Jump to content

What is the state of the FM as of today?


bkthunder

Recommended Posts

Just reading the forums and watching the changelogs, I can't really get a clear picture.

On release it was widespread opinion (confirmed by SMEs etc) that the FM was basically spot on.

Now it seems a lot of patches / changes have been made. So, as of this post, is it more realistic / imporved compared to release day or is it in a state of flux? WHy have so many changes become necessary if the FM was so good upon release?

 

I don't fly constantly enough to notice all the minor changes when playing...

 

Thanks.

Windows 10 - Intel i7 7700K 4.2 Ghz (no OC) - Asus Strix GTX 1080 8Gb - 16GB DDR4 (3000 MHz) - SSD 500GB + WD Black FZEX 1TB 6Gb/s

Link to comment
Share on other sites

  • Replies 108
  • Created
  • Last Reply

Top Posters In This Topic

There's an ongoing thing that changed making her draggy or heavier than normal that came with... dont know, few patches back. Without speculating, it's not as good as it was. Best not to attempt more detail than that because there's some assumptions gonna get folded in A current hypothesis seems to be the weight and drag doesn't reset from launch weight and drag.

___________________________________________________________________________

SIMPLE SCENERY SAVING * SIMPLE GROUP SAVING * SIMPLE STATIC SAVING *

Link to comment
Share on other sites

The FM is in an overall good spot, with the exception of a few areas that are currently being worked on. We stand by what was said at release, but please do not forget that it is also early access and we are still fine tuning based on SME feedback. The entire basis of the FM comes from very accurate wind tunnel data that got us 90% of the way there, but the Tomcat is a very complex FM that still requires a large amount of work to get right, especially in the handling qualities area to which SME feedback is critical.

 

We use an FM testing system that resides entirely outside of DCS. We use this testing system to give us feedback on changes to the FM using tests derived from the F-14A+ performance manual. We like to test as close to a “clean” aircraft as possible since we don’t have much control over stores drag in DCS. The manual’s lowest drag configuration is a 4xAIM7, 4xAIM9, no tanks loadout. For these tests we use hardcoded stores drag data that was provided to us by ED during initial FM development. Additionally, these tests are mostly static tests or tests related to aspects of aircraft performance in specific areas of the flight envelope. These tests run every time we make a commit, and the results have been largely unchanged since EA release with the exception of a few which I will address for transparency’s sake.

 

At release we felt (and tests indicated) that we were slightly overpowered in level flight acceleration. In response to this we made even more FM tests against specific excess power across the entire maneuvering envelope. These new tests indicated we were massively overpowered, so I reduced engine thrust operating on the assumption that the drag data from the wind tunnel was correct (we still believe this). The result was an underpowered engine for 1-2 patches while we tried to figure out what was wrong with the excess power tests and why they disagreed with the level flight acceleration tests (both of these sets of tests essentially test the same thing in different ways, and we cross-check them with in-game handflown testing, too). I ultimately ignored the excess power tests and started the development of a new more advanced testing system which I am still working on, which brings us to now. The FM performance is more or less back where it was at EA release according to our current FM test system.

 

Unfortunately, the entire FM’s foundation actually rests upon the shifting sands that are DCS, not our external FM test system.

 

The interface between our FM code and DCS physics is primarily only forces and moments. DCS handles the equations of motion as well as a few other significant areas, including:

 

  • Aircraft gross weight and mass properties (we only have control over empty weight). An incorrect gross weight calculation will seriously affect performance. DCS handles all aircraft mass properties beyond just the empty weight.
  • External forces and moments. Think the forces and moments generated by interactions with the ground via landing gear or airframe contact, or even contact with other objects. All of the non-aerodynamic forces and moments applied to the aircraft.
  • Stores drag. DCS provides the forces and moments generated by stores drag. We have little insight or control over these forces and moments.
  • Atmosphere model. DCS provides temperatures, pressures, and density at the ownship used for all aerodynamic calculations.

 

This ultimately means that if any of the above areas have changed, these changes are no longer reflected in our FM testing environment. If something on the DCS side has changed that now affects the Tomcat’s performance in-game, we have no real way of knowing short of ED telling us what changed or catching it while hand-testing in game. We suspect some things have changed and other things may be inaccurate, but we’re not sure what. We’re working with ED to try and solve these issues.

