10. New Status: Certified - Quality Assurance testing... - Test all other cars in set for similar improvements: ... - Test another HQ track with representative car for similar improvements: ... 9. Problem: Road feel too light Theory: Steer Force Output Max is still overpowering road bumps. Decrease Steer Force Output Max; FFB steer force output max="0.758875", 0.7296875 ^ Continued from Step #7 8. Too much dampening/friction ...not sure this is true anymore... Theory: Reducing Steer Force Output Max in previous step lowered wheel weight which emphasized dampening/friction intensity so reduce dampening/friction to compensate. FFB steer friction coefficient="0.1041675", 0, 0.05208375, 0.078125625, 0.091146563 , 0.084636094+... ^ Continued from Step #2.1 FFB steer damper coefficient="0.244125", 0, 0.1220625, 0.18309375, 0.213609375, 0.198351563+<, 0.190722656+<, 0.084636094+ ^ Continued from Step #2.1 7. Steering feels too heavy; DBR9 seems to clip ffb just briefly in high ffb scenarios; Theory: Try decreasing Steer Force Output Max to reduce steering force wheel weight. Heavy wheel force is coming from steering forces not road bumps so reduced this setting should lighten up the wheel but keep road feel. FFB steer force output max="0.93400", 0.902+<, 0.886, 0.443+>, 0.6645+>, 0.77525, 0.93400, 0, 0.467+>, 0.7005+>, 0.81725+<, 0.758875+?... ^ Continued from Step #2.1 Also going to track changes above to Dampening/Friction because they're correlated: FFB steer friction coefficient="0.1041675", 0.078125625, 0.0911465625, 0.08463609375... ^ Continued from Step #2.1 ^ Note: No +/- notation, just tracking Steer Force Output Max above FFB steer damper coefficient="0.244125", 0.18309375, 0.213609375, 0.1983515625... ^ Continued from Step #2.1 ^ Note: No +/- notation, just tracking Steer Force Output Max above ******************************************************************************** * * From Step 6 onward: Default Wheel Lock (15deg for Aston Martin DBR9) * ******************************************************************************** 6. Adjust for default wheel lock as it has a moderate but noticeable impact on ffb intensity of steering forces and road bumps; I've tuned to 22deg so far and it feels better than the defaults 15deg/18degs on most FiaGT 2005 cars. Theory: Try decreasing Force Exponent to compress forces closer to center; Higher wheel lock means less wheel travel lock-to-lock and that means steering forces are compressed into less wheel travel which makes it feel more intense; In order to lower wheel lock back to default, and so increease wheel travel lock-to-lock, we need to maintain forces near the middle but decreasing Force Exponent. FFB steer force exponent="0.84649", 0.64649+, 0.84649+ (+ at 22deg; now tune for 15deg), 0.83649+<, 0.81649+<, 0.77649+<, 0.69649?<, 0.53649->, 0.61649+>, 0.65649+>, 0.67649+>, 0.68649+>, 0.69149+>, 0.70149+?, 0.72149+?, 0.76149+<, 0.74149+>, 0.75149... ^ 0.64649: That seems to have done, maybe a smidge too tight, but it feels like 22deg before at 15deg now ^ I didn't like getting too far away from where I was because I felt the FFB response was great before this great and felt like I degraded it by such large steps. So I've started the search over. ^ 0.69649: Hint of excessive push back after oversteer snapback ^ 0.53649: Oscillation ^ 0.61649: Oscillation ^ 0.65649: Oscillation ^ 0.67649: ~ ^ 0.68649: ~ ^ 0.69149: ~ ^ 0.70149: ~ (very minor, could be okay) ^ 0.72149: ~ ^ 0.76149: No oscillation ^ 0.74149: Minor oscillation ^ 0.75149: Osccillation effectively gone Result: Default wheel lock ffb now feels pretty much like what I was tuning to, 22deg. 5. Problem: Not feeling slip angle enough; Not feeling wheel weight loss due to grip loss enough in mid corner. Theory: Try increasing Grip Weight to increase wheel weight loss during grip loss. FFB steer force grip weight="0.38227", 0.34473662109375+<, 0.325968310546875+<, 0.316584155273437+>, 0.321276232910156+>, 0.323622271728516... ^ Continued from Step #4 4. Problem: Wheel weight loss due to grip loss feels a little exaggerated. Theory: Try decreasing Grip Weight as wheel weight loss feels slightly higher than maybe it should be. This might have been impacted by the the changes with Steer Force Ouput Max FFB steer force grip weight="0.38227", 0.3072+ Result: Felt good. More natural wheel weight loss in grip loss situations and improved wheel weight for steering forces. Will keep it as-is for now until I find a situation where it's not optimal. 