[MVRC] Mantium Virtual Racecar Challenge 2017

Post here information about your own engineering projects, including but not limited to building your own car or designing a virtual car through CAD.
koss
0
Joined: Tue Dec 19, 2017 3:57 pm

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by koss » Sun Dec 31, 2017 6:30 pm

LVDH wrote:
Fri Dec 29, 2017 4:22 pm
Yes, apart from sometimes answering emails and this forum, I was busy with other stuff. You should now have your details. Good luck. Keep in mind there is no plan yet about next year. It will be an excellent exercise to build a car to current regulations and see how you would compete with the top dogs here.
Thank you very much! Looking forward to the next season.

RicME85
58
Joined: Thu Feb 09, 2012 12:11 pm
Location: Derby

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by RicME85 » Mon Jan 01, 2018 1:55 pm

CAEdevice wrote:
Thu Dec 28, 2017 11:11 pm
A small update about the prize: all European guys should have received the book (except Robert/JJR: the signed book has been shipped only today).

People from outside Europe (Australia, South America) will receive the book in January (two central weeks).

People from North America will receive the book in a couple of days.

Please let me know if something went wrong.
Once again, thank you for this amazing gesture Matteo, its a beautiful book.
I went on Audible the other night and picked up the audiobook version so that I can keep it looking good but use it for reference when something interesting in the audiobook pops up. Thankfully Newey doesnt do the reading, as interesting as he is I couldnt listen to him for 5+hours, his soft, slow, laboured way of speaking makes me really anxious when he is being interviewed on TV :lol:

CAEdevice
22
User avatar
Joined: Thu Jan 09, 2014 2:33 pm
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by CAEdevice » Mon Jan 01, 2018 2:32 pm

RicME85 wrote:
Once again, thank you for this amazing gesture Matteo, its a beautiful book.
Honored to have been a sponsor of the challange. I hope I can do better next year, possibly in the company of other sponsors.

Happy New Year!

schwepes
0
User avatar
Joined: Mon Apr 01, 2013 9:01 am

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by schwepes » Mon Jan 01, 2018 11:08 pm

Very interesting book! Maybe a bit too short ;)... just 3 days to read from cover to cover.

CAEdevice
22
User avatar
Joined: Thu Jan 09, 2014 2:33 pm
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by CAEdevice » Tue Jan 02, 2018 11:02 am

I think I am the one that only looked at the pictures. I am a bit busy with some (boring) projects completely different from CFD and racecars.

schwepes
0
User avatar
Joined: Mon Apr 01, 2013 9:01 am

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by schwepes » Tue Jan 02, 2018 5:26 pm

I found it very interesting to read about the background of some situations that I remember watching years ago when Benetton, McL and than Ferrari was dominating in F1. Also some sudden loss and gains of performance of F1 cars are now much more understandable for me.

machin
167
User avatar
Joined: Tue Nov 25, 2008 1:45 pm

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by machin » Tue Jan 02, 2018 11:10 pm

A day late... but it took me a while to get the flow under the wing just right to spell the words... ;-)

Image
COMPETITION CAR ENGINEERING -Home of VIRTUAL STOPWATCH

etsmc
4
Joined: Wed Apr 04, 2012 12:20 pm

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by etsmc » Wed Jan 03, 2018 4:42 am

What people thought about not changing the rules if there is a 2018 season?
I was thinking about it a bit and there were a few small changes during 2017 and the HX rotation which was big in my opinion.
There was also only 1 race where everyone achieved 100% engine power so I feel there is a lot to be developed still, also some of the lower teams are still only producing half the amount of downforce as the top contenders.
If this were a real series and had a 107% rule then over half the field would not have raced in the last race, I think we could get the racing closer and the time closer with more development.
Also keeping the rules the same newcomers would be able to look at previous cars and results to see where they need to be..

schwepes
0
User avatar
Joined: Mon Apr 01, 2013 9:01 am

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by schwepes » Wed Jan 03, 2018 9:43 am

For me it would be good to see if I can make enough developement to go up a bit in the classification with current rules. Nevertheless I'm not opposing against new ones. It will be interesting to compete with any rules.

CAEdevice
22
User avatar
Joined: Thu Jan 09, 2014 2:33 pm
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by CAEdevice » Wed Jan 03, 2018 10:32 am

I have no personal preferences about the reference layout (prototypes, open wheels, GT ...): in my opinion the only fixed point is the realism of the rulebook and of the simulation environement.

I agree with etmsc and schwepes: brand new rules would be an advantage for expert designers and professional with big computational resources.

I am planning to release an "open source" LMPx prototype, based on the car that won 4 out of 5 races in 2017, with a cl/cd ratio above 5.5 "out of the box". It will be indipendent from the CAESES online modeler and ready for the "top class" rules. Everything between the wheel axles will be frozen, but there will be parametric features in the front and rear areas for the teams who use Solidworks and two/three aero kits for the teams who simply import the STEP. It would be great (and it would help new and small teams) to have a constructors challenge.

I will probably do the same with a completely new rulebook (open wheels, GT) but I am not sure (it depends on the time between the release of the rules and the first race).

machin
167
User avatar
Joined: Tue Nov 25, 2008 1:45 pm

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by machin » Wed Jan 03, 2018 1:40 pm

I tend to agree with keeping the rules the same: it makes life a lot easier for Andre and Chris for starters, and also it means that no changes are required to the CAESES app which now includes the new diffuser.