 

For people who perceive an in-game FM performance loss, our FM tests have not indicated any loss in performance over the past month. If anything, the Tomcat's performance has only been increased in the past 4-6 weeks based on intentional FM changes I’ve made that the testing reflects.

 

Future work

 

There are a few outstanding FM items I want to address, beyond normal maintenance that will pop up:

 

  • Subjective handling qualities:

    Flaps, Gear, Speedbrake pitch moment and drag effects. Working to better tune these based on SME feedback.

    Misc handling accuracy improvements, all minor and on the edge of the envelope.


  • Performance:

    Tune for specific excess power across the whole envelope. It’s close, but definitely needs more investigation and better testing.

    Tune performance of slats and flaps while using maneuver flaps.


 

As you can see this list is not very long, but many items are just time consuming.

Tuning the handling qualities is very time consuming, as it has to be done by hand-flying in DCS and is hand-verified by SMEs since no handling qualities tests are available for the Tomcat. Some aspects of performance are more easily verified by automated tests and tuning can be done more quickly, assuming the tests are good.

 

Once we think the FM is as “done” as it can possibly be, I’ll make an FM status post.


Edited by fat creason

Systems Engineer & FM Modeler

Heatblur Simulations

Link to comment
Share on other sites

Thanks for the transparency.As an avid simmer and aeronautical engineering myself I am very pleased you are looking at this simulation with such minutiae.

 

It is too bad you can't create 'ingame' automated tests including the overall simulation. It must be very frustrating to understand what changed on the parts you have no control on.

 

Br

 

Sent from my VTR-L09 using Tapatalk

Link to comment
Share on other sites

Snip: Amazing info.

 

This is the part about DCS and all of its modules that nurtures my apprehension about 3rd party modules. Of course I have the Tomcat because... well... duh but this is the same issue modders and scripters have trying to keep up with subtle changes.

 

You guys (3rd party devs) smash modules out of the ballpark in terms of quality and then ED changes "something" and now you've literally compromised an entire flight model- at least insofar as its relationships with the core program. And there's literally nothing "proactive" that you can do about it. ED does their thing, whatever that is- changing an 0 to an O for example in some code somewhere, and then you guys are left scrambling trying to clean up your own module.

 

I'd expect better coordination between the 3rd party developers and the developers but apparently that's impossible.

"ENO"

Type in anger and you will make the greatest post you will ever regret.

 

"Sweetest's" Military Aviation Art

Link to comment
Share on other sites

Watching a racehorse fed on hay is a sad thing.

HB, you did and do a great job. I feel for you guys. I know this is a super complex sim, even before we get to 3rd party and ED communication topics, but I can't help but feel you could have gone all the way with enough time and just made a sim. You deserved that kind of dream.

ED has to look at the core game and it's partner relations and communications very intently and ask if this is a sustainable foundation for brilliant work, or the equivalent of building a skyscraper on a bedrock of mud and jelly.

___________________________________________________________________________

SIMPLE SCENERY SAVING * SIMPLE GROUP SAVING * SIMPLE STATIC SAVING *

Link to comment
Share on other sites

The FM is in an overall good spot, with the exception of a few areas that are currently being worked on. We stand by what was said at release, but please do not forget that it is also early access and we are still fine tuning based on SME feedback. The entire basis of the FM comes from very accurate wind tunnel data that got us 90% of the way there, but the Tomcat is a very complex FM that still requires a large amount of work to get right, especially in the handling qualities area to which SME feedback is critical.

 

We use an FM testing system that resides entirely outside of DCS. We use this testing system to give us feedback on changes to the FM using tests derived from the F-14A+ performance manual. We like to test as close to a “clean” aircraft as possible since we don’t have much control over stores drag in DCS. The manual’s lowest drag configuration is a 4xAIM7, 4xAIM9, no tanks loadout. For these tests we use hardcoded stores drag data that was provided to us by ED during initial FM development. Additionally, these tests are mostly static tests or tests related to aspects of aircraft performance in specific areas of the flight envelope. These tests run every time we make a commit, and the results have been largely unchanged since EA release with the exception of a few which I will address for transparency’s sake.

