulsoga mentions about the gear, is the fact that the gear shown was something our system saw the player wearing at the time the M+/raid was scanned. So it is still real gear that this player has used in the past, but there is a chance of error due to the fact that the scans can only pickup data about a player at logout... and the longer from the actual event (like an M+ run) to when we scan the character, then the larger the opportunity is for the player's gear to have changed.

We do think it's useful to try our best to show this type of information, but we may not be making this limitation (intrinsic to the Blizzard API) as clear as we ought to on the site... so we'll think of how we improve on that.

[Alvito]

I did want to clarify that the gear *is not* a "guess" in the sense that we are making things up. The uncertainty that Ulsoga mentions about the gear, is the fact that the gear shown was something our system saw the player wearing at the time the M+/raid was scanned. So it is still real gear that this player has used in the past, but there is a chance of error due to the fact that the scans can only pickup data about a player at logout... and the longer from the actual event (like an M+ run) to when we scan the character, then the larger the opportunity is for the player's gear to have changed.

We do think it's useful to try our best to show this type of information, but we may not be making this limitation (intrinsic to the Blizzard API) as clear as we ought to on the site... so we'll think of how we improve on that.

I wasn't trying to imply you were just randomly filling in blanks. Nor am I trying to dump on the site I am all for something that tracks mythic + better then what we have now. Just pointing out some information not matching up since you are still working out the kinks in the site. I imagine this will happen often with hybrid toons and other pure classes. I am also certain that it will happen less at higher levels and ultimately that is what will be scanned more so it may not matter.

[Deleted]

Unfortunately heroic rankings are a tricky thing. Blizzard provides us with an API for getting the exact date and time for any Mythic kill... but they don't give us this same luxury for Heroic or Normal kills. What Blizzard DOES provide is the date of your last Heroic or Normal mode kill. That means that if your guild wasn't added to our system until say... last week, our first detected kill might be from last week. This is a really unfortunate thing for scanning, and Wowprogress has cope with it as well.

Didn't know Blizzard API was so bad, never worked with it at all. I wish you really good luck on the succes of this webpage since it is a neat work overall I myself will add it to markers to check out when ToS comes out. Cheers on the webpage and thanks for litsening the users

[nastiest]

Pleeeeeaaaaase when you do M+ rankings do not include depleted runs into those. For me that is the main reason why the M+ rankings on wowprogress suck.
They currently list a depleted +18 run higher than someone 3chesting a +17 for example.

Hey Reloe, good seeing you here (watched your bear challenge video from the first iteration, nice work)

I'm an extremely active m+ player myself and I see where you're going with this and don't agree fully.

1. As soon as you insert a keystone, it becomes a depleted key. Next time you enter a dungeon and start one, take a look at it in your bags. So technically, we're always running on a depleted key. In this case, is it even possible to determine which are depleted from the start?

2. You never know the circumstances behind a depleted key. Was it a DC? Was it a rage quit? I think players get the short end of the stick often enough to where in this case, it's best to give the group the benefit of the doubt versus assuming that they just weren't good enough to finish it in the first place.

3. Even a depleted key timed to what would have been a 3 chest will upgrade the key +3. Basically, the penalty is already so there is no loot. Why double dip where external sites are not giving credit?

4. This is a matter of opinion and subjective experience, but I don't believe that a 3 chested level x is definitely more difficult than 0 chest level x+1, especially when dealing with the curvature of scaling and the 9-to-10 thresholds. For example, we had a group 1 chest a 19 EoA tyrannical and spent 2 hours on the ensuing 20 EoA (same group). Not only do all mobs and bosses receive the additional percentage, but the tyrannical multiplier is also applied to that percent increase. This is an even calculation across all mobs, so the difference is significant.

Aspir/Ulsoga,

I love the idea of incremental scoring. That is something that wowprogress lacks and would definitely spark a culture shift in how dungeon completions are approached.

Sent from my SM-N920P using Tapatalk

[Reloe]

Hey Reloe, good seeing you here (watched your bear challenge video from the first iteration, nice work)

I'm an extremely active m+ player myself and I see where you're going with this and don't agree fully.

1. As soon as you insert a keystone, it becomes a depleted key. Next time you enter a dungeon and start one, take a look at it in your bags. So technically, we're always running on a depleted key. In this case, is it even possible to determine which are depleted from the start?

2. You never know the circumstances behind a depleted key. Was it a DC? Was it a rage quit? I think players get the short end of the stick often enough to where in this case, it's best to give the group the benefit of the doubt versus assuming that they just weren't good enough to finish it in the first place.

3. Even a depleted key timed to what would have been a 3 chest will upgrade the key +3. Basically, the penalty is already so there is no loot. Why double dip where external sites are not giving credit?

4. This is a matter of opinion and subjective experience, but I don't believe that a 3 chested level x is definitely more difficult than 0 chest level x+1, especially when dealing with the curvature of scaling and the 9-to-10 thresholds. For example, we had a group 1 chest a 19 EoA tyrannical and spent 2 hours on the ensuing 20 EoA (same group). Not only do all mobs and bosses receive the additional percentage, but the tyrannical multiplier is also applied to that percent increase. This is an even calculation across all mobs, so the difference is significant.

