I can't download the 0.5.1 from Scummvm.org please could someone Mail that Version?

@ThePlayer

You can download it here.

Thanks for the Link Toto and for the Mail Bladerunner. But i have still the problem that i can't load my save games.

My fault it works!

@Bladerunner

Okay, my mistake, it is really Lucas Arts Masterpiece Extra-ordinaire

Cunning and witty dialogues, hard but logical puzzles, and totally not lineair, oh no.

Hi Bladerunner,

So sound of MonkeyIsland 1 and Zak McKracken doesn't work because they are Amiga versions?
Ok, that explains why I only can choose for EGA and VGA resolutions for MonkeyIsland 1!

This version 0.5.1 can save and load, cool!

@Toto:
Thanks for the link!!

Now of to play MonkeyIsland2

Bye,

Spidey

@Stevo

Are you trying to be ironic? Just it's an ask...

Hmm,

A small glitch with Monkey Island 2 so far:

The room where the Voodoo lady sits is dark, you see only Guybrush there. But look closely and you'll see the outlines of the whole screen (and the texts at the bottom).

I've played it too on the pc, there this part of the game looks good.

Bye,

Spidey

Allright, found my Curse of the Monkey Island cd's! And it seems to work...with some problems though.

1) When playing from cd, the intro music seems to "get behind" and desperatly tries to catch up. On the other hand, when playing from hd (after copying the whole cd to my hd) the intro music seems to be +/- 1 sec. "in front" of the rest of the game. I tried the "tempo" command, but that didn't seem to do anything. Ingame (with Wally in the cannon room) the sound/graphic syncronisation seems fine both ways, but...

2) The game crashes because of a sound buffer overflow, while looking out of the porthole. Haven't tried it with the "music=off" option yet.

Anyone suggestions and/or further advice on playing monkey3 on ScummVM mOS 5.1? Thanx.

ps.

@Timofonic

I was not trying to be ironic, honest...but I might have been a little sarcastic though...

The problems in The Curse of Monkey Island are fixed in 0.6.0. I suggest you wait for a newer port of ScummVM or play the game in GNU/Linux.

Well, that is a bit of a last resort option, isn't it?

There should be a way around the problems since Frostwork wrote:
Quote:

So, Frostwork, if you are around (or anyone else who solved the problems) please enlighten me

I would like to hear how he solved the cannons as I was unable to complete that part in the MOS port due to the bug.

Quote:

The most recent version for MorphOS had this fixed for sure.

Hi,Ho!
Yes, the last version worked better at that place, but
I'm sorry to say, that it wasn't completely fixed anyway... (Sorry, that's not against your port!!!).
I had to switch to window-mode, and fought a little
bit with the mouse (GRRRRRRR, it HAS to work ;) ) ->
it worked!
After cannon-shooting: SAVE! (Puuuh) ;)
bye
frostwork

Hmmm, maybe we should start each post on this thread saying "Tomjoad, we really, really, really love the port, we are unworthy, but..."

Anyways, *assuming we are all talking about the cannon problem at the beginning of the game, with a crying Wally in the background* I did get past that "sound buffer overflow" problem by blowing up the pirate sloops and then going for the skeleton arm really quick, without wasting any time talking to Murphy (wich is a shame, because Murphy has got to be one of the best co-actors ever ). Phew, made it, saved instantly...and I am on the roll! Great stuff!

As I've mentioned before, I tried porting a CVS release of ScummVM to MorphOS but failed because I needed something called eTude. I searched around for this but couldn't find any relevant information. What is this eTude and where can I find the SDK?

Hi dholm,

Just read this thread again
As I understand Etude is Amidi 2.
It isn't finished and some parts of that (music)program is put in the ScummVM program.....or so

Bye,

Spidey

Hello,

I spent a couple of hours into ScummVM 0.6.0 and here my final words about it:

a) I got it compiled against powersdl from itix.
b) The compile process often broke down because GCC 2.95.x couldn't handle the C++ code properly so I had to hack a few lines to get it working again.
c) The compile process continued and finished till end.
d) I then hit this final error:

undefined reference to 'Common::Singleton<Common::ConfigManager>::singleton_

Which continued a couple of times. After asking the ScummVM developers I was told that this is a template error inside the GCC 2.95.x compiler. I was advised to update to GCC 3.x which of course is a problem at the moment for MorphOS.

