Jump to content

[Merged] General Bug Fixes clarification



61 replies to this topic

#21 Oderint dum Metuant

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 4,758 posts
  • LocationUnited Kingdom

Posted 23 January 2013 - 02:40 PM

Quote

- Fixed some issues with CryEngine.
- Fixed a few memory leaks.


Those 2 are probably the most interesting 2, what issues with Cry Engine, and what memory leak (4fps bug?)

The only other thing not related to patch, is Xfire and SLI profiles, we've seen Hawken get theirs in pretty quick but we have yet to hear anything on the status of MWO.

Edited by DV McKenna, 23 January 2013 - 02:42 PM.


#22 Beo Vulf

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 739 posts
  • Google+: Link
  • LocationHalsey, NE

Posted 23 January 2013 - 02:44 PM

View Postsuperbob, on 15 January 2013 - 03:49 PM, said:

I think their time is better spent working on the game than typing up a long list of bugfixes.


There is someone in most companies who's job it is to track changes and document them. In some cases it is even a whole office or department. If PGI has no one doing this they are setting themselves up to fail as a company.

#23 Matthew Craig

    Technical Director

  • 867 posts
  • LocationVancouver, BC

Posted 23 January 2013 - 02:56 PM

The CryEngine fix refers to what most users are seeing in the reduced lag shield. It reduces some latency that the engine was adding in the network layer essentially delaying the sending of client packets. This means the lag shield issue 'feels' better as your getting more updates meaning the Mechs positions don't get time to drift as far.

The memory leaks were related to Mechlab, technically the main one isn't a leak. What was happening was that when switching between Mechs VRAM was being allocated and dereferenced when you switched to another Mech but it only gets truly freed upon a level load, so if you switched around a lot of Mechs in the Mechlab you could generate high VRAM consumption. We added a flush after every Mech switch to ensure it's cleaned up properly.

#24 Oderint dum Metuant

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 4,758 posts
  • LocationUnited Kingdom

Posted 23 January 2013 - 02:58 PM

View PostMatthew Craig, on 23 January 2013 - 02:56 PM, said:

The CryEngine fix refers to what most users are seeing in the reduced lag shield. It reduces some latency that the engine was adding in the network layer essentially delaying the sending of client packets. This means the lag shield issue 'feels' better as your getting more updates meaning the Mechs positions don't get time to drift as far.

The memory leaks were related to Mechlab, technically the main one isn't a leak. What was happening was that when switching between Mechs VRAM was being allocated and dereferenced when you switched to another Mech but it only gets truly freed upon a level load, so if you switched around a lot of Mechs in the Mechlab you could generate high VRAM consumption. We added a flush after every Mech switch to ensure it's cleaned up properly.


Thanks for those that is the sort of detail that is good to know, without being overly complex, specifically going back to the last one given that it flushed on a level load previously is it safe to assume this was unlikely to be connected to the 4FPS issue and that this one is still out there to be caught ( i can appreciate it's difficult to reproduce it)

Edited by DV McKenna, 23 January 2013 - 02:59 PM.


#25 Dakkath

    Member

  • PipPipPipPipPipPipPipPip
  • 1,980 posts
  • LocationG-14 Classified

Posted 23 January 2013 - 02:59 PM

View PostMatthew Craig, on 23 January 2013 - 02:56 PM, said:

The CryEngine fix refers to what most users are seeing in the reduced lag shield. It reduces some latency that the engine was adding in the network layer essentially delaying the sending of client packets. This means the lag shield issue 'feels' better as your getting more updates meaning the Mechs positions don't get time to drift as far.

The memory leaks were related to Mechlab, technically the main one isn't a leak. What was happening was that when switching between Mechs VRAM was being allocated and dereferenced when you switched to another Mech but it only gets truly freed upon a level load, so if you switched around a lot of Mechs in the Mechlab you could generate high VRAM consumption. We added a flush after every Mech switch to ensure it's cleaned up properly.



For a computer geek like me, that's pretty cool to hear. rock on brotha.

#26 ThunderOverWater

    Member

  • PipPipPip
  • 85 posts

Posted 23 January 2013 - 03:15 PM

View PostMatthew Craig, on 23 January 2013 - 02:56 PM, said:

The CryEngine fix refers to what most users are seeing in the reduced lag shield. It reduces some latency that the engine was adding in the network layer essentially delaying the sending of client packets. This means the lag shield issue 'feels' better as your getting more updates meaning the Mechs positions don't get time to drift as far.


That's kind of a big deal.

"- Fixed some issues with CryEngine."

"- First round of netcode improvements."


Fixed it for you. :P

#27 Matthew Craig

    Technical Director

  • 867 posts
  • LocationVancouver, BC

Posted 23 January 2013 - 03:21 PM

Yes sadly the 4 fps issue is still out there to be caught (still high priority), the cleanup work obviously has the joint benefit of reducing things that it could be while also increasing overall stability. What I'm working on currently is actually low level tools for tracking memory leaks that shouldn't only catch any outstanding memory leaks but ensure we don't incur them in the future (should be rapidly caught in test by the reports that get generated).

There's also more performance fixes coming that address likely culprits e.g. looking at lots of Mechs was incurring high animation costs, work has been done to lower these and we'll continue to monitor for the issue. There's also ongoing work on continuing to improve profiling data such that we can better catch this and similar issues. We had some reports from the community of expected ways to get it to happen faster (running windows XP, less physical RAM etc.) but none have panned out so far.

We'll keep narrowing it down though until it's resolved.

#28 Joe Mallad

    Member

  • PipPipPipPipPipPipPipPipPip
  • Rage
  • Rage
  • 3,740 posts
  • LocationMichigan

Posted 23 January 2013 - 03:29 PM

so cant you guys put this kind of info in the patch notes please!!!!! This is the kind of info we all would like to see. granted we all know you guys are fixing stuff but to be able to see in detail whats being fixed and how REALLY helps you guys. And those that want to complain a lot that you guys are not doing ANYTHING... can see just what you are doing or attempting to do. This keeps those people calm and in turn less apt to be complaining to you guys as much. Everyone wins then, we all get better detailed written info on whats being done and you guys dont hear all the complaining (at least not as much) lol


PLEASE keep up the awesome work by the way!!

Edited by Yoseful Mallad, 24 January 2013 - 10:59 AM.


#29 BunnyWabbit

    Member

  • PipPipPipPipPip
  • 104 posts

Posted 23 January 2013 - 03:52 PM

View PostYoseful Mallad, on 23 January 2013 - 03:29 PM, said:

so cant you guys put this kind of info in the patch notes please!!!!! This is the kind of info we all would like to see. granted we all know you guys are fixing stuff but to be able to see in detail whats being fixed and how REALLY helps you guys. And those that want to complain a lot that you guys are not doing ANYTHING... can see just what you are doing or attempting to do. This keeps those people calm and in turn less apt to be complaining to you guys as much. Every one wens then, we all get better detailed written info on whats being done and you guys dont hear all the complaining (at least not as much) lol


PLEASE keep up the awesome work by the way!!



+++

I'd rather have the expanded version in your post than the abridged version in the patch notes. Especially considering you even abridged that version.

Keep up the good work. Keep us all posted. ;)

#30 Volume

    Member

  • PipPipPipPipPipPipPipPip
  • The Privateer
  • The Privateer
  • 1,097 posts

Posted 23 January 2013 - 04:15 PM

View PostMatthew Craig, on 23 January 2013 - 02:56 PM, said:

The CryEngine fix refers to what most users are seeing in the reduced lag shield. It reduces some latency that the engine was adding in the network layer essentially delaying the sending of client packets. This means the lag shield issue 'feels' better as your getting more updates meaning the Mechs positions don't get time to drift as far.

