Cars Ford Sierra Cosworth RS500 Pack 3.11

[Kyuubeey]

Thanks for your feedback. It's always interesting to see another point of view. Modeling suspension is not an easy thing.
With many pictures and datas, we try to reproduce the suspension geometry in the AC model.
Maybe you shoud help when you'll be back...
Concerning the engine, I have datas from an RS500 Eggenberger benchmark. I know it's quite brutal but it is what it is.

There's just some technical differences which need to be taken into account with the suspension. Maybe if I update all my things and I have energy, I can take a look at this RS500 later. For now looking at some of my trailing arm cars can give a better idea. The reason it's different is because on trailing arms the camber/toe curve is determined by the sweep angle (Angle of the links from above) and the roll center is determined differently than in strut or DWB. You should pull a line along the axis of the trailing links' geometric connection and so on. Search for "semi trailing arm roll center".

Can you show that dyno? Dynos are not to be taken as gospel because for one there's inertia present, and especially at low RPMs there's increasingly more error due to a variety of factors, on a turbo engine simply because it doesn't take lag into account properly.

The curves often also have a big rise or fall near the beginning which should not be considered as truthful, and it looks like the RS500 has that rise incorrectly modeled into the curve.

 

[=LM=908GTS]

There's just some technical differences which need to be taken into account with the suspension. Maybe if I update all my things and I have energy, I can take a look at this RS500 later. For now looking at some of my trailing arm cars can give a better idea. The reason it's different is because on trailing arms the camber/toe curve is determined by the sweep angle (Angle of the links from above) and the roll center is determined differently than in strut or DWB. You should pull a line along the axis of the trailing links' geometric connection and so on. Search for "semi trailing arm roll center".

Can you show that dyno? Dynos are not to be taken as gospel because for one there's inertia present, and especially at low RPMs there's increasingly more error due to a variety of factors, on a turbo engine simply because it doesn't take lag into account properly.

The curves often also have a big rise or fall near the beginning which should not be considered as truthful, and it looks like the RS500 has that rise incorrectly modeled into the curve.

 

[Kyuubeey]

I think it would not be unreasonable to assume that big mound there at 3.6k~ is error, although the gear is 4th gear and not a very low gear.

If you want to trust the curve very literally, smooth it out at least. Smooth out all of the curves in fact.

 

[Dr. Death]

Has the interiors, specially the steering wheel center alignment, and the driver view alignment been improved? it was really strange seeing the wheel not move centered.

 

[Akinastyle]

@shadow118

The wrongly made suspension geometry probably doesn't help at all.

The strut doesn't have the steering axis going through the lower balljoint (MacPherson requires the virtual steering axis, not the strut tube) and the rear is completely wrong. It's not a strut; it's a semi-trailing arm. Take a look at my Singer, CTR-1, E30's to see how I did it. Singer is the best of those.

Although I really suggest to move the lower links farther away from each other on the Z on the trailing arms I make! There is a physics solver error if they get too close together and are too long (= too weak). Otherwise good.

Soon I'll update the E30, it will have a decent, similar rear geometry.

Also, I don't think the unsprung on the rear is quite so high; trailing arms are heavy but not *that* heavy. Maybe -10kg or so?
Also have serious doubts about the rear ARB; I REALLY don't think you'd be reaching 30k on any typical trailing arm ARB setup. E30 DTM with Delrin (Assumed 0.9x rate at wheel, but 1.0 is probably not far off either for any "solid" bushing and links) for example came out 3k to 9k for me.

Serious doubt about the power curve on the DTM car too. Smooth it out a bit at least.

I fixed the strut and put in my E30 rear into the DTM car, and it's better. Fix geo, smooth the power curve, calculate real rollbars and then you can probably run more realistic spring rates as well, not super front split like right now.

Making tires better would help, but okay, I can forgive that. I or someone else more knowledgeable would need to help with that.

I would help with the suspension but I need a break. However doing the above should produce a good result.

We developed it by feeling and not by numbers,
it seems to me I've already had a long talk about it can but not with you.
I think that the rear axle is more like a STRUT than a DWB for this car, especially on a DTM it's very hard to model finally.
If you think you can help contact us

 

[Kyuubeey]

Alright, well, you're objectively wrong about the rear being more like a strut because semi-trailing simply doesn't work like that, and you're objectively wrong about the strut geometry because strut simply doesn't work like that. So it'll never be right with feeling; or even with numbers. Even if you have the coordinates of the rear geo you need to engineer it correctly to work.

DTM also doesn't matter; DTM allowed subframe modifications while Group A didn't which means they could potentially have changed the sweep angle or subframe height, but those are quite easy to model if you know what your curves are. Not an issue at all; I did it for the E30 already.

Maybe in the future I will. RS500's a cool car. Right now I just told you what you should do to improve the car and make it more realistic.

 

[FrankN2O]

I don't see a needle in the rev counter (DTM and Group A, don't have the others installed).
Something missing or a CSP thing ?

It has never work since the 1.2 version.

