Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Riso posted:

Edit: and they're gone again because there's a problem with the cooling figures that affect reliability of the engine.

Ask me how I found that out :v: Cooling provided vs. cooling needed was something I had to enter in manually - there is no neat solution with the .luas to determine it automatically.

Adbot
ADBOT LOVES YOU

Riso
Oct 11, 2008

by merry exmarx
The game is quite dumb and stores the cooling value only once in the model and then updates it based on the last trim you opened. :wtc:

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

BRC 1976 trump cards are up now, if anyone wants to have a look.

https://drive.google.com/folderview?id=0B4x5S3wQzQFlV0tEY0p6WWk4WW8&usp=sharing

Also Automation has migrated their forums to Discourse, which is super annoying to read. Anyways, here's the new link for that (also updated in the OP):

http://discourse.automationgame.com/

Riso
Oct 11, 2008

by merry exmarx
Oh right, they ran the Brands Hatch race tuesday.

Here's the result. Spoiler: I did very well.
https://www.youtube.com/watch?v=_R2-R8Mo5gs

Riso
Oct 11, 2008

by merry exmarx
Practice for the 1000km Hockenheim race.

https://www.youtube.com/watch?v=6md5Vf25Pbc&feature=youtu.be

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

New dev update:

https://www.youtube.com/watch?v=voEjNtpsmUI

Talking about the status of the current open beta (we are about a week away from the next version of the beta, and likely another week from the next release), some discussion of a few new features in-game (scenarios work again!) and the new forums.


Interesting; my FR car finished in 32nd and the rear-engine car finished 59th (for those that didn't see, Riso finished P26 and P33). I think I'm going to drop the the rear-engine car, as like you I've found it's just not capable of going fast enough with these rules. Even with the revised version I've made, which should be a second a lap faster, it isn't enough.

As for the prime entry, the added horsepower didn't add much in terms of lap times, but sure added a hell of a lot of wheelspin, so I'm going to have to get that under control - not an easy task with an FR car under these rules.

Riso
Oct 11, 2008

by merry exmarx
You can't change cars now, only pit strategy. You might want to add another stop or two. Your top car lost like 18 seconds at the end because of bad tyres.

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Riso posted:

You can't change cars now, only pit strategy. You might want to add another stop or two. Your top car lost like 18 seconds at the end because of bad tyres.

Which is another thing I don't get - my tires were far from the worst, but the differential between my best lap and my last lap was almost 18 seconds...yet other cars, with worse tires, were in the 5-7 second range. What gives?

Riso
Oct 11, 2008

by merry exmarx
Based on camber I think: -3 camber cars lose a lot more time than -1.5. It's also influenced by driver errors IIRC. I might add another stop myself to have fresher tyres.

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Riso posted:

Based on camber I think: -3 camber cars lose a lot more time than -1.5. It's also influenced by driver errors IIRC. I might add another stop myself to have fresher tyres.

I backed off the camber on the prime and the secondary has less still; neither is above -1.5 for any axle. The prime's tire problem might be caused by massive wheelspin, and the fact it seems to work it's tires harder than most cars.

Also, I came up with what appears to be the most efficient pit strategy for both cars. Frankly, the result of my calculations was pretty damned shocking; let me put it this way...if you have chased efficiency at all cost, you've already lost the race. :saddowns:

Riso
Oct 11, 2008

by merry exmarx
I made some crude calculations.
If tyres and reliability weren't an issue, the top five would be between TheBobWiley, Absurdist and mattmr2.

I'd be #15, #26 and you #18, #53.
An additional pit stop of my first car however moves me behind you.

And that's my problem, designating pit stop window. Because it's set in minutes and not laps.
I know my cars did 19/20 laps in practice but how many minutes are that for me? Ugh.

Riso
Oct 11, 2008

by merry exmarx
I made a really spergy doc to predict race stops and results.
https://docs.google.com/spreadsheets/d/1Vw70Cep2K4V9syNde1mC72Vwbbn52kIe5AIOFSv160c/edit#gid=0

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

I've gone through your calculations and there are a few issues in them; mostly with some of the assumptions made in some of the calculations. After looking at my cars and a couple other entries , I found that there isn't a significant lap time degradation until the tires wear beyond 20-25%, at which point the lap time degradation becomes very severe...as in the case for my prime car, over a second per lap past that point. Collecting each lap time as it came across the ticker, I used that information to plot the total amount of time to complete the 148 laps based on how many laps per stint the car completed, for my car as well as a few others, just as a sanity check. The resulting curve of time vs. pit interval looks sort of like this (it varys considerably for each car; this is for my alternate entry);



On on the short interval side, you are wasting more and more time in the pits as you move to the left. As you move to the right, tire degradation begins to slow your car down overall. Anyways, from this plot you choose the minimum value, figure out approximately how long it takes the car to complete that many laps, then feed it to BRCTool.

