Misc ACC Results Companion 1.24.3

[omg.imba]

Yes, if you would either add a check box to "Exclude laps with missing sector times" or simply regard Race Lap1 as an out lap, this would work nicely.

 

[Doug Duthie]

Doug Duthie updated ACC Results Companion with a new update entry:

ACC Results Companion V1.12.4

• Rework leaderboard
  • Address performance issue
  • BUGFIX: Minor possibility of wrong track being reported
• BUGFIX: Player name not reported in single player (SRO driver was reported instead)
• Specify overrides for results summary pop-up
• Log settings moved to log page
• Filter on backups page
• Minor changes to leaderboard blacklist filtering

Read the rest of this update entry...

 

[Doug Duthie]

ACC Results Companion V1.12.4 Released (here)

This update is mainly about bugfixes/improvements rather than features

Leaderboards
The leaderboard was becoming very slow. With nearly 9000 entries on some tracks, this was becoming unmanageable. Previously, the fastest laps were extracted directly from the timing tables with a single query. While this had the advantage that any timing changes were instantly reflected on the leaderboard, it was incredibly inefficient for the large table sizes.

Now, leaderboards will be a static summary of the fastest laps only (updates only at session end/when leaderboard records are received). This makes both the leaderboard and circuits screens much faster to update. Now the bottleneck is just rendering the results in the page rather than the database query.

In addition, the previous query, which was horrendous, looks like it had a bug in a small number of cases. For some users, it would report times against the wrong track on the global leaderboard. For example, on the longer tracks (eg Spa) you may see the leaderboard headed by suspiciously low times. I believe this is a manifestation of the bug and the leaderboard records should be corrected when those users update, and their records will be corrected.

A final benefit is now leaderboards don't use the same tables as the regular sessions, the space requirements are much smaller. Once the DB maintenance task runs (once every 30 days normally), the DB may shrink by about 150MB.

Leaderboard Blacklists

A filter has been added to the leaderboard to let you show or hide blacklisted records (aimed at those bad records described above). You can also un-blacklist a record by right clicking on the row.

Results Summary Pop-up
I finally got fed up with the behaviour of this. It attempts to display in the foreground at the end of the session. However, at certain points ACC will also jump to the foreground. In an attempt to "win" the foreground, the pop-up repetitively retried for 5 seconds. This became really annoying. Now you can specify how many times the pop-up will try this, and the interval between retries.

Other Bug-fixes/changes

• Fixed bug (from 1.12.2) in SP events (Hotlaps, hotstints, practice, qualifying) where the virtual driver name was displayed rather than the player name
• Log settings have been moved to the log page
• Filter added to the backups page (only some of the filters work at the moment)
• The sessions Lap Record/Personal best filters have been changed on the sessions page as they had poor performance. At the moment the combined filter (Show/Exclude both LR & PB) is non-functional

 

[omg.imba]

Hi Doug,

just noticed in the Car/Track History that BrakeBias seems to be calculated wrongly. Session reports show correctly 63.2% for my Lexus, but the same laps on the history tab are shown with 77.2% BB. This 14 points difference matches the -14points offset from the telemetry. So probably the offset is not applied there.

 

[Doug Duthie]

Hi Doug,

just noticed in the Car/Track History that BrakeBias seems to be calculated wrongly. Session reports show correctly 63.2% for my Lexus, but the same laps on the history tab are shown with 77.2% BB. This 14 points difference matches the -14points offset from the telemetry. So probably the offset is not applied there.

Yes - you're right. The mapping wasn't applied on that page. This will be fixed in the next update

 

[Doug Duthie]

Doug Duthie updated ACC Results Companion with a new update entry:

ACC Results Companion V1.12.5

Fix blacklist - If entry from local race (SP or MP) blacklisted, it will blacklist all local results

Read the rest of this update entry...

 

[AmazingReflex]

Thanks mate,

You are doing an amazing job with this app.

When I get a minute I will give you a donation for all your efforts.

 

[Kylooooren]

Is it possible that there might be something broken for endurance and especially endurance when the main driver doesn't join first?
Example of what happened:
I'm 2nd driver and join first, then 1st driver joins, I give him the car he does Qualifying, it doesn't even show up in the ACR, the quali.
Then we do the race, first him, then me, then 3rd driver and then him again, ACR shows it as Quali (me), Race (me + 3rd driver + me)
Had to delete the two sessions cause first driver is a second at least faster than me and it would break my stats.
Technically easy to differentiate because it doesn't have any sort of tyre pressures etc.

 

[Doug Duthie]

Is it possible that there might be something broken for endurance and especially endurance when the main driver doesn't join first?
Example of what happened:
I'm 2nd driver and join first, then 1st driver joins, I give him the car he does Qualifying, it doesn't even show up in the ACR, the quali.
Then we do the race, first him, then me, then 3rd driver and then him again, ACR shows it as Quali (me), Race (me + 3rd driver + me)
Had to delete the two sessions cause first driver is a second at least faster than me and it would break my stats.
Technically easy to differentiate because it doesn't have any sort of tyre pressures etc.

Yes - I know about this. It is proving tricky determining when a new session starts and the mechanism I currently use doesn't detect team sessions where you aren't the first driver. If you are happy to be a guinea pig, I have a new mechanism in test. To use this, go to the settings page, click Test: Developer Mode then New Broadcast Connect Method. This may help detecting the new session. I haven't been able to test this fully, only that it doesn't appear to have any negative effects on other session types

Can you try this out and let me know what happens. If you or the other tester have positive feedback, I'll make this the default method

 

[themalse]

Hi I have done SP Practice session but it did not record anything in the app. I just escape and exit halfway during a lap. Is there a way I need to exit? I had been in and out of the pits and spent about 30 mins in the session. Cheers

Im using V1.12.5

 

[AmazingReflex]

Hi Doug,

I've donated you £5 for all your efforts.