Aspir/Ulsoga,

I love the idea of incremental scoring. That is something that wowprogress lacks and would definitely spark a culture shift in how dungeon completions are approached.

Sent from my SM-N920P using Tapatalk

I don't think you quite got what I was saying with that, I don't mean "running on a depleted key". I don't care if you tried 1 or 20times to get the dungeon done. I mean that if finishing the dungeons results in a depleted key then it shouldn't count aka not finishing in time.
and yes I know that the key is technically depleted once you insert it, but as I said we are discussing about the result.

Oh and another reason why non-intime keys shouldn't count at all:
If they do count you could shuffle around your group to give the score of that dungeon to multiple people with just one key by never making it in time and just inviting new people everytime, people have actually done that with the current system.

[nastiest]

I don't think you quite got what I was saying with that, I don't mean "running on a depleted key". I don't care if you tried 1 or 20times to get the dungeon done. I mean that if finishing the dungeons results in a depleted key then it shouldn't count aka not finishing in time.
and yes I know that the key is technically depleted once you insert it, but as I said we are discussing about the result.

Oh and another reason why non-intime keys shouldn't count at all:
If they do count you could shuffle around your group to give the score of that dungeon to multiple people with just one key by never making it in time and just inviting new people everytime, people have actually done that with the current system.

Ah ok. I misunderstood. Well, I don't agree with 0 score, but I don't believe completion should equal a similar score to a timed dungeon. I think we can at least agree on that piece. I do like scoring at a reduced rate for depleted keys.

We had a guy on my previous server who used to help his friends out with high keys the way you mentioned. Just ran people through the same one for points

Sent from my SM-N920P using Tapatalk

[enyesi]

needs to have ur own page, so that me, as a recruiter can find people

[Ulsoga]

May 4, 2017


- Mythic+ leaderboards (see write-up below for how scoring works)
- Increased mobile support for the M+ pages
- FAQ updated and given a place in the top bar navigation


So today we have implemented our new Mythic+ leaderboards. After a fair amount of discussion within the community we came up with a scoring system that we think works pretty well.


Here are the details on how we currently score Mythic+ and our philosophy behind these decisions.:


1) A score is made up of two factors: BaseValue and TimeBonus.

2) The BaseValue is calculated as MYTHIC_LEVEL * 10. This means a +8 Mythic level is worth a base of 80 points. A +22 run is worth a base of 220 points, and so forth.

3) The TimeBonus is then added to this value to adjust for how fast (or long) it took for the run to be completed. This TimeBonus swings between the range:

[Max(MYTHIC_LEVEL * -0.5, -15), MYTHIC_LEVEL]A positive bonus is calculated directly from how long your run took vs. the dungeon's base timer:

MYTHIC_LEVEL * (1.0 - RUN_TIME_MS / DUNGEON_TIME_MS)Our philosophy here is that the higher the M+ level the harder it is to complete it within the timer, so we wanted to offer more reward for those that do.

Conversely, we felt a long that takes a long time over the dungeon's timer should have a way to "down rank" in value, which would allow a really fast run from the previous level to potentially surpass it. This downranking cannot occur at rankings where there are affix changes.

4) However, TimeBonus can go negative. As the run takes longer and longer over the dungeon's base time, we apply a linearly interpolated penalty based on the specific time. This penalty is capped at a max of 10 (a full "tier" of rankings) which is achieved if the run has gone over twice the base time for that dungeon.

Again, our thinking here is that we want to have the opportunity for runs to essentially "down rank" if they have exceeded the timer by too much, but we didn't want players to be penalized much for just barely missing the timer.


Looking forward to seeing the new input we receive on this update. Thanks all!


Ulsoga
Raider.IO

[Bowshewicz]

I did want to clarify that the gear *is not* a "guess" in the sense that we are making things up. The uncertainty that Ulsoga mentions about the gear, is the fact that the gear shown was something our system saw the player wearing at the time the M+/raid was scanned. So it is still real gear that this player has used in the past, but there is a chance of error due to the fact that the scans can only pickup data about a player at logout... and the longer from the actual event (like an M+ run) to when we scan the character, then the larger the opportunity is for the player's gear to have changed.

We do think it's useful to try our best to show this type of information, but we may not be making this limitation (intrinsic to the Blizzard API) as clear as we ought to on the site... so we'll think of how we improve on that.

Is there any possibility to use log information to derive this info, or at least improve its accuracy in certain cases? I know WarcraftLogs shows a character's gear level at the time the data was logged.

[Ulsoga]

Is there any possibility to use log information to derive this info, or at least improve its accuracy in certain cases? I know WarcraftLogs shows a character's gear level at the time the data was logged.