 

At release we felt (and tests indicated) that we were slightly overpowered in level flight acceleration. In response to this we made even more FM tests against specific excess power across the entire maneuvering envelope. These new tests indicated we were massively overpowered, so I reduced engine thrust operating on the assumption that the drag data from the wind tunnel was correct (we still believe this). The result was an underpowered engine for 1-2 patches while we tried to figure out what was wrong with the excess power tests and why they disagreed with the level flight acceleration tests (both of these sets of tests essentially test the same thing in different ways, and we cross-check them with in-game handflown testing, too). I ultimately ignored the excess power tests and started the development of a new more advanced testing system which I am still working on, which brings us to now. The FM performance is more or less back where it was at EA release according to our current FM test system.

 

Unfortunately, the entire FM’s foundation actually rests upon the shifting sands that are DCS, not our external FM test system.

 

The interface between our FM code and DCS physics is primarily only forces and moments. DCS handles the equations of motion as well as a few other significant areas, including:

 

  • Aircraft gross weight and mass properties (we only have control over empty weight). An incorrect gross weight calculation will seriously affect performance. DCS handles all aircraft mass properties beyond just the empty weight.
  • External forces and moments. Think the forces and moments generated by interactions with the ground via landing gear or airframe contact, or even contact with other objects. All of the non-aerodynamic forces and moments applied to the aircraft.
  • Stores drag. DCS provides the forces and moments generated by stores drag. We have little insight or control over these forces and moments.
  • Atmosphere model. DCS provides temperatures, pressures, and density at the ownship used for all aerodynamic calculations.

 

This ultimately means that if any of the above areas have changed, these changes are no longer reflected in our FM testing environment. If something on the DCS side has changed that now affects the Tomcat’s performance in-game, we have no real way of knowing short of ED telling us what changed or catching it while hand-testing in game. We suspect some things have changed and other things may be inaccurate, but we’re not sure what. We’re working with ED to try and solve these issues.

 

For people who perceive an in-game FM performance loss, our FM tests have not indicated any loss in performance over the past month. If anything, the Tomcat's performance has only been increased in the past 4-6 weeks based on intentional FM changes I’ve made that the testing reflects.

 

Future work

 

There are a few outstanding FM items I want to address, beyond normal maintenance that will pop up:

 

  • Subjective handling qualities:

    Flaps, Gear, Speedbrake pitch moment and drag effects. Working to better tune these based on SME feedback.

    Misc handling accuracy improvements, all minor and on the edge of the envelope.


  • Performance:

    Tune for specific excess power across the whole envelope. It’s close, but definitely needs more investigation and better testing.

    Tune performance of slats and flaps while using maneuver flaps.


 

As you can see this list is not very long, but many items are just time consuming.

Tuning the handling qualities is very time consuming, as it has to be done by hand-flying in DCS and is hand-verified by SMEs since no handling qualities tests are available for the Tomcat. Some aspects of performance are more easily verified by automated tests and tuning can be done more quickly, assuming the tests are good.

 

Once we think the FM is as “done” as it can possibly be, I’ll make an FM status post.

 

Thanks, I was afraid this thread would turn into the ususal 50 pages of opinions with no facts, but you have made a perfect post clearly stating what the situation is, you are true professionals and I hope you keep up this kind of open interaction with your customers :thumbup:

 

This is the part about DCS and all of its modules that nurtures my apprehension about 3rd party modules. Of course I have the Tomcat because... well... duh but this is the same issue modders and scripters have trying to keep up with subtle changes.

 

You guys (3rd party devs) smash modules out of the ballpark in terms of quality and then ED changes "something" and now you've literally compromised an entire flight model- at least insofar as its relationships with the core program. And there's literally nothing "proactive" that you can do about it. ED does their thing, whatever that is- changing an 0 to an O for example in some code somewhere, and then you guys are left scrambling trying to clean up your own module.

 

I'd expect better coordination between the 3rd party developers and the developers but apparently that's impossible.

 

Fully agree with this and other posts above, I feel uneasy about 3rd parties and the 3rd party modules I have invested in, knowing how deeply dependent you are from a notoriously unreliable "partner" such as ED (when it comes to effective communication, fixing bugs and sticking to timelines). You guys have out-done anyone else on all levels when it comes to the F-14: graphics, FM, sense of immerison, attention to detail etc. Including ED themselves.

