I am very surprised to see this next patch focusing on new content when the current content is bug-ridden and desync still affects all.
Of course, I'm only one voice in many, but I, for one, feel that the notion of adding new content when there are things needing fixed is somewhat backwards and irresponsible when it comes to "giving the people what they want". If keeping the players happy so they spend more money is of interest to the staff, perhaps you might consider fixing the desync.
Shouldn't a working game be the first priority long before adding more stuff that's going to almost work correctly? After reading some more about desync, I'm inclined to alter my position of vehement refusal and perhaps take a more practical and pragmatic approach. So, instead of offering negative feedback, I will attempt a solution.
Why not allow the fan-base to decide whether they'd rather deal with desync or, instead, compromise on some game content? If the GGG staff is unable to fix it, why not post a sticky thread specifically dedicated to getting the fans involved with coming up with a solution to the problem?
EDIT: I opened up my first end-game map last night and died three times from desync, so three portals gone *POOF!* because there are abilities in the game that tax the system past its limits. I'd rather have these abilities removed than to have them available and broken.
I am curious why a company, who worked on a game for six years before the beta even came out, put so much content, love and effort into this amazing game (that many of us have been waiting for, in one form or another, for several years), would trip at the finish line. It seems to me that an online game's number one priority would be the network systems.
Tripping at the finish line is the wrong phrase as it implies an accident whereas PoE's network problem is neglect or simply the inability to cope with the network problem but you get the idea. Why so much effort just to have a mediocre game in terms of functionality?
IDEAS FOR A SOLUTION
Rather than be someone who simply points out the obvious and complains about it, I thought I'd offer an idea that might work or at least have potential for generating a new idea that will work. I've had a chance to read a bit about the desync issue posted by the devs as well as the thread regarding it.
Disclaimer: This would not be simple to implement but it may be worth the trouble in the long run.
We know a few things for sure about PoE. There are a limited number of classes and a limited number of skills, skill points and skill point combinations as well as a limited number of gems, gem combinations and gem levels. The idea is to create, by use of an algorithm, Skill Profiles which enumerate and classify the many, many combinations of skills points and skill gems and their levels. This algorithm can also recreate/update the profiles whenever a new skill gem is added to the game or the Passive Skill Tree is altered.
A Skill Profile would consist of Allocated Passive Skill Points, Skill Gems and theirs combinations and their associations with each other as well as the Skill Gems' levels.
For example, one of the (small, in terms of data) Skill Profiles would contain Attack Speed 6%, Attack Speed 4% and a Spectral Throw skill gem at level 1.
There would be a lot of them and every time a player altered, added or removed a skill gem or skill point on a character (which happens far less often than things like attacks or moving), that data would be sent to the server and a new and appropriate Skill Profile would be assigned to that character. The server and client would both assume that this Skill Profile would be active and unchanged until the player changed it through gem and point alteration. This way, the server would not need to send so much info about the ability that's being used or the attack that's being performed each time it occurs. As the server and client would have all the available Skill Profiles, the amount of data sent would be reduced because the ability or attack used would be represented by a Skill Profile ID which resides on the server and the client. Since there are also a limited number of assignable hotkeys and mouse buttons, these key profiles could also be sent to the server and assumed active until changed by the player. The idea is to send less data for each action. Not sure if this would work in practice but there it is. I'll leave the rest to you software engineers.