63 replies to this topic

[Jman5]

I wonder if you could out-run a streak missile going 200kph.

[Appogee]

Piipu, on 25 September 2013 - 03:04 PM, said:

This would be great. Only I think PGI will never actually get to fix the speed cap. Sad but true.

I share your pessimism.

Another option would be to rescale speed entirely. You can keep the numbers accurate, but the fastest Mech would still move at whatever is the fastest speed the game can stand, and slower Mechs would go progressively slower down from that.

At least that way there'd be some relative distribution of speed, and the tradeoff of engine size vs weapon and armor loadout would be more differentiated.

[Training Instructor]

Shadey99, on 25 September 2013 - 03:12 PM, said:

A Spider at 200 kph (120 mph) hitting an obstacle would not disintegrate... A 2 ton car doesn't disintegrate clipping the side of a building at that speed and it doesn't have 6 tons of armor or a 1.5 ton 'internal structure' of 'super advanced' materials like the Spider.

When I was a teenager, two guys from my high school hit a patch of small trees in the median of one of the bigger streets in town late one night. They were in a Nissan 300z, going well over 100mph. Pieces of the car, blood, and gore were scattered over about 200 yards of road. The car absolutely disintegrated when it hit the trees at high speed.

Blood for the Blood God.
Skulls for the Skull Throne.

Edited by Training Instructor, 26 September 2013 - 12:37 AM.

[culverin]

All you people talking about knockdowns, about a spider clipping the leg of an Atlas, that Atlas is going to get shredded just as bad.

[img][url="http://spiff.rit.edu/classes/phys200/lectures/ke_low/eqn_ke.gif"]http://spiff.rit.edu/classes/phys200/lectures/ke_low/eqn_ke.gif[/url][/img]

KE means a lot.
So that means if I clip the arm of an Atlas in a Jenner, the Atlas probably lose the arm and part of the toros while the armor is only gone on my jenner.

[Hexenhammer]

The golden triangle in MWO is speed , offense, and defense and that's the beauty of BattleTech (MWO). It's built in balance. Max speed? Sure! But it will cost something. Will it be weapons? Armor? Both?


And its a natural intuitive balance, unlike ghost heat.

Edited by Hexenhammer, 26 September 2013 - 12:56 AM.

[Oppi]

Shadey99, on 25 September 2013 - 03:12 PM, said:

A Spider at 200 kph (120 mph) hitting an obstacle would not disintegrate... A 2 ton car doesn't disintegrate clipping the side of a building[...]

The heavier you are, the heavier the blow when you hit a solid object that you can't move / break through. E_kin = mv² , where m is mass and v is velocity.

If the additional armor or any "super SciFi materials" would be able to absorb the additional energy ... who knows ? But a comparison to a 2 ton car is in no way adequate here.

[Shadey99]

Oppi, on 26 September 2013 - 01:34 AM, said:

The heavier you are, the heavier the blow when you hit a solid object that you can't move / break through. E_kin = mv² , where m is mass and v is velocity.

If the additional armor or any "super SciFi materials" would be able to absorb the additional energy ... who knows ? But a comparison to a 2 ton car is in no way adequate here.

Ok How about a CV90 which weighs 23-35 tonnes and can do 70 kph? Not quite a Spider in either speed or mass, but closer in form and function. Now what happens when a CV90 hits a office building at speed? Or 'runs into' a hill? The important things btw are the armor and frame, followed by the power of the engine...

More realistically Battlemechs in cities would do tons of collateral damage just brushing them without taking all that much damage themselves and the speeds in general are cut way down because turning at higher speeds takes more room. Hence Urbanmechs going as fast as modern Main Battle Tanks...

[Clit Beastwood]

Pyrrho, on 25 September 2013 - 03:12 PM, said:

Why does my AC/20 shot completely rock a 100t Atlas avatar yet only marginally disturb a 35t Jenner (at least until the shot causes it to die and flip into a glorious somersault)?

I love the somersault and the faceplant... Makes me laugh every time! (Even when it's me!)

[Kunae]

Appogee, on 25 September 2013 - 11:20 PM, said:

Another option would be to rescale speed entirely. You can keep the numbers accurate, but the fastest Mech would still move at whatever is the fastest speed the game can stand, and slower Mechs would go progressively slower down from that.

At least that way there'd be some relative distribution of speed, and the tradeoff of engine size vs weapon and armor loadout would be more differentiated.