This module is a work of true passion and I think we can all feel it.

I am looking forward to the day you will launch your own flight simulator, and in the mean time I hope ED has the ability and vision to proactively work with you (and other 3rd parties) to solve the issues they generate.

Windows 10 - Intel i7 7700K 4.2 Ghz (no OC) - Asus Strix GTX 1080 8Gb - 16GB DDR4 (3000 MHz) - SSD 500GB + WD Black FZEX 1TB 6Gb/s

Link to comment
Share on other sites

Hi guys, I just wanted to kindly ask you to please not point fingers too much and stay on topic with the FM. :) Let us not forget that DCS is a living environment, ever evolving and thus ever changing. This is of course necessary and naturally it will mean that parts of the sim need ongoing revision, be it by ED, 3rd party or even modders. In fact, this is really something that is quite normal and expected. Fat creason's post was meant to give you insight in how and why we have to go about the FM, but it was not meant to blame anyone or anything. Quite the opposite, we believe the insight that we can give you, can lead to a better mutual understanding of the ongoing processes and necessities in the background. It is meant to bring us all closer together.

 

In a way we all have to manage our expectations - but if you take a step back and look at it from a distance, we have come really far all together the past decade and more. Things are possible today that were unthinkable even a few years back. Some of us have been flying it for 16 years now, the Flanker 2.0 generation even longer. That is 2 decades of fun in continuity - I believe there is few sims and games that can match that. In the end, this has been possible because of all your amazing and continued support. Thank you! :thumbup:


Edited by IronMike

Heatblur Simulations

 

Please feel free to contact me anytime, either via PM here, on the forums, or via email through the contact form on our homepage.

 

http://www.heatblur.com/

 

https://www.facebook.com/heatblur/

Link to comment
Share on other sites

I'm confused. If your power excess numbers are off then how are your STR numbers?

 

Where are your level flight numbers exactly off? Sorry for the questions I'm just curious that's all. Is it in the transonic, supersonic range or subsonic?

 

 

Does DCS model ground effect? Speaking of atmosphere, what data is DCS using? The one from NASA about a decade ago?

 

Sorry again for the curiosity, having done flight simulations for the DOD understand what you're going through. I still remember one project we're at 30 days to fix 60 pages of issues with the product. In 28 days, I dropped it down to a half a page and it wasn't my area. The lead flight physics from Lockheed was overwhelmed and I was hired to fix it all. And the rest is history.... I did all the F14 variants.

 

Sent from my SM-N950U using Tapatalk

Link to comment
Share on other sites

I'm confused. If your power excess numbers are off then how are your STR numbers?

 

Where are your level flight numbers exactly off? Sorry for the questions I'm just curious that's all. Is it in the transonic, supersonic range or subsonic?

 

 

Does DCS model ground effect? Speaking of atmosphere, what data is DCS using? The one from NASA about a decade ago?

 

Sorry again for the curiosity, having done flight simulations for the DOD understand what you're going through. I still remember one project we're at 30 days to fix 60 pages of issues with the product. In 28 days, I dropped it down to a half a page and it wasn't my area. The lead flight physics from Lockheed was overwhelmed and I was hired to fix it all. And the rest is history.... I did all the F14 variants.

 

Sent from my SM-N950U using Tapatalk

 

 

Confirm DCS does model ground effect. Best way to see this is in the helos, esp the Huey.

 

Fat Creason. Thanks for the insight into your day to day world. You have brought up issues that likely no one on here would have ever thought of. Weapons drag being controlled by DCS for example, however I suppose that is fair, it levels the playing field for everyone, no matter the aircraft.

 

Just goes to show how amazingly complex, detailed and accurate all the devs are striving to make it.


Edited by Tinkickef

System spec: i9 9900K, Gigabyte Aorus Z390 Ultra motherboard, 32Gb Corsair Vengeance DDR4 3200 RAM, Corsair M.2 NVMe 1Tb Boot SSD. Seagate 1Tb Hybrid mass storage SSD. ASUS RTX2080TI Dual OC, Thermaltake Flo Riing 360mm water pumper, EVGA 850G3 PSU. HP Reverb, TM Warthog, Crosswind pedals, Buttkicker Gamer 2.

