As requested in my other thread, I decided to undergo a data collection challenge. 100 Matches of Machine Guns data, to see just how viable they are. After not 100, but 102 grueling challenges, using only a mix of MG + Flamers for the first 50 matches (before the recent 5th June Patch mind you) and MG + SLAS for the remaining 52 matches, I have finally managed to collate all the data.
Given I'm exhausted for now and dont have much time, I will put up the data for people to examine and pick at (including my inane comments), while I will conduct a full results and analysis of Machine Gun issues a bit later. The stats can be found here:
A brief overview (note, some figures are not 100% accurate, as some stats were lost to disconnects - but these are barely 2-3% of the total cases at most):
Total MG Damage for all matches: 17,353
Total MG Rounds hit: 220,860
Total damage over 100 rounds: 23,137
Total SLAS Damage for 50 matches: 4,497
Total number of components destroyed: 129
Total number of kills: 90
Total number of Assists: 190
Total number of wins: 51
Total number of disconnects: 26
Average match damage: 227
Average Component Destruction: 1
Average Number of Kills: 1
Average Number of Assists: 2
Average Total MG damage: 170
Average Per MG Damage: 28
Edit: I also have all the screenshots for almost every match sans 10%, but do not know where to upload the file, as it is 500+ MB. I will try to put this package up on Google drive so people can take a look at it if they wish to see general placements and corroborate with the excel file.
My Analysis at last:
As promised, OP shall deliver. Also editing this in the first post. Took me a bit of while due to a busy weekend, but here are my conclusions after playtesting:
A) Machine Guns, in Theory, are viable Ballistic Sidegrades to Small Lasers.
Before you go all nuts, let me show the following image:
As you can see, in theory, Machine Guns are very much similar to Small Lasers. They do almost the same amount of damage, but trade that tiny bit in exchange for no heat and a higher effective range (which can mean the difference between getting away or ending up a smoking hulk). Add to that their extreme crits and they become actually very decent alternatives. The reason they aren't a sidegrade is best illustrated in point B.
Machine Guns, in Practice, are only viable Ballistic sidegrades to Small Lasers in very few cases.
These cases include: having enough 50+ CT armor, having 3+ Ballistic slots, lower speed than 100KPH. The reason can be described by the following image:
As you can see from the reality of the situation, the ineffectiveness of Machine Guns currently stems, like LRMs before, from a bug.
Considering this, the average of my weapons - 170 MG damage per match - is actually very much less than the average it should theoretically be. Had I been doing full 0.8 damage, the average would have been much, much higher - around 210.8 per match. Component destruction would also be much higher, as the crits are dependent on fire rates - fire less, fewer chances for crits, less crit damage.
The fire rate bug has been proven time and again by many forum-goers and in the threadnaught, but the fire rate bug is only one of three glaring problems. These other two problems include:
1) The visual projectile speed is too low
An edit: many thanks to stjobe for corroborating this information and confirming. The visual effect of machine gun profiles do not in fact match up with their actual mechanics. This makes it very confusing on if you are in fact hitting something at especially high pings of 200+. It is also visually confusing and really needs to be fixed to better assist in aiming and more skillful use of the weapons. Visually fixing it would additionally allow for a better new player experience since it will at least not confuse newbies who aren't aware of the actual mechanics - especially since several trial mechs do have MGs as default loadouts.
2) The cone of fire needs to be removed.
I was arguing for a reduction a while back; but upon playing 100 games I've come to realize that Machine Guns can be very good high-skill weapons. The problem is that in their current iteration with the cone of spread, you are not rewarded for good aim. In fact, you can be dead on tracking a Raven's legs, but only half your shots will hit.
Small mechs with small bodies are actually very hard to hit - especially those with leg and arm joints close to each other, like in the case of the Commando. This is another thing that basically makes using Mgs on lights useless - given many fights end up you vs another light, any damage you do is spread out too far while the enemy can return fire easily to specific components. Add onto this the fact that projectile speeds are low and you can understand why the spider 5K is ineffective.
Conclusions on MGs:
These three problems of fire rate, projectile speed and cone of fire means that the only mech where it is viable is the Jager DD, which carries 6 of them, resulting in a DPS of .56 x 6 = 3.36; slightly higher than an AC5. It also explains why in my data collection, MG-Small Laser ratios were so varying. Machine Guns were only effective against large mechs with large central hitboxes, like Atlai, Dragons, Awesomes, Catapharacts, other Jaegers, etc, because all my fire could be easily focused - and it is against these opponents that you see MG-SLAS ratios converging.
Against smaller opponents like Jenners, Commandos and esepcially Centurions, where the CT was smaller or like Stalkers and Catapults, where its rounded and with the hitboxes of the side torsos nearby, you see far less damage because a lot of my shots simply went into thin air or scattered accross multiple components instead of being applied - in these cases, the ratios diverged. Additionally, through my 100 rounds of testing, I have had only a few cases of ammunition explosions. In fact, I can count the number on one hand. This suggests the crit mechanics aren't that useful and the chance of ammo to explode is far, far too low.
Fixing these three problems would finally, finally bring MGs in line to Small Lasers, making them actual viable weapons on Spider 5Ks and Cicadas. Two Ballistics slots and the investment of 2 tons would actually be viable, given you get two small-laser equivalent that can crit effectively and have decent engagement ranges.
Conclusions on Flamers:
I also played 50 matches using the flamers and found them to be, by far, the worst weapons in the game.
The reason is that Flamers raise up heat on an enemy mech with a cap of 90%. The problem is this cap is only on the enemy mech - not on you. Since the application of heat is very, very slow it means that by the time you get the enemy mech up to 90%, you will have shut down from it. The only use I found for flamers is as a way to cloud enemy vision completely - but to do this required getting in their front arcs, which means you can be hit with their weapons anyway. Their short range also made them worse than small lasers.
In addition, heat inflicted seems to be almost entirely independent of the current heat of the mech - i.e. instead of multipying a mech's current heat level or raising based on that, its an aditional effect based entirely on the flamethrower itself. This is flawed in many ways as it means it doesnt matter how many Alphas an enemy tosses on you, you only raise heat depending on what your flamers do. You dont even stop the cooling of another mech like you used to I suspect - but have no way to verify this.
I really hate to be the one to say this, but the entire flamer system eneds to be reworked from the ground up again. How, I have no idea, but the current iteration simply does not work.
Edited by Alternate22, 10 June 2013 - 04:32 AM.