Yeah... no.

We definitely do not need any mechs slowed down. When HSR is "fixed", the already vulnerable lights will get even more splatty. If you slow them down, rather than speeding them up, taking a light mech would be suicide.

Relative speed is not the issue. Actual speed vs average person's reaction time, is much more relevant.

[Kazma]

First of all they should fix hit registrations and lower ping. If this isn't done first Idon't think anyone is going to like these changes, light mech players aside.

[Joseph Mallan]

Fire Moth (All) 216 KpH top speed in burst with M.A.S.C. and 162 KpH without. Just sayin.

[Shadey99]

Kazma, on 26 September 2013 - 06:44 AM, said:

First of all they should fix hit registrations and lower ping. If this isn't done first Idon't think anyone is going to like these changes, light mech players aside.

Well the topic originally was started for three reasons:
1. Some people claim that even with 'fixed' HSR that upping the speed cap would be horrible for bigger mechs
2. People were talking about how crappy the Locust will be, as this shows if the cap was lifted the Locust could be much much faster without really losing much firepower... However this leads into #3...
3. People are used to lights all going the same speed, this shows without the cap how variable it would actually be...

It has the side effect of showing the way PGI gets the engine values for all machines...

Edited by Shadey99, 26 September 2013 - 07:01 AM.

[Mehlan]

Quote

First of all they should fix hit registrations and lower ping. If this isn't done first Idon't think anyone is going to like these changes, light mech players aside.

Pgi cannot fix 'PING', they might be able to upgrade their connecting to the 'net, but they cannot control the routeing and quality of connections beyond that between themselves and their provider. They have no influence or control over your isp, my isp etc... If you or one of the other players your matched with has a bad connection, a high latency....PGI cannot do squat about it.

[Appogee]

Kunae, on 26 September 2013 - 06:33 AM, said:

Yeah... no.

We definitely do not need any mechs slowed down. When HSR is "fixed", the already vulnerable lights will get even more splatty. If you slow them down, rather than speeding them up, taking a light mech would be suicide.

Relative speed is not the issue. Actual speed vs average person's reaction time, is much more relevant.

I was saying ''if actual speed cannot be increased''. I agree higher speeds is the preferable solution. But it may not be technically possible.

[Grits N Gravy]

Having played in closed beta when there were no speed caps I can personally tell you that the game was awful at that time. Net code with server side authentication, and wildly varying client pings can not sync up that much movement at rates greater than we have. Not as function of bad net code necessarily but the underlying physics of the operation.

Let look at a mech moving at 150 KPH with a ping of 50. At that speed he is moving .04 meters per millisecond. From one packets round trip to the next that mech has moved 2 meters. Over 10 packet trips the mech may be desync-ed ed by as much as 20 meters. As speeds and ping times increase, the dsync gets exponentially worse with linear rises in speed and ping time.

Lets look at a 200 kph mech with a 200 ping. At that speed the mech is moving .05 meters per millisecond. From one packets round trip he may displaced by as much 10 meters. Over 10 packet trips he may displaced as much as 100 meters. That's just the raw displacements. We get into all other kinds of issues when everyone has different pings. As the positional updates will all be occurring at different times between players with different pings. The relationship between the difference in ping becomes a factor too.

The primary answer to this dilemma is what in MWO is called HSR which amounts to caching the positions of the mechs, then using your ping and the time to determine if there was a hit. The faster mechs go, the worse the ping, the longer the flight time of the projectiles, the larger the cache size must be.

You quickly hit a set of limiting returns here. As the cache grows larger, it takes longer and longer for the server to process the rewinds. As rewinds eat up your processor cycle, everything else gets delayed and we get server lag on top of client lag. Compounding the problem and possibly setting up a feedback cycle which could crash a server. So there is a physical limit to how good HSR can be.

Currently you see anecdotal reports of spiders testing those limits via their small size and speed. So the real question is how well is HSR optimized, how much further can it pushed?

Those technical question don't even get into gameply issues either. How fun would conquest wack-a-mole be, as lights just ran from point to point? Would this cause the rise of circle strafe online? Aka go fast or go home. The 125 Kph hunchback is a force. I honestly don't see much of a benifit from raising the engine limits. Except for in specific cases for to increase the viability of that mech.

[Anastasius]

Without collision damage and the sketchy netcode/hit registration removing speed caps is a terrible idea.

How would you like to try and hit an Aussie in a maxed out speed SDR w/ a 600 ping?

