G-resistance of the virtual pilot - opinions and discussion

[Krisu]

I haven't read the paper(s) this model is based on, I don't understand the parameters that went into making this what it is, this makes it difficult for me to judge realism. 

 

Don't get me wrong, the implementation of this system was impressive; you can pull harshly and let go before your screen goes fully black but then it's already too late and you black out all the way because you applied at too high a rate. In the words of the devs, the G-load depends on its duration and rate of G applied and the delay comes from the "hemodynamics" of the cardiovascular system. Seems really legit and well researched....however when playing the game I wonder, is it really that hard to pull G. 

 

The FEELING is that the +G-loc visual onset seems a little too harsh, the recovery time seems a little too harsh; you do feel "restricted" in a dogfight, especially one that lasts. -G-loc needs to be more punishing I agree but... go a little easier on the pilot fatigue, I still get the feeling I should be able to pull off maneuvers I can't pull off. 

 

inb4 git gud (I know but these are my feelerinos)

[Floppy_Sock]

Some people have already linked to my previous posts regarding the comparison of the current IL2 physiology model to results published in the scientific literature. Though I fear it might seem a little repetitive at this point, I would like to provide a summary of my findings and compare them to the current model we have in the game right now.

 

First and foremost, I would like to point out that I think modeling pilot physiology has been a solid forward step for this sim and I am in no way advocating its removal - I simply would like the model to be as accurate as permissible. As such, I'd like to recommend two changes to the current model:

 

1. The inclusion of the "push pull" effect. It has been well documented in scientific literature that a exposure to -Gz reduces tolerance to +Gz. A few people have already mentioned this phenomena so I'm not going to dwell on it too long. 

 

Quote

All the subjects experienced a reduction in relaxed +Gz tolerance after the preceding exposure of -1 Gz for 5 s. The statistically significant average decrease of relaxed +Gz tolerance in the volunteer group and the pilot group was 0.87 and 0.95 G respectively.

[1]

 

Quote

relaxed +Gz tolerance was reduced by 0.77 G if preceded by -1 Gz for 2 s exposure

[2]

 

Quote

-0.5 Gz for 5 s and -2 Gz for 5 s exposure lowered the tolerance by 0.6 and 1.2 G, respectively, verifying that higher levels of -Gz exposure would aggravate the reduction in +Gz tolerance

[3]

 

Unfortunately, the literature's testing of this effect is limited comparatively "gentle" -Gz accelerations though it is clear that even very short exposures of mild -Gz can induce substantial tolerances reductions. 

 

2. An overhaul of the rapid onset rate (ROR) tolerances. For this section, I'd like to provide a little bit of context for those who are unfamiliar. If you do a bit of googling about g induced loss of consciousness, you will surely stumble onto this figure

 

This is a figure from a 1956 paper by AM Stoll titled "Human tolerance to positive G as determined by physiological endpoints." It compiles 40 events of which 13 are GLOC events and uses that to create the curve you see above. Remember those number as they're important. 

This was largely left unchallenged until the 2013 publication by Whinnery and Forster [4] which compiled 880 GLOC episodes spanning the years 1978 to 1992 at the USAF school of aerospace medicine and the Naval Air Warefare Center. This data is a combination of both protected ( g suit / reclined seat) and relaxed GLOC episides and a large variety of participants: volunteer research subjects, aircrew undergoing training to improve G tolerance, students in various aerospace medical disciplines, and aircrew undergoing medical evaluation.

 

An investigation of the entire data set's response to acceleration onset rate revealed the following: 

 

Maybe surprising to some, but above an onset rate of 1g/s, the time to GLOC (LOCINDTI) is pretty much constant. 

See the column outline in yellow in the following table. The data set above is binned and the averages are reported as follows:

 

Regardless of onset rate, the averages time until GLOC remains practically constant. The above data was then used to determine that there exists what is know as a "functional buffer period" (FBP) which is the approximate time the human brain can operate without a blood supply before consciousness is lost. This led to the following model 

Note the curve I showed at the beginning of this post is marked in red here and labeled Stoll. This is not to criticize Stoll for bad work but simply that he did not have the data sets that are available today when he proposed his model. The current model is marked in blue. 

