Jump to content
Yngwie

Race Coordinator (Software)

Recommended Posts

Hi Dave

Phil and I are talking about the same race night on the same track.

Might give you a clue as to what is happening.

I'll try to get you the files in question.

 

Andrew


A man without a woman is like a neck without a pain.

Share this post


Link to post
Share on other sites

Andrew / Phil...

 

I had a look throught the genereated xls results file from our Group C races, and I've noticed a potential issue...

 

In a few of the races, this scenario seemed to pop up with the winners...

Total Full Laps: 17

Fastest Lap: 7 seconds

Total time take for 17 full laps: 118

Total race time: 120 seconds

Calculated partial results: 17.9

 

What seems to be happening is that the last 2 seconds is somehow being calculated as 0.9 of a lap, instead of approx 0.35.

Not sure what is currently happening in the partial lap algorithm, but somethings not quite right in there...

 

Edit: This incorrectly calculated 0.9 is probably what's affecting the Average lap times too...

Edited by shadow_rusty

Share this post


Link to post
Share on other sites

Looks great Matt. Care to share your work!

 

Certainly, I just need to figure out how to attach a file to a forum post.

Share this post


Link to post
Share on other sites

Matt,

 

You might want to put view boxes around the leader board names in the upper right. That way you can control how much space they take up and guarantee there's room for the full name and score. Just a thought. Otherwise it looks great.

Share this post


Link to post
Share on other sites

Andrew/Phil/Rusty,

 

Are you guys using the "Allow Finish Single Lap (auto segments)" option? I can explain how it auto generates the segments, which can be confusing and not always very accurate, but I didn't think segments affected the average lap time. It would be a bug if it did and the only way it could do what you're saying is if your best lap and your average lap were very close together. Are you guys running super consistent lap times?

 

-Dave

Share this post


Link to post
Share on other sites

By the way Matt, if you really want to get cool, look into animation. You could crossfade in and out that bottom ribbon to change the data there, or animate it to the left making it a ticker... At one point I had a 24 driver race screen that cross faded out the driver heat data so you could see more information but it was too annoying to look up and not see yourself on screen if you happened to look at the wrong time so we ditched it.

Share this post


Link to post
Share on other sites

Yes Dave we are using "Allow Finish Single Lap (auto segments)"

 

And on many of the heat results it is recording 12.01 or 11.01, instead of say 12.095 or 11.255.


A man without a woman is like a neck without a pain.

Share this post


Link to post
Share on other sites

Andrew/Phil/Rusty,

 

Are you guys using the "Allow Finish Single Lap (auto segments)" option? I can explain how it auto generates the segments, which can be confusing and not always very accurate, but I didn't think segments affected the average lap time. It would be a bug if it did and the only way it could do what you're saying is if your best lap and your average lap were very close together. Are you guys running super consistent lap times?

 

-Dave

Several of the races run, the leading drivers had no offs, and as such were quite consistant.

 

After much testing with the "Allow Finish Single Lap (auto segments)" option, I can't get it to give anything other than .9's and .0's...

Allow Drift, setting a drift, changing number of sectors all have no impact on this.

 

If I use "Allow Finish Single Lap", the the software works correctly and the drivers finish in the correct order.

 

Also, I've noticed that the "Allow Finish Single Lap" will actually record the last lap to the .xls file whilst the "Allow Finish Single Lap (auto segments)" won't.

 

Note: For testing I've been using using the Demo Mode with the following main settings...

60sec time race

Allow Finish Single Lap / Allow Finish Single Lap (Auto Segments)

Round Robin

Ranking by Lap Count

Accumulate Heats

4 drivers

2 lane 20m track

Demo Minimum Reaction - 0.01

Demo Maximum Reaction - 0.5

Demo Minimum Lap - 6.5

Demo Maximum Lap - 8.0

Share this post


Link to post
Share on other sites

Looks great Matt. Care to share your work!

 

Certainly, I just need to figure out how to attach a file to a forum post.

I don't think you can, best option is to email it to John then we will be able to play with it

 

If your interested i have also modified a couple of one lane formats as well as having 60 places in the results section (great for viewing proxy races)


Quickly read this post before it is deleted or i turn grey again

Gary

http://www.facebook.com/Rallyproxy2017

Share this post


Link to post
Share on other sites

Hi Phil,

 

Same output with real race results, from our club race last Friday night for one heat with my manual calculations at the bottom for a comparison:-

 

Total Laps from RC V1.7.1,15.01,16.989,16.983

Lap Number,,,

1,8.432,8.242,8.017

2,7.423,7.257,7.21

3,7.433,7.537,7.468

4,7.389,7.357,7.302

5,7.659,7.285,7.155

6,7.524,7.489,7.361

7,7.378,7.497,7.192

8,7.15,7.352,7.259

9,13.386,7.528,7.081

10,7.723,7.268,7.268

11,7.749,7.215,7.29

12,7.571,7.308,7.054

13,7.581,7.168,7.021

14,7.741,7.199,7.093

15,7.731,7.053,7.069

16,,7.11,7.074

Manual Calculatios:-,,,

Time for all full laps,119.87,117.865,115.914

Average time/lap,7.991333333,7.3665625,7.244625

Time short a end of race,0.13,2.135,4.086

Estimated lap completed,0.016267623,0.289823103,0.564004348

Estimated total laps,15.01626762,16.2898231,16.56400435

Sorry the editor does not support tables, so had to paste as a CSV.

