Right, maybe LiveForIt when he have the time .. hope it will not too mutch difficult, but if that results in a future merge between the various efforts i think it will worth for sure!
Its easy to reintroduce bugs that has already been fixed, when updating to a newer official mplayer source code, while I'm working on trying to fix this one I do not want to do that.
Like Ka1e says the difference between a newer mplayer and the one I'm working on is not that big, if joerg commits his fixed vo_ahi_dev to code.google.com, I can easily back port it if needed. Or if he provide a diff file I can add the stuff that's missing or changed.
What did he fix anyway?
Joerg wrote: Quote:
using the original mplayerhq.hu SVN sources simply works
Some how I doubt that, Linux and Amiga are not compatible, more then likely I'm going to reintroduce bugs that has been fixed. (I think this is what happened last time, some one did it)
Joerg wrote: Quote:
without such a mess.
As for Joerg's vo_pip, its of no use to me, I intend to clean up the code.google.com version, move the stuff where it belongs and standardizing it.
Joerg wrote: Quote:
They just don't include much AmigaOS parts (a few are there).
True there are not many AmigaOS parts to begin whit, not every thing from Andrea Palmatè are included, the MorphOS version has more stuff, it has to be ported to AmigaOS4. (like I did whit vo_ahi_dev and vo_cgx_wpa), or taken from Andrea Palmatè.
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
Too bad since Joerg has already worked a lot on this new version, which is really fast and tearing free by the way.
It's not the point, the point is that what is in code.google.com is a mess when it comes to p96_pip, introducing changes to it now at this stage, makes work harder for me.
As now I'm in between two version of p96_pip (one that is stripped clean), and the one in the code.google.com repo, and GUI stuff has yet to be moved out so it can be used by all vo_output drivers.
If Joerg cleans up the repo first then he is more then welcome to do it.
Or I can replace my striped p96_pip on my hard-drive, and do the GUI clean up stuff later (I don't want to just remove it, I think they have put lot of work into the reaction GUI).
Quote:
Maybe you should focus on your video composite2d video output and improve it even further. Then you'll just have to use Joerg work for p96_pip
It was almost ready yesterday.
For now I think just upload my new mplayer whit out Joerg fixes.
Edited by LiveForIt on 2014/4/26 11:55:00 Edited by LiveForIt on 2014/4/26 12:03:00
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
The thing is that its not just about vo_comp has backend cgx_common.c, cgx_common.c is what handles the events, now the problem p96_pip has its own event code, its different so instead of being able to maintain it in one place you fix stuff in two places, this is what I do not like about p96_pip, and this way its not just about vo_comp its about cgx_common.c that I'm thinking should be the foundation for all AmigaOS (MorphOS) GUI's. MUI-MPLAYER depends on cgx_common.c for events, so this the logical please for all events stuff.
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
the MorphOS version has more stuff, it has to be ported to AmigaOS4.
From where you come ? Morphos version ported years ago, it is called mui-mplayer. And vo_comp about which i ask you, its for including to muimplayer (that morphos port). Strange that i need to explain it, it was clear from begining that muimplayer is port of morphos version ?
Quote:
(like I did whit vo_ahi_dev and vo_cgx_wpa),
Like you did ? I.e. taken our fixes for morpho's vo_cgx_wpa and call it as "i did" kind of suck. And you say it like its you first time do it. How you think muimplayer (that morphos port) works at all ?
Quote:
or taken from Andrea Palmatè.
? I can't get you really. Andrea's sources already on google-code page. That is those sources. But ao_ahi_dev and vo_cgx_wpa it is output drivers done for morphos, in morphos version. You just upload them and reuse in andrea's version. Reuase them in any version of mplayer for aos4 its deal of few minuts. There is nothing about what need to worry or mention.
I do not understand to be honest all that mess happens now. I mean, you do not need to fix anything in andrea's version, instead, you only need to make vo_comp.c available, and i can reuse it for muimplayer, and it will works as before for everyone, with normal gui, and without those bugs which google-code version have.
True amiga mess ! One more repo :)) We need 100 more , then 100 more with changed comment in one file, then 100 more with copy+paste of something from one to another. Then, all who works on different parts, should says that he do not want that, or that, or that and will make another 200 versions.
Quote:
For now I think just upload my new mplayer whit out Joerg fixes.
I feel you love to say "my new mplayer" ?:)) Ego time !
@Varti Why don't create it all on google-code page which already known by all ppls ? Why need one more repo ?
