46 replies to this topic

[mekabuser]

HIgh speed objects and cryengine 3 dont mix right?
SO , lets reduce all assets by 10 % or 20 and that would help alleviate the problem..

Of course we dont SAY we slowed everything down because nerds would rage, but it would vastly improve hit detection etc..
SO.. in the cockpit 50 kph is still 50 kph, but your just going 10% slower...

Would have to reduce lrm velocity by the same amount.
But i dont think this would be that difficult to implement, would , should have a large benefit for gameplay .
Thoughts?

[Fiachdubh]

I agree totally, the one problem I always had with the game was that it just felt too fast. It should feel a little more ponderous piloting a massive walking tank, seems more twitch fps than BT sim so far (and no I am not just getting old and slow, well I probably am a little but not that much).

[Skyfaller]

mekabuser, on 22 November 2012 - 05:49 AM, said:

HIgh speed objects and cryengine 3 dont mix right?
SO , lets reduce all assets by 10 % or 20 and that would help alleviate the problem..

Of course we dont SAY we slowed everything down because nerds would rage, but it would vastly improve hit detection etc..
SO.. in the cockpit 50 kph is still 50 kph, but your just going 10% slower...

Would have to reduce lrm velocity by the same amount.
But i dont think this would be that difficult to implement, would , should have a large benefit for gameplay .
Thoughts?

Any mech going faster than 100kph has lag shield issues. This is tested and confirmed in multiple mechs.

That should be the top speed attainable in the game.

Slow down the mechs to adjust accordingly and also reduce the torso turn speed accordingly that way there is no issues from mechs that are supposed to be faster suddenly slowed down and slower mechs be able to track them.

[iminbagdad]

I thought about this for a while and i like it. 50kph is an arbitrary number. You arent really going that fast so slow everything down. Speed in this game is the biggest issue so slow everything down and make it feel like big mechs.

[sycocys]

It might be worth a test, but I don't think the engine other than the netcode portion is really at fault here.

[CrowTRobot]

I really don't have any issues with speed, and prefer to have the speeds of the mechs remain roughly canon

[sycocys]

Occasionally I'll get some lag, or a tele-Jenner, but the jenner issue and the missing lights is getting better for me anyhow.

[Csypher]

yes lets slow it down so i can shoot everything perfectly. you dead faster works for me makes games quicker

Edited by Csypher, 22 November 2012 - 01:22 PM.

[MustrumRidcully]

How about making it turn-based?

I was always in the market for a turn-based shooter!

[Ryvucz]

mekabuser, on 22 November 2012 - 05:49 AM, said:

HIgh speed objects and cryengine 3 dont mix right?
SO , lets reduce all assets by 10 % or 20 and that would help alleviate the problem..

Of course we dont SAY we slowed everything down because nerds would rage, but it would vastly improve hit detection etc..
SO.. in the cockpit 50 kph is still 50 kph, but your just going 10% slower...

Would have to reduce lrm velocity by the same amount.
But i dont think this would be that difficult to implement, would , should have a large benefit for gameplay .
Thoughts?

My only concern with this, is that the scaling of mechs and terrain may need to be reduced to accommodate this. (More attention placed on re-working versus expanding)

Unless of course, they steered clear of actual size/scaling/movement speed. Then that would make my concern invalid.

[Random Numbers]

I posted about this before .. that a) The game play is way too fast atm it's too twitchy
and b ) if you're having massive lag shield issues with things moving to fast ... SLOW IT DOWN

Would make the game way more tactical as then an Atlas would need to decide EARLY which way he's going to move down.

Edited by Random Numbers, 22 November 2012 - 02:30 PM.

[sycocys]

MustrumRidcully, on 22 November 2012 - 01:45 PM, said:

How about making it turn-based?

I was always in the market for a turn-based shooter!

+1

[Rhent]

MustrumRidcully, on 22 November 2012 - 01:45 PM, said:

How about making it turn-based?

I was always in the market for a turn-based shooter!

It's called Mechwarrior Tactics. You are playing the wrong game. Please leave MWO and go to its sister game Mechwarrior Tactics.

[Khobai]

We dont need to slow everything down 10%. Just cap the max speed at 130 until the netcode issues are resolved.

[cmopatrick]

interesting concept, well thought out. not sure i wouldn't like the bugs fixed that keep the present from working, but this could work.