 

Summary:

Total Laps from RC V1.7.1, 15.01, 16.989, 16.983

 

Estimated total laps, 15.016, 16.2898, 16.564

I have also done tests with the V1.7.0 RC version with the same issues.

 

 

Regards from Phillip.

Edited by QRphil

Share this post


Link to post
Share on other sites

Checked v1.6.3.18 and Digital v1.0.0.0, and they seem to have the same bug.

All partials start with either a 0, or a 9.

Share this post


Link to post
Share on other sites

We were using this system as it seemed to be a good way of not having to waste time on adding manual laps but did away with it as we too found these errors you are finding.

Share this post


Link to post
Share on other sites

I was going to write up how auto segments work, but instead I'll say its complicated, confusing, and not always accurate (of course). Looking over the code one thing that can really mess it up is if you have a very small number of track sections configured for your track. For this option to even remotely work I recommend setting the Track Sections to either 0 (which allows me to do whatever I want, but doesn't allow you to use track sections yourself) or set it to 100 or higher. You can think of 100 as a percentage around the track...

 

The other thing to note is that if your lap times are not fairly consistent this calculation won't be accurate, especially if the driver deslots on their last lap. Essentially, if the time it takes for the driver to complete the last lap after the heat officially ends takes longer than the drivers average lap time, the driver will be awarded the smallest amount of sections possible (0.01 if the segments are set to 100) because RC thinks the driver just crossed the line when the heat ended.

 

@QRPhil, there's not enough information in your data for me to see what's going on. Now that I understand the problem a bit better, I think the best thing for me to do is to add some debug code that logs the math involved in the auto segment code... from that we can figure out what RC is doing and why, and if there's an issue here... I have to do a beta release very soon so if you're interested in trouble shooting this through my logs let me know.

 

Also, I thought the original problem was that sometimes your fastest lap time was larger (slower) than your average lap time? Under no circumstances should that be possible unless you have a lot of dropped fastest laps due to drifting which isn't likely....

 

-Dave

Share this post


Link to post
Share on other sites

Tried "partial laps" with sectors set to 0 and to 1000, and the same issue occurs.

 

In our club, there are a significant number of races where the leading cars do not have a deslot, so the average lap can be quite high when compared with the best lap, but as you say, should never beat it.

We also do not use a yellow flag situation, so drift is nil.

 

----------------------------

 

Had a bit of a think about it, and as we are looking for a percentage of a lap, would it make sense to do the following calculation for partials...

"Percentage Partial" = "Lap Time" / "Time Remaining"

 

If you wanted to calculate sectors instead, then the Partial could then be calculated into the correct number of sectors by using...

"Whole Sectors" = Round or Truncate ("Percentage Partial" * "No. Sectors") To the Nearest Whole Number

"Final Partial" = "Whole Sectors" / "No. Sectors"

 

You would use "Round" for calculating the closest sector line, and "Truncate" for calculating the completed sectors, which could even be a user setting...

 

eg.

Final Lap = 12 Seconds

Time Remaining = 4 seconds

Sectors = 40

Percentage Partial = 4/12 = 0.333

Whole Sectors = Round(0.333 * 40) = 15 (Or Truncate(0.333 * 45) = 14)

Final Partial = 15 / 40 (Closest Sectors), or 14 / 40 (Completed Sectors)

 

It would possibly require a setting for 'how long to wait for the final lap to finish'

Cars that don't finish in that time frame would get 0 partials...

Edited by shadow_rusty

Share this post


Link to post
Share on other sites

That's basically exactly how the auto segments work. Except that I use the Average lap time as a basis. Meaning if the driver goes for 8 seconds after the heat ends of a 10 second final lap, it's assumed by RC the driver did 20% of the lap before the heat ended so they 20% of the total segments (100 if you set it to 0). I use Average lap time to determine if they should get only a single segment. So if that 8 seconds is higher than their average lap time I assume they had just crossed the line when the heat ended and give them the smallest segment amount possible.

Share this post


Link to post
Share on other sites

By excluding laps that are longer than your average, you can severly penalise someone in the following situation...

 

Avg Lap - 7 seconds

Time remaining - 6 seconds

Last lap - 7.5 seconds

 

They should have gotten 0.8 as their partial, but end up with 0.01.

If that happens 3 times, once in each lane, they end up penalised by 2.4 laps.

 

 

Also, in about 20 test races, and 10 real races with 12+ drivers, I have never seen a driver awarded a 0.2 segment .

The segments added are always 0.0's or 0.9's.

Given the above, consistant drivers are being penalised with 0.0's, while drivers that crash a couple of times, are getting 0.9's...

This has resulted in several wins / places going to the wrong driver.

Share this post


Link to post
Share on other sites

I have never liked the option of any program Calculating the part laps.