Riso
Oct 11, 2008

by merry exmarx
We'll get more accurate brobot data in the future.
Lap times for each lap in practice will be output!

extreme_accordion
Apr 9, 2009
Next time I'll have to submit an entry and share specs with you guys in order to work the system a bit better to find an optimum for the given event.
A win is a win even if it isn't mine alone :D

Riso
Oct 11, 2008

by merry exmarx
I only wanted to see how it will possibly shake out. Unlike MrChips car, mine's not bad enough on tyres to have to find a strategy besides drive to max fuel range. :smug:

For 'normal' 90 min BRC I'll check for ideal number of pit stops every now and then but two stops is pretty much always slightly faster. I'll try to stop a lap or two earlier than most of the pack though to avoid traffic.

Still, the real magic is in car setup and design and there's some real wizards on the automation forums. Like best car in BRC 1966 is a nearly unbeatable FWD.

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Riso posted:

I only wanted to see how it will possibly shake out. Unlike MrChips car, mine's not bad enough on tyres to have to find a strategy besides drive to max fuel range. :smug:

Well in fairness, my car is naturally at a huge disadvantage to yours, what with being front-engine, rear-wheel drive and having almost 650 horsepower to put through these lovely, made-in-East Germany tires that somehow ended up being narrower than I intended. :v: The car absolutely roasts its tires, and I am more than a little worried about the possibility of rain...

Still, I am optimistic that I might have a bit of a leg up with my tire wear and stint length calculations. Either that or I got something horribly wrong and it's gonna blow up in my face. Regardless, it should be good for a laugh!

extreme_accordion posted:

Next time I'll have to submit an entry and share specs with you guys in order to work the system a bit better to find an optimum for the given event.
A win is a win even if it isn't mine alone :D

It's too bad you didn't make an entry for this! Even if it turned out to be dog-slow, it's a good learning experience...I know I've learned a lot, both about car setup and some interesting ideas I have for my upcoming challenges.

Riso
Oct 11, 2008

by merry exmarx
I fully expect my cars to not make it to the end.

Currently I am trying to build a decent turbo under the rules in the Porsche body but I think I need a front engine variant. FR are a lot easier to set up in some ways.

I also learned that currently in the open beta all cars seem to go faster.

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Qualifying!

https://www.youtube.com/watch?v=6h5JDLV3SBE

It's a disaster! :v:

Riso
Oct 11, 2008

by merry exmarx
A rainy qualifying for a six hour race. We will manage.

Riso
Oct 11, 2008

by merry exmarx
New update is finally out, as is a new dev preview: https://www.youtube.com/watch?v=zL8X6H2IdkA

http://store.steampowered.com/news/?appids=293760

quote:

Main Features
Engineering Projects & Engineers
Factory Production & Managers
V12 Turbos
Spaceframe Chassis
8 new car bodies, 27 new body variants in total

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

And in the next version, likely due out in July, we will have inline-3 engines, more fleshed out factories and finally, the first tastes of a campaign mode! After that, the game gets ported over to the Unreal Engine, which should make things a lot easier for everybody involved than fighting with the Kee Engine as they are now.

And since I have seemingly neglected it, an update on the upcoming SASC-89 challenge; I have come up with a functional dynamic weather simulation, which simulates not only precipitation, but also temperature as well, along with the effect that precipitation and cloud cover has on it. The tire wear and grip model has been rewritten to take this into account; for example, soft tires suffer high wear in high temperatures, and if you leave your car out on wet tires in the dry they will also suffer extreme wear too - a full wet in the dry will last something like 20% of a dry tire's life. I have tweaked the fuel economy model slightly; after further experimentation with the old model, some cars were slightly too high while others were slightly too low.