Strange. They works fine in previous v2.7:

 

[=LM=908GTS]

Strange. They works fine in previous v2.7:

View attachment 372195

We will correct that in an update, the needle is still there but it has moved during the work on the 3d, it's hidden behind. Sorry for that.

 

[Patrik Marek]

havne't got a chance to try it out yet, but your LOD files are wayyyyy too big, when exporting them through AC SDK , there is a option save car (no textures) , that's what's meant for LODs

 

[=LM=908GTS]

havne't got a chance to try it out yet, but your LOD files are wayyyyy too big, when exporting them through AC SDK , there is a option save car (no textures) , that's what's meant for LODs

Yes, I know, they are from the original mode, we will see in another update.

 

[Dirk Steffen]

Just a few issues I have seen since installing this newer version of the mod(tested only the DTM version of the car so far):
- tach needle (already mentioned above)
- engine sounds is much too low (compared to other sounds - especially environmental sounds or for example rain sounds when running wet races - normally these should be LOOOOOUD especially with the side exhaust ;-) )
- there is an LOD issue with headlights only showing in high quality LODs but vanishing in lower Q LODs (well visible in replays but also when watching other cars in rear view mirror)

I love the general feeling of the DTM car (suspension wise) - it feels VEEERY believable of what the real car could feel like (many, many, many years ago I owned and BLAAASTED a three door 2.0 LX Sierra with RS Cosworth interior pack and mods and I still remember the cars behavior on loose ground (snow, loose gravel forest service roads, wet road, ..) very very well.
This DTM version does show the character traits I remember the three door Sierra had and wish I had some garage space and be getting a 3 Door Sierra as a weekend racer track gar again - these were great cars indeed - you can keep your E30 M3s and 190s for yourself ;-)

I also have to comment that the steering feel and FFB is SUPERB on this one - big props for the guy who built the suspension physics - this is first rate stuff !

 

[Dirk Steffen]

PS - for anyone to have a quick fix for the invisible tach needle:
1) open car in Content Manager Custom Showroom
2) select the tach gauge cover glass
3) select the material editor (three dots next to the material name of the cover glass object
4) click "change values"
5) change Depth mode to "Read Only"

Now you will see the tach needle.

Now you will see that the tip of the tach needle is black instead of the correct color.
Easy fix:
1) as above but
2) select the tip of the needle
3) open the material editor of the needle tip object
4) edit the shader settings for the needle tip to reasonable values (you can look at the shader values of the needle body and use the exact same ones to perfectly match the needle tip to the body ;-)

Done

If you want to use the mod offline only and want a temp fix for the LOD headlight issue you could disable the LODs until a proper fix is in place with a future update ;-)

If you would like to have a backlit tach gauge and instrument needle as well so you can race the car in endurance events in bad lighting conditions (or just in a badly rainy weather DTM race) you could add this to the end of your lights.ini as well:


[LIGHT_3] ; tach gauge
NAME=boostlight_SUB0
COLOR=0.22,0.10,0.05

[LIGHT_4] ; tach needle
NAME=TURBO_SUB0
COLOR=0.60,0.40,0.20

[LIGHT_5] ; tach needle tip
NAME=TURBO_SUB1
COLOR=0.60,0.40,0.20


Ideally the "boostlight_SUB0" object would be split into the instrument and the instrument bezel for this to work best (currently the bezel will be lit as well, hence a very mild backlight as a compromise).

It would be nice if the other light related objects could be separated properly as well into individual objects, so the Bosch digital readout could get a nice backlight, direction lights, reverse light, fog light, etc could be made operational properly ;-)
This is especially nice when driving on online servers with full CSP functionality like the LA Canyon cruising and street racing servers which see regular people online, driving together.

 

[Kyuubeey]

^ Why owner feedback isn't very valuable most of the time.

 

[Dirk Steffen]

^ Why owner feedback isn't very valuable most of the time.

This comes over disrespectful and entirely unhelpful for readers of the forum.
You may have found ti funny in the moment when you typed it as you may know something I and others don't but getting over with sarcasm rather then sharing helpful information is not helping anyone.

I'd be curious to learn and I am sure everyone here would appreciate a helpful tone rather then snarky comments. At least I am not here to make fun of people but for sharing a great hobby with like minded people.

I know you have the skill and knowledge to share with others, why not be nice and not make others feel bad for a little chuckle you may get from a comment like this all by yourself?

 

[Kyuubeey]

This comes over disrespectful and entirely unhelpful for readers of the forum.
You may have found ti funny in the moment when you typed it as you may know something I and others don't but getting over with sarcasm rather then sharing helpful information is not helping anyone.

I'd be curious to learn and I am sure everyone here would appreciate a helpful tone rather then snarky comments. At least I am not here to make fun of people but for sharing a great hobby with like minded people.

I know you have the skill and knowledge to share with others, why not be nice and not make others feel bad for a little chuckle you may get from a comment like this all by yourself?

I already posted some stuff earlier.

I was illustrating the point that just because a driver says it's "very realistic" or "just like his car" or whatever, doesn't really mean that much. It just has to be >50% there, which is...not very helpful.