Link to comment
Share on other sites

I believe ground effect is modeled per module, not in the DCS environment as a whole. Prime example being the Hornet's inverted ground effect that took nearly a year for ED to officially acknowledge. I believe the Yak-52 is also missing ground effect. The helos (especially the Hip) exhibit ground effect very noticeably. The warbirds also to some extent.

Link to comment
Share on other sites

I was flying it the other day, and it felt very much like the first version that was released, albeit they now seem to be achieving it from the right performance modifiers. I do notice the controls seem a LOT lighter and more responsive than they were.

 

 

Its just generally a very nice kite to throw around now

Link to comment
Share on other sites

Ground effect and any other aerodynamic phenomenon is entirely up to the FM engineer to add. DCS does not inherently model any effects other than the ones listed in my previous post.

 

Where are your level flight numbers exactly off? Sorry for the questions I'm just curious that's all. Is it in the transonic, supersonic range or subsonic?

 

Our level flight acceleration is very close currently from 0-35,000 ft according to our internal FM test system. In-game, I'm not sure anymore.


Edited by fat creason

Systems Engineer & FM Modeler

Heatblur Simulations

Link to comment
Share on other sites

Hi guys, I just wanted to kindly ask you to please not point fingers too much and stay on topic with the FM. :) Let us not forget that DCS is a living environment, ever evolving and thus ever changing. This is of course necessary and naturally it will mean that parts of the sim need ongoing revision, be it by ED, 3rd party or even modders. In fact, this is really something that is quite normal and expected. Fat creason's post was meant to give you insight in how and why we have to go about the FM, but it was not meant to blame anyone or anything. Quite the opposite, we believe the insight that we can give you, can lead to a better mutual understanding of the ongoing processes and necessities in the background. It is meant to bring us all closer together.

 

In a way we all have to manage our expectations - but if you take a step back and look at it from a distance, we have come really far all together the past decade and more. Things are possible today that were unthinkable even a few years back. Some of us have been flying it for 16 years now, the Flanker 2.0 generation even longer. That is 2 decades of fun in continuity - I believe there is few sims and games that can match that. In the end, this has been possible because of all your amazing and continued support. Thank you! :thumbup:

 

Great work to the Heatblur team, and the information is very much appreciated. I still believe this is IMO the highest quality module I own!

Link to comment
Share on other sites

Great work to the Heatblur team, and the information is very much appreciated. I still believe this is IMO the highest quality module I own!

 

I mirror the opinion, the F-14 module is first class and so is the communication with the community in regards to addressing any bugs or performance discrepancies reported.

Link to comment
Share on other sites

I mirror the opinion, the F-14 module is first class and so is the communication with the community in regards to addressing any bugs or performance discrepancies reported.

 

Yeah, the team at ED itself should learn a thing or two from HB, especially when it comes to keeping the community informed and engaged

[sIGPIC][/sIGPIC]

Simming since 2005

My Rig: Gigabyte X470 Aorus Ultra Gaming, AMD Ryzen7 2700X, G.Skill RipJaws 32GB DDR4-3200, EVGA RTX 2070 Super Black Gaming, Corsair HX850

Link to comment
Share on other sites

fat creason, thank you for addressing the topic directly! I would like to ask a general question regarding the FM and my approach to learning high AOA / slow speed maneuvering:

 

- Since a few updates ago, I find the plane takes too much time to slow down when reducing power at slow speeds (trapping for instance). Full brakes out, gear down, throttle idle, it takes a lot of time to slow downwind

 

- Also at slow speed the increases/decreases in lift generated by changes in the wing geometry (flaps and man. flaps) seems exaggerated, sometimes I will float up like a balloon or drop like a stone very suddenly

 

- I used to think the ground effect was too pronounced, around 200-300 ft from the sea coming in at high AOA I would get pushed upwards and had a hard time following the glidescope, this seems to have been fixed now.

 

 

Some pilots (Victory205 for example) have thrown the towel even recommending we shouldn't spend anytime practicing case I until the FM is stabilized....

