-
Notifications
You must be signed in to change notification settings - Fork 3
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Potential slow down in the simulation #92
Comments
Yes, I will have a look at it. |
@tnguyenh I know the project is kind of over but maybe it make sense to have a look on the last commit(s) done before the opening while it's still fresh in your head? I am not sure exactly how to reproduce the behavior we had during the opening |
I am trying to find what's the problem here but it's pretty hard to track. So after one day (500k iteration) I start to see some lag (like sometimes one cycle out of 10ish would take like 1sec to execute) so we see a lag in the simu. The memory is not too crazy though (700Mo) so I really don't know what's happening. I will rerun a sim with Bikes to see if it comes from the last commit with the bikes |
Do you mean a sim without bikes ? |
without yes |
Good and bad news Good news: It semes to wok normally without bike If we can fix it quickly that will also faciliate the issue we have rigth now in the pavillon @tnguyenh do you think you can have a look at your last commits concerning the bikes? |
to test :
after creating the bike graph: bike_graph <- bike_graph use_cache false;
Le lun. 2 mars 2020 à 20:17, Arnaud Grignard <[email protected]> a
écrit :
… Good and bad news
Good news: It semes to wok normally without bike
Bad news: It means there is somethign wrong with the bike
If we can fix it quickly that will also faciliate the issue we have rigth
now in the pavillon
@tnguyenh <https://github.com/tnguyenh> do you think you can have a look
at your last commits concerning the bikes?
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#92?email_source=notifications&email_token=AALPWHPZBCGG4PHYOMVXEBDRFQA3ZA5CNFSM4KWR5OJKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOENQSSXI#issuecomment-593570141>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AALPWHMQCXZHENFIINTTJB3RFQA3ZANCNFSM4KWR5OJA>
.
|
on l'a déjà fait non ? line 518
|
Question pour la reproduction du bug : cela arrive même quand on laisse la
simulation sur la situation "d'avant" ou il faut que je passe à la
situation d'après (avec plus de vélos) ?
Petit problème chez moi (non lié au modèle) : j'ai un bug GAMA/windows (du
à une mise à jour de windows ou de ma carte graphique ?) qui fait que GAMA
crash parfois (tout le temps quand je le laisse tourner longtemps) quand
j'utilise opengl, donc, je dois faire mes tests en java2D (mais je ne pense
pas que cela ait un impact sur ce bug).
Le mar. 3 mars 2020 à 14:20, Arnaud Grignard <[email protected]> a
écrit :
… on l'a déjà fait non ? line 518
loop j from: 0 to: stateNumber - 1{
bike_graph[j] <- directed((as_edge_graph(bikelane where (each.allow_bikes[j])))) use_cache false;//with_weights weights_bikelane_sp;
bike_graph[j] <- directed(bike_graph[j]);
}
````
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#92?email_source=notifications&email_token=AALPWHLXBXKQYM7WPZCGTLTRFT7ZLA5CNFSM4KWR5OJKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOENTONRA#issuecomment-593946308>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AALPWHNIRR2JA572WDMQT53RFT7ZLANCNFSM4KWR5OJA>
.
|
Il est très chiant ce bug, je dirai que le fait de changer existat/vision devrait le favoriser mais j'ai du mal à le reproduire... |
De mon côté je n'arrive pas à le reproduire... |
Très très chiant ce bug. J'arrive à le reproduire la au bout de 150 000 iterations et je vois clairement une baisse dans le fps surtout au niveau de la transition mais j'ai fait pas mal de swithc entre avant et apres et activer/desactiver les layers... |
Je vais réessayer, mais jusqu'à présent, je n'ai pas réussi à reproduire ce
bug.
Le ven. 6 mars 2020 à 00:43, Arnaud Grignard <[email protected]> a
écrit :
… Très très chiant ce bug. J'arrive à le reproduire la au bout de 150 000
iterations et je vois clairement une baisse dans le fps surtout au niveau
de la transition mais j'ai fait pas mal de swithc entre avant et apres et
activer/desactiver les layers...
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#92?email_source=notifications&email_token=AALPWHL2R3ONZWKJHUG77HLRGA2LRA5CNFSM4KWR5OJKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEN7DVMA#issuecomment-595475120>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AALPWHNW5LYJSE62UJ4KLALRGA2LRANCNFSM4KWR5OJA>
.
|
J'ai reussi à le reproduire un peu cette nuit en laissant tourner sur le mode vision la j'en suis a 280 000 iterations et il y a une petite pause dans la simu de temps en temps tout les dizaines de cycles mais ce n'est pas constant... Aussi j'ai les deux displays image et simu peut etre que ca aurait une influence? |
During the opening night, we have noticed a slow down in the speed of the simulation after few hours of running. It might be due to the last commit done in the 1 or 2 days before the opening (especially the one about the bike)
@tnguyenh before we forget it would be great to see if we can track this bug?
The text was updated successfully, but these errors were encountered: