I’m assuming this is unintentional, but who knows?
5 Likes
Same here. 
3 Likes
This is why a conjoined client for Era is a terrible idea.
No one who can think even 30mins into the future of possibilities would ever use a conjoined client and Database.
5 Likes
You’ve made the mistake of assuming any of these changes/bugs were either intentional or unintentional.
They’re just doing stuff without considering the consequences of post-effects at this point.
2 Likes
Sadly this seems to have been happening for a while.
This sort of random thing keeps leaking in. Basically, I agree with Cezar.
The most annoying part is when someone who wants Era changed to fit exactly what they want comes here and talks about how many changes have already happened and no one cared so why not change the thing they want too, even when people do mention not wanting even these smaller changes. Then they’ll make fun of the people who notice the smaller things.
3 Likes
While Blizzard is making it hard sometimes, I do remain in support of a consolidated client. The very argument in support of it is the same one you made…
I think what largely takes away from the public perception of this though is…
- Many people believe the Era client should never be touched and therefore there is no benefit to having a conslidated client.
- We don’t get exposure to the positive aspects of this approach. We only see the negatives.
The core issue with both of these is that we don’t have insight into what’s happening behind the scenes. Blizzard is notorious for not engaging with the community. Every now and then someone decends down from on high and gives us a nugget of insight, then they disappear forever.
Personally I find this extremely frustrating. I think the commonly held perception is that the community can be the absolute worst sometimes and so Blizzard gets a bit gun-shy with communications. If that’s true, I would hope that they would remember that there are always going to be bad eggs but that many of us just like to know what’s going on, even if we don’t agree. For an example, my favourite post around here was when they gave us insight into their plans for WoW Classic at the start of WotLK. It was remarkably detailed and insightful! In a later patch they changed up their stance on it (ie, RDF) but it really would have been nice to see a follow up post on why the change of heart.
For the topic of consolidated clients, there absolutely are benefits going on that we don’t see. Fixes that needed to be applied to Era (that aren’t patch notes worthy) that required refactor that didn’t need duplication due to the consolidated client. Again, folks may not always agree but it’s just nice to hear sometimes.
The other side of the coin though is that this bleed through is happening enough that I’m reasonably sure at this point that Blizzard lacks a backbone of automated tests to validate core gameplay behaviour expectations. These kinds of tests, usually the functional ones, are certainly expensive to create/maintain but in the end they save a lot of time and good will because they catch stuff going into a release. A test verifying that you can mount on a boat should be fairly easy to create, assuming the engine has the relevant test hooks in place (which for a 20 year old engine, I would expect them to), and then it just gets run in an automated suite every time an appropriate build milestone is kicked off.
The only reason to not have this is because Blizzard doesn’t believe WoW Classic (not just Era) is worth the investment, which to me is absolutely mind boggling. According to ironforge.pro there are over 200k active raiders. While some are certainly alts, this doesn’t account for the non-raiders as well and they make up a reasonably high percentage of any server. Like… it’s been 5 years. People are loving Classic, Blizzard. You can stop looking at it side-eyed like it’s going to bite you at any moment.
Given this understanding, and your own acknowledgement that they don’t communicate either here or in proper patch notes, why would you then go on to continue giving them the benefit of the doubt when they’ve routinely demonstrated they aren’t worthy of that.
This is exactly the reason that so many “believe the Era client should never be touched”.
2 Likes
It’s not Blizzard I’m giving the benefit of the doubt here, it’s the approach itself. This is just based on personal experience, having worked on projects that did it both ways.
I do understand the frustration that people are feeling here, but I think we’ve latched onto the wrong thing. It’s not the consolidated client that’s the issue here, it’s Blizzard’s QA practices, specifically in the realm of Classic and Classic Era.
1 Like
QoL feature
1 Like
What fixes?
Well Era is stuck with SoD forever now. I can’t see this as a positive because every outcome is bad.
- they easily move SoD characters to Era → the worst I would unsub forever
- they create SoD Era and now Era players forever have to download SoD bloat
- they add token or other really bad features to SoD and now can easily push to Era → also = unsub.
Even for the developers, they have to tip toe around while making SoD changes trying to not break Era, and it is near impossible to write tests for this unless they already existed, which they clearly do not.
The only upside I can see of consolidated approach is players can swap between Era and SoD without having to log into a different client.
1 Like
This is pretty much where I’m at also.
Surely at this point, the consolidated client has become much more of a headache for the developers than a positive.
If SoD and Era were two separate clients then devs could simply do whatever they wanted in the SoD ecosystem and never have to retroactively bug-fix everything that it messed up in Era.
2 Likes
I can’t do half the stuff I used to.
who needs a map or skill bars on your screen.
Except I will never need to swap from era/ hc to join season of dog ish server lolol
There are always going to be folks (which may or may not include yourself) who will harumph and say that Era shouldn’t get touched ever, but that’s rather short sighted. Every piece of software requires maintenance, there aren’t really any exceptions to this.
The particular block you quoted though was in reference to Blizzard being more transparent about the positive aspects of the consolidated client. It might be nice if they did this.
They shouldn’t have to. I’d imagine the whole point is for the client to be entirely data driven and for SoD and Era to be separate data sets. There are obviously issues here but remember, even back in 2019 we started on what I can only assume is a fork of the Legion client. You’re not actually playing the 2006 version of Vanilla and you never were.
The benefit is supposed to be so that you don’t have to apply fixes on multple clients. Imagine a scenario where a game breaking exploit is discovered that allows you to dupe gold due. Said exploit was present in the Legion client and therefore exists on all versions of WoW.
If all clients were separate, Blizzard would have to make that fix on Era, HC, Cata, SoD, and Retail. As each of those code bases would have diverged from the common client at varying times, there is no guarantee that the implementation of the fix would have the same comparative cost and may require significant refactor on each client to integrate. With a consolidated client, you only need to make that fix in a single stream. The fix is deployed faster to all clients and it’s more stable because you only did the work one time. You do still have to test all the clients though…
Now this is just an example. Will such an exploit be uncovered? It might, but probably won’t. However, as I said, no piece of software is perfect and they all require some kind of maintenance. We can’t predict the kinds of things a playerbase will uncover, especially one so bent on breaking the game in any way they can to squeeze out a few more seconds on a raid clear.
Also, as I pointed out, a lot of the benefits from this type of approach just aren’t being published. Any time there’s a bug the forums screams to high heaven so everybody knows it, but nobody knows what goes on behind the scenes.
Please understand though that I’m not defending Blizzard. I’m defending a software development architectural approach. There’s a difference and I’ve been pretty vocal that I believe there’s room for improvement here.
I understand this. I just struggle to think of an example which, after the solution is discovered for one place, would be so difficult to duplicate on another.
In contrast, we now have dozens of examples of unintended impacts from SoD on Era and greatly reducing stability of hardcore. Clearly it isn’t done on purpose, it is because the interaction between these things are complex and the people changing them aren’t the ones who originally designed them. And honestly, if I were to guess, the people who originally designed most of what they are working on - would of just made a copy for SoD and built it on that.
Separate architecture fixes a major source of bugs that we are currently facing, which is changes in one version causing odd things in another. This sort of bug is extremely hard to design a test for, because if you knew the issue it would be so easy to avoid causing the bug in the first place.
“We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil. Yet we should not pass up our opportunities in that critical 3%.” -Donald Knuth
Imo a ‘mixed client’ is not a critical 3% but is harming dev productivity and causing a bad experience for players. Whereas a simple split would have made it much easier for the team to move fast with confidence that they won’t break era or kill hardcore players.
I say this all blindly though just based on what I can see and deduce and with bias towards adding friction to changes on era.
yes, this is where I am also discussing. If I were leading SoD I would have it on a separate architecture so the team could do whatever they wanted without having to worry about impact Era / hardcore.
