Can't Seem to Fully Disable ABS

[Griffith500]

The balance adjuster 0-5 is a pretty fine adjustment. Max rear rear brake (+5) does not mean that you have rear braking only and no front braking, it just means you are asking the rear brakes to do more stopping at +5 vs +1 or 0. Cars always have more front brake balance than rear because of dynamic weight transfer to the front tires. When you move the brake balance adjuster you are simply changing the percentage of front vs rear braking never eliminating one end of the car or the other. Let's not also forget that brake balance adjustments are largely for braking under maximum stopping. If the ABS in game works like real world cars once you start getting close to maximum stopping power the ABS is going to step in and limit the braking at whichever axle has the higher braking pressure so at that point the brake balance adjustment kind of goes out the window, I dont know for a fact but that is probably the reason that you cant run BB at +5 and use the rear tires to steer the car, ABS takes over and keeps rear brakes from locking. Add on top of that that it appears that we can not completely eliminate ABS in game regardless of our settings.

I still often find I need to be between settings. It's not as bad as having to compute ratios on the fly like with GT5 Prologue, but still not ideal.

The bias setting has little effect once ABS is enabled, for the reason you stated - this was always done per wheel, and was the reason GT understeered so badly on the brakes, since the outside front wheel had the most grip, it got the most braking force and actively yawed the car out of the corner you were trying to enter. PD seem to have introduced an adjustable limit to that effect as well, which is actually a benefit as far as driving feel goes.

If they ensure the 1:1 operation of the brake control and the braking force (at least as it was in previous games with ABS off), but allow us to see and set the actual bias setting in real percentage terms (and perhaps adjust total force for race machines), that would be fine.

 

[LeGeNd-1]

Nobody's talking about the brake fluid. It's clear it goes no further than a torque applied at the wheels. But that is determined by the car parameters, and the "bias" setting just determines how much of that figure is achievable per axle.

The main issue with the bias setting has always been that it does not have a fine enough range, because PD refuse to let us set the actual bias figure directly.

You think we think PD "modelled" brake fluid distribution? Do you think we think PD "modelled" fuel flow rate (into the engine) too?

C'mon now guys, I was being overly sarcastic in my post. I know what you're trying to say here. IRL let's say you have maximum of 100% brake force if you apply it 50:50. But if you want to move the bias to the front then you can only reduce rear brake pressure, while keeping front the same. And vice versa for moving bias to the rear. So anything than 50:50 bias, you will not use 100% of braking force, and that is what you think the red bar is representing in GT Sport. Am I correct?

Problem is, while that may be true for rear bias in GTS, using front bias in GTS allows you to reach full white bar and lock up the front wheels. So the argument that the red bar represents loss of total brake pressure is moot.

My humble opinon, either the bias setting in GTS isn't linear (if 0 is 50:50, -5 is 80:20, then +5 is 45:55 for example), or PD just doesn't want to turn off ABS entirely for rear bias to make it easier for casuals (which I think is the likely, and simplest, explanation).

Honestly, PD should just use 50:50 terminology like with every other decent sim. And while they're at it, also use actual physical values for dampers, ARB and downforce instead of arbitary integers which nobody knows what it represents.

 

[GTWelsh]

I thought it was accepted that 0 BB isn't necessarily 50/50? It's simply an adjustment from standard, which could be anything.

 

[XSquareStickIt]

Problem is, while that may be true for rear bias in GTS, using front bias in GTS allows you to reach full white bar and lock up the front wheels. So the argument that the red bar represents loss of total brake pressure is moot.

This isn't true. As already noted by DJShadeUK's post, the red bar does show up even with front brake bias.

Hence why I believe the entire bar is a representation of total braking force available, which I termed as "brake fluid in master cylinder". Of course I don't think it's fully simulated, but rather just represented.

I thought it was accepted that 0 BB isn't necessarily 50/50? It's simply an adjustment from standard, which could be anything.