@All Its only me, or i just see there true-amiga-ego ? I mean, no one want to work normally with each others, instead, everyone want to make "his own version" , reacreate a wheels, do not follow logic, and make everything, but just to have users be in fear and in mess when someone will say "mplayer" ?
But where is the repo, all you did what give me two files and direct me to code.google.com, the repo I have worked on.
Didn't you read that fab do not want to put his sources on any repos ? Then it is : fab do not want to put his code on repo. And if you are respect ppls wish who share code you will know that uploading their sources without their permission are suck. All what i do, its just port his code to aos4 with help of Deniil. I can't make any repo from it, because its suck to do it when author says he do not want he code be in repo. And because port the stuff done by others its not by which one should be proud of or even create repos of it (to make original author cry and do facepalms).
Quote:
so where is your repo? I was hoping you commit your changes so we work in on the same thing.
What repo ? I din't want to create any more new repos. Its enough of mess already when there is google-code version, now there is another "sandbox" version, now there is 2 more builds from you and joerg. Want another one ?
And how can i create anything, if fab says he do not want his code be on any repo ?
You works on andrea's version, just to have some code to test your vo_comp, as fab's version can't be shared on repo (as he do not want this). You no need to fix those bugs in that mess, all you need just make vo_comp works fine and we can put it to muimplayer. We (at least me) only point our on google-code version before just to have you something on which you can make vo_comp and test it, as explaining how to build muimplayer can be mess.
I can share aos4 fixes for morphos port, but that can be in no use for us in general, as we can't just upload it to repo. Maybe we can use it as reference, but then fab's code already open, and i share those aos4 changes in cgx_wpa and co. Do we need anything else from ?
Imho, all what should be done for "repo", its just that what Varti do now in one more new repo, make that on google-code page, not on new repo. All that old shit what we have no in repo should be put to the "old_branch" , new version of mpalyer svn code put there, then all the os4 native parts put from that old_branch back piece by pice. Then add there joerg as commiter too.
Why there needs to create new page, why there needs now to create 3 more different mplayers, i just can't get. After we will have new latest code of mplayer on google-code, then probably joerg will be happy too and can share his new vo_p96_pip.c (as seems there again stop factor to sharing it).
Its just so suck ,that so many devs now spend their time on the same , recreating wheels, and do again the same and same. Every version have some different bugs, which didn't happens on another versions, or was fixed. Sad mess for sure. Sadly we can't merge fab's code on repo, but so ok, let's there be at least only 2 versions : fab's port, and our shared one. But its already too much to have 2 different version, but as unnecessary evil : ok. But having 3,4,5 versions that just crazy.
Edited by kas1e on 2014/4/26 12:51:20 Edited by kas1e on 2014/4/26 12:54:18 Edited by kas1e on 2014/4/26 12:55:53
Also just to make the mess more mess we should found a normal method to reintroduce many of the old feature from the old afxgroup version
Joerg started from an old version (so many things are missed) but it's the same also with the newest port from LiveForIt, even if he started from the code in repo, i don't understand why ..
Another repository, for who and for what ?? Please ask him if he can just update MPlayer to the last revision and then put this in current repo, we need a recent version that should be atleast in sinch with the latest snapshot from Joerg (r37148) so with some luck we will be able to merge your port with the Joerg one ... just to start reducing the confusion ..
I do not want to disrespect FAB or You for your work, this is GPL code, its supposed to be shared.
ao_ahi_dev was in the original Andrea's mplayer as well. FAB put it out there and source code is under GPL, and I have NOT removed credits, I have full respect for FAB, he knows what he is doing.
I the end we have more to gain from cooperation, and to making code reusable is one way do that.
Quote:
Why we need different versions ?
We don't, we just need every one to agree, to use one REPO, that includes FAB and Andrea also, and maybe a few AROS developers, as well as once who has already joined code.google.com
I expect FAB not wanting to join seeing the lack of commitment to the Mplayer project from different developers to commit and work on the same repo.
Quote:
True Amiga mess ! One more repo :)) We need 100 more , then 100 more with changed comment in one file, then 100 more with copy+paste of something from one to another. Then, all who works on different parts, should says that he do not want that, or that, or that and will make another 200 versions.
Joerg asked Varti if he can do merge whit newest mplayer, I guess that's what he is doing.
SandBox = private repo. My guess is that he is going to commit his changes when its done, but I wait for his comment.
