GPWizard F1 Forum
F1 News & Discussions => General F1 Discussion => Topic started by: John S on January 10, 2010, 12:01:27 PM
-
So the big teams will still have the advantage they just have to get by with less team members. I'm personally glad that the one size budget must fit all has been binned, like the cars in F1 over the years team individuality has been the biggest key word driving the sport.
According to Lawrence Butcher, of IPC magazines RC Engineering, FOTAs ‘Resource Restriction Agreement' that member teams will have to abide by for the 2010 and 2011 seasons will lead mainly to limiting team personnel numbers.
From January 1, 2010, teams will be limited to 250 personnel and a maximum spend of 50 million Euros. In 2011 this number will decrease further to 180 personnel and 30 million Euros. However the budget cap excludes salaries.
Our source quoted ' A major motorsport publication was saying how Peter Sauber has saved Sauber with his brilliant business plan and reducing staff to 250 people, but everyone has to do it!'
The restrictions will cover everything excluding salaries, marketing and engine costs because these are considered beyond control. For example to expect Mecedes Benz, as an engine manufacturer, to spend the same as Force India would not be feasible.
-
€30m? That's a huge restriction! I would argue that personnel numbers are far from the only thing restricted by a resource restriction that severe. It's little things like a supercomputer costing a third of the annual budget to buy and then run for a year. It could also have an effect on the willingness of teams to produce stuff in-house - expect more sharing arrangements like Force India's with Mercedes for hydraulics and gearbox to emerge in the next two years.
-
€30m? That's a huge restriction! I would argue that personnel numbers are far from the only thing restricted by a resource restriction that severe. It's little things like a supercomputer costing a third of the annual budget to buy and then run for a year. It could also have an effect on the willingness of teams to produce stuff in-house - expect more sharing arrangements like Force India's with Mercedes for hydraulics and gearbox to emerge in the next two years.
For just about any business in the world salaries are the majority cost. When you consider that an F1 team is basically building 10 cars, 30 million covers the raw materials, electricity, and amortised equipment costs. A passenger car, including the engine and windows, has less than 1000 in raw materials. Even given that an F1 car has more exotic materials, 30 million is plenty.
-
That's a bit of an exaggeration. Those raw materials go through millions of dollars of machines before they can be mounted by employees. Same thing for a F1 car. Like it was already mentioned, a supercomputer running simulations could eat up a lot of that budget already, throw in a few autoclaves, and a wind tunnel budget (for running it as well as depreciating it) and I'm sure you're getting near the $30m.
-
The really expensive bits about a F1 car are the design and running. The design because it takes a lot of time and effort to figure out what to tell the computers, get the computers up and running (and the supercomputers are generally on overnight for months at a time, so the power bill is considerable) and also get them tested to satisfaction (scale wind tunnel testing is only one of several performance checks a component can be expected to pass before being put into production).
The running because not only does a F1 car typically do about 4 mpg, but a component needs changing every 20 minutes of running on average. That's a lot of componentry.
To put this into perspective, in 2005 a chassis cost around £250,000 to make and about 10 times that to develop. In fact, even Force India will need to cut its budget to meet the $30m because its estimated budget was around $80m for 2009.
-
To put this into perspective, in 2005 a chassis cost around £250,000 to make and about 10 times that to develop. In fact, even Force India will need to cut its budget to meet the $30m because its estimated budget was around $80m for 2009.
How much of the 80 million was for engines, and how much for personnel, advertising, transportation? None of that counts to the 30 million cap, and does account for the majority of the F1 budget.
The problem is going to be capital expenditures... the part where F1 gives back to the economy and engineering science by buying outrageously expensive and experimental machines.
My guess... F1 will start partnering with universities, and letting donors cover costs of 'research' equipment. What engineering programme wouldn't want access to McLaren's shop in the off season? (Our American friends will recognise an NCAA type problem here.)
-
The problem is going to be capital expenditures... the part where F1 gives back to the economy and engineering science by buying outrageously expensive and experimental machines.
My guess... F1 will start partnering with universities, and letting donors cover costs of 'research' equipment. What engineering programme wouldn't want access to McLaren's shop in the off season? (Our American friends will recognise an NCAA type problem here.)
That would be an incredibly brilliant and sensible idea...which is why the FIA will never let it happen.