Its bad enough as it is now with no collision and a speed cap. Removing the current cap could very well be game breaking.

[Mehlan]

Quote

Without collision damage and the sketchy netcode/hit registration removing speed caps is a terrible idea.

How would you like to try and hit an Aussie in a maxed out speed SDR w/ a 600 ping?

Its bad enough as it is now with no collision and a speed cap. Removing the current cap could very well be game breaking.

Sorry, but I'm to rant a tad here... READ a thread before your going to post in it. The first post

Quote

Hit registration and issues with server performance (which are the stated reasons for the cap) aside, would any of these really break game balance...?

Edited by Mehlan, 26 September 2013 - 08:38 AM.

[AlexEss]

No speed cap + max engine + Capture accelerator = troll mech of the year =P

[Mehlan]

AlexEss, on 26 September 2013 - 08:45 AM, said:

No speed cap + max engine + Capture accelerator = troll mech of the year =P

People would have to learn to adjust tactics.... much like the complaints of cap's in assault. Yea while it's irritating, it is a valid tactic. The more people that thinking about that and adjusting the better the game might turn out.

[Shadey99]

I'm going to address these in a different order...

Grits N Gravy, on 26 September 2013 - 07:52 AM, said:

Those technical question don't even get into gameply issues either. How fun would conquest wack-a-mole be, as lights just ran from point to point? Would this cause the rise of circle strafe online? Aka go fast or go home. The 125 Kph hunchback is a force. I honestly don't see much of a benifit from raising the engine limits. Except for in specific cases for to increase the viability of that mech.

The only 'well armed' (If you call ~4 tons 'well armed') 'high speed' lights would be the Locust (182-200 kph) and Jenner (160-176 kph), the Spider would come in third in that category because a 340 XL is unrealistic unless you want to run 200 kph with a BAP and no guns or 3xSL with no JJs. Few other lights would gain much if anything. All 3 would be a limited threat and while difficult to hit they are going to take a very long time to do significant damage. They would be better put to use as scouts and flankers identifying targets and keeping the enemy force off balance.

Btw the Locust is the one that needs the cap raised most, it's fragile as a stock Spider (3.5 tons) because of the armor limit by tonnage (~4.5 tons for a 20 ton mech). Speed is it's only real hope on staying competitive with the stock 4 ballistic+1 energy hard point configuration. It has alot of badly used space because it cannot mount larger engines.

Oh and current MWO would look nothing like Closed Beta. These are not 50 ton mechs. Also there are caps to engines even without the speed cap. Hence all those figures in my original post.

Grits N Gravy, on 26 September 2013 - 07:52 AM, said:

Having played in closed beta when there were no speed caps I can personally tell you that the game was awful at that time. Net code with server side authentication, and wildly varying client pings can not sync up that much movement at rates greater than we have. Not as function of bad net code necessarily but the underlying physics of the operation.

You quickly hit a set of limiting returns here. As the cache grows larger, it takes longer and longer for the server to process the rewinds. As rewinds eat up your processor cycle, everything else gets delayed and we get server lag on top of client lag. Compounding the problem and possibly setting up a feedback cycle which could crash a server. So there is a physical limit to how good HSR can be.

Currently you see anecdotal reports of spiders testing those limits via their small size and speed. So the real question is how well is HSR optimized, how much further can it pushed?

We could debate possible solutions to the issue, my field actually is networking and data processing. Like big real time transaction systems (Which should in a sane universe look very much like the backend of a networked gaming system that utilizes servers like MWO does).

However it's not really the point of the topic. This thread is more to look at what 'solving' the hard limit on the '150 kph' (140 kph in game) speed cap would actually mean for changes to the games balance and dynamic. Differentiation of light mechs is actually a big issue at 150 kph as Khobai points out any time it comes up. Giving lights a purpose in the game is of key importance to balancing the game in general.

Anastasius, on 26 September 2013 - 08:05 AM, said:

How would you like to try and hit an Aussie in a maxed out speed SDR w/ a 600 ping?

At a 600 ms ping you couldn't play a light, you'd run into everything. Lately at 'night' (8-10 pm EDT/EST) my ping will jump to around 120 ms and I find it hard to play my lights at that ping (to much warping in a fast mech or sudden massive changes in position of other mechs). I don't fear 600 ms ping assuie's in 150 kph Spiders, let alone 200 kph Spiders with almost no guns (or no armor).