 

Finally,  I want to quote a section of the abstract which summarizes the findings succinctly: 

Quote

The two new G-LOC curves differed significantly from previous curves in temporal characteristics and key aspects underlying neurologic response to acceleration. The new acceleration onset rate curve reveals that for onset rates ≥ 1.0 G/s, G-LOC will occur in a mean time of 9.10 s and is independent of the onset rate. The new +Gz-level curve demonstrates that G-LOC will occur in a mean time of 9.65 s for rapid onset rate exposures to +Gz levels ≥ +7 Gz. The minimum +Gz-level threshold tolerance was defined as +4.7 Gz. When +Gz onset rates are gradual, ≤ 0.2 G/s, G-LOC occurs in a mean time of 74.41 s. G-LOC did not occur earlier than 5 s for any acceleration exposure.

 

Note the last sentence - nowhere, regardless of onset rate, did a subject lose consciousness before 5 seconds. 

 

To further hammer home this point, compare the above findings of minimum LOC time to the equivalent LOC time from a variety of studies which also induced a rapid LOC

 - these are direct quotes from [4] on page 8 and 9. See those for sources if interested. Time to LOC is bolded and underlined at the beginning of each test.

 

6.41 - 6.91 s - Acute arrest of cerebral circulation in 74 humans using a cervical pressure cuff  

 

7.9s (range 6.3s to 12.2s) - Mean time from vascular neck restraint to LOC defined by onset of eye fixation in 24 healthy volunteer police officers 

 

10.4 ± 3.0 s (range 8 to 18 s) - Time to loss of consciousness in 14 filmed human hangings

 

9.2 and 10.7 s   - The mean LOCINDTI values from completely unprotected (no anti-G straining maneuver or anti-G suit) G-LOC episodes for two types of ROR runs (onset rates (1.0-1.5 G/s and 2.2 G/s – 3.0 G/s respectively) to a mean of +6.1 Gz (This is especially relevant to the figure I have shown below)

 

12.65 s - mean LOCINDTI in a  large US Navy Pensacola study, taking 935 individuals to GLOC with a mean +Gz level of +5.3 Gz and a mean onset rate of 0.8 G/s

 

9.40 ± 4.10 s - the limiting time to complete visual blackout with abrupt onset of rapidly applied external eye pressure equal to or greater than systolic ophthalmic artery pressure in 10 normal individuals

 

Contrast this to the physiology model in IL2 right now. Below is the g load experienced with a fresh pilot during a max performance horizontal break turn in a Spitfire mk9. Here I can achieve GLOC in 6 seconds reliably while loading at approximately  ~ 1.5 g/s. This is in stark contrast to the results I showed above. I should reiterate, this is a mint pilot whose g tolerance decreases substantially as the current model renders him "exhausted" but even fresh it does not match the relaxed g tolerances shown above. The green line denotes the turn initiation and the red denotes GLOC.

The recording corresponding to the chart can be found here: https://youtu.be/Rt4_5vT_BUo

 

And then a test to see how fast I can GLOC - around the 3.6 second mark is pretty consistent - the 2 red lines denoting a band where blackout occurs. Sorry for the change in chart axes - these are from two different post processing iterations. In the figure below the right axis in orange is the important one unlike above. 

 

[1] Xu, Y., Li, B., Zhang, L. et al. A centrifuge simulated push–pull manoeuvre with subsequent reduced +Gz tolerance. Eur J Appl Physiol 112, 2625–2630 (2012). https://doi.org/10.1007/s00421-011-2234-3

 

[2] Lehr AK, Prior ARJ, Langewouters G, Ullrich B, Leioner H, et al (2001) Previous exposure to negative Gz reduced relaxed +Gz tolerance. Aviat Sp Environ Med 63:405

 

[3] Wright H, Buick E (1998) The +Gz-tolerance limits of the push–pull phenomenon. Aviat Sp Environ Med 69:202

 

[4] Whinnery, T., Forster, E.M. The +Gz-induced loss of consciousness curve. Extrem Physiol Med 2, 19 (2013). https://doi.org/10.1186/2046-7648-2-19 https://link.springer.com/article/10.1186/2046-7648-2-19 (free to download here)

 

Edited September 2, 2020 by Floppy_Sock

[216th_Jordan]

15 hours ago, DerSheriff said:

However more recent research shows, that especailly the G onset (the speed with which the forces set in) play a smaller role than this model suggest.

