Wow that was fast. I'm gonna give yours a try now and give some feedback a while later.
Edit: I've tried it, check your PM.
Announcing SMS project
...just fixed a problem with SMS loaded from MC. It seems that the problem lies in ps2sdk (iomanx.irx module). Current version returns empty names when I try to read partition list from the hard disk (using fileXioDopen/fileXioDread). With "old" version (that which uses ps2link) it works good. If anyoune is interested to test it - PM/e-mail.
Best regards
Eugene
Best regards
Eugene
BTW, did you PM Oobles or Neovangelist on that forum to ask for SVN account ? :)
pixel: A mischievous magical spirit associated with screen displays. The computer industry has frequently borrowed from mythology. Witness the sprites in computer graphics, the demons in artificial intelligence and the trolls in the marketing department.
From the above I take it that you are having some problems coming to grips with the new repository. So did I too at first, but I think I'm past those problems now, so perhaps a few tips about what I use might help you as well.EEUG wrote:This is from README that will be available as soon as SVN/CVS will work:
First of all, you need a good SVN client, and if you use Windows I strongly recommend the one I found (tipped by Oopo), which is TortoiseSVN. It is a bit unusual in that the entire user interface is based on context menus, but once you get used to it that does come across as very natural. Many of the operations you do on repositories and checked-out modules are of the type where you work mainly by clicking or dragging stuff in explorer windows, and then the use of context menus makes perfect sense.
You can find it at:
http://tortoisesvn.tigris.org/download.html
Another thing that tricked me at first is that SVN can be implemented with different protocols, and the URL you must use depends on the protocol. At first I tried to 'check out' "http://svn.ps2dev.org/ps2" with variations at the end of that string, but I kept getting error messages. Then I read some more in the help texts, and checked out Oopo's new toolchain script, and then I realized that the proper URL was "svn://svn.ps2dev.org/ps2".
Now here's what I did to get a properly checked-out copy of the entire repository.
1: On a drive with plenty free space I created a folder named "SVN", to hold ALL future SVN stuff...
2: In that folder I created a subfolder named "ps2dev.org", to hold all stuff from this particular repository group
3: In this folder I created a subfolder named PS2
4: Next I rightclicked the newly created folder to get a context menu
5: Then I chose the entry "SVN Checkout..." to open the checkout dialog
6: In the dialog I entered the URL "svn://svn.ps2dev.org/ps2" and clicked OK
7: The transfer process then started, showing progress until completed.
8: Then I repeated steps 3..7 for the repositories "ps2ware", "psp", "pspware" as well
(NB: there will be an error message for a bad folder name in "ps2ware")
At the end of this process I had my own copy of each of the four repositories, each in its own subfolder. As it worked for me, so it should for you too, though you will have to study up on how to 'Commit' stuff as well. This is something I have not had occasion (or rights?) to try yet.
Best regards: dlanor
This problem seems identical to one we have with LaunchELF, which also needs an old version of the iomanx module to work reliably. With newer ones the partition list can show pure garbage...EEUG wrote:...just fixed a problem with SMS loaded from MC. It seems that the problem lies in ps2sdk (iomanx.irx module). Current version returns empty names when I try to read partition list from the hard disk (using fileXioDopen/fileXioDread). With "old" version (that which uses ps2link) it works good.
(Not always though, it depends on various circumstances.)
Best regards: dlanor
EEUG wrote:1. no. Only CD at the moment ;);
2. no. Source code modifications are required to change PAL/NTSC
Hi EEUG,great job with the new Ps2 SMS ............
We know It can media can be read from hdd,lan,cddafs.
Others has implement usb support.
Can I know whether SMS can read normal avi burn to a cd?
Something like ps2mp3SE/PS2reality media player can do.
Cause most of my collection is normaly burn it such a format....
additional ps2 hdd is really hard to get now days !!!!...
EEUG wrote:...my PS2 is not modded, so the answer is "no". Maybe someone else can implement this. Sorry ;). Note, that if you burn CDDAFS disk, you can still copy files back from it...
thanks EEUG for the reply!!!
Maybe someone can implement this........... : )
Yeah I know we can still copy files from it .......
but the times it takes to convert it back.... ." : )
anyway here the screenshoot of the new version of the player!!!!