-
How much of the 80 million was for engines, and how much for personnel, advertising, transportation? None of that counts to the 30 million cap, and does account for the majority of the F1 budget. {Jericoke - 2 comments ago}
It's not clear how much was for engines, because Force India get engines (which are not in the cap), gearbox and hydraulics (which are in the cap) from the same deal with one combined price and no number has been mentioned for the total let alone the constituent parts. Said number would probably be wrong anyway because originally KERS was part of the price but now won't be, unless they're doing some serious pre-planning for the final year of the contract.
Force India has a staff of 280 people, which will need to go down by 100 to meet the resource restriction requirement of 180 staff. Note that Minardi employed aroud that many when Red Bull bought it at the end of 2005 and they were paying their staff a combined wage of $12.6m at the time. Add, say, 3% inflation per year and you have an estimated current bill for that staff level of $14.181411m & 6/10ths of a cent (€9,784,677,382.16). By the start of 2011 it will be $14.606853.86 and 18/100ths of a cent (€10,078,218.07 at current exchange rate).
Transportation should be covered by "Bernie Air" (a service which transports a certain number of people, considered the minimum for competitiveness, plus three chassis, their engines and 2 tonnes of other car parts to every race free) except for excess baggage, sea transport and people transport. As a guideline, Minardi took 60 people to races in 2006 (13 more than can enter the paddock under the resource restriction guidelines) and paid $2.47m (€1.7m) for their travel and accommodation expenses in a season. It had Bernie Air priviledges, albeit not perhaps as generous as they might be now. I'd guess at a ballpark figure of €3 million for the sum total of Force India's transportation expenses, noting that accommodation on Grand Prix weekends has gone up by rather more than 3% per year (much to the chagrin of spectators).
Advertising is impossible to calculate because it's not even clear how much of Force India's advertising is coming from the team and how much from the wider comglomerate that owns it at this stage (there have been some combination advertising going on and the team itself is effectively considered an advertising board for Vijay Mallya's companies and values). I would expect Force India's advertising bill to skyrocket in the next two years as money previously spent by UB Group is now spent by the team on advertising... ...but that doesn't help me figure out how much it's spending on it now!
I would dispute those three items being the main expenses in a F1 team's budget - at least in the case of those teams not producing their own engines. In 2006 (the year for which I have figures in front of me) engines cost $180m for Toyota in 2006 (36% of its budget) but Jordan, who bought those same engines, only paid $15m (14.42% of its budget). Advertising isn't given in the figures I've got, but it does include corporate entertainment, which took up $11.5m of Toyota's $499.05m budget (2.30% of its budget). Jordan spent $1.4m (1.34% of its budget). Travel is the only item out of the three that is roughly proportional to budget (in fact it's an inversion; Toyota spent $12.97m/2.60% of budget while Jordan spent $4.86m/4.67% of budget on travel).
The expensive bits of running a small F1 team that isn't an engine producer are operating cars at races and tests, along with R&D costs. These are very much included in the budget caps. Wind tunnels used to also be a major item before the 2009 restrictions came in.
So the short version is that I can't answer your question, Jeri, but I'm sure at least $13.078m of the current $80m won't count towards the 2010 or 2011 budgets... ...and it's nowhere near the $50m Force India needs to remove by the latter year.
-
WOW Ali ! This post is awesome. How did you get all the data points ? :D
-
The initial data points are in the March 2006 edition of F1 Racing and the Force India staff stat came from an article in the November 2009 edition of F1 Racing. I used Yahoo!'s search engine to figure the exchange rates and my calculator to do inflation.
-
:D Good one. But actually technology can reduce the cost of design and development in F1. Its just that some of the teams in the interests of secrecy or whatever dont fully use technology, simulation, CFD's and so on.
I know companies which design aerospace parts using CFD completely and they pass all the physical test by Boeing, Airbus and so on. And moroever today we dont need supercomputers any longer. The Average pentium chip has immense processing power and massively parallel computing and Cloud Computing are changing development cycles all over the world.
For a sector which prides itself on innovative usage of technology and science, their reluctance to harness all the advances are a bit of a paradox.
-
Supercomputers still have much more processing than an average (or even exceptional) Pentium chip and although cloud computing has potential, it's currently rather slow because information can only be sent across the cloud at broadband speed. It's possible to make a small supercomputer out of five graphics cards for around £3000 these days, but even small supercomputers are a tad limited in what they can contribute to CFD. The new technologies make certain kinds of R&D cheaper, but are of limited utility in a F1 R&D context at the moment.
-
Point Taken Ali ! But today's Super Computers do work in Geo Physics, Earth Mapping, Nuke explosion simulations and so on.
Dont think designing F1 cars really require Super Computers Ali.
-
Point Taken Ali ! But today's Super Computers do work in Geo Physics, Earth Mapping, Nuke explosion simulations and so on.
