[RFC] Project needs a new maintainer

[edubart]

As many of you already noticed I don't really maintain this project anymore, although it's used by many otserv based games in the wild, if anyone is willing to continue improving otclient with the focus of making it better and more compatible with a widely used otserv distribution such as https://github.com/otland/forgottenserver I am open to to allow access to the repository for new contributors or leader once trust is gained through a few reviewed and approved pull requests by me and the community

I don't think that's probable to happen, but anyway I thought you should all know and will leave this open for comments here

[HeavenIsLost]

@edubart Como uma ultima contribuição, você poderia compartilhar com a gente algumas das estrategias para otimizar o OTClient? Boa parte dos lags do OTclient provem de códigos mal feitos(como o caso da battlelist que eu já ouvi de mais de um desenvolvedor falar que laga o client todo caso existam muitas criaturas na tela) mas a principal(pelo menos pra mim) razão do frame rate baixo é a quantidade massiva de texture binds que o OTClient faz. Eu vi que tanto o client antigo da cipsoft, o flash client e o novo client(tibia 11) aparentam usar uma estrategia de Z order, onde eles desenham na tela todas as sprites iguais para evitar a mudança de textura e assim otimizar o processo de desenho das sprites na tela. O único problema com esse método é que até onde eu pesquisei não é possível ter sprites transparentes e usar essa estrategia de Z Order. Eu até pensei em uma nova estrategia que usasse uma textura com as sprites que estão sendo usadas no momento assim evitando bindar varias texturas para desenhar a tela, o problema é que que parece que cada placa de video tem um limite máximo para o tamanho das texturas e o próprio processo de montar essa super textura também poderia ocasionar lag/baixo framerate que é contra o proposito de tudo isso.

[edubart]

@HeavenIsLost please only english here, what you said is also not related with this RFC, but there is a long topic regarding that in otland and at this page I have discussed details about further optimizing there, including some you said. Also don't generalize that the client have poor performance because on recent machine it has better performance than tibia or flash tibia, it performs poorly only on old hardware, tibia was coded decades ago so makes sense them supporting better old hardware, while otclient is much more recent and was coded in a recent hardware environment.

[HeavenIsLost]

Thanks for your in input, i totally forgot that topic(i was also the one that have made a code to proof how much we would gain if we get ride of texture binds). I think what you said in that post is a good enough material/ideas and it would take months to implement them. I have use cipsoft clients only as an example of which drawing strategy cipsoft has choose for their clients.

[andersonfaaria]

@edubart I don't really like github but I'm willing to maintain the project. Could you assign me as contributor?

[edubart]

I voting to @diath be the project maintainer for now if he want, as he have done some quality PRs recently.

[diath]

Unfortunately I don't think I can maintain otclient long term.

[EgzoT]

@edubart I suggest @conde2 to be new collaborator. His commits was be helpful, now he is 8-th contributor in terms additions.

@diath your experience is very helpful for this project and may be more helpful if you will be collaborator, even you don't may do it in long term.

[JSkalskiSBG]

@edubart There are 27 PRs waiting for acceptation. If there is no better person to verify them, I can do it.

// Gesior.pl at work

[okk3]

We have plenty of people to verify them, me included.

[EgzoT]

@edubart I would like to recommend @JSkalskiSBG and @okk3, they are known and respected users. Now we need persons to verify commits, manage issues (verify and close fixed/empty/outdated issues).

Or adding something who may help to faster verify commits, for example use option reviewers to help verify commit by suggestions added by known members (approve changes).

Other idea is that we may create additional OTClient repo who may to be something like beta version/early access/test version, and then tested systems/fixed etc. add to main OTClient repo.

[Riverlance]

I recommend these: @EgzoT @diath @okk3 Based on contributions I saw at this repository recently.

[peonso]

@edubart I'm vouching for @EgzoT and @okk3. I volunteer myself also. Despite my limited knowledge I can manage issues and milestones. I have been doing it with https://github.com/TwistedScorpio/OTHire

[ranisalt]

There's otland/otclient fork now, bringing devs and possible interoperability with TFS to the table.

[vfjpl]

@ranisalt is it still maintained?