23 replies to this topic

[Piousflea]

TL;DR SUMMARY:
Based on extensive math, there is an optimal range of engine sizes for each weight class of 'Mech. This isn't a hard and fast rule, but 'Mechs will generally feel stronger if they fall within the optimal speed range.

These tables are expressed as an optimal engine size range with base speed in parentheses. This does not include Speed Tweak. The upper speed limit assumes that you are using EndoSteel to save weight. The modified "NoEndo" upper speed limit does not.

* * *
OPTIMAL STD ENGINE SIZE:

20t: STD170 (137.7kph) - N/A
25t: STD190 (123.1kph) - STD270 (174.96)
30t: STD225 (121.5kph) - STD270 (145.8)
35t: STD240 (111.1kph) - STD285 (131.9)
40t: STD245 (99.2kph) - STD295 (119.5)
45t: STD260 (93.6kph) - STD300 (108)
50t: STD260 (84.2kph) - STD300 (97.2)
55t: STD275 (81.0kph) - STD305 (89.8)
60t: STD280 (75.6kph) - STD305 (82.3); NoEndo = STD300 (81.0)
65t: STD280 (69.8kph) - STD325 (81.0); NoEndo = STD310 (77.3)
70t: STD285 (66.0kph) - STD340 (78.7); NoEndo = STD335 (77.5)
75t: STD285 (61.6kph) - STD340 (73.4); NoEndo = STD335 (72.4)
80t: STD285 (57.7kph) - STD340 (68.9); NoEndo = STD335 (67.8)
85t: STD285 (54.3kph) - STD340 (64.8); NoEndo = STD340 (64.8)
90t: STD285 (51.3kph) - STD355 (63.9); NoEndo = STD350 (63.0)
95t: STD295 (50.3kph) - STD360 (61.4); NoEndo = STD350 (59.7)
100t: N/A (0.0kph) - STD360 (58.3); NoEndo = STD350 (56.7)

OPTIMAL XL ENGINE SIZE:

20t: XL245 (198.5kph) - N/A
25t: XL255 (165.2kph) - XL300 (194.4)
30t: XL260 (140.4kph) - XL310 (167.4)
35t: XL265 (122.7kph) - XL320 (148.1)
40t: XL280 (113.4kph) - XL350 (141.8) ; NoEndo = XL320 (129.6)
45t: XL280 (100.8kph) - XL350 (126.0) ; NoEndo = XL325 (117.0)
50t: XL295 (95.6kph) - XL360 (116.6) ; NoEndo = XL350 (113.4)
55t: XL295 (86.9kph) - XL365 (107.5) ; NoEndo =  XL360 (106.0)
60t: XL300 (81.0kph) - XL365 (98.6) ; NoEndo = XL360 (97.2)
65t: XL330 (82.2kph) - XL380 (94.7) ; NoEndo = XL370 (92.2)
70t: XL340 (78.7kph) - XL380 (87.9) ; NoEndo = XL370 (85.6)
75t: XL340 (73.4kph) - XL380 (82.1) ; NoEndo = XL375 (81.0)
80t: XL340 (68.9kph) - XL390 (79.0) ; NoEndo = XL380 (77.0)
85t: XL345 (65.8kph) - XL390 (74.3) ; NoEndo = XL380 (72.4)
90t: XL360 (64.8kph) - XL390 (70.2) ; NoEndo = XL380 (68.4)
95t: XL360 (61.4kph) - XL390 (66.5) ; NoEndo = XL380 (64.8)
100t: N/A (0.0kph) - XL395 (64.0) ; NoEndo = XL395 (64.0)

* * *

First of all, let's define what "Optimal Engine Size" means:

I) A 'Mech is "Too Slow" for its tonnage if you could go up in weight class and upgrade its engine to maintain equal or greater speed, without losing any payload.

Let's use a stock HBK-4G as an example. It's a 50-tonner that moves at 64.8kph. The total payload is 50 tons minus 5 tons of internals minus 13.5 tons of engine and required heatsinks = 31.5 tons.

Compare this to a stock GHR-5H, a 70-tonner that also moves at 64.8kph. The total payload is 70t - 7t internals - 22t engine = 41 tons.

This means that the GHR-5H can carry 9.5 tons more armor+weapons+equipment while having exactly the same speed and agility as a HBK-4G (ignoring quirks).

