FA On vs. FA Off Data

[Mohrgan]

So, at long last I wrestled the process to the ground, and I came up with these numbers for a FAS with t3 Clean Drive Thrusters. I think the differences in Top Speed across the motions has stuck in my craw. Has anyone else seen this?

Vectors	Top Speed	FAon	FAoff
Fwd	270	27.47	35.39
Fwd+Up+Rt	270	27.27	151.26
Ret	163/270	17.4	22.0
Up+Rt	270	21.14	26.81
Ret+Up+Rt	270	23.83	37.35
Up	217/270	14	14
Ret+Up	251/270	19.61	31.77
Fwd+Up	270	20.04	51.53

 

[Truesilver]

So, at long last I wrestled the process to the ground, and I came up with these numbers for a FAS with t3 Clean Drive Thrusters. I think the differences in Top Speed across the motions has stuck in my craw. Has anyone else seen this?

Nice data, Mohrgan, thank you.

Concerning your craw! - obviously 100% of top forward speed in all directions with FA-off is kind of the point of the thing and although it leads to a heavy prevalence of 4-0-2 reverski it's a well-established part of the game's flight model.

Concerning the FA-on speed caps, the 60% cap on FA-on reverse (as opposed to 100% with FA-off) is afaik a universal across all ships and builds.

The 80% FA-on speed cap on up thrust alone I'm not sure about but it seems plausible. It's interesting that it rose to 93% with reverse+up before capping at 100% with forward+up.

I vidded all my vectors last night in a g5 dirty enhanced drives iCourier and will get on with working out my times and posting.

 

[StiTch]

I vidded all my vectors last night in a g5 dirty enhanced drives iCourier and will get on with working out my times and posting.

If I recall you use a relatively bulky iCourier loadout....do you not get better results out of the G3 DDs?

 

[Truesilver]

If I recall you use a relatively bulky iCourier loadout....do you not get better results out of the G3 DDs?

I think it's very middle-of-the-road rather than bulky, in iCourier circles - size 3 bi-weave, 3 x size 1 HRP, no SCB's, everything else stripped down. Tends to be about 635 max boost with 3 x medium weapons.

(I don't like the fat prismatic + SCB's build.)

If you mean g3 for better optimal mass on the thrusters, I (think??) I have a really good roll on my current g5 dirty enhanced giving me the best of both worlds. But I can post more specs tonight.

 

[StiTch]

I think it's very middle-of-the-road rather than bulky, in iCourier circles - size 3 bi-weave, 3 x size 1 HRP, no SCB's, everything else stripped down. Tends to be about 635 max boost with 3 x medium weapons.

(I don't like the fat prismatic + SCB's build.)

If you mean g3 for better optimal mass on the thrusters, I (think??) I have a really good roll on my current g5 dirty enhanced giving me the best of both worlds. But I can post more specs tonight.

Heh yeah to me, that's relatively bulky. My iCourier is rocking G3 drives at present because at first I used my mats on bigger ships, but then realised using Coriolis that the move to G5 would actually hinder me without the intervention of a favourable secondary effect.

If you got lucky on said secondary effects then good on ya, just piqued my interest as I noticed the move to G5 shafting the iCourier at a fairly early mass point using hypothetical min/max stats.

On topic, I would be interested to know more specifics about deceleration. I need to sit down, throw my ships around a bit and observe deceleration patterns at various velocities when I start turning with FA-Off. The early days of flying my iCutter taught me the folly of underestimating rear thrusters as a deceleration tool, which I realised could be applied to smaller ships in a slightly different way...can't afford to turn down more lessons like that.

 

[Aris_]

I would like to hint a small detail about the lead aim reticle and faoff , if we assume that we have a skin radar to calculate the lead for fixed wpn targeting then by using faoff and rotate the nose around the flight path should change the position of aim reticle also.
The reason is in real life in order to calculate the next position is not only the main speed vector needed which in faoff (affected bigger at least in acceleration) is the resultant from the use of thrusters thus the position of the nose needed -carrier of the engine thrust- so i think this is not taken into consideration for arcade reasons because will render fix non laser wpn pretty much useless.

 

[Truesilver]

OK, here's my first batch of data. I will add more over the next few days, in particular the FA-Off results.