The memory leaks were related to Mechlab, technically the main one isn't a leak. What was happening was that when switching between Mechs VRAM was being allocated and dereferenced when you switched to another Mech but it only gets truly freed upon a level load, so if you switched around a lot of Mechs in the Mechlab you could generate high VRAM consumption. We added a flush after every Mech switch to ensure it's cleaned up properly.


This is a fantastic post, thanks for the insight.

#31 Taizan

    Com Guard

  • PipPipPipPipPipPipPipPip
  • 1,692 posts
  • LocationGalatea (NRW)

Posted 23 January 2013 - 04:32 PM

View PostMatthew Craig, on 23 January 2013 - 03:21 PM, said:

We'll keep narrowing it down though until it's resolved.


Keep it up! GJ! Now read this before every bug fixing session. Possibly someone who can impersonate Al Pacino would be best.

Quote

Well, you'll find out life is a game of fixes, so is MWO. Because in either game - life or MWO - the margin for error is so small. One half a step too late or too early and you don't quite make it. One half second too slow, too fast and you don't quite catch it. The fixes we need are everywhere around us. They're in every break of the game, every minute, every second. On this team we fight for that fix.

On this team we tear ourselves and everyone else around us to pieces for that fix. We claw with our fingernails for that fix. Because we know when we add up all those fixes, that's gonna make the f**king difference between winning and losing! Between living and dying! I'll tell you this, in any fight it's the guy whose willing to die whose gonna win that fix. And I know, if I'm gonna have any life anymore it's because I'm still willing to fight and die for that fix, because that's what living is, the six inches in front of your face.

Now I can't make you do it. You've got to look at the guy next to you, look into his eyes. Now I think ya going to see a guy who will fix that bug with you. Your gonna see a guy who will sacrifice himself for this team, because he knows when it comes down to it your gonna do the same for him. That's a team, gentlemen, and either, we heal, now, as a team, or we will die as individuals. That's MWO guys, that's all it is.

Edited by Taizan, 23 January 2013 - 04:33 PM.


#32 p00k

    Member

  • PipPipPipPipPipPipPipPip
  • 1,661 posts

Posted 23 January 2013 - 05:04 PM

nice to see some actual info about the patch changes

View PostMatthew Craig, on 23 January 2013 - 02:26 PM, said:

We appreciate there are those that may want to know the gritty details though for most users it would just be jargon or worse just creates a lot of confusion and arguments on the forums.

the people who consider it just jargon will skim over it and ignore it. there's no harm in having more detailed patch notes; if anything the ninja patch changes were one of the big reasons pgi often seems so out of touch.

it doesn't take a genius to figure out the appropriate detail to be informative but not confusing. game balance issues are easy enough, and there really shouldn't ever be a game balance change not explicitly noted in patch notes. game engine changes, sure few people will care exactly how many lines of code were inserted where, but a short sentence about what was changed to address what end-user experience issue suffices nicely. for example

View PostMatthew Craig, on 23 January 2013 - 02:56 PM, said:

The CryEngine fix refers to what most users are seeing in the reduced lag shield. It reduces some latency that the engine was adding in the network layer essentially delaying the sending of client packets. This means the lag shield issue 'feels' better as your getting more updates meaning the Mechs positions don't get time to drift as far.

The memory leaks were related to Mechlab, technically the main one isn't a leak. What was happening was that when switching between Mechs VRAM was being allocated and dereferenced when you switched to another Mech but it only gets truly freed upon a level load, so if you switched around a lot of Mechs in the Mechlab you could generate high VRAM consumption. We added a flush after every Mech switch to ensure it's cleaned up properly.


is great, the kind of info people wish they had every time a patch is released, but is probably too much to put for every change in the patch notes. but a simple "cryengine fix to reduce latency issues and reduce lagshield" or "memory flush to address excessive memory use when switching mechs" tells us a lot more than "- Fixed some issues with CryEngine"

#33 flackee

    Member

  • PipPipPipPipPip
  • Survivor
  • Survivor
  • 109 posts

Posted 23 January 2013 - 05:48 PM

View PostMatthew Craig, on 23 January 2013 - 02:56 PM, said:

The CryEngine fix refers to what most users are seeing in the reduced lag shield...