Dont think designing F1 cars really require Super Computers Ali.
Actually... a simple arithmetic exercise shows that they do.
A modern desktop computer does approximately 2 billion operations a second. A Supercomputer does more than a trillion.
The point of the CAD is to see how a car performs in real world conditions... a computer game might get away with a couple hundred polygons making a car, and a vector simulating wind, but to do that in F1, to make the world a set of vague rules, instead of real conditions, is a waste of time. Basically, for real time results, you can make a super computer 1000 times more detailed, which is nice if you're trying to adjust a winglet a millimetre.
However, they don't have to do real time. They can let a programme run for two weeks, and then see what happens. Just try letting a home computer run one intense programme for for two weeks and see what happens. Suddenly a supercomputer can provide results that are literally a billion times more accurate than you could hope to do on a PC.
And this technology does 'trickle down' to real world car design. So while it's not needed per se, it does make the world a better place.
-
There will always be big gap between top and back grid teams. Force India has severe financial problems but some bigger teams have less. Now when economics situation is difficult locally, also bigger and richer teams lose their sponsors like Renault will lost his own this year.
-
Point Taken Ali ! But today's Super Computers do work in Geo Physics, Earth Mapping, Nuke explosion simulations and so on.
Dont think designing F1 cars really require Super Computers Ali.
Actually... a simple arithmetic exercise shows that they do.
A modern desktop computer does approximately 2 billion operations a second. A Supercomputer does more than a trillion.
The point of the CAD is to see how a car performs in real world conditions... a computer game might get away with a couple hundred polygons making a car, and a vector simulating wind, but to do that in F1, to make the world a set of vague rules, instead of real conditions, is a waste of time. Basically, for real time results, you can make a super computer 1000 times more detailed, which is nice if you're trying to adjust a winglet a millimetre.
However, they don't have to do real time. They can let a programme run for two weeks, and then see what happens. Just try letting a home computer run one intense programme for for two weeks and see what happens. Suddenly a supercomputer can provide results that are literally a billion times more accurate than you could hope to do on a PC.
And this technology does 'trickle down' to real world car design. So while it's not needed per se, it does make the world a better place.
Brute processing power is not what is required for most applications. CAD was originally meant to cut down on development time and costs by doing things faster than humans. For instance take the case of an architect designing a house without the aid of computers. It would take between 6 to 8 weeks just to get the design, some more time to test it and than finally iterations when it is actually being built.
Most CFD's run SG machines and so on. Super Computers are really used to input millions of variable data and and simulate a real life situation.
For instance the reason countries like US, UK and a few others want a ban on Nucleur Testing is because they have massively parallel processing super computers capable of simulating a bomb and analysing the results. They no longer need actual tests to do that. This means they save millions of dollars on testing and these machines are not available to a lot of countries, so they have to test.
Car design and simulating car performance is really not an application requiring super computers in the conventional sense. Moreover todays super computers are blade servers running parallel activities and delivering results.
A Good Blade server can be purchased for $3000.
-
To give an idea of what processing requirements designing a F1 computer these days are, Williams' CFD supercomputer does 10 billion calculations to calculate a car part (according to the October 2009 edition of F1 Racing) - and they still complain that they have to assume a simplified turbulence calculation because it's not cost-effective to get the upgrade to the computer that would tell them if the calculation was accurate (because it's 100 times bigger than what they have). So at the moment F1 design requires a better computing power than the teams can afford to get. Because the power the computer systems produce is nowhere near enough, there are still major assumptions being used.
-
To give an idea of what processing requirements designing a F1 computer these days are, Williams' CFD supercomputer does 10 billion calculations to calculate a car part (according to the October 2009 edition of F1 Racing) - and they still complain that they have to assume a simplified turbulence calculation because it's not cost-effective to get the upgrade to the computer that would tell them if the calculation was accurate (because it's 100 times bigger than what they have). So at the moment F1 design requires a better computing power than the teams can afford to get. Because the power the computer systems produce is nowhere near enough, there are still major assumptions being used.
Hence McLaren, and others, feel the need of their magic paint amongst other things.
-
Ali ! I think you should leave it after I have made a comment. BecaUSE i KNOW WHAT i AM TALKING ABOUT HAVING SPENT 22 YEARS IN COMPUTERS
-
Huh, thought we were all allowed to make a comment on here raindancer, don't you know it's considered rude to shout.