I don't think BB0 is 50/50 either,, but I don't think there's a way to test for and verify this. From my personal experience, leaving the brake bias at 0 with ABS off in most cases locks the front tyres way even before the rears even start to squeal. Then there's the issue of how most cars seem to stop in shorter distances with rear brake bias (see: some of my previous posts on this thread).

I will concede that BB0 is most likely not an even 50/50 split between the front and rear, since the brakes on the front of a car are almost always larger than the ones on the rear. Should this be the case, I think BB0 would show some red bar, since an even split of the braking force would mean that the front brakes aren't utilised to their full potential. I'm thinking that BB0 is something like a 70/30 split favouring the front, with BB1 being something like 63/33 or something like that, with each increment reducing front by 10% and increasing the rear by 10%, for example, resulting in the "red bar of lost braking force". It's kinda hard to put into words, but it's all my speculation anyway.

 

[fastone371]

This isn't true. As already noted by DJShadeUK's post, the red bar does show up even with front brake bias.



Hence why I believe the entire bar is a representation of total braking force available, which I termed as "brake fluid in master cylinder". Of course I don't think it's fully simulated, but rather just represented.



I don't think BB0 is 50/50 either,, but I don't think there's a way to test for and verify this. From my personal experience, leaving the brake bias at 0 with ABS off in most cases locks the front tyres way even before the rears even start to squeal. Then there's the issue of how most cars seem to stop in shorter distances with rear brake bias (see: some of my previous posts on this thread).

I will concede that BB0 is most likely not an even 50/50 split between the front and rear, since the brakes on the front of a car are almost always larger than the ones on the rear. Should this be the case, I think BB0 would show some red bar, since an even split of the braking force would mean that the front brakes aren't utilised to their full potential. I'm thinking that BB0 is something like a 70/30 split favouring the front, with BB1 being something like 63/33 or something like that, with each increment reducing front by 10% and increasing the rear by 10%, for example, resulting in the "red bar of lost braking force". It's kinda hard to put into words, but it's all my speculation anyway.

It doesnt make what brake bias is on zero, it's just a reference, slide bar to left for more relative front brake bias and slide to right for more relative rear bias. The actual percentages dont matter, only that they make it adjustable. Same with other suspension settings, the number doesnt matter, it's only for reference, it would be nice if PD included a wider scale of parameters.

 

[LeGeNd-1]

This isn't true. As already noted by DJShadeUK's post, the red bar does show up even with front brake bias.



Hence why I believe the entire bar is a representation of total braking force available, which I termed as "brake fluid in master cylinder". Of course I don't think it's fully simulated, but rather just represented.



I don't think BB0 is 50/50 either,, but I don't think there's a way to test for and verify this. From my personal experience, leaving the brake bias at 0 with ABS off in most cases locks the front tyres way even before the rears even start to squeal. Then there's the issue of how most cars seem to stop in shorter distances with rear brake bias (see: some of my previous posts on this thread).

I will concede that BB0 is most likely not an even 50/50 split between the front and rear, since the brakes on the front of a car are almost always larger than the ones on the rear. Should this be the case, I think BB0 would show some red bar, since an even split of the braking force would mean that the front brakes aren't utilised to their full potential. I'm thinking that BB0 is something like a 70/30 split favouring the front, with BB1 being something like 63/33 or something like that, with each increment reducing front by 10% and increasing the rear by 10%, for example, resulting in the "red bar of lost braking force". It's kinda hard to put into words, but it's all my speculation anyway.

That's interesting. Never thought of testing it at standstill, thanks 👍 Some very wonky stuff going on under the hood for sure.

From the little that I've driven with ABS off though, setting front bias I always see the white bar go all the way up. While setting to rear bias the red bar is very noticeable. Hence my initial assumption. Clearly that video has proven the red bar shows up in front bias as well.

Anywhoo, who knows what's going on with PD's physics. Regardless of what the red bar represents, I still feel ABS Off is wonky in this game and unless you're a masochist everyone should just use Weak/On. I'm just sooo annoyed that they got this right in GT6 and went backwards.