As for the rules, I have determined that trying to do a proper homologation, where you are told how many cars of that type must be built in a year, is just too much of a mess, so we will be sticking with cost limitations for the time being. Also, there will be relatively few engine restrictions in this challenge; basically, the only restriction will be a fuel burn limit - each class will have a maximum fuel burn per hour that must not be exceeded. Don't worry, a calculator will be provided so that you can easily figure out where you stand.

There are a lot of other things I need to do before I am ready to put all of this stuff together in the new model, but I assure you, progress is being made!

Riso
Oct 11, 2008

by merry exmarx
Now you just have to add a visual display of the race and you have your own brobot :v:

You should probably add a production unit restriction and/or perhaps a time limit for r&d. Values dropped a lot so money limits aren't good enough.

Like my Hockenheim car now costs 1700$ less but it takes like engine/car 160/153 prod units and 324/420 months engineering time.

Comparison my 1966 car:
-2300, 130/113 prod, 201/259 time.

Btw I ran my SA-GTE car for a laugh after the patch. It's now two sec faster on Le Mans. According to the race on Nurburg it's now 13sec faster there...
I dont have the old costs but its 970/447 prod and 1458/1254 time.

Riso fucked around with this message at 11:20 on Apr 19, 2016

Duuk
Sep 4, 2006

Victorious, he returned to us, claiming that he had slain the drought where even Orlanth could not. The god-talkers were not sure what to make of this.
I see I've missed MrChips running (and winning) a successful 1975 challenge. That's alright I suppose, I was getting burnt out on the idea after all those false starts.

The BRC youtube videos are shockingly cool. I hope they implement something like that into the actual game. Any clue on when a new round of brobot races would be starting? I went and registered on the forums, but haven't had a chance to look any closer.

Riso
Oct 11, 2008

by merry exmarx

Duuk posted:

Any clue on when a new round of brobot races would be starting? I went and registered on the forums, but haven't had a chance to look any closer.

No idea. First we will finish BRC 1966 and then that single 1976 endurance race on Hockenheim.

So it will take one, two months at least? There will be a big announcement on the forums but until then just join interesting sounding competitions to design some cars.

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

OK, the work on the new simulation model for this and other challenges continues apace, so let's have a bit of a discussion about a new feature; the driver simulation and some of the things that go with it. Bear in mind that at the moment this is only a preliminary take on how this will work, and could change dramatically between now and the beginning of the challenge.

Basically each car will have a team of three unique drivers, each of whom will have unique traits, skills and weaknesses. Things like, how fast are they in wet or dry conditions, are they hard or easy on their car or their tires, do they drive efficiently or wastefully. Some of these traits, like wet and dry speed, will be on a scale of 0 to 10, with 5 being no effect. Other traits (or weaknesses) will be a simple bonus/malus to some aspect; for example, a driver tagged as "Aggressive" will be more likely to be involved in an accident than a driver who isn't. And weaknesses will be unavoidable; even the highest-rated drivers will have something to give you something to think about.

On the subject of driver ratings, this affects how much money it cost to hire them and run them for the season...high-rise drivers will be the most expensive obviously, while low-tier drivers will be cheaper; in fact, more often than not, they will be pay drivers, in that they actually bring money to the team in the form of sponsorship rather than cost you money. Before the season begins, I am thinking what we will do is hold a draft of sorts, where each entrant gets to choose their driver lineups from the pool of available drivers. Before you fret about the number one entrant picking the six best drivers and that's it, first of all it would be prohibitively expensive to do, and second of all, the drivers will be streamed into tiers, from which you will be allowed to choose only a limited number of drivers from each. For example, with a three-tier system, each entrant can choose a maximum of one top-tier driver, two middle-tier drivers and unlimited bottom-tier drivers.

This also beings me to the next subject; the economic simulation that will be baked into the SASC-89 Challenge. For now, the details are up in the air as to how this will work, but the basic idea is that each team will have a fixed budget to develop and build their cars, with add-ons for each additional class entered; the add-ons will not be the full basic amount either, so commonality between classes, in as much as possible, might be a good idea! Additionally, there will be sponsors (chosen by draft as well, similar to how the driver draft will work), TV money and merchandise sales that go up with success and down with poor results for revenue. On the cost side, transportation cost will be factored in (so each team will have to declare the city and country where they are based, so it might not be a good idea to be based in North Korea or on Pitcairn Island). Additionally, accidents and failures will incur more costs, as will design revisions.

