Aller au contenu

Photo

Turian Havoc got a significant buff, will you be using him now?


177 réponses à ce sujet

#151
Kenadian

Kenadian
  • Members
  • 5 031 messages

Arctican wrote...

Kenadian wrote...

Do you guys think this strengthened the melee build much? As it is, I can make it work but not that I'm particularly effective with it.


I recall the old dual strike combo would leave some enemies with just a little bit of health left due to martial artist expiring after a period of time. With this new buff, it should be able to kill some of those enemies in one dualstrike without the help of martial artist.


Might be worth muckng about with then.

#152
UnknownMercenary

UnknownMercenary
  • Members
  • 5 547 messages
Seconding the request for a video with this guy.

#153
Feneckus

Feneckus
  • Members
  • 3 076 messages
I'm trying to make a Havoc video, it's a really great class now, but 90% of the time I end up in a game in progress and the other 10% I end up with spawn-nuking noobs.

I wish I could host :crying:

#154
stysiaq

stysiaq
  • Members
  • 8 480 messages

Feneckus wrote...

I'm trying to make a Havoc video, it's a really great class now, but 90% of the time I end up in a game in progress and the other 10% I end up with spawn-nuking noobs.

I wish I could host :crying:


I'm up for some Havoc wrecking later this day. I can host, and well, you know I'm bad enough to miss the proper spawns. :D

#155
Chealec

Chealec
  • Members
  • 6 508 messages
Hmmmm - may need to respec for a test, I had him specced out of Havoc Strike and used him as a kind of mobile Destroyer with Cryo Blast for debuff.

#156
Cohen le Barbare

Cohen le Barbare
  • Members
  • 1 789 messages

Feneckus wrote...

I'm trying to make a Havoc video, it's a really great class now, but 90% of the time I end up in a game in progress and the other 10% I end up with spawn-nuking noobs.

I wish I could host :crying:

Feneckus, I'm French and have a pretty good connection, so if you need a lame host that can let you demonstrate the uberness of the Havoc, let me know.

#157
Feneckus

Feneckus
  • Members
  • 3 076 messages

Cohen le Barbare wrote...

Feneckus, I'm French and have a pretty good connection, so if you need a lame host that can let you demonstrate the uberness of the Havoc, let me know.


Cool, sent you a friend request.

#158
stysiaq

stysiaq
  • Members
  • 8 480 messages
What about a video with 4 Havocs on Jade (nades & Jungle) pawning everything in the video with HS and Claymores with some obvious music as the accompaniament:?

#159
Cohen le Barbare

Cohen le Barbare
  • Members
  • 1 789 messages

Feneckus wrote...

Cohen le Barbare wrote...

Feneckus, I'm French and have a pretty good connection, so if you need a lame host that can let you demonstrate the uberness of the Havoc, let me know.


Cool, sent you a friend request.

Just accepted it. Let me know when you want to shoot your video.

#160
Blitzkrieg_33

Blitzkrieg_33
  • Members
  • 1 432 messages
He is an absolute beast with the claymore. But then again, everything is.

Glad I waited to do my havoc waves until now.

#161
Cohen le Barbare

Cohen le Barbare
  • Members
  • 1 789 messages

stysiaq wrote...

What about a video with 4 Havocs on Jade (nades & Jungle) pawning everything in the video with HS and Claymores with some obvious music as the accompaniament:?

No, this music only works with valkyrie sentinel with valkyrie assault rifles ^_^ 

The "soundtrack" for 4 havocs would have to be a person reading Shakespeare's Julius Cesar. /mrknowitall :P

#162
stysiaq

stysiaq
  • Members
  • 8 480 messages

Cohen le Barbare wrote...

stysiaq wrote...

What about a video with 4 Havocs on Jade (nades & Jungle) pawning everything in the video with HS and Claymores with some obvious music as the accompaniament:?

No, this music only works with valkyrie sentinel with valkyrie assault rifles ^_^ 

The "soundtrack" for 4 havocs would have to be a person reading Shakespeare's Julius Cesar. /mrknowitall :P


But can't you imagine the glory of four Turians in a row, deploying Cryo Blasts, boosting themselves with stimpacks and finally launching their jetpacks one after another, with Wagner in the bacground? It's almost like a postcard from Apocalypse Now.