Thanks again.

 

[Doug Duthie]

Hi Doug,

I've donated you £5 for all your efforts.

Thanks again.

Cheers mate - much appreciated

 

[Doug Duthie]

Hi I have done SP Practice session but it did not record anything in the app. I just escape and exit halfway during a lap. Is there a way I need to exit? I had been in and out of the pits and spent about 30 mins in the session. Cheers

Im using V1.12.5

It sounds as though the connection to ACC may not be configured. Can you export diagnostics (instructions here) and I'll have a look

 

[themalse]

It sounds as though the connection to ACC may not be configured. Can you export diagnostics (instructions here) and I'll have a look

UserID:
-bx4wAhOlKx1FqzcJz59p2RF6n_p5klg_Td4UBP5Fng

The practice session was started around 16:30 on 8/6/2021 if that helps.
Track was Laguna Seca with the 911ii GT3 R

Also, just letting you know brake bias values are showing 62% in your app results, but it was around 56% or 57% in game.

Thanks Doug

 

[Doug Duthie]

UserID:
-bx4wAhOlKx1FqzcJz59p2RF6n_p5klg_Td4UBP5Fng

The practice session was started around 16:30 on 8/6/2021 if that helps.
Track was Laguna Seca with the 911ii GT3 R

Also, just letting you know brake bias values are showing 62% in your app results, but it was around 56% or 57% in game.

Thanks Doug

I'd missed that Kunos had changed the brake bias offset for the 911ii as well as the other Porsches - I'll fix that

On your missing session, ARC connected ok to ACC on the broadcast interface, but ACC didn't send information for the laps. I've seen this with a few other users and I've got a support request open on the Kunos ACC forum.

There's one thing in common with the other cases. You've had ARC open for > 1 day and I'm presuming ACC was also open for > 1 day. This might be irrelevant, but there might be a resource leak on either side.. However, in most cases it does seem to recover.

Until I get an answer from Kunos, the only thing I could suggest is to restart ARC and ACC every day.

 

[MB190E]

Hi Doug,

I have an odd thing that's happened. Usually I don't have any issue with session info correlating between panels in ARC, but this time...

I had a race, and I had a clean lap that showed a new best time for me on the 2020 Hungaroring track. (I know it's 2020 as I'm the server admin). I can also confirm that the game itself indicated this was a "Best lap" for that race, which it would not do if it was one with a cut.

The session tab recorded this correctly:

However, when I go to the Circuits:

That is my previous personal best, in the same car on the same track. Should it not update it here too, like it used to?

 

[Doug Duthie]

Hi Doug,

I have an odd thing that's happened. Usually I don't have any issue with session info correlating between panels in ARC, but this time...

I had a race, and I had a clean lap that showed a new best time for me on the 2020 Barcelona track. (I know it's 2020 as I'm the server admin). I can also confirm that the game itself indicated this was a "Best lap" for that race, which it would not do if it was one with a cut.

The session tab recorded this correctly:

View attachment 481160

However, when I go to the Circuits:

View attachment 481161

That is my previous personal best, in the same car on the same track. Should it not update it here too, like it used to?

I had to change the way this all worked. Previously, this dynamically aggregated the fastest laps from all laps stored. This was starting to become a big performance hit, especially when filters were applied. The leaderboard table was extremely slow and the circuits table was also beginning to show troubling signs.

Now, fastest laps are stored in a separate table and only updated at the end of sessions. This should make things a lot faster. However, I have seen one use case where the fastest lap wasn't updated correctly (it was in the old portion of a resumed race). There may be other cases where the fastest laps aren't carried over. Can you export the diagnostics and I'll check.

Also, in the next release, there will be a button to re-build the fastest laps just in case something is missed

 

[Kylooooren]

Yes - I know about this. It is proving tricky determining when a new session starts and the mechanism I currently use doesn't detect team sessions where you aren't the first driver. If you are happy to be a guinea pig, I have a new mechanism in test. To use this, go to the settings page, click Test: Developer Mode then New Broadcast Connect Method. This may help detecting the new session. I haven't been able to test this fully, only that it doesn't appear to have any negative effects on other session types

View attachment 480099

Can you try this out and let me know what happens. If you or the other tester have positive feedback, I'll make this the default method

Did the new broadcast method.
Quali first me then 2nd driver, was all good.
Then new server with the race, started with practice, I swapped to 2nd driver with &swap 2 and 2nd driver started the race, then me, then 2nd driver again.
But the ACR flagged the first stint as practice and only "started with the race" when I got in the car.
Haven't had a chance to try the session first as 2nd driver though.

 

[MB190E]

Sent a PM with the file info

 

[Doug Duthie]

Did the new broadcast method.
Quali first me then 2nd driver, was all good.
Then new server with the race, started with practice, I swapped to 2nd driver with &swap 2 and 2nd driver started the race, then me, then 2nd driver again.
But the ACR flagged the first stint as practice and only "started with the race" when I got in the car.
Haven't had a chance to try the session first as 2nd driver though.

Apologies for being slow in replying - I needed to concentrate on some other things first. Can you export the diagnostics (instructions here) and I'll check how it is behaving

Thanks