Damm!!! it fast to load compared to previous version...
Great job!!!!
New version is in SVN
...pop the trunk, read README and get it ;). Enjoy!
Best regards
Eugene
Best regards
Eugene
Btw, I don't know if I said it enough, but, nice job :)
pixel: A mischievous magical spirit associated with screen displays. The computer industry has frequently borrowed from mythology. Witness the sprites in computer graphics, the demons in artificial intelligence and the trolls in the marketing department.
...well, all avi management happens in two files: SMS_AVI.c and SMS_AVIPlayer.c. SMS_AVI.c contains actual demuxer (no avi 2.0 support yet) routines, and SMS_AVIPlayer.c is a player itself, which uses routines from SMS_AVI.c to validate file and to obtain data packets. I've made a mistake making these routines global. Probably it's much better to make them as function pointers inside 'SMS_AVIContext' structure, so other demuxers will just substitue them by their own functions that perform validation, read packets etc. (like C++'s polymorphism). To tell the truth, I never thought that I'll be able to do what I did :), so the dirty code can be seen everywhere. Anyway, I'll be really glad to help anyone who has interest. It's also possible to use PC to develop and debug since almost everything works under Windows also (MSVC6).
Best regards
Eugene
P.S. I'm now implementing AC3 audio (very ssslllooowww progress with it, but research phase is done, I've got sound on PC using integer only calculations, so I hope Station will handle it).
Best regards
Eugene
P.S. I'm now implementing AC3 audio (very ssslllooowww progress with it, but research phase is done, I've got sound on PC using integer only calculations, so I hope Station will handle it).
Maybe I should start by trying to cleanup the avi support - like you said, make it a set of function pointers. Then additional files can be added that implement other containers like MKV. When a file is selected, the type could be determined and the pointers set.
Yeah, it's amazing how far along this player has come. It's great and getting better all the time. :)
I still think the VU is capable of handling the FP calculations needed for MP3 and AC3 (and others). One of these days, I REALLY need to sit down and try it. DMAing a packet of code and frame data to the VU and DMAing decoded data back has got to be faster than having the EE decode it. Even if you just brute-force calculate the data using a single element of a vector (all data in the x coord for example), it's still plenty fast for audio - we're only talking 48000 samples (times 2 for stereo or more for more channels like 5.1) per second.
Yeah, it's amazing how far along this player has come. It's great and getting better all the time. :)
I still think the VU is capable of handling the FP calculations needed for MP3 and AC3 (and others). One of these days, I REALLY need to sit down and try it. DMAing a packet of code and frame data to the VU and DMAing decoded data back has got to be faster than having the EE decode it. Even if you just brute-force calculate the data using a single element of a vector (all data in the x coord for example), it's still plenty fast for audio - we're only talking 48000 samples (times 2 for stereo or more for more channels like 5.1) per second.
...it all your's :). Well, I thought about VU also. The problem I've encountered so far is that AC3's algorithms use auxillary tables for imdct calculations, so all these tables + audio data itself won't fit in the VU memory. Though, calculating values of these tables 'on the fly' is possible, but all these 'sin', 'cos' and bessel functions will eat up a lot of VU cycles. Maybe it will be possible to process audio data by smaller chunks, I don't know yet. "Let's get loud" first, then we'll see what is possible.
Best regards
Eugene
Edit: you can also take a look at MP3 code. There's one function namely '_compute_antialias_float' which deals with floating point calculations and which is not used (it's replaced by '_compute_antialias_integer'), but it could be a perfect candidate for VU1 :).
Best regards
Eugene
Edit: you can also take a look at MP3 code. There's one function namely '_compute_antialias_float' which deals with floating point calculations and which is not used (it's replaced by '_compute_antialias_integer'), but it could be a perfect candidate for VU1 :).
fringo wrote:alwsome!!! cd/dvd iso suppor & host support !!!
Now I can read it from normal cd/dvd iso compilation!!!!
NoW !!! What do I need to burn a bootable SMS cd ?
Sms.elf,audserv.irx(where do I get one that u have fix on hissing sound issue ?),netsupport(what file do I need to include?),usbd.irx,system.cnf