 

 

Thats basically my only criticism right now - For example one has to be extremely careful when initiating snap rolls above 400 kph to not G-LOC and die. 

[Soilworker]

20 hours ago, mattebubben said:

First im happy to see a poll instead of just making a change based on  repeated complaints from a small amount of players.

 

Since this poll is a very clear example of the Silent majority Vs the Loud Minority.

 

And second.

 

Im mostly happy with the G-Resistance as it

only changes it would like made is regarding the Negative G (like many others have also talked about).

 

So i want it to stay on the Intermediate level but with some continuing tweaking since the Negative Gs dont feel quite right.

 

My sentiments exactly. 

[Real_NBD]

With small tweaks discussed in this thread (onset rate and "bobbing"), the model could be pretty much perfect I think. Increasing the max G could be something the server admins/map designers could choose just so we could try out the balance consequences. Don't know if that is technically feasible, and I do think the current limit is nice gameplay-wise and that's what really matters in the end (190s should get extra 0.3g or so due to the seating position tho, it would really suit their high-speed nature).

[Chief_Mouser]

I'm happy with it as it as - as long as the AI suffer exactly the same effects. I can't say that I've ever noticed an AI behaving as if he's unconscious, but I would expect a lack of new manouvres whilst they are.

[blue_max]

I'd really want an 'ace' option, halfway the two options now. And I feel like the current realistic seeing is a tad too agressive in blacking you out with a turn at high speed. 

 

Thanks to the devs for putting this poll up!

[=IRFC=SmokinHole]

I voted the last option.  I experience moderately high g's (+7, -3) on a regular basis during the season.  G-tolerance is as much a part of one's innate physiology as it is something gained through experience.  And the "experience" aspect of G-tolerance has more to do with your exposure over the last month than it does a lifetime of pulling and pushing.  The "Ace" coming back from a week in Paris is probably no more tolerant than the newbie who's been busting his butt since arriving at the squadron 3 days ago.  My other issue is that the "g-thing" seems to differ from plane to plane and with speed.  G is G.

Edited September 2, 2020 by SmokinHole

[US63_SpadLivesMatter]

If this game let us customize pilot builds I bet we'd see a bunch of 4 foot 9 power lifters in every cockpit. 

[Voyager]

Honestly, changing the threshold for everyone would leave everyone at the same spot we are at now; the same.

 

I'm happy with it the way it is. I just need to learn the plane's and pilot's limits and respect them. 

[=IRFC=SmokinHole]

Quote

If this game let us customize pilot builds I bet we'd see a bunch of 4 foot 9 power lifters in every cockpit.

I am 6'4" and 169 pounds.  The seated position on my Pitts is straight up.  6 or 7 G's for as many seconds is mildly uncomfortable with some vision loss at the periphery.  Being short and stocky is more helpful for sustained g over longer periods.  Others here have more experience than me I am sure but I believe that the advantage provided by a thick neck and short extremities benefits the F-16 pilot way more than the guy flying props.

Edited September 2, 2020 by SmokinHole

[BraveSirRobin]

33 minutes ago, US63_SpadLivesMatter said:

If this game let us customize pilot builds I bet we'd see a bunch of 4 foot 9 power lifters in every cockpit. 

Short and stocky is better for withstanding g’s.

[=IRFC=SmokinHole]

Yes I know.  But that is only a benefit over a long sustained pull.  Modern jets have the power to allow for such lengthy g-loads.  For the 10 seconds that a P-51 pilot might experience under 6+ g's, keeping blood in the brain is not as much of a problem.  So being short and stocky isn't really an advantage.  What is more important is that he doesn't smoke, eats properly, drinks in moderation (if at all) and gets regular light exercise.  Since these things can't be simulated, I submit that all pilots should be the same.  But as my "team" in the poll above numbers only in the bottom 12 percent I know that I am fighting a loosing battle.  We only know what we know.

[SCG_Tzigy]

12 minutes ago, BraveSirRobin said:

Short and stocky is better for withstanding g’s.