3. Problem: Backend feels a little loose / Catching backend slides feels a little too lively to catch. Theory: Try increasing Grip Factor towards the front now that I've been tweaking Steer Force Output Max and Damper and Friction settings. It feels a little loose on the backend maybe I can reduce that towards the idal Grip Factor 1.0 setting. FFB steer force grip factor="0.99998", 1.0+ Result: Tested dry and wet and it's hard to feel the difference but at least it feels natural so I'm going to stick with 1.0 until I find a situation where it needs to be looked at again. ******************************************************************************** * * From Step 2 onward: No Vsync/No Frame Limiter (Was 62FPS NVIDIA Frame Limiter before) * ******************************************************************************** 2. Problem: Soft-clipping during high ffb strength corners overwhelms with steering force and can't feel road bumps. 2.1 Theory 1: Steer Force Output Max (recommended 0.8 to 2.0) might be referring to steering forces as opposed to just 'all forces' so if you lessen this you might get more road feel. FFB steer force output max="1.8", 0, 0.9,1.35,1.125,1.124,1.122,1.118,1.11,1.094,1.062,0.998+<, 0.87+>, 0.934+... ^ Yes, this worked well. This is good. I can see in the AccuForce Torque Meter it's no longer pegging at the limit (which is less than max AF torque thus it's soft clipping / in-game clipping). I can now feel my way through corners. It's subtle but there. Maybe lowering even more will help. ^ Damper and Friction become much more apparent and needs to be tuned. ^ May also be affecting Grip Factor vs Grip Weight meaning Grip Weight effect feels greater when lowering Steer Force Output Max. FFB steer friction coefficient="0.13889", 0.069445+>, 0.1041675... FFB steer damper coefficient="0.3255", 0.16275+>, 0.244125... 2.1.1 Experiment: Try half of Steer Force Output Max to see if road bumps come through more clearly because I don't want to go too low and lose steering forces but I also want to know if road bumps should be coming through more pronounced... FFB steer force output max="0.934", 0.467 ^ Results: Steering forces much lower but road bumps still coming through at the same low intensity as the previous value. So, it appears we're at the mercy of track implementations for intensity of bumps through FFB or maybe there's some other setting I can find... 2.2 Theory 2: soft-clipping might be alleviated by Force Exponent; that might introduce oscillation again - may have to look at other fixes for that. ^ Probably don't need this if the first theory pans out. 1. Problem: Not feeling grip loss on corner exit Theory: Increase Grip Weight to emphasize wheel weight loss during grip loss. FFB steer force grip weight="0.30720", 0.40720+<, 0.3572+>, 0.3822+>, 0.3947+<, 0.38845+<, 0.385325+<, 0.3837625+<, 0.38298125+<, 0.382590625+<, 0.3823953125+<, 0.38229765625+<, 0.382248828125?>, 0.3822732421875+... 0. Testing Conditions: Date Started: 2018-03-08 Car Set / Mod: 2005 FiaGT Mod Track: Silverstone GP 2005 Conditions: Dry, Ambient 20C, Track 20C Car / Setup: Aston Martin DBR9, Medium Tires, Default 15deg Wheel Lock (22deg Steps 1-6) Wheel: SimXperience AccuForce V1, SimCommander 4 v4.1.0.0, Reset to Defaults, Wheel Mode High, 60% Overall Intensity, Degrees of Rotation 540, Engine RPM Disabled Graphics: Windowed (Borderless Gaming App), No Vsync/No Frame Limiter (was 62FPS NVIDIA Frame Limited Steps 1-6) Initial FFB Setup: HQ_Cars_and_Tracks_Certified_PLR.txt Mod-specific AI Tuning: 1. AI 110 carried over from DTM Classics, don't think I'll be able to catch these guys ^ Q13 +8.705s 2. AI 100: Q10 1:38:320 +3.345s 3. AI 90: Q10 1:41.656 +4.073s - Up to here was on 2005 FIA GT 2.7 Beta 2, after this was Beta 3 4. AI 80: Q1 1:41.424 +0.000s - Q1 1:41.424 +0.000s - Splits: Q3 +2.481s / Q6 +4.901s / - BAD: 0.3841s per AI 1% (vs AI 90) 5. AI 86: Q2 1:40.165 +0.087s - Q1 1:40.079 +0.000s - Splits: Q3 +0.450s / Q6 +3.1134s - BAD: AI-Time Gain per AI 1%: Top 3 -0.3385s / Top 6 -0.29793S 6. AI 90: Q3 1:40.039 +1.975s - Q1 1:38.063 +0.000 - Splits GT1: Q3 +1.975s / Q6 +3.476s - Avg AI GT1 Lap Time: 1:40.326s (Top 6 AI) 7. AI 80: Q1 1:40.650 +0.000s - Split GT1: Q3 +3.719s / Q6 +6.153s - Avg AI GT1 Lap Time 1:45.795 (Top 6 AI) - Avg AI GT1 Lap Gain per AI 1%: -0.5469s (vs Step #6) 8. AI 100: Q12 1:40.375 +6.455s - Split GT1: Q3 +0.396s / Q6 +3.836s - Avg AI GT1 Lap Time 1:35.629 (Top 6 AI) - Avg AI GT1 Lap Gain per AI 1%: -1.017s (vs Step #7) Mod Feedback 2018-03-08 for Beta 2 - SENT: - seats too high - little more road feel would be nice - Some front outers overheating (TVR) - P. Peter too fast by 10s vs anybody else (AI 110) - Can't see wheels in Saleen from cockpit - Porsche RSR outer rears overheating Mod Feedback 2018-03-17 for Beta 3 - SENT: - DBR9 front outers overheating - DBR9 backend a little loose - AI strength changed from beta 2 to beta 3: AI 100 I was +3.345s off pace, beta 3 AI 90 I'm off pace by 4.073s....? Mod Feedback 2018-03-25 for 2.7 Official - SENT: - Great work on the look and feel of the 2005 tracks. They appear like a really nice visual improvement. - Weather: I Was thinking some of these desert/tropical places like Dubai should be hotter but they all these tracks start off at 20C and rise to 30C. Don't tracks have different weather/climate variables? - This is a really handy page for temps highs and lows by month and I'm sure each location has one on wikipedia we could use for temperature in GTR2: https://en.wikipedia.org/wiki/Climate_of_Dubai - Dubai 05/Instanbul 05 track FFB intensity feels better than say Silverstone 05/Bahrain 05 but they're both smooth-ish tracks, I believe. Seems to me that these tracks could use increased FFB intensity at the track design level. ^ I bet track designers are running into the same problem I had where it doesn't feel good when the track FFB intensity is high due to Steer Force Output Max soft clipping but if that were fixed by tuning like I di then track ffb could be improved. ^ Do you contact track owners when you include it in your mods? Could you ask them about that? ^ Example: First straight in Magny-Cours 05 my car is bouncing like crazy but I'm only feel the slightest FFB of that. That's when I know the track level ffb intensity needs to go up. ^ There's a limit to how much FFB you can "bring out" of the track. Reducing Steer Force Ouput Max only lowers the steering force part of the calculation, it doesn't increase the road bumps part of the in-game FFB calculation. If it's super subtle FFB all you can do is turn up your wheel strength and I have my AccuForce (13Nm) at 100% and it's still subtle. Track designers have to bring that out more. - Dense areas of some tracks have frame rate issues even on a high spec system like mine (ie. can't keep a 60fps minimum) - View distance: I've tried increasing LOD but still see pop-in on tracks. Is this a track-level issue? - Mirrors: Mirrors don't display shadows and maybe other high quality details. Is there a fix for that or is that track or car specific? - Hard to keep tires warm at Spa 05 < 25C on mediums. Not sure if that's realistic or not. - AI Speed: Why can I out-accelerate AI on a straight (AI 92)? That implies they have less power than me. Is this an AI tuning thing? Do they have less power the lower the AI level is? - Grippiness of kerbs a tuning thing? Because they seem overly grippy at Spa 05 and Zuhai 05. - Big FPS dips near pits at Zuhai 05. FPS issues on that straight from T8-9 where you see the rear of the front straight buildings. Probably the same issue. - What frame rates do you get? What video settings do you use? I'm just wondering what I should really expect. My graphics are maxed out and I'm testing with 33 AI so I'm just not sure if that's too much for my system or not. Mod Feedback 2018-03-25 (2) for 2.7 Official - SENT: - Ferrari 360 feels very unstable compared to DBR9, C8, Viper; Kerbs really throw it; Backend feels very loose. - 360 GT/GTC feel more stable, better able to handle kerbs, backend a little bit tighet - 550 GTC feels good but suspension feels overly bouncey on relatively smooth Silverstone GP 2005 - Lotus Exige FFB feels lightish, a little too dampened for a race car - Maserati MCI2, like 550 GTC, suspension feels on the bouncey side - Maserati Trofeo Light mirror are all very dark at night, like can't really see cars behind very dark - Silverstone GP 2005 kerbs have no feeling. I expect a little bit different feel than the pavement, a little bit of bumps, but I don't feel anything. - Clipping problem at just approaching Bridge there are textures showing through the pavement as you drive over it. It's suble but happens every lap. - Bertolini seems overly quick by 1.5s compared to every other AI Mod Feedback 2018-03-31 (2) for 2.7 Official - TODO: ...