From a personal (selfish) point of view I would also like to use the results of the Le Mans race as the first step of my 2018 development plan; changing the rules now would require going back to square one...
COMPETITION CAR ENGINEERING -Home of VIRTUAL STOPWATCH

CAEdevice
22
User avatar
Joined: Thu Jan 09, 2014 2:33 pm
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by CAEdevice » Wed Jan 03, 2018 3:00 pm

Since we are talking about proposals for 2018, here are my ideas:

1) Five races are ok, but three (with longer pauses and better quality of simulation/showcase, at least 2 of them with efficiency tracks) would be ok too

2) Increasing points (maybe LeMans could bring double/triple points) in order to keep the final result uncertain but also beacuse the bast car in last race is usually the best car/team, since we don't have other variable parameters than aerodynamics (no drivers errors, no strategy). See TF 2016 for example (the best car but not the challenge winner)

3) "Intro class" (but I would prefer to name it "customer class") with same engine as "top class". The difference would consist of a less heavy mesh of the solutor (es."mvrc_fast" vs "mvrc", in order to stay into 8gb RAM) and of the use of standard cars provided by the top teams, with everthing included between front and rear axles frozen. I would like that the best "intro class car" could be on the podium sometimes.

4) For the "top class" only: more detailed cooling (ex. front brakes cooling, oil radiator)

5) I confess this is just for aesthetics: a slighty shorter rear overhang (see the image below)

6) No complete CFD images during season, but a more detailed 3D view ( ex. http://3dbin.com/ ), excluding underfloor (as it happens in real races).

7) No Robin masks on the cars during the preview ;)

A cloud computing service provided by a sponsor and transient simulations would be a dream, but I do not want to ask too much : )

Image

machin
167
User avatar
Joined: Tue Nov 25, 2008 1:45 pm

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by machin » Wed Jan 03, 2018 8:46 pm

I'm not sure about 3; for me the Intro class car should always be about learning about aerodynamics... Not an easy step to the top of the championship. I think we should encourage people to design their own car (why bother if they can just use your car as a basis)... Of course no one can stop you from releasing your car to anyone who wants to use it... But a championship full of 12 near identical cars at the front and a few other different ones at the back wouldn't be so cool in my opinion...

6, Personally I would like to see MORE CFD images; this is a CFD competition and sometimes it's not easy to see that; how many CFD images are in this message thread or on the challenge web site? Not many. I don't think surface pressure plots, even underside surfaces, give too much away since simply copying the look of a specific car doesn't yield good results...

I think an isometric surface pressure plot from above and below of the top 3 cars (at least) should be released before the race results are released; it would provide a much needed talking point IMO...

I wish there was something more we could do to make the race results more exciting.... The problem is that the fastest cars are pretty evident after the first corner or two (since there is no chance of a driver mistake), so even a really cool full animation of the lap wouldn't have much suspense past the first few corners (although I admit this would still be cool, and ultimately would be a nice thing to have).
COMPETITION CAR ENGINEERING -Home of VIRTUAL STOPWATCH

CAEdevice
22
User avatar
Joined: Thu Jan 09, 2014 2:33 pm
Location: Erba, Italy

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by CAEdevice » Wed Jan 03, 2018 9:55 pm

Point 3: I agree with you. I think the to design everything is in front of the front wheels axle and behind the rear wheels axle is a hard job. I also hope to not be the only one to release a "customer" edition of the car ;)

Point 6: I would not be very happy about seeing surface pressure plots (not realistic), but you are right: it would be much more acceptable than a velocity plot, so it would be ok for me.

PS: I studied an alternative to AssettoCorsa: Torcs (worse graphics but great customizable AI and Open Source code). The problem is the same: too much time is needed to import the car geometry.

variante
50
Joined: Mon Apr 09, 2012 10:36 am
Location: Monza

Re: [MVRC] Mantium Virtual Racecar Challenge 2017

Post by variante » Fri Jan 05, 2018 3:46 am

etsmc wrote:
Wed Jan 03, 2018 4:42 am
What people thought about not changing the rules if there is a 2018 season?
It would be a valid option. Yet, there are other valid options to consider.

First, I'd like to talk about why it would not be a goog option, since you already pointed out the pros:
- the technical know-how gathered on this kind of cars by the top teams is so big that not many newcomers would be able to catch up in a reasonable period of time.
- there is still some performance gain to be made, so we can expect the top teams to grow further more, keeping the performance gap as big as it is now.

So what about a new formula?
- many design concepts to explore, which may shake up the situation quite a bit, as well as making the design challenge even more entertaining. But yes, as CAEdevice pointed out, those with bigger computational resources would have an advantage.
- potentially simpler cars to design and develope (I find it very time consuming to design or modify car of the current layout. An open wheeler, for example, is quite a bit better by this point of view). I think this characteristic would encourage a good number of people to join the challenge.
- potentially less time consuming for the guy who checks rules compliance.

CAEdevice wrote:
Wed Jan 03, 2018 3:00 pm
2) Increasing points (maybe LeMans could bring double/triple points) in order to keep the final result uncertain but also beacuse the bast car in last race is usually the best car/team, since we don't have other variable parameters than aerodynamics (no drivers errors, no strategy). See TF 2016 for example (the best car but not the challenge winner)
I don't fully agree. The winner of the first race deserves a lot of credit too, for example. I would rather keep the points constant. Possibly with a different distribution (i.e. 10pts for 1st, 8 for 2nd,...), as we don't have to encourage any overtake of spectacular action "during the race".

So... my proposal looks like this:

1 - only simple and essential rules >> easy to check, easy to interpret, less useless limitations

2 - linear points distribution (10, 8, 6, ...) >> closer to reflecting actual preformance gaps, championship open untill the end

3 - CFD images (as remarked by machin) >> catchy for the eye, more interest, more oportunities to learn

4 - intro class with no power penalty >> enough penalty from the frozen bodywork, equal dignity

5 - more schematic rulebook, with clear titles >> easier to read and catch infos

6 - more linear rules templates and quotes >> easier design phase