14 replies to this topic

[M R T]

...when you're searching for matches? Or fool around in the mechlab for that matter? Why do you get locked out of everything when searching for a match? Not so much a problem in QP, but in FP and GQ, when finding a match can take a very long time, it would be nice to do something else while waiting. Is it an engine thing? If so, that's kind of strange since you still receive messages in the chat window when it's in the background.

[JediPanther]

It's probably the engine which almost no one now days knows anything about since pgi heavily altered it for online play. About the only way to "chat" with some one is to use a voip program such as teamspeak or discord which were the two most used ones long before the generic in game voip was added.

A lobby system would be nice but its always an either or case of pgi being able to do it (and taking the time since they make no profit) or the game engine is or isn't able to do it.

[Prototelis]

Because the UI they developed doesn't support it.

[Nesutizale]

Wasn't the UI added on top of the engine and was Java or something of an today antik system? It was definitly not part of the engine as far as I remember.

I think PGI should concider, when MW5 runs well, to port MWO over. It might solve a lot of their problems with the older engine.
Problem might be the license, if it even allowes them to port MWO. When MWO is bound to the cryengine...

[Feral Clown]

Most people serious about cw have discord and teamspeak installed so this issue really only effects pugs....who should use this quiet time to reflect on why they are doing this and why they don't have any friends.

[LordNothing]

accretion. developing games out of a bunch of black boxes that once dubbed complete by the developer never see another coder ever again. they just build up on the base engine like cancerous growths. its easier to just make a new system and stick it on the pile than it is to fix a broken old system that does more or less 90% of the same stuff. woe to he who ever needs to make 2 of those subsystems talk to each other in a way that was never planned for in the original design of either/both. this is an unfortunate side effect of rent an engine studios.

i prefer the total systems integration approach where once things are developed they are maintained and improved over the life of the product. you dont run into the problems like "we cant add more teams" and "we cant add ammo types" because its always possible to go back and add features. of course to do this you need full time programmers with experience in the existing codebase and good documentation. added bonus if the programmer making those changes is the same one that wrote the documentation. of course with all the shortcuts modern game devs take, nobody can afford to do this approach and stay competitive. really the only ones who do it are engine houses that make most of their money selling engines and if they put out a game its pretty much a marketing tool so they can sell their engine (unreal is a good example of this).

Edited by LordNothing, 27 April 2019 - 03:38 AM.

[VonBruinwald]

Morte Nilsum, on 26 April 2019 - 01:53 PM, said:

...when you're searching for matches? Or fool around in the mechlab for that matter? Why do you get locked out of everything when searching for a match? Not so much a problem in QP, but in FP and GQ, when finding a match can take a very long time, it would be nice to do something else while waiting. Is it an engine thing? If so, that's kind of strange since you still receive messages in the chat window when it's in the background.

You can chat while searching Solaris/FW as of the latest PTS. Can even chat while map banning in Solaris if you have it open before a match is found.

This works on the live server for Solaris, not sure about FW though.

[Jonathan8883]

LordNothing, on 27 April 2019 - 03:32 AM, said:

accretion. developing games out of a bunch of black boxes that once dubbed complete by the developer never see another coder ever again. they just build up on the base engine like cancerous growths. its easier to just make a new system and stick it on the pile than it is to fix a broken old system that does more or less 90% of the same stuff. woe to he who ever needs to make 2 of those subsystems talk to each other in a way that was never planned for in the original design of either/both. this is an unfortunate side effect of rent an engine studios.

i prefer the total systems integration approach where once things are developed they are maintained and improved over the life of the product. you dont run into the problems like "we cant add more teams" and "we cant add ammo types" because its always possible to go back and add features. of course to do this you need full time programmers with experience in the existing codebase and good documentation. added bonus if the programmer making those changes is the same one that wrote the documentation. of course with all the shortcuts modern game devs take, nobody can afford to do this approach and stay competitive. really the only ones who do it are engine houses that make most of their money selling engines and if they put out a game its pretty much a marketing tool so they can sell their engine (unreal is a good example of this).

To be fair to PGI as well... our corporate office (Multi $100M subsidiary of a $1.5B+ private co) struggles with this problem as well. The base system is a web-based port of a 2 decade old Telnet system, and the best plan right now is to come up with a new layer to place on top of all the old systems to unify them, and then eventually change the now-out-of-sight old stuff.
I think the timeframe on that is 2-3 years, and the budget is well into the 7 figures.