* * *

II) A 'Mech is "Too Fast" for its tonnage if you could go down one weight class and downgrade its engine while maintaining equal or greater speed, without losing out on any payload.

Let's use the infamous BNC-3E as an example, a 95-tonner that moves at 64.8kph. The total payload is 95t - 9.5t internals - 48t engine = 37.5 tons.

Amazingly, the Assault BNC-3E carries less payload than the Heavy GHR-5H! It's no wonder the stock 'Mech is so bad. The Banshee is too fast. If you wanted to go that fast you should run a lighter 'Mech. If you want to run a 95-tonner, you will need to go slower.

* * *
III) For any tonnage class of 'Mech, the Optimal Speed Range is defined as the entire range of engine sizes that do not meet the criteria of "Too Slow" or "Too Fast".

In order to produce a generous range of speeds, we will use the lowest possible estimate of "too slow" and the highest possible estimate of "too fast":

If we count payload as "weapons and equipment only" (not including Armor), 'Mechs will have to slow down because they have less tonnage to split between engines and payload. Therefore, the lower-bound Slow estimate assumes that all 'Mechs equip maxed armor (~20% of gross tonnage).

If we count payload as "armor, weapons, and equipment", and we allow 'Mechs to equip Endo-Steel internals, we will allow the greatest possible tonnage to mount large engines. This produces the highest possible "fast" estimate.Therefore, the upper-bound Fast estimate ignores armor and assumes that all 'Mechs use EndoSteel.

A modified upper-bound estimate may be used for heavier 'Mechs since they are fairly unlikely to have room for Endo-Steel. The Semi-Fast estimate ignores armor but does not use EndoSteel.

I made a spreadsheet to do all of these calculations and you can find the sheet at the end of this forum post.

* * *
IV) Light Discussion

Based on the math, 20- and 25-ton 'Mechs have an optimal XL-engine size that is greater than the maximum allowed engines in game. This makes a lot of sense when you look at Light 'Mechs in game. A 145-kph Arctic Cheetah outclasses a 154-kph Locust because going up from 20 to 30 tons gives you so much more room for Armor and Weapons, and within the 140-150kph range the extra tonnage doesn't slow you down. Locusts are incredibly heavily quirked because they'd otherwise be useless.

In order to fully utilize the move-speed advantage of the 20-ton weight class, one would need to mount a XL245 and run at 200kph. While there are tabletop 'Mechs that can move at these insane speeds (12/18/0 and up), in MWO lag and desyncs would be a big problem at such speeds.

* * *
V) Exceptions to the Rules

Optimal speed ranges are certainly not hard-and-fast rules. There are many times when you may want to build a 'Mech slower than its optimal speed range. Some 'Mechs simply cannot mount engines large enough to put them in the optimal speed range, including all 20- and 25-ton 'Mechs in the game.

Otherwise, downgrading an engine is usually done to take advantage of hardpoints or Quirks. While a spreadsheet may say that a 70-ton, 60-kph Archer cannot carry as much payload as a 90-ton, 60-kph Mauler... we all know that an Archer has 9 missile hardpoints and a Mauler does not.

There are also plenty of "Turret Builds" (such as Urbanmechs) that intentionally move very slowly and carry as much firepower as possible. Since Optimal Engine Size is used to maximize 'Mech speed for a given amount of weapons payload, it really doesn't apply to 'Mechs that are intentionally slow.

On the other hand, there is no good reason why you would ever want to build a 'Mech faster than its optimal speed range. You can't take advantage of quirks or hardpoints if your 'Mech has no tonnage left over for weapons, like the BNC-3E. If you're really committed to moving at 64.8kph, downgrading to a lighter 'Mech will actually give you more payload tonnage. Amazingly enough, dropping 10 tons from a Banshee to Battlemaster allows you to mount 5 more tons of weapons and armor while keeping 64.8kph (4/6/0) mobility.

* * *
References:
http://mwomercs.com/...dsheet-warrior/
http://www.ci-n.com/...imalweight.html

Excel Spreadsheet at:
https://drive.google...iew?usp=sharing

Edited by Piousflea, 25 March 2016 - 01:08 AM.

