Pretty much.

Quote from amp88 :

Increasing timing precision to thousandths of seconds has been discussed a bit before. The problem is that to have an accurate position every 1000th of a second the physics loop would have to run at 1000Hz (an increase of 10x the current rate). This would further increase the burden on the CPU, which is already fairly heavy. The increase in precision (which would definitely be useful in certain scenarios especially) is desirable, but not if it comes at the expense of requiring much more powerful PCs to run LFS.

I'm no computer science major, but what if you made a modification to the timing function which only recorded in 10^-3 sec increments when it was deemed necessary as per scenario (e.g., during qualification, if the program estimates that the final difference in drivers' recorded times will be less than some threshold (say, 0.02sec) begin recording in thousandths. Of course, this potentially lag-inducing procedure would only be initiated in the last 0.03sec, or so, before crossing the finish line. If lag were to result, the time interval over which it occurs would be too short to be of any consequence to driver experience/technique).

Anyway, that's my two cents on the problem--no idea if it's feasible. Like I said, I'm not really a programmer.

Also, here's another idea--something I think would be really handy:

It seems kinda standard (among racing simulators/games) to include a historical database display of personal track times.

Inclusion of a 'Results' or 'History' page with various data fields would be great--display:
- driver
- track ID
- track times (lap times and overall times)
- number of laps
- conditions (e.g., wind, number of players/AI, setup, etc.)
- penalties incurred

This would allow the user to look back at all past performance, and note their personal progress over time (hopefully :razz

Additionally, maybe these stats could be used in multi-player arenas which require certain experience levels (determined by track times, as is often the case) for driver participation.

I think this would be a lot better than using an external program, or someone's clunky post-release add-on.

Quote from DerRouge :

I'm no computer science major, but what if you made a modification to the timing function which only recorded in 10^-3 sec increments when it was deemed necessary as per scenario (e.g., during qualification, if the program estimates that the final difference in drivers' recorded times will be less than some threshold (say, 0.02sec) begin recording in thousandths. Of course, this potentially lag-inducing procedure would only be initiated in the last 0.03sec, or so, before crossing the finish line. If lag were to result, the time interval over which it occurs would be too short to be of any consequence to driver experience/technique).

Anyway, that's my two cents on the problem--no idea if it's feasible. Like I said, I'm not really a programmer.

While it seems like a good idea initially, I am sorry but it is a terrible idea in practice. I'll (hopefully) explain why.

First, the lag that gets caused, even breifly can - and inevitably WILL effect the player. There are times when I've forgotten to shut an IMer program off, and someone IM's me locking my computer up, causing LFS to lag for a single moment, yet I totally miss a turn because of it. Also, and the worse part about doing this, there is an issue that you have to play catchup. I will try to explain, but it is a technical issue with timing.

Say you simulate 0.01s every 'update'. This doesn't necessarily mean you simulate 0.01s every frame. Some frames you might not run the simulate portion, others you might run it 2 or 3 times. Which isn't terribly noticable. However once you get behind, and the framerate drops it gets exponentially harder for the computer to catch up. There are some ways around this - and LFS may have some protection already but it is a problem to consider. Another bad thing is the car could behave odd for that small time slice when nearing a split, finish line etc for your more accurate timing.

I am not quite convinced that computing the exact time, by using the results of the acceleration through-out the simulated time is inaccurate. (in LFS's case 0.01s). As I said above I agree it could be slightly less accurate if the car could accelerate during that hundredth of a second, but if it could do that the simulation loop would need to run faster anyways, giving more precision. Slightly off my reason for posting . . .

I don't want to pop the bubble, but I'm actually more interested now in the other suggestion, the setup storage one (basically because timing has been already deeply discussed).

For someone who has been active for a couple of years it's very common to have tons of setups for every car. I really like the idea of allowing subfolders into the setups folder so we can have them sorted as we like. We could create folders to our desire in "{LFS}\data\setups", and then in the car setup screen we would see:

- Relative path showing the current folder.
- An entry for the parent folder ("..")
- Folder names on top with a different font colour.
- Setups contained in the current folder.

When you create a new setup it is stored in the current folder.