-
raindancer, please calm down. I did not mean to cause any offence, simply fill in some details that explain F1's current position with regard to CFD (note I put in a reference from an article quoting an actual F1 team, who would presumably be in a better position to explain their computing requirements than either of us). F1 rigidly follows the Pareto Principle, not only with regard to time, but every other resource it can get its hands on. Computers happen to be just another resource for F1 teams to exploit in their eyes.
-
Ali, how would you reckon Williams' supercomputer would compare to someone like McLaren's, given the differences in budgets between teams? Also, do BMW Sauber get to keep that massively expensive one BMW brought in?
-
Calm down raindancer,everyone is entitled
to their own opinion.
-
Also, do BMW Sauber get to keep that massively expensive one BMW brought in?
I suppose if its at Hinwell then Sauber gets it, mind you with advances in computing power still continuing apace Albert II the BMW super computer may now be less cutting edge than a couple of years ago.
-
Very true, yesterday's super computer is today's desktop. Didn't seem to do BMW a lot of good though, given how much they hyped it up!
-
Ali, you go ahead and carry on posting, if I have a question on something it's you who I'll ask. :good:
-
Ali, you go ahead and carry on posting, if I have a question on something it's you who I'll ask. :good:
True, if I have a specific question about something, I more often than not ask Ali in the post for an answer!
-
Ali ! I think you should leave it after I have made a comment. BecaUSE i KNOW WHAT i AM TALKING ABOUT HAVING SPENT 22 YEARS IN COMPUTERS
If you can't see what Ali is talking about Raindancer perhaps you will agree Adrian Newey knows what he is talking about, the following is a quote from a press conference at the RBR 2010 car launch. Unless I have this all wrong he seems to quite clearly say that proper CFD work in F1 requires a awful lot of computer power, time, and possibly a highly intensive program as well.
"CFD is an electronic simulation of a real environment, but it still has pitfalls - not least that every single run in CFD for a given attitude of the car, or ride height, or whatever it might be, is a discreet run. Whereas in the wind tunnel, what we call a normal run, will have 20 or more data points in it. In other words, that is equivalent to 20 runs in the CFD. That is a limitation of size really, so your CFD cluster has to be that much bigger to do that many runs."
-
Never meant shout Apologies. YOu wont believe it but my caps lock key malfunctioned at that precise time. I got a new keyboard now.
-
Folks Relax this is not about Ali or me. All I was saying is most people talk about brute force when talking computers and that is what I was refering to and not contradicting anyone least of all Ali.
Ali I was only commenting on comparing billions of operations /sec as computing speed. Floating point operations / Second or FLOPS as they are know signify the number crunching ability of the computer in short brute force.
There are different types of computers and they today use processors paralelly to deliver on multiple tasks. Gigaflops, Tera flops, Peta , Exa and so on only go on to tell you the number of FLOPS a computer can do.
Moreover in building computers various technologies and architectures are being used compared to the olden days of CISC or RISC machines. SPARC and massively parallel computing has taken root in the world and application usage requires more than purely number crunchers. The ability of the computer to run programs, handle video , Audio and so many other inputs are crucial for building Super Computer architectures.
I Agree with Adrian who is talking more from a Wind Tunnel perspective.
-
And for everyone who encouraged Ali to post ! I have not tried to discourage her , all I meant was such info is easily avaiable over Google if someone cares to search instead of we having an argument continously on computers.
-
Ali, how would you reckon Williams' supercomputer would compare to someone like McLaren's, given the differences in budgets between teams? Also, do BMW Sauber get to keep that massively expensive one BMW brought in? {Williamsfan - 9 posts from end of page 2}
Williams has one of the smaller supercomputers in F1, though they have seriously upgraded it within the last two years (before that Force India had the same capacity as they did). So Williams no longer has the smallest in-team computational capacity for the established squads (Force India have less, but they've recently invested in some distributed computer capacity and Toro Rosso was sharing with Red Bull and thus probably has less capacity than either Williams or Force India in-house).
I'm not sure what McLaren's level is, except that it is somewhere between Williams' and BMW's. I'm not sure Albert2 was even given to Sauber by BMW, but if it was, there may be restrictions on the amount that can be used. (As raindancer said in his post, supercomputers are much more modular than the olden days, so "downgrading" is possible, if extremely inconvenient for the IT department).
-
:yahoo:
There you go Ali. You are an informatiopn pwerhouse. Mean it sincerely.