 

Can you confirm if this is mostly 'in my head' or if the current FM changes heavily affect theses high AOA / slow speeds situations?


Edited by METEOP
Wrong Member adressed

METEOP

 

i5-6600K OC@4.5Ghz, GTX 1070 OC, 32Gb RAM, M.2 NVMe SSD

Warthog HOTAS, Saitek Rudder Pro, Trackhat Clip, 1080p projector, Custom touchscreen rig, Ikarus touchscreen panel, Voice Attack, ReShade, Simshaker Aviator

Link to comment
Share on other sites

 

- Since a few updates ago, I find the plane takes too much time to slow down when reducing power at slow speeds (trapping for instance). Full brakes out, gear down, throttle idle, it takes a lot of time to slow downwind

 

- Also at slow speed the increases/decreases in lift generated by changes in the wing geometry (flaps and man. flaps) seems exaggerated, sometimes I will float up like a balloon or drop like a stone very suddenly

Can you confirm if this is mostly 'in my head' or if the current FM changes heavily affect theses high AOA / slow speeds situations?

 

Was actually just wondering this myself. Haven’t been trapping for quite a while now and when I took a few runs today I was wondering just this. One example, doing the turn, gear and flaps down, added some power not to drop and when I came out of the turn, I reduced power like usual but I felt like a balloon just going up. My speed was under 140-45 and on AoA before and during the turn.

 

Its probably just me forgetting how to trap properly but it felt different from what I remember

Link to comment
Share on other sites

Was actually just wondering this myself. Haven’t been trapping for quite a while now and when I took a few runs today I was wondering just this. One example, doing the turn, gear and flaps down, added some power not to drop and when I came out of the turn, I reduced power like usual but I felt like a balloon just going up. My speed was under 140-45 and on AoA before and during the turn.

 

Its probably just me forgetting how to trap properly but it felt different from what I remember

 

Yes one of those instances where it seems out of whack is when you are leveling from a high aoa turn the lift picks up like crazy.

 

Also going beyond the 22.5 deg. roll (I believe) mark and lift just ceases you drop like the proverbial rock.

METEOP

 

i5-6600K OC@4.5Ghz, GTX 1070 OC, 32Gb RAM, M.2 NVMe SSD

Warthog HOTAS, Saitek Rudder Pro, Trackhat Clip, 1080p projector, Custom touchscreen rig, Ikarus touchscreen panel, Voice Attack, ReShade, Simshaker Aviator

Link to comment
Share on other sites

Future work

How is the carrier-traping FM coming along?

Right now hook-skipping isn't modeled & you can trap on crazy fast/slow AoA's.

Compered to F-18, in the F-14 it's a child's play to catch a wire.

i7 8700k@4.7, 1080ti, DDR4 32GB, 2x SSD , HD 2TB, W10, ASUS 27", TrackIr5, TMWH, X-56, GProR.

Link to comment
Share on other sites

This thread brings me to the question if there's a document that explains what parts of the sim (esp. the physical variables) is controlled by the DCS enviroment and what is controlled by the developer... and how they interact with each other...

--- blue plasma ---

Link to comment
Share on other sites

This thread brings me to the question if there's a document that explains what parts of the sim (esp. the physical variables) is controlled by the DCS enviroment and what is controlled by the developer... and how they interact with each other...

 

 

 

 

Check Fat Creason's reply earlier, it lists exactly what is and what isn't. going into single variables is neither feasable, nor would it tell you much imo, pulled out of context. it is also something we cannot share, sorry.

Heatblur Simulations

 

Please feel free to contact me anytime, either via PM here, on the forums, or via email through the contact form on our homepage.

 

http://www.heatblur.com/

 

https://www.facebook.com/heatblur/

Link to comment
Share on other sites

How is the carrier-traping FM coming along?

Right now hook-skipping isn't modeled & you can trap on crazy fast/slow AoA's.

Compered to F-18, in the F-14 it's a child's play to catch a wire.

 

Our SME has brought up this issue as well, we intend to address it in the near future. Currently trying to finalize handling qualities around the boat, this is a slow process that requires lots of hand-flying by our SME to assess.


Edited by fat creason

Systems Engineer & FM Modeler

Heatblur Simulations

Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...