All inputs were digital, with 4 pips to Eng, using this iCourier:

Spoiler

I'm a bit pressed for time so will just give results in seconds for now, i.e. seconds taken to change speed by 100 m/s.

Where I say "Decel" I mean that I'm applying the direct opposite inputs to the row above to bring the ship back to stationary. In this instance therefore 100-200 becomes 200-100 and 0-100 becomes 100-0.

Where I give a "?" for decel this means that the way the throttle shows speed change in multiple vectors makes it too difficult to state the outcome.

All with Flight Assist ON:

INPUT	0-100 (or decel 100-0)	100-200 (or decel 200-100)	Top Speed
Forward	2.8	1.3	465 (100%)
Decel	2.2	2.3
Reverse	3.2	2.3	279 (60%)
Decel	1.2	1.3
Up	3.5	2.7	372 (80%)
Decel	0.8	0.7
Right	3.3	2.7	372 (80%)
Decel	0.8	0.8
Up + Right	2.8	1.9	465
Decel	0.7	0.8
Forward + Up	2.8	1.2	465
Decel	?	2.3
Forward + Right	2.8	1.2	465
Decel	?	?
Forward + Up + Right	2.7	1.2	465
Decel	?	?
Reverse + Up	2.7	1.8	431 (92.5%)
Decel	?	0.7

Note that where I show "2.8" or "2.7" (for example) this almost certainly just means the same outcome, i.e. more like "2.75".

o7

Truesilver

 

[Aris_]

@Truesilver

+52% optimal multiplier ?
How many rolls?

That is impressive

 

[Truesilver]

@Truesilver

+52% optimal multiplier ?
How many rolls?

That is impressive

It's a good roll, particularly with the very high optimal mass also but I think that's actually just an enhanced drives thing ... I think (iirc) you have to adjust down by 15% ... so it's really like +37%. Because enhanced begins at +15%. Or something.

Anyone remind us of the details?

 

[jgm]

It's a good roll, particularly with the very high optimal mass also but I think that's actually just an enhanced drives thing ... I think (iirc) you have to adjust down by 15% ... so it's really like +37%. Because enhanced begins at +15%. Or something.

Anyone remind us of the details?

The nominal optimal multiplier of EPTs is 1.15, yes.

Interestingly, the EPTs have different multipliers for acceleration, rotation and speed. They're actually 1.1 for acceleration, 1.1 for rotation and 1.25 for speed. 1.15 number comes from an average of the three.

To my knowledge, the EPTs are the only thrusters that have different values for these multipliers. Normal thrusters have a single value for all three stats.

 

[Frenotx]

The nominal optimal multiplier of EPTs is 1.15, yes.

Interestingly, the EPTs have different multipliers for acceleration, rotation and speed. They're actually 1.1 for acceleration, 1.1 for rotation and 1.25 for speed. 1.15 number comes from an average of the three.

To my knowledge, the EPTs are the only thrusters that have different values for these multipliers. Normal thrusters have a single value for all three stats.

I remember reading the same thing. The idea was to give EPTs a lot more speed, but without making their maneuverability worlds better. To me, it just makes ships feel... weird. The higher speed to thruster power ratio makes the ship's movement feel a little unresponsive to me.

 

[Truesilver]

The nominal optimal multiplier of EPTs is 1.15, yes.

Interestingly, the EPTs have different multipliers for acceleration, rotation and speed. They're actually 1.1 for acceleration, 1.1 for rotation and 1.25 for speed. 1.15 number comes from an average of the three.

To my knowledge, the EPTs are the only thrusters that have different values for these multipliers. Normal thrusters have a single value for all three stats.

Thanks as ever, JGM.

I remember reading the same thing. The idea was to give EPTs a lot more speed, but without making their maneuverability worlds better. To me, it just makes ships feel... weird. The higher speed to thruster power ratio makes the ship's movement feel a little unresponsive to me.

Yeah, the disproportionate buff to top speed was Mike & Co's plan, I believe.

I wonder, Mr Otter Fox, what ships you've tried 'em in? My biggest bar to fully RNGineering a second PvP ship after my Courier used to be da grind. Now da grind has been righteously nerfed, the biggest bar is that I simply hate every single second flying any other ship than my g5 enhanced Courier ... mainly because I love the manoeuvrability, as the top speed I hardly ever use!