 

And so is untreated HTN; i bet lot would qualify here   ?

[Noisemaker]

37 minutes ago, SmokinHole said:

What is more important is that he doesn't smoke, eats properly, drinks in moderation (if at all) and gets regular light exercise.

Well that pretty much rules out most WW2 aviators. 

[unreasonable]

33 minutes ago, SmokinHole said:

Yes I know.  But that is only a benefit over a long sustained pull.  Modern jets have the power to allow for such lengthy g-loads.  For the 10 seconds that a P-51 pilot might experience under 6+ g's, keeping blood in the brain is not as much of a problem.  So being short and stocky isn't really an advantage.  What is more important is that he doesn't smoke, eats properly, drinks in moderation (if at all) and gets regular light exercise.  Since these things can't be simulated, I submit that all pilots should be the same.  But as my "team" in the poll above numbers only in the bottom 12 percent I know that I am fighting a loosing battle.  We only know what we know.

 

From my reading, and knowledge of that generation, I would guess that most WW2 pilots smoked, ate poorly (except our American friends), drank like fishes in the mess or in pubs to relieve the strain: but they did get regular light exercise. Although I am not sure that mess rugby was good for you. 

 

The game should really model a wheezing, undernourished young man with a hangover.

[US63_SpadLivesMatter]

1 hour ago, BraveSirRobin said:

Short and stocky is better for withstanding g’s.

 

Yup.  That was the joke.

[FTC_DerSheriff]

1 hour ago, unreasonable said:

 

From my reading, and knowledge of that generation, I would guess that most WW2 pilots smoked, ate poorly (except our American friends), drank like fishes in the mess or in pubs to relieve the strain: but they did get regular light exercise. Although I am not sure that mess rugby was good for you. 

 

The game should really model a wheezing, undernourished young man with a hangover.

Its basically the same question if the game should model the average build quality of aircraft. And it does not. All planes are in pristine factory condition. 

[JG7_X-Man]

1 hour ago, BraveSirRobin said:

Short and stocky is better for withstanding g’s.

 

I mentioned back awhile ago and go laughed at!

4 minutes ago, DerSheriff said:

Its basically the same question if the game should model the average build quality of aircraft. And it does not. All planes are in pristine factory condition. 

 

Yeah I am not sure we would what the developers to induce any aircraft condition but "pristine factory condition" as anything else would be subjective.

 

I remember back in the IL-2 1946 days the developers tried to factor in the Luftwaffe late war poor aircraft/pilot skill quality - that was a disaster as if the Luftwaffe was the only one with such issues at any point during the war.

[Noisemaker]

21 minutes ago, DerSheriff said:

Its basically the same question if the game should model the average build quality of aircraft. And it does not. All planes are in pristine factory condition. 

I'd be all for random failures due to wear and tear, drunk ground crews, and the like.  You already had that in one of your videos where you were missing wheel covers or something after a re-arm and repair.

Hell, in one of my SP careers, my pilot reported wounded after landing, taxiing and stopping the engine.  I mean, if my pilot can get injured shutting of the engine, then surely we can expect our airplane's repairs to be somewhat suspect as well.

I mean how real can it really be unless we see the duct tape starting to peel off the left wing and have to worry about that next turn to starboard?

[Stig]

On 9/1/2020 at 12:43 AM, =VARP=Ribbon said:

Ingame we're not newbies, by experience a much closer to aces, that would be much more realistic than newbie g-resistance.

I'm all for realism, but give us experienced pilot fatique and abilities!

 

That experience has been gained with infinite virtual lives; comparing to real life aces  that gained experience and g-resistance without havimg the option to 'refly' over and over doesn't  make much sense to me.

[JG7_X-Man]

5 minutes ago, Stig said:

 

That experience has been gained with infinite virtual lives; comparing to real life aces  that gained experience and g-resistance without havimg the option to 'refly' over and over doesn't  make much sense to me.

 

Well it is a game after all so the G tolerance should be that of an experienced pilot with a brand new aircraft. If you disagree with me - you are wrong, so don't LOL 

[Stig]

Fair enough, but then maybe we should keep the refences to what the real pilots did in their aircraft back in the real war out of the discussion. 