If any of you have ideas that you would like to see, by all means suggest away! Nothing is set in stone just yet, and these are just the ideas that I have come up with.

Riso
Oct 11, 2008

by merry exmarx
Sounds like you would make Edward Grabowski proud MrChips.

On the car and engine designs I recommend you set money and production unit limits. Just money isn't good enough any more. Also are you going to use the in-game drive/sport info or the track tameness and sportiness settings from the lua?

Oh btw, fun bug in public release: engine variant designs (say 2016) reset to the base model (say 2009) on game exit!

The last and final BRC 1966 Race on Donington.

EDIT: CANCELLED. Will happen some other time

Riso fucked around with this message at 15:39 on Apr 29, 2016

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Riso posted:

Sounds like you would make Edward Grabowski proud MrChips.

On the car and engine designs I recommend you set money and production unit limits. Just money isn't good enough any more. Also are you going to use the in-game drive/sport info or the track tameness and sportiness settings from the lua?

Oh btw, fun bug in public release: engine variant designs (say 2016) reset to the base model (say 2009) on game exit!

The last and final BRC 1966 Race on Donington.

EDIT: CANCELLED. Will happen some other time

I am leaning towards the lua values personally because it's easier and I'm lazy, but I haven't made a decision just yet which will be used to interface with the weather simulation. I will say this much though; cars with higher driveability/tameness will have a lower spread between dry and wet laps, at the very least.

Also, there will be a production unit limitation on both engine and chassis; I'm not a fan of it because it isn't all that well balanced at the moment, but it's still the best we have nonetheless. The money limitation in this rule set will be the final production value too, after factories are taken into account and whatnot. I haven't found a clean way of doing that just yet though either.

extreme_accordion
Apr 9, 2009
Seems cool enough but we are then raising the bar on entry without some good explanation to those new to it. As an example: The BROBOT stuff seems pretty impressive with the pit ins, visual simulation, lap count, etc. What threw me off is no easy way to calculate fuel consumption in order to determine your pit strategy. After reading several linked posts I still couldn't make heads or tails of how to calculate fuel consumption within the confines of the BROBOT sim so I just did not submit my entry. Looking back on it I would have had a low top tier to middling entry but without being able to determine pit and fuel who knows. Supporting sheets or tools would go a long way to help determine what a team would need as a driver and pit strategy (assuming you are adding this to your sim) given the added variable of: driver.

Additionally if the '89 specs are posted we could begin building cars to throw at the previous version of your sim in some fashion with the three variants of: FR MR RR to help you develop your model.

AKA I'm bored and want to start building cars to a spec.

Riso
Oct 11, 2008

by merry exmarx

MrChips posted:

The money limitation in this rule set will be the final production value too, after factories are taken into account and whatnot. I haven't found a clean way of doing that just yet though either.

Might be best to just ignore the factory stuff. It's pretty much pointless. Setting a market rating would be better but that also has some weird behaviours and suffers from a lack of competitor cars.

extreme_accordion posted:

Seems cool enough but we are then raising the bar on entry without some good explanation to those new to it. As an example: The BROBOT stuff seems pretty impressive with the pit ins, visual simulation, lap count, etc. What threw me off is no easy way to calculate fuel consumption in order to determine your pit strategy. After reading several linked posts I still couldn't make heads or tails of how to calculate fuel consumption within the confines of the BROBOT sim so I just did not submit my entry. Looking back on it I would have had a low top tier to middling entry but without being able to determine pit and fuel who knows.


In brobot the only real way to gauge fuel consumption is to check pre-race practice on each track. My 1966 car can range from 0.74 to 1.38l/lap depending on track.

quote:

Supporting sheets or tools would go a long way to help determine what a team would need as a driver and pit strategy (assuming you are adding this to your sim) given the added variable of: driver.

MrChips thankfully was able to provide for 75 GTE a fuel calculation google docs so that was nice.

quote:

Additionally if the '89 specs are posted we could begin building cars to throw at the previous version of your sim in some fashion with the three variants of: FR MR RR to help you develop your model.

AKA I'm bored and want to start building cars to a spec.

No challenges to your liking on the official forums?

Oh, btw. That flatnose not-porsche 930 was updated with a cabrio and front engine option.

extreme_accordion
Apr 9, 2009

Riso posted:

No challenges to your liking on the official forums?