 

[Morbad]

If maneuverability scaled linearly with speed for EPT, any one with good thruster controls could orbit most of the medium ships in the game without ever being exposed to fire.

It may feel weird, but it's probably for the best.

 

[Frenotx]

I wonder, Mr Otter Fox, what ships you've tried 'em in? My biggest bar to fully RNGineering a second PvP ship after my Courier used to be da grind. Now da grind has been righteously nerfed, the biggest bar is that I simply hate every single second flying any other ship than my g5 enhanced Courier ... mainly because I love the manoeuvrability, as the top speed I hardly ever use!

The courier and sidewinder, actually. They felt very drifty to me. Hopped back in my VIV fairly quickly. XD

- - - Updated - - -

If maneuverability scaled linearly with speed for EPT, any one with good thruster controls could orbit most of the medium ships in the game without ever being exposed to fire.

It may feel weird, but it's probably for the best.

Sounds good to me.

 

[Truesilver]

For those that have not seen there is currently some very hot discussion over whether Frontier have implemented a stealth change to the FA-Off / FA-On characteristics ... specifically whether the previous massive advantage to lateral deceleration conferred by FA-On may have been changed (possibly via a buff to FA-Off thrusters).

Please see my quote below and the linked thread in which the debate is going on pending more testing:

OK, I have been back to Windows Movie Maker, for science.

Here is my direct 21st February 2017 comparison of acceleration and deceleration, with Flight Assist On or Off, in my 631 mps iCourier with a medium weight build and g5 dirty enhanced drives. The full ship characteristics are in the spoiler:

Spoiler

http://i.imgur.com/CwytrRD.png

Here's the table. Note that I'm not here giving an acceleration value, just the time taken in seconds. The ratios are of course the same either way, only the means of expression is different:

INPUT	Flight Assist?	0-100 mps	100-200 mps	200-100 mps
Right	FA-On	3.3 secs	2.7 secs	0.8 secs
Right	FA-Off	2.7 secs	2.7 secs	3.4 secs
Up + Right	FA-On	2.8 secs	1.9 secs	0.8 secs
Up + Right	FA-Off	2.1 secs	1.9 secs	2.3 secs

NOTES

Note that the 200-100 decel figure was using reverse inputs, e.g. down + left.

There was no forwards/reverse motion in this phase of testing.

Although it is difficult to compare 100-0 decel (because of the way the HUD shows speed) with FA-off and even with FA-on in some planes, wherever a clear comparison can be made, 200-100 appears to be identical to 100-0.


OBSERVATIONS

Now, some observations, on how the game's flight model was back in February 2017:

1. In all cases FA-On and FA-Off acceleration were identical 100-200 mps.

2. However, FA-Off was superior in 'getting started': FA-Off acceleration is higher than FA-On for 0-100 mps.

3. The colossal difference comes in deceleration, which has become the main topic of discussion here. With FA-On the ship reduced its lateral speed from 200-100 mps by 4.25 times more quickly with a single lateral thruster in play (left thrust arresting prior right thrust) and by 2.875 times more quickly with two lateral thrusters in play (down + left thrust arresting prior up + right thrust).

This is the 'magic decel' that some of us have not only documented and tested but actually built flying styles around. I have previously stated for good reason that I have found a good way to make a plasma guy miss is to accelerate in one plane with FA-Off, then put FA back on and head in the opposite lateral direction, causing him to miss in the original direction of travel due to magic FA-On decel. You can't - or, at least, couldn't - do the same thing if you kept FA-Off because the decel would be far less forceful.

What I have documented above in my Courier is precisely what Frentox documented in his Viper IV and what Tannik Seldon (@SushiCW) documented in his Keelback, what Morbad has observed and what Alexander the Grape has observed. The latter was actually commenting on precisely this topic the other day on a sub-reddit, where he was saying that this is why he still toggles FA back on from time to time.

Now, from @Ziljan's Corvette test results earlier in this thread, combined with @Morbad's immediate and urgent attempts to confirm, it sounds as though a profound change has been made and when I duplicate this testing in the same Courier, I will find that the massive FA-On decel discrepancy no longer exists.

This will be interesting...

o7 all,

Truesilver