There are just too many variables to be taken the equation.

The only real way to do this is to have sensors at every 10th lap.That will reduce the error.

 

 


Phil

 

Hobart Miniature Car Club

 

Tassie Resins

 

Email

 

 

 

 

 

 

Logos%2016_17.small_zpswkcwjf0q.jpg

Share this post


Link to post
Share on other sites

I agree with you Kal, but I had a lot of requests for this feature so it became an option....

 

Shadow, the lap time after the heat has ended has to be longer than the average lap time for RC to knock it down to the lowest possible segment counts. Grab the latest beta release and email me. At the last second I threw in some debug logs that output all the numbers in the calculations. All you have to do with the latest beta is turn debug logging on (set it to 0) and run your heat. Exit the RMS and send me the logs as soon as you feel it was wildly wrong and tell me what happened and why you think its wrong. You can find the messages in the logs yourself, but with the logs I can track through and tell you exactly what happened in code.

Share this post


Link to post
Share on other sites

New beta release posted.

 

****If you use phidget to do lap counting/refueling you'll want this release. There was a subtle but pretty major problem with that particular part of the interface.

 

There's also also new support for the "Titus/Befpe" track interface. If you have that hardware and want to try a different RMS vbglossarlink.gif give RC a shot. A few other changes/fixes in this release. It's been so long now I can't remember ):

Share this post


Link to post
Share on other sites

Thanks Dave.

 

I'll give it a go, and let you know...

 

 

ps.

This is great software, with a nice interface, which is extremely flexible...

Certainly not wanting to discredit the app, as over all it is brilliant...

Share this post


Link to post
Share on other sites

No worries here Shadow... by the way I forgot to address one of your other posts directly so here's some comments below. Your original post is marked by ">" and my comments are marked by "DPA:" One more thing, the auto segments is never going to be very accurate. There's a lot of things that can make it wildly wrong. If your drivers never deslot so their min lap time and max lap time are very close together, thus making the average lap time close to those two values, the calculation should be fairly accurate, but the more deviation in those numbers and the more deviation the driver has in that final lap, the less accurate the calculation will be. This is why I personally don't like the feature. The good news is that you can evaluate it's accuracy each heat and if you want to you can adjust it by hand...

 

On to the other comments:

 

>After much testing with the "Allow Finish Single Lap (auto segments)" option, I can't get it to give anything other than .9's and .0's...

>Allow Drift, setting a drift, changing number of sectors all have no impact on this.

DPA: This sounds like a bug or something. My new beta which logs the math behind the auto segments will help figure that out.

 

>Also, I've noticed that the "Allow Finish Single Lap" will actually record the last lap to the .xls file whilst the "Allow Finish Single Lap (auto segments)" won't.

DPA: That's correct. The last lap in a an "Allow Finish Single Lap" is a lap that occurs after the heat ends. RC leaves track power on so you can finish the lap and it can try to create the partial laps. But the lap absolutely should not count in the drivers total.

 

>Note: For testing I've been using using the Demo Mode with the following main settings...

>60sec time race

>Allow Finish Single Lap / Allow Finish Single Lap (Auto Segments)

>Demo Minimum Reaction - 0.01

>Demo Maximum Reaction - 0.5

>Demo Minimum Lap - 6.5

>Demo Maximum Lap - 8.0

DPA: This is a pretty fast heat (60 seconds) with a fairly large difference in lap times. It might work, but the results will be hard to predict. A better way to do it might be to set the min/max lap times to say 5.0 (both of them). Then set the heat time to like 12 seconds. In theory you should get 2 laps in and the third lap will have 2 of the 5 seconds run before the heat ends. The problem is that demo mode might not be that accurate to the timing. So even this might not be right. The key to all the tests though is to accurate observe how much time elapsed from the last counted lap to when the heat ends. From there, the debug logs will show the math which should be in the ball park of what you observed.

Share this post


Link to post
Share on other sites

Ok here goes...

I did a bunch of testing tonight with both the older 1.6.3 and 1.7.0 versions before upgrading to, and testing the 1.7.1 beta version.

I used both the demo mode, and real racing on my 2 lane scaley track.

 

Firstly, the new version is great... Lots of small tweaks that make the whole lot nicer.

Secondly, the partial laps bug is still there... (Not really suprising as only data logging was added for this)

 

So, whilst looking at the log data, I believe that I've found the source of the problem...

When calculating the partial, it appears that the software is dividing the last lap by the average lap. This is why the partial is usually a high number.

It should be dividing the time left (from the previous lap to the time cutoff) by the last lap.

 

I can send some logs through if you would like, but it's very easy to replicate with the demo mode...

It's also easy to see what the partial should have been... First, save the race results to the xls file.

Subtract the "reaction time" and all the individual "lap times" from the "race time" to give the "time remaining"

Divide the last lap time (only in the logs) by the "time remaining"

 

Is it possible to have a cutoff time limit instead of using the average laptime?

Also, is it possible to have the final lap (which is really a partial lap) time added to the race results?

 

Hope all that helped...

Thanks

Michael

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...