[CSW_Hot_Dog]

I vote for a little increase of G-tollerance. Not because of "everybody be like Ace pilot", but because everybody be intermediate pilot in combat, not relaxed proband in centrifuge. Combat means pilot full of adrenalin with heavy workload, clenched muscles and subconsciously reacting to his own feeling during pulling high G-loads and preventing himself from blacking-out. In the game right now, we have data acquired from relaxed pilot in centrifuge not controling G-Load himself and therefore resultin to weaker G-Tollerance than real pilot in warbird during fight. Furthermore behind PC we dont have such good and intuitive feedback like in real plane...

Edited September 2, 2020 by CSW_Hot_Dog

[US63_SpadLivesMatter]

"Relaxed"

[69th_Mobile_BBQ]

I voted keep it as it is.  I can hold almost any fighter at near-blackout for 20+ seconds with careful control.  Except the Tempest.  That thing onsets Gs so fast sometimes that it only takes 1 or 2 seconds to go to sleep.  

[TX-Zigrat]

I am glad that the developers are soliciting  input from the community, but I do have a little concern about response bias (https://en.wikipedia.org/wiki/Response_bias) based on the phrasing of the survey questions. I am admittedly biased myself by I think that we might have very strong responses to these three questions:

 

Q1) Do you think the addition of the G-resistance model was a big improvement for IL-2? (forecast:  survey says yes)

 

Q2) Do you think the current model  is good, but could be improved by better accounting for negative G effects and the rapid "push-pull maneuver" which is oftentimes currently used as an evasive tactic? (forecast: overwhelming yes)

 

Q3) Do you think the current model is good, but could be improved by better modeling the "functional buffer" as described by Floppy_sock on this thread? This would decrease the "rapidness" of initial GLOC onset for a "fresh" pilot but retain most of the other effects of the current model. (forecast: majority yes)

 

The concern is that due to the phrasing of the current survey, a significant majority appear to indicate that the current model is ideal and could not use some subtle refinement.. but I think some of this survey result is due to the way in which the questions were asked...

 

[Yogiflight]

11 hours ago, CSW_Hot_Dog said:

In the game right now, we have data acquired from relaxed pilot in centrifuge not controling G-Load himself and therefore resultin to weaker G-Tollerance than real pilot in warbird during fight.

No, that is not correct. In these tests the speed of the centrifuge and therefore the Gs for the pilot inside, gets increased slowly, so the pilot is very well aware of what is coming.

[Soilworker]

4 hours ago, TX-Zigrat said:

I am glad that the developers are soliciting  input from the community, but I do have a little concern about response bias (https://en.wikipedia.org/wiki/Response_bias) based on the phrasing of the survey questions. I am admittedly biased myself by I think that we might have very strong responses to these three questions:

 

Q1) Do you think the addition of the G-resistance model was a big improvement for IL-2? (forecast:  survey says yes)

 

Q2) Do you think the current model  is good, but could be improved by better accounting for negative G effects and the rapid "push-pull maneuver" which is oftentimes currently used as an evasive tactic? (forecast: overwhelming yes)

 

Q3) Do you think the current model is good, but could be improved by better modeling the "functional buffer" as described by Floppy_sock on this thread? This would decrease the "rapidness" of initial GLOC onset for a "fresh" pilot but retain most of the other effects of the current model. (forecast: majority yes)

 

The concern is that due to the phrasing of the current survey, a significant majority appear to indicate that the current model is ideal and could not use some subtle refinement.. but I think some of this survey result is due to the way in which the questions were asked...

 

 

As follow up questions I think it's a good idea. 

 

I also agree that the current questions aren't particularly well written (no offence Han), I already voted for the 1st option and only after reading through the posts did I realise that the 3rd question relates to negative Gs. 

[FTC_Zero]

It should be increased because the g.lock window is for many manouvers too compressed and gsuit of later us pilots or the unique flat leg out position of bf109 and fw190 pilots should matter compared to traditional sitting positions. That is important as well as engine performance or maschinguns and part of aircraft design quality. 

[CSW_Tommy544]

I think it's unfortunate that it is impossible to change your vote in the poll after you've voted, as there have been quite a few replies in this thread with some very good information and references to newer researches that could give another perspective to people participating in the discussion here. I believe that the 1st option in the poll would otherwise have considerably smaller lead.