If the project is ever proven and stable, then it may make sense in the future to merge them together. But there’s no rush to do that since it will evolve in shape quite a bit.
1 Like
Again, this is only something that I can relate to through experience. For a product I used to work on, I had to develop a fixes to issues that had an impact on multiple versions that the company still supported. In my case, on the modern client the fix was very easy to integrate. Partly because I just understood the modern client architecture better since it’s what I worked in every day, but also because of a lot of refactor work devs before me had undertaken to improve a lot of those pain points. There was a few times where the integration of the fix was significantly different, even though fix itself was mostly the same.
Pain points like this are exactly what drove my company to move to a consolidated client.
I’m not saying this is the case for WoW, but the system I worked on was a 20+ year old piece of software as well and so there are bound to be similarities. Heh, our rendering API used to be a strong competitor to DirectX… in the DirectX 1.0 days 
I think that’s the wrong conclusion to draw here. Moving to a consolidated client takes a lot of work to do. A company like Blizzard would not understake such a thing if it wasn’t going to save them money in some way shape or form. Again, you’ve highlighted that we’ve seen a few dozen or so issues come out of this. Most of them have been quickly resolved. What we don’t see are the scenarios where both Blizzard and the players see the benefits.
I’m not happy about issues leaking into Era either, I just maybe have more insight through personal experience as to why a consolidated client is a good thing. That doesn’t mean we can’t expect Blizzard to do better, but we need to stop barking up this client tree.
That’s just not how it works. Once those two code streams diverge it becomes extremely difficult to merge them back together. Once a company had done the legwork to create a client that could support multiple versions, they would be unlikely to allow them to diverge again… rightly so!
The reason it would be so hard to merge together is because they have different requirements and each will solve those in ways that make the most sense. Those ways are unlikely to be easily merged together.
Thus, by merging from the start - you are simply paying that cost in tiny amounts every single function you add. Which is what I mean by it harms the dev experience, instead of focusing on building SoD they have to worry how to make new things fit within the current system without breaking things.
This is what I mean by mixed architecture being premature optimization - save a few hundred hours a quarter of porting bug fixes to era, in exchange for thousands of hours of having to constantly deal with making SoD fit era schemas.
Think of the WoW client more like a game engine. It’s given the capability to interpret data and perform functions. The data that drives that is different for SoD and Era, but the engine that interprets that data is the same. Kind of like Unity or Unreal. The C# script or blueprints that compose your game may be different but the underlying code that executes it is the same (for a given version of either of those engines).
I’d imagine the SoD is running into issues in either (or both) of the following…
- The engine (ie, the WoW client) doesn’t currently do something they want SoD to do, so they have to build the capability into the engine to support it. Doing so break something that already exists, leading to a bug. This bug likely exists in both SoD and Era.
- The data driving a particular feature is common between SoD and Era. Blizzard changes some data to modify the behaviour of SoD, not realizing that this is common data and will impact SoD.
What you’re asking for is for Era to stay using an old WoW engine and not stay up to date. You can do this but you have all the probems I mentioned before when you want to migrate fixes.
That’s not the case though. They aren’t losing thousands of hours to save a few hundred on bug fixes. For the most part, these bugs they introduce that everybody gets upset about are quietly fixed within hours. I think there’s a few that have hung around, but they aren’t game breaking.
(inb4 Duckling mentions the Tarren Mill forge!)
There is a reason classic players are playing a 20 year old game. Leave it alone.
THIS IS ACTUALLY HOW THE GAME WAS BACK THEN!
This is restoring a 2004 behavior! I’m not even joking.
People saying this is awful for Era, this actually is how it was back in 2004 so it’s more Era than ever!
Video proof. Everyone gets on the boat and as the server catches up they dismount.
The boat originally dismounted you just like Zeps did in OG vanilla.
3 Likes