[John1352]

That is quite an elegant way of showing the gargoyle and ice ferret spend too much tonnage on engines. I'm a bit surprised by aspects like the 330 XL on a 65t mech being the minimum before it is better to take a 70 tonner.

In practice mechs are often best with engines smaller than this table shows. More guns with smaller hitboxes has benefits.

Did you account for the minimum 10 heat sinks with the smaller engines?

[aaykeem]

Piousflea, on 25 March 2016 - 01:02 AM, said:

On the other hand, there is no good reason why you would ever want to build a 'Mech faster than its optimal speed range. You can't take advantage of quirks or hardpoints if your 'Mech has no tonnage left over for weapons, like the BNC-3E. If you're really committed to moving at 64.8kph, downgrading to a lighter 'Mech will actually give you more payload tonnage. Amazingly enough, dropping 10 tons from a Banshee to Battlemaster allows you to mount 5 more tons of weapons and armor while keeping 64.8kph (4/6/0) mobility.

The analysis and data are very interesting, I had noticed something very similar while building mechs, but I didn't try to explain it as a relation between tonnage and engine size.
The only objection I have is that you are placing weapons and armor under the same "payload" umbrella. I'd actually put max armor on all mechs and only take into account the space left for weapons and equipment. The reason for this is that, when you say "dropping 10 tons from a Banshee to Battlemaster allows you to mount 5 more tons of weapons and armor", it's actually misleading because the BLR will have a lower armor cap than the BNC, so even if it has 5 more tons for "payload", it will not be able to use it to close the armor gap.

[Piousflea]

John1352, on 25 March 2016 - 02:23 AM, said:

That is quite an elegant way of showing the gargoyle and ice ferret spend too much tonnage on engines. I'm a bit surprised by aspects like the 330 XL on a 65t mech being the minimum before it is better to take a 70 tonner.

In practice mechs are often best with engines smaller than this table shows. More guns with smaller hitboxes has benefits.

Did you account for the minimum 10 heat sinks with the smaller engines?

Yes. All engines rated 245 and less include 1 ton for each missing heatsink.

aaykeem, on 25 March 2016 - 02:33 AM, said:

The analysis and data are very interesting, I had noticed something very similar while building mechs, but I didn't try to explain it as a relation between tonnage and engine size.

The only objection I have is that you are placing weapons and armor under the same "payload" umbrella. I'd actually put max armor on all mechs and only take into account the space left for weapons and equipment. The reason for this is that, when you say "dropping 10 tons from a Banshee to Battlemaster allows you to mount 5 more tons of weapons and armor", it's actually misleading because the BLR will have a lower armor cap than the BNC, so even if it has 5 more tons for "payload", it will not be able to use it to close the armor gap.

Actually, the "minimum" speeds do subtract Armor from the total payload. The "maximum" speeds do not.

Without armor subtraction, the Banshee has 37.5t payload while the Battlemaster has 42.5t payload, a difference of 5t.
Subtracting max armor of 20% gross tonnage, the Banshee has 18.5t payload while the Battlemaster has 25.5t payload, a difference of 7t.

As you can see, armor subtraction actually makes the big fast 'Mech look even worse.

[Boulangerie]

Really great analysis here! Good solid approach and I've not noticed any flaws in the theories so far. Posting here so that I remember to compare your data to my mechs. Thanks for all the hard work!

[Mole]

John1352, on 25 March 2016 - 02:23 AM, said:

That is quite an elegant way of showing the gargoyle and ice ferret spend too much tonnage on engines.

I dunno, I do really great in both of these mechs that you mentioned and I don't see anything wrong with them. I'd even dare say I would consider bringing my Ice Ferrent into a FW match if I played clan FW. I wouldn't bring my Gargoyle though because, even though I do extremely well in it, my Warhawk is by far a better 'mech for extended engagements like what is seen in FW. I eat solo queue face in my Gargoyle though.

[L1f3H4ck]

This becomes a lot more rigid when you apply tabletop rules, where the size of the engine is always weight times walking speed. This also made certain weights better than others, like 55 and 75 tons, since they could carry the most in their respective speed brackets.

The whole calculation becomes more realistic, and therefore more useful if you assume maximum standard armor value. The bigger and slower your Mech, the more armor it needs to survive. Armor an Atlas like a Firestarter and see who dies first. It is also the base from where you begin construction, being reduced as a sacrifice to gain even more available carrying capacity at the cost of survivability.

[Nyte Wolf]

Good work . I remember doing this for table top way back in 1987 or so before xl engines or endo steel. and i believe all engines had 10 heat sinks minimum. I used graph paper lol.

[sycocys]

"Too fast for your tonnage" --- There is no such thing.

[Bluttrunken]

Interesting and not entirely off the mark but it disregards the fact that the higher weightclass still has more armor. While the speed and the payload might be the same, the mech is still more durable. There is still a trade-off.

The table might present itself as a nice starting point while building a mech but given other constraints it's more reasonable to choose your engine on a chassis by chassis basis.

Edited by Bluttrunken, 31 March 2016 - 07:17 AM.

[Fiona Marshe]

305 is an absolutely AWFUL engine to fit, due to the way overall tonnage works.

Gyro goes from 3t to 4t for no significant gain.

[Bud Crue]

See:

http://mwomercs.com/...e-and-speed-is/

For CS's original thread on this topic.

[Hrett]

So, curious what people think about a Zeus with an XL? I know its not an ideal mech at least based on MetaMech, but its my favorite chassis so far and I do reasonably well in it with the STD 320 engine because zombie.

But I hear it is "XL Friendly" for an assault, so was going to try a XL370. The 9S Gets me 80kph with 3LL, 3MPL and an SRM6. How fast is too fast for an assault? Im 2 weeks new and still getting the torso twist shielding thing down, but the Zeus seems rather forgiving so far. Im worried that the XL will make it not so, and the engine costs more than any mech in my stable, so Id rather not make that rather expensive mistake this early.

As an aside, the "Doomfist" 6T Zeus is great fun!

Thanks in advance.

[Bud Crue]

Hrett, on 12 April 2016 - 01:44 PM, said:

So, curious what people think about a Zeus with an XL? I know its not an ideal mech at least based on MetaMech, but its my favorite chassis so far and I do reasonably well in it with the STD 320 engine because zombie.

But I hear it is "XL Friendly" for an assault, so was going to try a XL370. The 9S Gets me 80kph with 3LL, 3MPL and an SRM6. How fast is too fast for an assault? Im 2 weeks new and still getting the torso twist shielding thing down, but the Zeus seems rather forgiving so far. Im worried that the XL will make it not so, and the engine costs more than any mech in my stable, so Id rather not make that rather expensive mistake this early.

As an aside, the "Doomfist" 6T Zeus is great fun!

Thanks in advance.

Yes, Zeus is XL "friendly". I actually think of it as closer to XL "necessary". Without an XL it just doesn't feel like it caries enough weapons to be an "assault".

As to doom fist. Agree. A must have.

[ImperialKnight]

ideally go for engines of multiples of 25. each 25 increasing in rating either gives you more internal DHS, or an extra slot for DHS. which saves you crit slots

[Hrett]

knightsljx, on 12 April 2016 - 04:20 PM, said:

ideally go for engines of multiples of 25. each 25 increasing in rating either gives you more internal DHS, or an extra slot for DHS. which saves you crit slots

Ahh this is helpful. So XL 375 it is!

[Snowbluff]

This is like the coolest thing ever. Like an energy-maneuverability theory but for mechs with guns. :0

[Gasboy]

sycocys, on 31 March 2016 - 06:09 AM, said:

"Too fast for your tonnage" --- There is no such thing.

The Charger disagrees.

[Snowbluff]

knightsljx, on 12 April 2016 - 04:20 PM, said:

ideally go for engines of multiples of 25. each 25 increasing in rating either gives you more internal DHS, or an extra slot for DHS. which saves you crit slots

Not necessarily. The 25 is nice, but some engines weight the same as others. Check the engine weights- sometimes you can get an extra 5 power over one of the ones with a heatsink slot for no weight.

[Boulangerie]

Snowbluff, on 03 May 2016 - 04:39 PM, said:

Not necessarily. The 25 is nice, but some engines weight the same as others. Check the engine weights- sometimes you can get an extra 5 power over one of the ones with a heatsink slot for no weight.

Yes, they are all XL engines, and due to the rounding when calculating weight of an XL Engine. One example is XL275/XL280.