[[DBS]Browning]

If someone provides good enough data, I'm sure the Devs will take note of it, regardless of the poll results.

 

[JG7_X-Man]

57 minutes ago, CSW_FMF_Tommy544 said:

I think it's unfortunate that it is impossible to change your vote in the poll after you've voted, as there have been quite a few replies in this thread with some very good information and references to newer researches that could give another perspective to people participating in the discussion here. I believe that the 1st option in the poll would otherwise have considerably smaller lead.

 

See this guy is who ALL (...me included) of us should strive to be exactly like!

• He made a decision based on the information he had at the time
• He took the time to read other opinions both for and against his initial position with an open mind
• He then switched his opinion based on facts and data he was presented

@CSW_FMF_Tommy544You are my hero of the day bro!

[busdriver]

20 hours ago, CSW_Hot_Dog said:

In the game right now, we have data acquired from relaxed pilot in centrifuge not controling G-Load himself and therefore resultin to weaker G-Tollerance than real pilot in warbird during fight.

 

For the record, while in the centrifuge the test subject (some were WSOs, GIBs, backseaters) had to hold the sidestick aft simulating pulling on the stick. If he/she let go of the stick the centrifuge would slow down. I know this, because I was one of them. I was able to spin up to 6 G without straining, but I had flown a bunch of BFM sorties in the previous 60 days.

 

For the record, test subjects were generally well rested, well fed, didn't smoke, athletically fit and were not getting shot at compared to WWII fighter pilots that smoked a lot (calms the nerves), partied hard, and got shot at on a daily basis. Contemporaneous training material for the P-51 suggests pilots avoid pulling more than 4 G to avoid GLOC.

 

[RedKestrel]

2 hours ago, busdriver said:

 

For the record, while in the centrifuge the test subject (some were WSOs, GIBs, backseaters) had to hold the sidestick aft simulating pulling on the stick. If he/she let go of the stick the centrifuge would slow down. I know this, because I was one of them. I was able to spin up to 6 G without straining, but I had flown a bunch of BFM sorties in the previous 60 days.

 

For the record, test subjects were generally well rested, well fed, didn't smoke, athletically fit and were not getting shot at compared to WWII fighter pilots that smoked a lot (calms the nerves), partied hard, and got shot at on a daily basis. Contemporaneous training material for the P-51 suggests pilots avoid pulling more than 4 G to avoid GLOC.

 

WSO is a Weapons Systems Officer...what is a GIB? All I can think of is "Guy In Back"?

[Ribbon]

On 9/2/2020 at 9:00 PM, Stig said:

 

That experience has been gained with infinite virtual lives; comparing to real life aces  that gained experience and g-resistance without havimg the option to 'refly' over and over doesn't  make much sense to me.

So we should also fly like a newbies every "refly" and not use learnt skills?

 

Not going to discuss opinions/taste/wishes!

 

S!

Edited September 3, 2020 by =VARP=Ribbon

[Livai]

On 8/31/2020 at 5:40 PM, Han said:

In general it looks like the majority of our community wants to have abilities and stamina not as regular, but as top-ace pilots. This desire is understandable.

 

Lets skip the intro - When can we start? - to have our own abilities and stamina in the game, from every player around the world..........

[Gambit21]

 

 

Quote

Anyway, thread was made to gather opinions and i posted mine, not going to arguee with bunch of Karens who take single sentence out of the context!

 

Who’s the one taking things out of context again? Might want to re-think that one.

 

Also you’re the only one throwing around pejorative terms - so yes maybe it’s best that you’re done. 

 

 

[LukeFF]

1 hour ago, RedKestrel said:

WSO is a Weapons Systems Officer...what is a GIB? All I can think of is "Guy In Back"?

 

Yep, Guy in Back - it's a highly technical USAF term. ? 

[busdriver]

2 hours ago, RedKestrel said:

WSO is a Weapons Systems Officer...what is a GIB? All I can think of is "Guy In Back"?

 

Correct 

 

Bugger... @LukeFF beat me to it. 

Edited September 3, 2020 by busdriver