Thankyou! This helps tremendously not only in explaining what I've been seeing all these months, but this is exactly the kind of information beta testers need to help solve some of these issues. Keep it up please!

#34 msqueak

    Member

  • PipPip
  • Legendary Founder
  • Legendary Founder
  • 33 posts

Posted 23 January 2013 - 05:52 PM

View PostMatthew Craig, on 23 January 2013 - 02:26 PM, said:

We appreciate there are those that may want to know the gritty details though for most users it would just be jargon or worse just creates a lot of confusion and arguments on the forums.

I think we'll see what we can do in the future to have a full version somewhere, for now if there's something specific you'd really like more info on just ask I'll try and clarify.

That is good to hear. I do not know how useful every change can be, but it is a beta. If something gets past all the internal testing then perhaps something in the change notes could entice one of us beta testers to discover it.

#35 Davers

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,886 posts
  • Facebook: Link
  • LocationCanada

Posted 23 January 2013 - 06:28 PM

Thanks for the info! Lot's of people think bug fixes are just looking at the code and seeing 'errors'. They don't realize you have to design tools to find the problems first!

#36 Perihelion Jack

    Member

  • PipPipPip
  • Death Star
  • 60 posts

Posted 23 January 2013 - 07:54 PM

Mr. Craig,

I feel the need to thank you for your time in elaborating on some of the issues that have been addressed. As a long-time gamer with virtually no programming experience, I feel the decision to be vague in the official patch notes was acceptable, because I wouldn't understand the particulars of the changes anyway. This patch, as another user has stated previously, is one of the best in recent memory in spite of the lack of detail given in the notes.
I also applaud your willingness to explain these changes in more detail within these forums, as many of us are fluent in techno-speak, and may be able to assist in finding some solutions to ongoing issues. I believe you will find that the tone of the forum users will improve with continued, regular, updates on the progress you all are making. I think that the trend of Devs actively responding to some questions/comments is EXTREMELY heartening, as many of us want to help improve the game in any way we can; but prolonged silence breeds contempt from the very group of passionate fans this game was made for.

Thank you and your colleagues again for your hard work on the game itself, as well as the time you spend in these forums. Keep up the good work!

#37 Budor

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,565 posts

Posted 23 January 2013 - 08:14 PM

Either i am hallucinating or a lot of the cockpits got updated without a mention in the notes to.

#38 Pyrrho

    Member

  • PipPipPipPipPipPipPip
  • Bad Company
  • 854 posts

Posted 23 January 2013 - 08:32 PM

This thread makes me squee.

Thank you, Dev team, you are all doing a great job with a difficult task.

#39 Sprouticus

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • 2,781 posts
  • LocationChicago, Il, USA

Posted 24 January 2013 - 06:41 AM

I think we would all love to hear about what you guys actually fixed that is causing better hit detection, and what some people have characterized as 'paper thin armor'


There are a lot of theories going around, many of which are pure speculation. Some fact would help minimize the rumor mill.

#40 Protoculture

    Member

  • PipPipPipPipPipPip
  • 428 posts

Posted 24 January 2013 - 06:50 AM

View PostSprouticus, on 24 January 2013 - 06:41 AM, said:

I think we would all love to hear about what you guys actually fixed that is causing better hit detection, and what some people have characterized as 'paper thin armor'


There are a lot of theories going around, many of which are pure speculation. Some fact would help minimize the rumor mill.


Hit detection was because of Cryengine memory leak fixes that were holding back packets. I read it yesterday from Matthew whatever the Technical Director, I'm sure you can find it, but it was buried in a thread somewhere. I haven't heard anything about paper-thin armor but probably it's because people can actually land hits on lights better now if I had to guess.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users