One less forum to read or watch :P

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

extreme_accordion posted:

Seems cool enough but we are then raising the bar on entry without some good explanation to those new to it. As an example: The BROBOT stuff seems pretty impressive with the pit ins, visual simulation, lap count, etc. What threw me off is no easy way to calculate fuel consumption in order to determine your pit strategy. After reading several linked posts I still couldn't make heads or tails of how to calculate fuel consumption within the confines of the BROBOT sim so I just did not submit my entry. Looking back on it I would have had a low top tier to middling entry but without being able to determine pit and fuel who knows. Supporting sheets or tools would go a long way to help determine what a team would need as a driver and pit strategy (assuming you are adding this to your sim) given the added variable of: driver.

Additionally if the '89 specs are posted we could begin building cars to throw at the previous version of your sim in some fashion with the three variants of: FR MR RR to help you develop your model.

AKA I'm bored and want to start building cars to a spec.

Don't worry, there will be as many supporting tools for this challenge as I can possibly make. There will be a calculator that will tell you whether your engine is over or under the fuel burn limitiation, as well as graphs (or possibly even calculators as well) that will show you the effect of temperature, moisture and driveability on your car's lap times, and the effect that reliability will have on your ability to finish the race.

As for the timeline, I will actually be releasing one part of the rules very, very soon - the prototype challenge car is basically ready to go, I'm just writing up the "Customer Guide" for it!

Riso
Oct 11, 2008

by merry exmarx
Alright the final race of BRC season was Wednesday.
I nearly forgot about it as soon as I read the announcement on Tuesday evening but I just got there to watch the race.

And what a race it was, with the commentators at the edge of their seats watching a wet Donington.

https://www.youtube.com/watch?v=tRYM-FMAoiQ

https://www.youtube.com/watch?v=8Ir9Sxh0lBc

extreme_accordion
Apr 9, 2009

Riso posted:

Alright the final race of BRC season was Wednesday.
I nearly forgot about it as soon as I read the announcement on Tuesday evening but I just got there to watch the race.

And what a race it was, with the commentators at the edge of their seats watching a wet Donington.

https://www.youtube.com/watch?v=tRYM-FMAoiQ

https://www.youtube.com/watch?v=8Ir9Sxh0lBc

2nd in points for the season?

Riso
Oct 11, 2008

by merry exmarx
That's correct, I managed to be the first of the losers for the whole season.

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

New dev update!

https://www.youtube.com/watch?v=I4HI0usE_I8

Talking about campaign mode and inline-3 engines. Also, they are putting in a huge number of new bodies as well for the next update.

Riso
Oct 11, 2008

by merry exmarx
The crazy bastards actually did it.
Brobot endurance race on Hockenheim with commentary.

https://www.youtube.com/watch?v=DNAe2Gmjjdo

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Riso posted:

The crazy bastards actually did it.
Brobot endurance race on Hockenheim with commentary.

https://www.youtube.com/watch?v=DNAe2Gmjjdo

Well, that was bittersweet for me; my prime car failed after only 16 laps with a fuel system failure, and my secondary entry, which ran as high as P16 for quite a while, had an exhaust failure on Lap 48.

Riso
Oct 11, 2008

by merry exmarx

MrChips posted:

Well, that was bittersweet for me; my prime car failed after only 16 laps with a fuel system failure, and my secondary entry, which ran as high as P16 for quite a while, had an exhaust failure on Lap 48.

The displayed errors are just randomised fluff. You had rather low engine reliability so I wasn't surprised.

I finished higher than I thought but the tyre changes around 70% wet were a mess. I left everything at default but it worked a lot better at the lower end around 25%.

Adbot
ADBOT LOVES YOU

MrChips
Jun 10, 2005

FLIGHT SAFETY TIP: Fatties out first

Riso posted:

The displayed errors are just randomised fluff. You had rather low engine reliability so I wasn't surprised.

I finished higher than I thought but the tyre changes around 70% wet were a mess. I left everything at default but it worked a lot better at the lower end around 25%.

The one car was low, but not outrageously so - the average across the field is something like 30 reliability or so - and the second car was slightly over average at 33. They should have a ~75% and ~90% chance of finishing this race respectively based on the data that Der Bayer posted early in the design phase.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply