[teknochild]

Originally Posted by shiv@vishnu

There isn't much we can do with bad data (wheel spin) except somehow try to filter it out. Perhaps someone can think of a maximum rate of acceleration and ignore anything that exceeds that (dismissed as wheelspin). In the datalog I've emailed to everyone, you can see some of it in first gear as well as a tiny bit shifting into 2nd. Looks like little fuzzy spike Any other ideas would be greatly appreciated!

-shiv

you cant really do that either, during that wheel spin the car is still accelerating pretty rapidly, if you just throw that section out youll get the opposite effect as leaving it in, a slower/lower estimate, i guess you could take the numbers from after the wheels spin subtract the ones from before it, and disperse the leftover in the discarded area, yea actually that would probably work best

would only matter if your going for a very exact 1/4 mile estimate (the way i was thinking wouldnt really be an estimate )

[RiXst3r]

How about a web based app? (php, flash)... it could still run off a local disk, but could also be hosted on the vishnu website... I could do that pretty easy... but normally i would charge more than $1300...ohwell, as long as its for a good cause... did anyone else already start development???

[DrM]

Originally Posted by RiXst3r

How about a web based app? (php, flash)... it could still run off a local disk, but could also be hosted on the vishnu website... I could do that pretty easy... but normally i would charge more than $1300...ohwell, as long as its for a good cause... did anyone else already start development???

I don't know if anyone has actually received the CSV file yet. Shiv said that he was going to email it out. Without the data and a more clear definition, I don't believe that anyone could actually start working.

I had the same thought as you. I do a lot of work in Ruby/Rails and thought that a web based app would be better, but I am unsure of the exact goals of the project.

I am sure that one of us in the E90 group will be able to do it.

I noticed in one of his recent posts that Shiv was waiting on "software" before shipping the units out. I would be HAPPY to build the thing if it meant I would actually get my Xede sooner...

[OpenFlash]

Hi guys,
there are already a few people that have begun development of the dyno/drag emulation software. In fact, I already got an early release that does 80% of the drag race sim. If I forgot to send the datalog info to anyone, shoot me your email addy and i'll get it out to ya!
-shiv

[AlexGn]

Here's what I got so far... First graph is from raw data and the second one is corrected for wheelspin. And of course there is a nice GUI on top of that.

[DrM]

Well, it looks like we are in good hands with AlexGN!

Although, from looking at the graph, I have to ask:

How did you run 13.30 through the quarter with the speeds dropping off so badly and ending at 85mph?
Are the calculations correct? Surely the scale of the graph is off and the placement of the labels is off. I would think that the quarter mile ends at the peak of that graph.

[OpenFlash]

No, the data is fine. Those logs were taken on the street and I got on the brakes before the end of the 1/4 mile. A stop sign was approaching really quick

[AlexGn]

Originally Posted by DrM

Well, it looks like we are in good hands with AlexGN!

Although, from looking at the graph, I have to ask:

How did you run 13.30 through the quarter with the speeds dropping off so badly and ending at 85mph?
Are the calculations correct? Surely the scale of the graph is off and the placement of the labels is off. I would think that the quarter mile ends at the peak of that graph.

That's exactly what I thought when I first saw the data

Now, the easy stuff is done, calculating hp / torque is less trivial.

[BMWPower06]

wow thats awesome, cant wait to see the graph of an actual 1/4 run. 13.3 and u slammed on the brakes, thats impressive

[RiXst3r]

Originally Posted by DrM

I don't know if anyone has actually received the CSV file yet. Shiv said that he was going to email it out. Without the data and a more clear definition, I don't believe that anyone could actually start working.

I had the same thought as you. I do a lot of work in Ruby/Rails and thought that a web based app would be better, but I am unsure of the exact goals of the project.

I am sure that one of us in the E90 group will be able to do it.

I noticed in one of his recent posts that Shiv was waiting on "software" before shipping the units out. I would be HAPPY to build the thing if it meant I would actually get my Xede sooner...

I have started working some in ruby on rails, but I mostly just code php and java... I was looking at this for the graph side of things, free, and cool looking...

http://www.maani.us/charts/index.php...y&submenu=Line

and this...

http://www.maani.us/gauge/index.php?menu=Gallery

[CoralRedPlz]

Originally Posted by AlexGn

Here's what I got so far... First graph is from raw data and the second one is corrected for wheelspin. And of course there is a nice GUI on top of that.

omg. 0-100 in 9.3sec??

is that time accurate? if so, wow.

[Hegemony]

If I'm reading that graph correctly, you were in 3rd gear at 110. Confirm?

[OpenFlash]

Originally Posted by Hegemony

If I'm reading that graph correctly, you were in 3rd gear at 110. Confirm?

Not quite, the wheelspeed calibration was off slightly. I think 3rd gear only goes up to 105mph or so. That was just a sample file I had in my laptop. Don't mistake it for a proper performance test as the road wasn't even flat!

cheers,
shiv

[timn]

Shiv,

I emailed you about the csv file, but am still waiting for it. That is, if you're still sending it out.

[Terry335]

I agree, sounds like a 40-60 hour project. Hardly worth the chance of getting $1300. Maybe you can outsource it to India?

[teknochild]

i dont know where you guys are coming up with 40 hours, im not even out of college and i know i could write one w/o the graphing that could spit out plots in a few hours, 10 at the absolute max

[sdiver68]

Originally Posted by teknochild

i dont know where you guys are coming up with 40 hours, im not even out of college and i know i could write one w/o the graphing that could spit out plots in a few hours, 10 at the absolute max

You will learn young grasshopper, you will learn.

[Hegemony]

Originally Posted by sdiver68

You will learn young grasshopper, you will learn.

Yes
You will
learn that when
getting paid
by the
line
to write
like this
!

Originally Posted by shiv@vishnu

Not quite, the wheelspeed calibration was off slightly. I think 3rd gear only goes up to 105mph or so. That was just a sample file I had in my laptop. Don't mistake it for a proper performance test as the road wasn't even flat!

cheers,
shiv

I was just looking at the flat portions of the graph and figured that must have been shift points. I was going to rib you a little for taking a half second to shift

[Jonn335]

I was going to have a go at this, but there are so many people well ahead of me its not really worth it. I will pass on the idea i have for making the dyno-sim part of it much more accurate. That is to do a deceleration run over the same speed range out of gear (only in the MT car).this will give data for the rolling and air resistance. with this taken into account the only variable not accouted for is transmission losses, which is the same as on a chassis dyno. It means you need 4 sets of data per run, accelerate and declerate over the same road in diferent directions.
A nice long strait road will be needed as it will take some time for the car to decelerate.

[teknochild]

Originally Posted by sdiver68

You will learn young grasshopper, you will learn.

w/e

[BoLaS]

wow u guyz are freaking FAST! jeeze... :rocks:
So who here thinks they can add on some code for our NAV to play normal DVDs....?

is it possible?

[FunRevn]

Love the Java GUI.....