What is clear is that this project lacks coordination, and communication.
Edited by LiveForIt on 2014/4/26 13:24:38 Edited by LiveForIt on 2014/4/26 13:46:30 Edited by LiveForIt on 2014/4/26 14:46:00
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
[snip] Long ranting like two year old who hasn't got his way[/snip]
Quote:
Sadly we can't merge fab's code on repo, but so ok, let's there be at least only 2 versions : fab's port, a
Yes you can, mplayer is GPL2 you can do what you like with the code.
If fab gets upset then he shouldn't work on GPL software.
Really, it's not so hard to cooperate on a project, merging various ports is obviously tricky and so would be best done against the latest mplayer repo (hence I'd guess why Vartis is experimenting in seperate repo the word sandbox is a clue there. (why he didn';t make branch in the same repo I don;t know, maybe he didn't know how)
Put all the stuff in the same repo and make a proper AmigaOS port. It doesn't need to be n sunc with the Morphos effort but it would be cool if it was, that's up to fab really.
I just noticed it, while looking for projects related to “AmigaOS” on code.google.com. I don't have him on Skype or FB.
You can contact him via PM or per mail
Quote:
Its easy to reintroduce bugs that has already been fixed, when updating to a newer official mplayer source code
Yep i can understand this, however according to Joerg this might be the first step to let him (and you) working togheter
Quote:
Like Ka1e says the difference between a newer mplayer and the one I'm working on is not that big, if joerg commits his fixed vo_ahi_dev to code.google.com, I can easily back port it if needed. Or if he provide a diff file I can add the stuff that's missing or changed.
It's not only the fix on vo_ahi_dev, his version is different than yours and both versions (yours and the newest one from him) doesn't even include most of the old feature of the afxgroup, so pratically we have 3 different versions instead of just one !
Add to that we have also a separated MUI version from kas1e and you will see how confusing it is
But beside the MUI version (that is another story), atleast the other 3 versions should be merged normally --> using the current repository
Quote:
What did he fix anyway?
Don't know in details, but the important things from his new version seems to be:
- Updated to the latest 37148 revision - The new fixed (and very faster) p96_pip driver
But his newsest version (like yours) even if introduced some things at the same time they both missed most of the old feature of the original version from Andrea ..
Varthall will not update it, as he busy for last years, and didn't do anything with it.
He is already working on it, currently he is testing it in his sandbox repository since he didn't do such a vendor branching/merging yet and doesn't want to mess up the mplayer-amigaos repository in case he does something wrong, after everything worked correctly he will do it in the mplayer-amigaos repository.
The VO speed is, your results are more than twice as fast as the ones from my first version.
Quote:
I'm wondering if there is something wrong with my setup.
I'm using lavdopts = fast=1:skiploopfilter=all nosound = 1 in my conf/mplayer.conf
Since I'm only working on the VO, nothing else in MPlayer, I don't care about the VC speed, sound, or anything else in it, and use these options to make it faster.
using the original mplayerhq.hu SVN sources simply works
Some how I doubt that,
It does in my MPlayer executables
Quote:
Linux and Amiga are not compatible, more then likely I'm going to reintroduce bugs that has been fixed. (I think this is what happened last time, some one did it)
if joerg commits his fixed vo_ahi_dev to code.google.com, I can easily back port it if needed. Or if he provide a diff file I can add the stuff that's missing or changed.
What did he fix anyway?
It didn't set ao_data.format to a supported format (with current MPlayer versions you IIRC get floatbe), there was no IAHI variable, IExec->DropInterface(IAHI) was missing in the cleanup code and it tries to use external functions which don't exist in MPlayer (maybe in some AmigaOS parts in the code.google.com version, but it doesn't make sense to use external functions for SWAPLONG() and SWAPWORD() anyway, something like that should always be done inline, with local inline functions or macros): http://pastebin.com/EsYZp2CA
News feed
OS4 Feedback forum
OS4Depot Feedback forum
Amigabounty forum
OpenAmiga forum
AmiCygnix forum
ABC shell forum
AmiKit forum
Cinnamon Writer forum
CodeBench forum
Digital Universe forum
Dopus 5 forum
E-UAE forum
Gnash forum
Ibrowse forum
JAmiga forum
Odyssey forum
OWB forum
Qt forum
SmartFileSystem forum
Timberwolf forum
TouchDevice forum
TuneNet forum
Unsatisfactory Software forum
Donate