CrowTRobot, on 22 November 2012 - 11:21 AM, said:

I really don't have any issues with speed, and prefer to have the speeds of the mechs remain roughly canon

i think the OP point is that if the game scales speed differently, it might help the netcode respond a little closer to useable. i don't think any of us actually have measured distances on the maps to be exact to the meter (if you have, you are a lot better tester than i am), so our only frame of reference is what the speed bar says our speed is.

if ALL speeds (movement, missile fire, twist, arm speed) are rendered into the map at a slower pace, then there is technically no change in canon speed because the frame of reference is not there to say it is wrong... you have not "speedometer check" map sections to test that you are going 100kph (or whatever), so what the equipment tells you, as long as it applies completely proportional to everything else, is what you are doing. in other words, the idea is a clocking duration change. (added:) think of a maestro's baton in a concert... his timing is what the musicians use... the score is identical, but different conductors may easily use different pace for the same music without changing the music in any other way and unless you are a musician with a metronome in front of you, you will never tell the difference.

btw, capping mech speed (as some suggest above) would not alleviate the missile version of several problems (like the 4-7 fps mess that river city can become when they are fired)... fixing the netcode would and the OP's idea might.

Khobai, on 22 November 2012 - 04:48 PM, said:

We dont need to slow everything down 10%. Just cap the max speed at 130 until the netcode issues are resolved.

i have had a 113 kph Raven be accused of lagging... if that is true, the selection of 130 would not work.

Edited by cmopatrick, 22 November 2012 - 05:12 PM.

[Skyfaller]

You people need to get it through your thick skulls that CANON is IRRELEVANT when it comes to FPS gameplay based on a tabletop game.

Canon is already thrown out the window with the weapons firing more than once per 10 seconds. 'nuff said.

What we have now is a TECHNICAL issue that is ROYALLY giving the Jenner and other 100kph+ mechs a very unfair advantage. If what they do was caused by a third party program it'd be called an exploit.

So, slow down ALL mechs across the board so the top speed is 100kph. Then reduce the twist rate and leg turn rate of all mechs accordingly.

There. Now there is NO such thing as a light mech being at a disadvantage because heavier mechs can track them on the run as it'd be IDENTICAL to what we have currently.

The difference? No failnetcode lag shield.

Its a very easy fix to do and it should at least be TRIED. A temporary solution to a technical problem...until the problem is solved.

[Jadel Blade]

Skyfaller, on 22 November 2012 - 07:53 PM, said:

What we have now is a TECHNICAL issue that is ROYALLY giving the Jenner and other 100kph+ mechs a very unfair advantage. If what they do was caused by a third party program it'd be called an exploit.

So, slow down ALL mechs across the board so the top speed is 100kph. Then reduce the twist rate and leg turn rate of all mechs accordingly.

There. Now there is NO such thing as a light mech being at a disadvantage because heavier mechs can track them on the run as it'd be IDENTICAL to what we have currently.

The difference? No failnetcode lag shield.

And yet it is extremely easy to kill ravens, commandos, jenners, cicadas.

If its IDENTICAL to what we have now what is the advantage?

Rhent, on 22 November 2012 - 03:27 PM, said:

It's called Mechwarrior Tactics. You are playing the wrong game. Please leave MWO and go to its sister game Mechwarrior Tactics.

Point missed completely, by two of you lol. But since you brought it up, how would you go play Tactics when its not released? How is it "a sister game"?

Back to OP.

Not a bad idea I guess. Maintain the ratios but reduce the overall speeds. I dont have an issue with lag shield or whatever you like to call it though. Never had much problem killing lights. I do feel like the game is a bit fast though.

[Skyfaller]

Jadel Blade, on 22 November 2012 - 08:38 PM, said:

And yet it is extremely easy to kill ravens, commandos, jenners, cicadas.

If its IDENTICAL to what we have now what is the advantage?
...
Not a bad idea I guess. Maintain the ratios but reduce the overall speeds. I dont have an issue with lag shield or whatever you like to call it though. Never had much problem killing lights. I do feel like the game is a bit fast though.

It would be identical when it comes to the ability of heavier mechs to track the 100kph max speed lights if the mech's torso twist speed is also reduced.