I think the general issue with log info is the risk of people being able to tamper with it before they upload it. Armory information is untainted and comes directly from Blizzard, so there is no risking of "cheating" in that sense. Even so, I think it's certainly possible to glean useful information from logging (WarcraftLogs is a great example) so it is definitely something to keep in mind

[Emancptr]

...

I may be dumb, but I don't see the leader boards that you are talking about. I simply see the highest M+ dungeon completed for each dungeon. I do really like the stars next to the M+ level to show the number of chests though that is really well implemented (although rip'd in 7.2.5)

Am I missing where to find them?

Having read what your plan is, I like the way that you're handling it but would like to see it in practice to fully understand.

While I'm here i'll add a few other comments because I really like the direction that this is going. Some of these you've already heard before and are probably thinking about implementing.

1) Login/Account Registration - I would like to be able to "claim" my character and either put LFG or a short message on my characters profile. Claiming characters can be done by requiring them to log out with specific pieces un-equipped (wowprogress and other sites require similar things)

2) Guild Recruitment Functionality - I don't know wowprogress' traffic patterns, but I would guess a significant portion of them are driven by searches for guilds and/or character recruitment. Personally 90% of the reason I go to wowprogress is to see what guilds are recruiting.

3) Region/World Rank on Realm Pages - If I'm on a page for Kel'thuzad realm it doesn't show me the World/Region rank of the guilds (it merely ranks them based on the server (See image below). I would like to know if the Realm 1st Guild on Kel'thuzad was Region 10th or 100th or whatever without having to click on that individual guild. You have a bunch of space that I think you can fit it in given the below screenshot (Region/World/Server Ranks would work or do it after the guild name).

[Bowshewicz]

I think the general issue with log info is the risk of people being able to tamper with it before they upload it. Armory information is untainted and comes directly from Blizzard, so there is no risking of "cheating" in that sense. Even so, I think it's certainly possible to glean useful information from logging (WarcraftLogs is a great example) so it is definitely something to keep in mind

Oh true, I just opened up a log file and it's just plain text. I'm wondering whether logs are often tampered before being uploaded to WarcraftLogs now

[Ulsoga]

I may be dumb, but I don't see the leader boards that you are talking about. I simply see the highest M+ dungeon completed for each dungeon.

Am I missing where to find them?

Having read what your plan is, I like the way that you're handling it but would like to see it in practice to fully understand.

If you look at the top of the main page you'll see 4 large tabs, 3 of them are the current raid instances and 1 of them is for the M+ leaderboard. Once you click that you have a couple of filters in the sub nav to play with (specific dungeon and region). Here is a link for you to get there directly

http://raider.io/mythic-plus/black-r...d/leaderboards

- - - Updated - - -

Oh true, I just opened up a log file and it's just plain text. I'm wondering whether logs are often tampered before being uploaded to WarcraftLogs now

It is definitely possible, and I wouldn't be surprised if it has happened. I suspect WarcraftLogs might have some checks in place to look for fishy uploads. In the end I think it just doesn't matter too much because most people want an honest representation of how they are doing.

[Emancptr]

If you look at the top of the main page you'll see 4 large tabs, 3 of them are the current raid instances and 1 of them is for the M+ leaderboard. Once you click that you have a couple of filters in the sub nav to play with (specific dungeon and region). Here is a link for you to get there directly

I see all of that, but I thought with your talk about the points calculations there would be a separate leaderboard that ranks individual characters based on their M+ completions.

i.e. The summation of my dungeons given your scoring system would be 2,200 points or whatever and that would be compared to other individuals. Or i'm grossly misunderstanding your first post.

I guess looking at it more in-depth I see that your evaluation system is in place for individual dungeons (some +22's will beat out really bad +23s). So I understand where your scoring system is in place in the current leaderboard system that you have. I would just take it the next step and apply it to characters. Then create a leaderboard of individual character's M+ score after you've calculated their scores.

While the leaderboards you have now are good, most people are more interested in the collective score for their own character or other players M+ than each individual dungeon's leaderboard.

[Ulsoga]

I see all of that, but I thought with your talk about the points calculations there would be a separate leaderboard that ranks individual characters based on their M+ completions.

i.e. The summation of my dungeons given your scoring system would be 2,200 points or whatever and that would be compared to other individuals. Or i'm grossly misunderstanding your first post.

Ahh! Yes! We do plan on having individual character leaderboards as well, we just don't have them implemented yet. Hopefully we will have all of that worked out soon so that you can see at a glance what sort of experience a person has running M+. The scoring will be done a little differently than the group scoring. We are still working out those details and will have more to share on that soon.

Regarding your other feedback - valid point on adding the World ranking to the realm pages! We plan on implementing accounts and guild recruitment features, but both are fairly hefty updates that might take us some time to complete. They are on the to-do list though!

[Halym]

Functionality of the site is amazing. I can see and feel the progress as the weeks go by, but one thing has been bothering me since I found out about this a while ago ...

Visually the website is very pedestrian.
Any visual upgrade forthcoming? Something a bit more modern and sleek?