Even if the layout is completely wrong, or executed in an improper manner and the numbers match up like 30% at best, if it has two axles and four wheels with the engine in the right spot it's gonna fool a bunch of people as long as it just visually looks like the car they're supposed to be driving.

 

[AccAkut]

View attachment 372020

that's a weird curve, you should flatten the bump at ~3800rpm. It could but anything from wheel slip, clutch slip, overreacting wastegate or just imprecise software

 

[Dirk Steffen]

I already posted some stuff earlier.

I was illustrating the point that just because a driver says it's "very realistic" or "just like his car" or whatever, doesn't really mean that much. It just has to be >50% there, which is...not very helpful.

Even if the layout is completely wrong, or executed in an improper manner and the numbers match up like 30% at best, if it has two axles and four wheels with the engine in the right spot it's gonna fool a bunch of people as long as it just visually looks like the car they're supposed to be driving.

See, much nicer this way ;-)

I fully agree with your post 100 %, yet my comment above was meant as just that - a comment how a very typical behavior in driving dynamics of the three door Sierra's behavior and this trait being well mirrored in this mod.

The 3dr Sierra does have a very distinct cornering behavior - mild under-steer on corner entry (which is well masked in the mod (or even in the actual race car) due to it's racing slicks and aggressive alignment on much lowered and stiffened suspension).
It has additionally a very fragile F-R balance mid corner when on the edge of the rear tires lateral grip (VERY well mirrored in the mod) and on corner exit the rear end is very sensitive to throttle input (grounded in the Sierra's unfortunate very light rear end as of its passenger car packaging with priority on utility (large boot and an actual usable rear bench, resulting in a less ideal balance for performance driving).

The 3dr Sierra also has a VERY predictable behavior how it changes from under-steer to neutral to massive over steer.

All of this reminded me immediately on how I remember my road car behaving (which back when I was a young lad I drove haaaaaaard on all surfaces on different road and performance tires (not on slicks though).

Now of course this behavior could just be coincidental or by design - one would have to verify evidently ;-)

For now, I massively enjoy the mod as is, both as it reminds me a bit of my young years and as I very much love the motor sport history of the Sierra RS500 to bits - it is a very important car in motor sports history (which I wish Kunos had at some point included in the original AC content) but this mod does very well despite it's flaws (some of which surely can be polished).

 

[=LM=908GTS]

PS - for anyone to have a quick fix for the invisible tach needle:
1) open car in Content Manager Custom Showroom
2) select the tach gauge cover glass
3) select the material editor (three dots next to the material name of the cover glass object
4) click "change values"
5) change Depth mode to "Read Only"

Now you will see the tach needle.

Now you will see that the tip of the tach needle is black instead of the correct color.
Easy fix:
1) as above but
2) select the tip of the needle
3) open the material editor of the needle tip object
4) edit the shader settings for the needle tip to reasonable values (you can look at the shader values of the needle body and use the exact same ones to perfectly match the needle tip to the body ;-)

Done

If you want to use the mod offline only and want a temp fix for the LOD headlight issue you could disable the LODs until a proper fix is in place with a future update ;-)

If you would like to have a backlit tach gauge and instrument needle as well so you can race the car in endurance events in bad lighting conditions (or just in a badly rainy weather DTM race) you could add this to the end of your lights.ini as well:


[LIGHT_3] ; tach gauge
NAME=boostlight_SUB0
COLOR=0.22,0.10,0.05

[LIGHT_4] ; tach needle
NAME=TURBO_SUB0
COLOR=0.60,0.40,0.20

[LIGHT_5] ; tach needle tip
NAME=TURBO_SUB1
COLOR=0.60,0.40,0.20


Ideally the "boostlight_SUB0" object would be split into the instrument and the instrument bezel for this to work best (currently the bezel will be lit as well, hence a very mild backlight as a compromise).

It would be nice if the other light related objects could be separated properly as well into individual objects, so the Bosch digital readout could get a nice backlight, direction lights, reverse light, fog light, etc could be made operational properly ;-)
This is especially nice when driving on online servers with full CSP functionality like the LA Canyon cruising and street racing servers which see regular people online, driving together.

Thank you for your help, I have not looking for resolving this issue since the release.

Surely, the mod still need improvements and we will try to update it.

 

[=LM=908GTS]

that's a weird curve, you should flatten the bump at ~3800rpm. It could but anything from wheel slip, clutch slip, overreacting wastegate or just imprecise software

Yes, I have submitted this to pro and he says that it should be flatten. Thanks for your comment.

 

[=LM=908GTS]

This comes over disrespectful and entirely unhelpful for readers of the forum.
You may have found ti funny in the moment when you typed it as you may know something I and others don't but getting over with sarcasm rather then sharing helpful information is not helping anyone.

I'd be curious to learn and I am sure everyone here would appreciate a helpful tone rather then snarky comments. At least I am not here to make fun of people but for sharing a great hobby with like minded people.

I know you have the skill and knowledge to share with others, why not be nice and not make others feel bad for a little chuckle you may get from a comment like this all by yourself?

He has a lack of diplomacy but surely has a great knowledge.