Older versions of ScummVM compile perfectly but this is because the people were using non complex constructs in C++ and the final compile and linking didn't caused such big problems. But during the process from ScummVM version 0.5.1 to 0.6.0 a lot of code got changed and they were using much more complex stuff.

So far this either needs a total expert who is able to work around these problems or we need a new compiler pretty damn soon which deals with more complex C++ code properly. So far it was no problem to port ScummVM due to powersdl from itix. It is more a compiler hit the limit problem here.

The Frieden brothers got ScummVM 0.6.0 ported to OS4 because they used GCC 3.x for this. GCC 3.x handles C++ code much more properly and generates better code.

I hope I was helpful with my feedback here.

greetings,

oGALAXYo

Galaxy:
Well, I hope that tomjoad will continue the work
as he did before, without using the sdl lib!
The reason is simple, His direct port is much faster then the SDL one. To be honest, nearly every SDL App so far i had on my peg is noticabel slower than its direct counterpart. This is the reason why even the current Classic amiga OS Port is again based upon romjoads work (he did it before using sdl)
And yes I expect also the Friedens Bros Port to be slow, compared to the MOS Version without SDL.

@Galaxy:

. You may ask "fab" (irc/#morphos): he used to cross-compile some morphos stuff on linux using gcc3.x: myabe he could compile scummvm for you

. Ask CISC (irc): he's supposed to have a fixed GCC2.9x

(Maybe we'll have to wait for gcc4 to be released to have a gcc3 for morphos )

Regards,
Leo.

@Bladerunner

The problem here isn't SDL I was also able to fix the native MorphOS code so far (in the backends/morphos) subdir of ScummVM.

The problem here is that the ScummVM 0.6.0 code causes a lot of problems with GCC 2.95.x this is by far the bigger problem. I think that Tomjod either uses a different compiler GCC 3.x or simply crosscompiles it on a different machine.

@Leo

Yea, I was having a conversation with FAB last evening and handed him out the scummvm-0.6.0.lzx I made, he hit the same problem like I was hitting. Need to talk to him later on, when he is around again.

Maybe I can apply some RedHat or Debian patches to GCC 2.95.x and then apply CISC's patch over it and re-compile GCC 2.95.x that way but I am not sure whether this all results in usable stuff.

Quote:

Nice somebody appreciates the effort

Quote:

For the last time, Gabriele Greco's SDL port can be made somewhat faster with little effort; said this numerous times before but nobody's interested Just start up Exult (click your way through the menus) and measure the time from the last menu screen until the opening scene. How long does this take? With the original SDL it would have taken > 8 minutes (no kidding).

Although it depends on the program; this is a very extreme example. It was so darn slow that I first thought it would throw endless hits ...

As for the compiler problems, well we'll see. If there's a workaround I'll probably find it . Guess why at work (sw developer) I do always get the tasks on my desk where three other devs have failed before .

>aybe I can apply some RedHat or Debian patches to GCC 2.95.x and then apply CISC's patch over it and re-compile GCC 2.95.x that way but I am not sure whether this all results in usable stuff.

Having GCC3.x on morphos would solve a lot of problems I guess.

Well, I replaced the MorphOS SDK GCC 2.95.x compiler with the CISC version of GCC 2.95.x compiler which then worked perfectly on ScummVM. Somehow the version in the SDK is a bit buggy.

I was able to compile ScummVM 0.6.0 against Greco's SDL without problems and it works somehow but unfortuantely that SDL version is kinda buggy.

Anyways if you want to test my version you can grab it from my site.

ScummVM-0.6.0.lzx

But don't expect to much. FAB is already working on the native version of it but I can't tell much about it since I don't know how far he is or if he ever finishes it. So at the end we still depend on Tomjoad's coperation.

@Tomjoad

It would be nice if you could commit your changes back to the original CVS of ScummVM so someone else can work on it in case you are not available but please remove the __MORPHOS__ ifdefs inside it or replace them with something else. The reason why I ask for this is in case someone releases a working version of SDL one day so he can either compile it perfectly against SDL only or use the native version only. Right now __MORPHOS__ seems to include a bunch of native stuff whereass the SDL version was wanted. I had quite some work to do to filter out the parts.

I compiled the morphos backend (native port, without sdl) and it runs DOTT correctly for example.

However It seems to throw some hits sometimes, and soundmixer sometimes fails in an assertion. I'll try to see why.

Moreover the Etude stuff doesn't seem to be public, so midi and cd playback are disabled.