 

[Griffith500]

I contend that as long as you stay away from locking up (by working to a threshold), ABS off still feels more authentic in transitions etc, since there is none of the dynamic proportioning between all four wheels that GT's "ABS" applies, always - even when set to "weak".

The way GT's ABS uses the brakes to yaw the car is pretty advanced stuff that isn't fitted to many (any?) vehicles, at least not as a performance-oriented system. ABS (and stability control, and all other variants of independent brake force redistribution) is still primarily used (and tuned) as a safety device, first and foremost. So that's definitely not more authentic.

The ABS itself is much improved in Sport in that regard, but the driving experience is still not as lively, communicative and interesting as when it is turned off. If racing "seriously", I can understand that the way the tyre model affects locking / re-gripping behaviour (in combination with any drivetrain inertia weirdness) can be a real boar without ABS. But, otherwise, as things now stand with the game, PD have done nothing spectacular to the way the braking works, once you account for the actual brake bias itself.

 

[spacecowboy73]

Anywhoo, who knows what's going on with PD's physics. Regardless of what the red bar represents, I still feel ABS Off is wonky in this game and unless you're a masochist everyone should just use Weak/On. I'm just sooo annoyed that they got this right in GT6 and went backwards.

I agree about how braking is worse in GTS than GT6 without ABS. It kills my motivation to play this game, I don't like using computer aids. I hope they fix it but the game has been out for so long at this point I doubt they will.

 

[LeGeNd-1]

I contend that as long as you stay away from locking up (by working to a threshold), ABS off still feels more authentic in transitions etc, since there is none of the dynamic proportioning between all four wheels that GT's "ABS" applies, always - even when set to "weak".

The way GT's ABS uses the brakes to yaw the car is pretty advanced stuff that isn't fitted to many (any?) vehicles, at least not as a performance-oriented system. ABS (and stability control, and all other variants of independent brake force redistribution) is still primarily used (and tuned) as a safety device, first and foremost. So that's definitely not more authentic.

The ABS itself is much improved in Sport in that regard, but the driving experience is still not as lively, communicative and interesting as when it is turned off. If racing "seriously", I can understand that the way the tyre model affects locking / re-gripping behaviour (in combination with any drivetrain inertia weirdness) can be a real boar without ABS. But, otherwise, as things now stand with the game, PD have done nothing spectacular to the way the braking works, once you account for the actual brake bias itself.

Well the problem is finding the locking point with ABS Off in GTS is like playing roulette. It's just luck. You can't build up to it, and you can't do it consistently unless you're really really braking very conservatively, to the point that you'll be uncompetitive. I can set fairly similar laptimes in GT6 and Assetto Corsa driving with ABS On/Off, and tweaking the brake balance properly. In GTS ABS Off is just unusable in competitive situations.

I agree that ABS in GT also includes intrinsic EBD, but the effect in GTS is much less than in previous games. With ABS Weak the car does feel more dynamic (well, good enough for a console "sim" in my book). Obviously I'm not expecting Assetto levels of simulation in the braking system, all I ask is to revert back to GT6 ABS Off physics. That feels right, this doesn't. And for me, the feeling of what's "right" is what matters.

 

[Griffith500]

Well the problem is finding the locking point with ABS Off in GTS is like playing roulette. It's just luck. You can't build up to it, and you can't do it consistently unless you're really really braking very conservatively, to the point that you'll be uncompetitive. I can set fairly similar laptimes in GT6 and Assetto Corsa driving with ABS On/Off, and tweaking the brake balance properly. In GTS ABS Off is just unusable in competitive situations.

I agree that ABS in GT also includes intrinsic EBD, but the effect in GTS is much less than in previous games. With ABS Weak the car does feel more dynamic (well, good enough for a console "sim" in my book). Obviously I'm not expecting Assetto levels of simulation in the braking system, all I ask is to revert back to GT6 ABS Off physics. That feels right, this doesn't. And for me, the feeling of what's "right" is what matters.