-
Personally I wouldn't advise any member to question Ali's knowledge on F1. She really is a true fountain of F1 facts and figures. I have also worked in computers all my life, but if I wanted to know about the computers used in F1 - I would ask Ali. :good:
-
Personally I wouldn't advise any member to question Ali's knowledge on F1. She really is a true fountain of F1 facts and figures. {Wizzo - previous post}
:-[ :-[
(By the way, it is OK to question me. I admit to not knowing everything...)
-
My guess... F1 will start partnering with universities, and letting donors cover costs of 'research' equipment. What engineering programme wouldn't want access to McLaren's shop in the off season? (Our American friends will recognise an NCAA type problem here.)
Interesting discussion here, I found the above quote of particular relevance.
Back in the days of my Aeronautics degree at Southampton (mid 90s) Williams were using the University wind tunnel for a lot of their work. This brought the Uni some much needed cash I imagine, but more than that, the buzz going round the department after Mansell's 92 championship victory was quite significant. Sadly shortly after that all the team started building their own wind tunnels. Also, the fact Newey is a Southampton graduate helped...
I think getting universities more involved in F1 again would be great step for the sport to give something back to the engineering community at large.
-
Ali ! I think you should leave it after I have made a comment. BecaUSE i KNOW WHAT i AM TALKING ABOUT HAVING SPENT 22 YEARS IN COMPUTERS
I thought you were a Mergers and Acquisitions consultant? Not picking a fight - just asking?
-
My guess... F1 will start partnering with universities, and letting donors cover costs of 'research' equipment. What engineering programme wouldn't want access to McLaren's shop in the off season? (Our American friends will recognise an NCAA type problem here.)
Interesting discussion here, I found the above quote of particular relevance.
Back in the days of my Aeronautics degree at Southampton (mid 90s) Williams were using the University wind tunnel for a lot of their work. This brought the Uni some much needed cash I imagine, but more than that, the buzz going round the department after Mansell's 92 championship victory was quite significant. Sadly shortly after that all the team started building their own wind tunnels. Also, the fact Newey is a Southampton graduate helped...
I think getting universities more involved in F1 again would be great step for the sport to give something back to the engineering community at large.
Williams (and I am sure others, but I know Williams more specifically) do offer some kinds of schemes for the wider engineering world still, but I guess it is a lot harder to source out in the same way as had been previously seen. Here is some blurb from the Williams site:
2. Education
Williams has long supported a range of education initiatives, from vocational to academic and from primary to tertiary. The support has been equally varied, from providing financial bursaries (for instance for PhD students at Cranfield University) to assisting with curricula development (Oxford Brookes University) as well as annual vocational placement schemes for secondary school pupils and an active apprenticeship scheme.
In 2009, Williams formed a joint venture with Cambridge University Press, one of the world’s leading and longest established educational publisher. With the combined attractiveness of Formula One to school children and the sector expertise that Cambridge University Press has in educational software, the partnership created Race to Learn, a Formula One-themed teaching product for 7- to 11- year olds.
The product, which uses Formula One as an exciting point of engagement for primary school children, provides cross-curricular learning for Year 5 & 6 pupils providing 12 half-days of teaching support in many subject areas. It uses a range of multimedia content including real Formula One footage to promote individual and group working. Designed for interactive whiteboards, Race to Learn covers key curriculum topics – Science, Maths, Literacy, Geography, PSHE, Design & Technology, Physical Education – with each one linked to the relevant National Curriculum/Framework objectives.
Six months after its domestic launch in the UK, Race to Learn won a major industry award, the British Education & Technology Training (BETT) Digital Content award in the Primary category. It was commended by the BETT award judges as containing “superb activities that are highly engaging for children” and “with helpful teacher introduction and age-appropriate activities, Race to Learn is a well thought through support for cross-curricular learning.”
In addition to industry commendation, Race to Learn has also been wholeheartedly endorsed by the teaching profession and the education media. Reviewed recently in the leading trade title, Teach Primary!, the product scored 10/10 and attracted commendations that included “the activities have been carefully thought out and the children loved them.”
The initial success of Race to Learn has encouraged Williams F1 and Cambridge University Press to expand the project overseas, with foreign language versions currently in development. For more information, see www.racetolearn.org
-
(By the way, it is OK to question me. I admit to not knowing everything...)
Dang, Ali. I was about to ask you about some Lottery numbers. ;)
Seriously, you never fail to shed light on the murkier areas of the sport. Thank you very much!
-
I was about to ask you about some Lottery numbers. {cosworth151 - previous post}
2, 5, 9, 12, 35, 42...
...probably won't be the winning numbers on Saturday ;) Or maybe they will be - what do I know?!?
And you're very, very welcome. It's a pleasure to share and you teach me things sometimes as well.