See, the problem is not player aiming or player dodging skills or the how the mech is driven. Its the netcode chronically failing to register hits when objects move faster than 100kph.

Case in point, get two jenners or 2 cicadas with their fastest possible engine and have circle fight so that they are always seeing each other at 90 degree angles. In order for either mech to hit each other with LASERS.. LASERS of all damn things... you need to aim slightly in front of the mech..in thin air... so that the hit will register. If they aim at each other's exact middle the game engine only applies a quarter or so of the damage you 'see' should be applied because in the engine's view, your laser clipped the rearmost part of the mech and the rest of the laser stream missed...but you saw your laser hit the guy square in the middle for the entire duration of the laser. Normal or pulse lasers it doesnt matter.

Now, have them slow down to 95kph and try again. The aiming at the center suddenly starts to register a lot, LOT more damage. Why? Because the game engine is actually seeing the entire laser stream/pulse hit the mech not just a quarter of it.

So what you have here is a technical issue that is giving the 100kph+ capable mechs near immunity from any form of fire except that which comes from straight in front of their movement vector or from straight behind. Add to that the fact that light practically always circle fight and that heavier mechs have slower torso twist speeds and you get a perfect combination for near-impunity exploit. You can be hit but the damage you take is cut down to 25% or less.

The Raven is an excellent example. Normally the Raven does not mount an engine powerful enough to speed it past 100kph...hence why it is relatively unpopular light mech. Its near 100kph (avg 80-100 kph depending on engine) speed does not give it the benefit of the 'lag shield' so it dies relatively quickly when multiple heavy weapons are trained on it.

I agree with you that the game overall is too fast. It would benefit from a general slow-down ...switch from click-reflex warfare to one that resembles tactics and strategy. I play my Catapult with an XL 145 engine and I always am mindful of cover, firing angle and my team-mates. When I load the standard 260 and play spazticreflexclickmechQUAKE SSRM with it ...it just feels like im too busy clicking and reacting rather than thinking and fighting and making use of the game features.

[MustrumRidcully]

Rhent, on 22 November 2012 - 03:27 PM, said:

It's called Mechwarrior Tactics. You are playing the wrong game. Please leave MWO and go to its sister game Mechwarrior Tactics.

Nah, MW:T isn't a shooter, it's a strategy game, you control multiple units.

But obviously, I am not really suggesting making MW:O turn-based. But I am not sure 10 % would really be sufficient to deal with the lag issues. I've seen Ravens lag-shielding, and they aren't that fast (especially not the Raven I did see lag-shielding).

I am afraid they simply must find ways to improve their net code. Make the prediction of mech, torso and arm movements better for client and server, so they tend to yield the same results more often. Give the server more leeway in accepting client data (so it, say, accepts a 10 % error margin or whatever would be necessary). Give us a server-cross-hair. Make some magical thing that I don't understand because I don't know their code that makes things better.

While my primary interest is usually weapon balance, I fear that if they cannot improve their net code significantly, this game will fail. Lagshielded mechs are not a sign of a quality game. It's at best a sign of a single player game with a mediocre but workable multiplayer mode. A game people bought for the single player campaign but the enthusiasts also use to fight some duels against each other.

Edited by MustrumRidcully, 23 November 2012 - 02:58 AM.

[Sen]

It would be VERY sufficient to deal with these issues. You would effectively turn the game into the "race to the atlas".

I'm sorry, but I require speed to fill my role on the battlefield. 50MS ping or less, so don't complain at ME about hiding behind lag shield. Netcode. . well. . . Fix convergence and put collisions back in , you'll get lucky enough to consider the problem MOSTLY fixed ; )

But don't QQ when your assault mech can't hit me. it wasn't meant to. You want me off your rear armor, get your team to STEP UP AND DO THEIR JOB: Where's YOUR light pilot? Off somewhere chasing moonbeams? he should be peeling me off you. . and, YES, that is part of my job. . and I take my role VERY seriously.

ok, only because it's so much FUN . . . but I digress.

In other news, NERF ASSAULT ARMOR!!!! i can't kill them with 3 alphas to front CT!!! They shouldn't be able to TAKE that kind of punishment! a 10% reduction should do it. . . . .


(ya catchin' my point yet?)

Edited by Sen, 23 November 2012 - 04:42 AM.