Well that's exactly what I meant by driving to a threshold, and is why I said it's only workable if you're not racing "seriously". Others will be able to hit it more frequently; I don't have that much trouble, even on a controller, but then I don't race so my tolerance is likely to be higher. I do agree it felt better in GT6, despite the bias setting complexity / annoyance.


But I maintain this is an issue with the tyre model (and drivetrain model, including non-driven wheels!), rather than the brakes. PD have a habit of "fixing" the thing we notice a problem in, rather than the thing that causes the problem - i.e. covering up the effect, rather than eradicating the cause. The drivetrain model is one of the weakest links in the series at the moment, its effects are pervasive, pernicious and subtle, but it all adds up across several aspects.

The weirdness built into the tyre model is the best example of the drivetrain's insidious impact; the tyre model has slip in places that slip should not be occurring and it struggles against high inertia and damping in the drivetrain, both of which are necessary because the drivetrain model is fundamentally unstable.

 

[LeGeNd-1]

Well that's exactly what I meant by driving to a threshold, and is why I said it's only workable if you're not racing "seriously". Others will be able to hit it more frequently; I don't have that much trouble, even on a controller, but then I don't race so my tolerance is likely to be higher. I do agree it felt better in GT6, despite the bias setting complexity / annoyance.


But I maintain this is an issue with the tyre model (and drivetrain model, including non-driven wheels!), rather than the brakes. PD have a habit of "fixing" the thing we notice a problem in, rather than the thing that causes the problem - i.e. covering up the effect, rather than eradicating the cause. The drivetrain model is one of the weakest links in the series at the moment, its effects are pervasive, pernicious and subtle, but it all adds up across several aspects.

The weirdness built into the tyre model is the best example of the drivetrain's insidious impact; the tyre model has slip in places that slip should not be occurring and it struggles against high inertia and damping in the drivetrain, both of which are necessary because the drivetrain model is fundamentally unstable.

Yes, sorry, for some reason I skipped the last paragraph of your last post. I think we are on the same page in that regard.

Interesting take on the drivetrain issue, I've not heard of that theory before, could you care to elaborate? I know PD's physics is wonky in some areas. The infamous backwards aero and ride height in GT5 leading to the wheelie glitch. The overdamped suspension model until GTS. Relatively poor longitudinal tyre grip compared to lateral grip. Heck, until GTS I don't think toe/camber settings even have a consistent effect. And let's not get me started on how no one can agree what LSD Initial does. Generally PD physics works 90% of the time. It's when you start pushing hard or trying to correlate with real world values then everything breaks down. And I agree, even after 20 years and multiple major revisions, PD is still trying to play band aid when it comes to fixing problems with their core physics instead of rebuilding everything properly. At the same time though, despite the evolution throughout the years I always feel a common thread all the way from GT1 to GTS. It's a nice comfortable feeling if you've been with the series for so long, I've learnt to put up with the deficiencies. I have many other sims if I want better physics accuracy.

 

[Griffith500]

I can't be specific, in terms of the exact mathematical approach used, only my interpretation of the externally visible effects.

And it's more of what you describe, in my opinion: legacy approach being held over past its best usefulness. I expect there is little wrong with its overall behaviour, but it's when you start to look at special cases and the finer, more subtle behaviours (where sims are meant to shine) that it falls apart.

So that's things like the unstable model (rev limit bounce is the clearest indication of that), which has to be overdamped accordingly. In older games, as recent as GT6, a lot of the concept cars would shudder and warble because of the instability in its non-physically derived drivetrain parameters (they're just guesses, the car doesn't exist; suggesting the model can't fare well outside of a narrow performance window).