#163
Megakoresh

Megakoresh
  • Members
  • 610 messages
I was always able to effectively use the Havoc soldier against Geth, Reapers and to certain extend Collectors. This buff it good, because now I can use is against Cerberus as well, but the main problem with it was and still is the bug that shuts down your ability use. It is fixed with heavy melee and happens on this particular kit so often that it is rare to have at least 1 wave without such bug. This is the reason I don't play him as much, he is a fun character otherwise and a good tank.

Modifié par Megakoresh, 30 janvier 2013 - 02:13 .


#164
Cohen le Barbare

Cohen le Barbare
  • Members
  • 1 789 messages

stysiaq wrote...
But can't you imagine the glory of four Turians in a row, deploying Cryo Blasts, boosting themselves with stimpacks and finally launching their jetpacks one after another, with Wagner in the bacground? It's almost like a postcard from Apocalypse Now.

Oh I can see that, but I would be wiping the floor way too often with him ^_^

#165
Arkman279

Arkman279
  • Members
  • 123 messages
Probably not. Personally I'd like to see havoc strike replaced with another power it just isn't really useful in many situations and then theres the glitch where it won't work and u have to heavy melee. this probably isn't an issue that can be fixed. Some glitches are here to stay because the means too fix them is too time consuming or too great a change. So why not just replace it with something else.

Modifié par Arkman279, 30 janvier 2013 - 02:56 .


#166
SilentStep79

SilentStep79
  • Members
  • 3 702 messages
never use him.
tried last night after seeing the buff he got.
won't be using him again.
however, this was after seeing a player from here *ahem Kalas321* ABSOLUTELY DESTROY with the havoc earlier in the evening! so i know he's effective, but i can't do a thing with him.

#167
Blitzkrieg_33

Blitzkrieg_33
  • Members
  • 1 432 messages
For those that hate using him, try this playstyle:

CQC character, with shotgun.
Use Havoc strike to quickly close the distance on enemies/also stagger
Shotgun to the face
Pump stimpacks as needed to keep shields up

Rinse and repeat.

I do not see why people don't like this character :/

#168
Cohen le Barbare

Cohen le Barbare
  • Members
  • 1 789 messages

Blitzkrieg_33 wrote...
I do not see why people don't like this character :/

Because it's very frustrating to charge and see that it didn't strike. At least, when I use biotic charge, it locks my target. Alos, even on silver, a debuff cryo>area charge (4a/6b) won't kill a dragoon.

#169
jackjj

jackjj
  • Members
  • 189 messages

Crimson Vanguard wrote...

Barge6000 wrote...

nah original turian soldier all day



#170
Shampoohorn

Shampoohorn
  • Members
  • 5 861 messages

Cohen le Barbare wrote...

Feneckus wrote...

Cohen le Barbare wrote...

Feneckus, I'm French and have a pretty good connection, so if you need a lame host that can let you demonstrate the uberness of the Havoc, let me know.


Cool, sent you a friend request.

Just accepted it. Let me know when you want to shoot your video.


An all Havoc team in the vid would be great, both for demonstration and coolfactor.  Ideally, pattern their colors after the Asian Giant Hornet. ;o)

I'd love to join in too, but I'm GMT -8.  (Origin: Zerfidius just in case it works out.)

Modifié par Shampoohorn, 30 janvier 2013 - 05:39 .


#171
.458

.458
  • Members
  • 2 113 messages

Dunvi wrote...
...but it's not solvable at this late point in the game's lifespan to any reasonable degree (in fact, it would have been nearly unsolvable even before release. Damn architectural issues...), and chances are I know more about that than you do...


Without the weaknesses in how the UDP network was used, with a failed "reliability" layer to mimic TCP but without the latency, there would be no issues. This particular issue is not solvable in the near future, and is an issue faced since the beginning of the internet.

There are, however, a few smaller changes that could be made which would dramatically improve the game. Unlikely BW will fix this without a financial reason. Sweeping improvements could be made with some small changes in how events are sent over the network. Some very simple changes could eliminate missile glitching, could eliminate volus shields locking other players' weapons, etc. BW probably doesn't even realize they have most of the code already, they just haven't used it in the right place.

#172
Dunvi

Dunvi
  • Members
  • 4 841 messages

.458 wrote...

Dunvi wrote...
...but it's not solvable at this late point in the game's lifespan to any reasonable degree (in fact, it would have been nearly unsolvable even before release. Damn architectural issues...), and chances are I know more about that than you do...


Without the weaknesses in how the UDP network was used, with a failed "reliability" layer to mimic TCP but without the latency, there would be no issues. This particular issue is not solvable in the near future, and is an issue faced since the beginning of the internet.

There are, however, a few smaller changes that could be made which would dramatically improve the game. Unlikely BW will fix this without a financial reason. Sweeping improvements could be made with some small changes in how events are sent over the network. Some very simple changes could eliminate missile glitching, could eliminate volus shields locking other players' weapons, etc. BW probably doesn't even realize they have most of the code already, they just haven't used it in the right place.


I'm not sure though, that you can say that changing how events are sent would be small changes. You're talking about changing both the sender and the receiver for most? or even all? network communicated actions, and then deploying this in a patch. That's more programming energy than BW is going to expend at this point in the game's lifetime, not to mention whether it's possible to fit that much change into a patch (I believe the consoles have very limited patch size allotments). What I meant by unsolvable isn't "impossible" or "too hard", just simply "not gonna happen".

Magnum, we might also be agreeing with each other. If we are, I apologize. I blame my exhaustion - I keep on forgetting to sleep at night.

ETA: ... incidentally, dat patch...

Modifié par Dunvi, 31 janvier 2013 - 12:07 .


#173
AaronEh

AaronEh
  • Members
  • 415 messages

Shampoohorn wrote...

I'd love to join in too, but I'm GMT -8.  (Origin: Zerfidius just in case it works out.)


lol, that's you.  I think we've been Origin friends for a while but I can't recall a game we've played together.

And for the OP, used him before - will repsec and continue to use.

Modifié par AaronEh, 30 janvier 2013 - 10:50 .


#174
Shampoohorn

Shampoohorn
  • Members
  • 5 861 messages

AaronEh wrote...

Shampoohorn wrote...

I'd love to join in too, but I'm GMT -8.  (Origin: Zerfidius just in case it works out.)


lol, that's you.  I think we've been Origin friends for a while but I can't recall a game we've played together.

And for the OP, used him before - will repsec and continue to use.


A few months back, with Nicla.  A string of Platinum games.  Good times.

Modifié par Shampoohorn, 30 janvier 2013 - 11:07 .


#175
.458

.458
  • Members
  • 2 113 messages

Dunvi wrote...

.458 wrote...

Dunvi wrote...
...but it's not solvable at this late point in the game's lifespan to any reasonable degree (in fact, it would have been nearly unsolvable even before release. Damn architectural issues...), and chances are I know more about that than you do...


Without the weaknesses in how the UDP network was used, with a failed "reliability" layer to mimic TCP but without the latency, there would be no issues. This particular issue is not solvable in the near future, and is an issue faced since the beginning of the internet.

There are, however, a few smaller changes that could be made which would dramatically improve the game. Unlikely BW will fix this without a financial reason. Sweeping improvements could be made with some small changes in how events are sent over the network. Some very simple changes could eliminate missile glitching, could eliminate volus shields locking other players' weapons, etc. BW probably doesn't even realize they have most of the code already, they just haven't used it in the right place.


I'm not sure though, that you can say that changing how events are sent would be small changes. You're talking about changing both the sender and the receiver for most? or even all? network communicated actions, and then deploying this in a patch. That's more programming energy than BW is going to expend at this point in the game's lifetime, not to mention whether it's possible to fit that much change into a patch (I believe the consoles have very limited patch size allotments). What I meant by unsolvable isn't "impossible" or "too hard", just simply "not gonna happen".

Magnum, we might also be agreeing with each other. If we are, I apologize. I blame my exhaustion - I keep on forgetting to sleep at night.

ETA: ... incidentally, dat patch...


The indirect changes which would help a lot are small, but I don't think they know that. Fixing networking itself would be a very big fix. I do this for a living, and looking at how they are using events among components, I believe they already have the pieces and would simply rearrange it slightly. Not a major change.

The reason why events are so important is because they are going across a network which was intended to be completely reliable. It isn't TCP. Everything a player does looks like a single event, but is actually a series of smaller events where the state of something changes each sub-event. If the sequence gets interrupted or corrupted, nasty things can happen. Sometimes those things are unintentional because of something like a network dropping data, at other times it is intentional by means of a player taking advantange of the timing of his/her own actions versus the round trip time from off-host to host. Suppose you change weapons...a number of things happen, including an animation, change of amount of ammo, change of damage type, change of damage amount, change of related audio, notifications of changes to other players so they also see this, so on. While using a striker certain characteristics come out, e.g., fire rate, magazine size. While using a cobra missile, other characteristics come out. It isn't that two don't mix, because they are both a series of separate characteristics told what to set up as by messages from smaller sub-events. There is a control system though, which knows that if you are switching to a striker to send out the sub-events which match that weapon; or the sub-events of a cobra if switching to that. The set of events will not mix in most cases. In single player the messages occur so quickly that a human couldn't interrupt it by mashing buttons down on the keyboard or controller. In multiplayer though, some of those messages go across the network; in the case of an off-host, they go to the host to validate, then the host sends a response back...not instantly. Missile glitching is a case of starting a series of sub-events which will get a reply with noticable latency, and then intentionally hitting a different set of sub-events such that the delayed network reply for a different weapon will intercept in the middle. All of a sudden the sub-events of changing to a cobra missile are injected in the series of sub-events that make up a striker. One of the components (or more) of the striker do just as they are told and set up as a cobra, while the rest of the sub-events set up as a striker (like magazine size).

The network and its UDP is functioning 100% correctly by dropping some packets without notice. This is the intended purpose of UDP. Unlike TCP, when data is dropped, there is no buffer and no re-send. Knowledge of being dropped is typically also missing. UDP is a known protocol, and every internet device at every hop on the route will summarily drop a packet if the IP stack is full. TCP would remember and retry, UDP will not. So you have to build your own system of reliability to deal with it. Gets worse when you realize that instead of just dropping packets in UDP, that packets can also be truncated without notice. Perhaps the receiving end gets part of the data and doesn't realize part of it is missing. What happens to the truncated portion? Undefined. Some devices will continue sending the truncated data, others will discard it. So if it does send it, a real world network does not define which fraction will arrive first. Entire UDP packets can arrive out of order; fragments can arrive out of order; pieces can go missing; entire packets can go missing. All on a network which is 100% lossless and perfect. Networking can easily cause issues where a sub-event is dropped or sent out of order.

So you just got to your match, and your weapon is using the wrong sound. Sub-event corrupted. Sound is entirely missing for you and only you. Probably lost data to do the proper initializing, else it would go bad in single player. Your weapon jams when the volus uses shield boost...we all know they are completely unrelated, but being in the middle of a series of sub-events for one main event and suddenly having sub-events of something else shoved in the middle can do unexpected things. Enemies are invisible for just you? Maybe visibility is broken, but why does it work most of the time? More likely visibility code works, and something forgot to tell your client that the enemies are visible...visibility is a sub-component someone didn't turn on.

I'm reminded of threaded programming, and how deceptively simple it seems. You can make a call to a function or method and it looks like you're safe to do so, but that single call really consists of multiple assembler calls...which CAN be corrupted by a second thread. Network events using the UDP setup this game uses need to be treated similar to a data object which is protected by a mutex or other scheme. Only there is no mutex, what you have instead is the idea that a single event ID, if it traverses, really is atomic. Multiple sub-events for one major event are very dangerous, there is no UDP guarantee that they will all traverse, that they will traverse in order, that they will not be truncated, that they will not be broken up in pieces with time delays. There is no guarantee that data from another packet won't hit right in the middle.

The game HAS code on every client designed to traverse through a series of sub-events (provided it is set up correctly and not interrupted) to do the right thing for changing to a cobra, to do the right thing for changing to a striker. The sender of that data could send the major event instead of the sub-events, and that one event would be atomic, and non-divisible. Network fragments of one could not inject into the middle of another. The code for doing this is already in every game...it just isn't used correctly. All those sub-events should be coalesced into a single major event before going across the network, and then the guy at the other end could re-expand that back to sub-events. A benefit to this would be that instead of sending half a dozen events a single event could be sent...essentially compressing the data going across the network and reducing traffic while guaranteeing pieces of one event won't inject into another, and guaranteeing that when something happens (like throwing a grenade), that it will always complete (if you threw that grenade and only a single event went over the network, it would always explode too...same for OPS packs...wouldn't it be nice if every time you heard the sound and used an ops pack that you actually were guaranteed it would work?).

Modifié par .458, 31 janvier 2013 - 02:15 .