Nathan, will you list the steps to test my newly tuned sims. I used the second week at the meet for the tuning and I want to test against the third week.
I saved everything as per your tuning sims post.
How do I get SPAT to test this third week for the tuned settings an also for the untuned settings?

Will,

That's a darned good question -- there is a way. But, it could be a better way (I'll present). I almost implemented it today to do some testing of my own -- but I'm sure you folks want V3.5 before next year.

Here's what I would do:

Base testing of TUNED vs. UNTUNED

Keep two databases. One with tuned results, the other with non-tuned results.
To implement:

• Turn "Use Simulation biases" (Sim menu).
  
• Load card(s)
  
• Connect to TUNED DB, run all sims, store to db.
  
• Turn "Use Simulation Biases" off. Connect to NON TUNED DB, re-run all sims, store to DB.
  

This SHOULD work dandy without closing the cards and re-opening them when you switch between "use biases".

For your particular question, I would simply load the "week 3" cards and go through the steps above. That way you'd be applying Week 2 biases to week 3, and also getting Week 3 results with no biases. I'm sure you can also imagine a way to do this in an on-going database. The un-tuned DB would be easy. The tuned DB would reflect how often you tuned.

TO COMPARE:

Now, start TWO instances of EquiSim. One should connect to the TUNED DB, the other to the NON TUNED db. Put them side by side. Go to the SPAT in each, compare visually.

---------------
A possible "better" way. There's no way to get around the fact that the Database can only store one set of simulation results. Well, there is, but you don't want me to do that. So, if you want to test TUNED vs. UNTUNED, the best method is to create two databases. EquiSim has several facilities for generating databases on the fly (store by folder, multi-load cards and store them all quickly, etc), so this isn't a big deal even if you're comparing massive data sets. V 3.5 makes it even easier. You can delete the extra database files when you are done with them if you don't have a huge hard drive (cheap these days).

So, we're still on track with 2 databases -- one TUNED the other UNTUNED. The problem I see is comparing them. I think it would be nice to have a command that says "compare two databases". It would go in and give a multi-level report on the differences between the two databases. At the highest level, you'd see differences between Win Perc, ROI, for every race. At lower levels, the breakdown for each race type. At the lowest level, a race by race breakdown.

Still haven't thought this one through -- the happy medium is always a balance between the amount of work you have to do and the amount of information you get in return. Any thoughts welcome.

-Nathan