[LordNothing]

Jonathan8883, on 27 April 2019 - 05:21 AM, said:

To be fair to PGI as well... our corporate office (Multi $100M subsidiary of a $1.5B+ private co) struggles with this problem as well. The base system is a web-based port of a 2 decade old Telnet system, and the best plan right now is to come up with a new layer to place on top of all the old systems to unify them, and then eventually change the now-out-of-sight old stuff.
I think the timeframe on that is 2-3 years, and the budget is well into the 7 figures.

its not strictly a pgi problem, its more of a software engineering problem. making well documented, well commented, maintainable code takes a lot more expensive programmer time than just slapping something together which meets the current minimum design requirements. it does seem to be the way to go if you are designing a system that needs to last a decade+, like mwo, but its entirely possible that is outside the scope of what pgi can do with its budget.

Edited by LordNothing, 27 April 2019 - 02:11 PM.

[Nesutizale]

Didn't PGI said at some point that the guys who wrote a lot of the original changes are gone and that it would be to complex to figure out what that gone person did and instead they just add as little as possible without breaking things completly.

I think the same was said from the StarTrek Online game devs quite some time ago.

And I know that from some projects I worked on, also its not programming, working yourself into the though process of someone else can be quite bothersome.
I remember where someone interlaced what I wanted to change into another project file that was getting its data from another one that had in itself again some interlaced stuff. It was driving me mad at first because I changed something and weird things started to happen and I didn't know what was going on...even as all layers where at least named correctly but no where it was noted what was linked to each other.

Also from what I have seen a big problem is nameing conventions. So even IF someone made notes, it doesn't even mean you understand them. That is at least my impression as someone looking at it from the outside of codeing.

[LordNothing]

Nesutizale, on 28 April 2019 - 03:01 AM, said:

Didn't PGI said at some point that the guys who wrote a lot of the original changes are gone and that it would be to complex to figure out what that gone person did and instead they just add as little as possible without breaking things completly.

I think the same was said from the StarTrek Online game devs quite some time ago.

And I know that from some projects I worked on, also its not programming, working yourself into the though process of someone else can be quite bothersome.
I remember where someone interlaced what I wanted to change into another project file that was getting its data from another one that had in itself again some interlaced stuff. It was driving me mad at first because I changed something and weird things started to happen and I didn't know what was going on...even as all layers where at least named correctly but no where it was noted what was linked to each other.

Also from what I have seen a big problem is nameing conventions. So even IF someone made notes, it doesn't even mean you understand them. That is at least my impression as someone looking at it from the outside of codeing.

i hear a lot of third hand stories about pgi's original mwo programmer. but programmers should be able to read someone else's code or they cant call themselves programmers. even poorly maintained and commented code whos comments are in some foreign language that only a million people know can still be worked with. if they still have trouble its not like a real programmer cant slap together a script to scan the code files for comments, translate them with machine translation software and replace them with comments in english. some people working with really old closed source blobs can read disassembled machine code with no comments what so ever and make enough sense out of it to make meaningful changes to the software.

i think the problem is that pgi are either unwilling or unable to keep a programmer on staff who can maintain the codebase full time. and its better they be in bed with the company, no less than maybe 3rd in command and not on some kind of contract job. because thats how you get hard to maintain accreted code that no one understands. all the best games come from studios where the head programmer is the ceo. in cases where your ceo is a trump-like buffoon who love to go half cocked with ideas that will never work but because they run the show everyone under them has no choice but to comply, then dont expect much. companies that care more about making money than making games are what is killing gaming today. sure money is important, its what enables devs to pursue the creation of games. but if you loose sight of the fact that making games is what you are in this business for, then whats the point, you arent pushing the artform and you certainly arent doing anything new.

[Burning2nd]

i wish we where back when the UI change was a big deal.. that seems so trivial now

[LordNothing]

Burning2nd, on 30 April 2019 - 04:42 PM, said:

i wish we where back when the UI change was a big deal.. that seems so trivial now

ui work will always be trivial. solaris for the most part was ui work. the meat and potatoes of the mode, 1v1, was already supported by the core game. sure they made some maps and got duncan fisher, but that was the only substance that solaris had.

another good example was rearm and repair. we probably wanted what you had in mw3/mw4/living legends, live in game repair and rearm, what we got was a ui that just deleted cbills.

even fp had this problem, they put more effort into the inner sphere map than the incursion mode. phase 3 saw more ui work and several other things that didnt work. the new features fp will have, other than the new match maker, is mostly ui.

i for one want more meat and potatoes and less whitewash.

Edited by LordNothing, 30 April 2019 - 05:20 PM.

[Burning2nd]

[BlueStrat]

Another reason why PGI may have no interest in any sort of global chat is that since MWO is international in scope, the laws in many places would require them to have chat moderators to kick/ban people for language/hate speech/etc. Just too much hassle and expense to be worth it from PGI's perspective, and I can't really blame them.

Edited by BlueStrat, 01 May 2019 - 12:09 PM.