The over damping probably affects things like the LSD, particularly how it should affect lift off oversteer and so on, since, although engine braking clearly works in the game, the more subtle and varied effects it is supposed to cause don't materialise in the expected way. The overdamping and generally high inertia means that traction is initially difficult to break, but when it does, it's harder to get back, again affecting the LSD. iRacing's Dave Kaemmer once spoke of the issues they ran into with stability and damping when they introduced torsion to their already fully modular drivetrain model. The two outputs of the LSD would fight each other and oscillate continuously, very interesting.

In that respect, the individual drivetrain components are not fully modeled in GT as to their inertia, which also affects LSD operation, and probably also braking / locking behaviour. This bulk inertia was certainly true of GT6, although I think they had added independent wheel (rotational) inertia by then - but the model was not initially conceived to work in this way, so it would not surprise me if some of the weirdness comes from an imperfect coupling of the two distinct inertia models. Ironically, it might actually have been the suspension model improvements that indirectly helped brake simulation here.

A final note would cover things like the gear changes, instantaeous rpm speed changes with no effect at the contact patches (suggesting energy is not conserved; might explain the insistence on rev matching) and very, very slow revving smaller displacement engines (again, possibly a stability concern). Some of this might play into the clutch problems as well in subtle ways.

 

[LeGeNd-1]

I think I get the gist of what you're trying to say, but being no engineer a lot of that goes way above my head

Regarding the instantaneous rpm change with shifting, I notice they have introduced drivetrain wobble in GTS. The tach doesn't just drop instantly to an exact value and "hovers" a bit after an upshift/downshift. I like it. Though whether this is just an added "effect" or whether it's actually simulated in the drivetrain I don't know.

Agree also that smaller displacement engines are terrible to drive. Lightweight cars with bike engines (e.g. Caterham, Suzuki GSXR4) are particularly notable for this. Electric cars are also never quite properly simulated, especially with launching from standstill. Diesel torque is also not apparent - if you compare the BMW 120i and 120d in GT5/6 you'll know what I mean.

Clutch simulation, don't get me started. It's basically an on/off switch and until they fix this I will not bother with clutch in GT.

 

[Griffith500]

Oh I forgot they added that annoying rev rise during gear changes. It's not "wobble" (flex), for sure. GT5 had a "wobble" effect that only played during replays. This is at least a change to the scripted controls, but it's still just an effect really and masks the underlying problems.

The clutch has always been fine, it's again how it's implemented: specifically they added a requirement to have throttle fully off before a gear is allowed to be selected and there was even a timer on second gear in earlier games. Clutch slip works really nicely on its own.


This all hints that the core of the simulation needs work before the other stuff, such as the ABS' behaviour, will make a sensible difference if improved.

 

[LeGeNd-1]

I did try slipping the clutch back in GT5P. As you say there is slip from a standstill, but when you're driving at speed the requirement to be fully off the gas to engage a gear essentially makes it an on/off switch. Very unnatural and it makes heel and toeing a huge pain. As such I never bothered to use it again. Compare that to Assetto for example, where things are simulated properly and driving old cars with clutch and h-shifter is a joy.

 

[Greycap]

The thing is, PD can do the clutch right. In the earliest versions of GT5P the requirement for zero throttle wasn't there and it was very easy to use but being able to power shift gave such an advantage to clutch users that the outcry from people not having such equipment forced a change.

 

[LeGeNd-1]

The thing is, PD can do the clutch right. In the earliest versions of GT5P the requirement for zero throttle wasn't there and it was very easy to use but being able to power shift gave such an advantage to clutch users that the outcry from people not having such equipment forced a change.

I only got GT5P towards the tail end of its life (late 2008), so I missed that whole part. IMO, if PD models engine/transmission damage it wouldn't be a problem since flatshifting would just detonate the engine and/or destroy the clutch after a few laps. Secondly, driving with a clutch and h-shifter is much harder coordination wise. If you can do that and still set top world wide times, chances are you will be at the top no matter what, so I don't mind the advantage they get. I'd much prefer not having the current awkward throttle off requirement.