Our ReVolt Pub

Just a little news.
I had to rewrite the memory patcher part of WolfR4 practically from scratch, because my first try was working, but was really not robust enough in case of issues.
If every part takes such a long time, it's likely that this project will progres slowly, at least in the begining...

Just I little progress but a lot of time involved... Here are two screenshots to give you an idea of the current state of development:



For now, I'm just able to launch revolt... but with command line support . Great, no?

I think you will appreciate the nice design, and the wonderful colors...

Good to see progress!

And it is SOOOO beautiful lol, not because of the nice colours, but because of what it does .

Great to see what you're working on Jig. I will gladly test things out, whenever you think it's time for a test. (;

jigebren @ Feb 11 2010, 07:33 PM wrote: I think you will appreciate the nice design, and the wonderful colors...

Oh man that will use to much memory!
Use something like...

This?

Thanks, guys. I can't tell for now when the firt tests phase can begin. When you're in the beginning of a development, there is such a lot of things to think about, and sometime I spend a lot of time on a very little feature while some bigger stuffs can be made more quickly... It's quite unpredictable.

I'm trying to implement the first memory patches, and it looks like it's working.
But I have to remember all the various things I tried in the previous days, and the notes I've written down are a bit messy/lacunary, as usual...


@nero
Unfortunately, I can't compile for DOS, but I think that you can use the debug.com tool to achieve some interessting results.

Some news about the current WolfR4 state.

The memory patcher seems to work, but is currently incomplete (ie: I have to restart WolfR4 everythime I restart revolt to clean up the values in memory).

I've already implemented two bugfixes (bug on loading nhood1 under vista and bug on loading nhood tag battle), but I haven't thoroughly tested the result.
I've also added the support for DirectPlay Lobby launcher (eg. RVHouse), I hope it will work.

The custom models support is still far from being done...

I'll soon need a beta tester that has the GoGo bug and Bugatux bug. I worked on this again, and I maybe have hope to fix it, but I can't test anything as I don't suffer from this bug myself. If you feel you can do the job, tell me.

Well, I can't really say it's a rush of beta testers thronging around here to help, but here is a little news anyhow, if anybody but me still feel concerned by this project .

I have redesigned / improved a bit the memory patcher part of WolfR4. At first view it seems to work properly now. I have discovered new programming stuff, and it helps me specifying patches informations/action in a more consistent way (ok, maybe that kinf of gobbledygook doens't look very attractive).

I still have some bugfixes/tweaks to work on and implement before I start working on the custom models support for tracks.

The above request about bugatux is still topical (in fact, more than ever).

jigebren @ Feb 13 2010, 05:28 AM wrote: I'll soon need a beta tester that has the GoGo bug and Bugatux bug. I worked on this again, and I maybe have hope to fix it, but I can't test anything as I don't suffer from this bug myself. If you feel you can do the job, tell me.

sorry cant help you. at least a pointless reply is better than no reply at all...

I can probably test if it fixes the Go bug. I'll just have to make sure again that it happens without compatibility mode.

I tested (offline) and even the GO bug doesn't happen for me (compatibility mode was off). So I guess I can't be much of a help testing those .

I dont know what it is?

rickyd @ Feb 14 2010, 05:55 PM wrote: I dont know what it is?

The GO GO bug is the GO startup logo appearing at regular interval during a race, and I think the time is reset at the same time (not sure).
The bugatux/tuxabug is when time at the end of a (multiplayer?) race are inconsistent for some players.
These bug seems to be related to the processor used, and mainly with hyper threading and multi-core cpu.
But I'm not 100% sure of this, as I don't have any of these bugs myself.

Sorry, like Zipper, I don't suffer from these.

Geoff

I might be able to check my old Windows XP for the GO bug, I remember having it on that computer. I don't know if I can help with the tuxabug, though. Also, I don't have acces to my Windows XP all the time, so could you tell me what you need to know more exact?

O yes I had that dumb go in the track.
And yes the time was all screwed up.
I just put it in windows 98 comp. mode
and no more go and the time was right.
All fixed up here.

@rickyd
Ideally, I'd prefer having someone for which the Win98 comp. mode trick doesn't work, but if you're sure that deactivating it on your system brings you the 2 bugs, it's likely that you can do the job.

What I need is:
- you run revolt to be sure you have the Go Go and tuxabug bugs.
- then you launch revolt with a supplied WolfR4, with option to correct the GoGo bug activated, and you check if the go go bug disappear, and check if bugatux is still here.
- then, same operation but with just the option to correct bugatux bug.
- then same operation with both options activated in the same time.

Nothing very hard in fact, you just have to be rigorous...

I will go and turn off the compatibility and run re-volt.
I will see if the go go thing comes back.
Then I will get back to you.

The go go thing is back.
The time is messed up again.
So I guess Im set.

Ok, nice, rickyd. I have some stuffs to polish before I can send you the file, but I hope I'll make it soon. Maybe tomorrow... I'll PM it to you when it's done. Thanks, man.

No prob.
happy to help.
Talk to you soon.

I have implemented the following bugfix in WolfR4:
- slowdown or hang on level/frontend loading (compatibility mode deactivated)

And I worked on a tweak/fix to prevent revolt from eating all the cpu in some specific cases, for example, when you run revolt in windowed more, and the revolt window is not currently active.
That's an idea I had when I try to understand why revolt still use so much cpu even on quite modern system (ok, my P4 2.40GHz isn't very modern, but it is for revolt). I can't find any real issue when the game is running with vsync activated, it looks like everything is normal, and the fact is that it use a lot but not 100% of the cpu in that case.
But I was able to fix the abnormal cpu consumption in some other cases, like the one I said above. I've also implemented this fix in WolfR4.


@rickyd
I think it's clean enough/workable now, so I've send you the WolfR4 file for testing the Go Go and tuxabug bugs. Instructions are in a previous post. Thanks.

Testing now.

Ok from what I can see.
The fix bugatux dont work if it is to fix the time problem.
But the fix go go fixed the go go and the time problem.
The go go and the time problem are the same.
So the go go fix fixed both problems.
I dident see where the fix bugatux did any thing.
I hope this helps you. This is a cool program.
Let me know if I can help more.
Talk 2 U

I'm very glad to hear that. At least, the go go bug is fixed. Thanks rickyd for the help.

The go go bug and timer bug are closely linked, so the result is not very surprising.
At the begining, I thought there is one bug and that the fix 'go go' would be enough.
But after some discussions here, it appear that we could be in presence of two bugs...

So I tried to implement a fix for each bug, but develloping without being able to test (as I don't have the bugs) is not easy, as I can't try different things to see which work.

But in fact, I'm wondering if you (rickyd) really have the bugatux on your system, I mean, the real bugatux, the one that messes with the time result during an online race, and that could be due to mutli-core cpu.

So, I'm still looking for someone who have this tuxabug...

I have a p4 at 3.1 ghz.
Not duel core.

Is this program going to be like the phoenix program
that rst made but wont share?
I hope it is!
If so then all will be set right for rst.
A nobody.
LOL

Without dual core, I think you can't have the bugatux (except maybe if you have hyperthreading stuff), so the results from your test are normal, as my fix deals with multicores.

About the phoenix comparison, I can't really tell. But for now, the last big thing that lack to WolfR4 is the support for custom models. I have already done it as a standalone little testing tool, so I know how to do it. But I have to rewrite it all again for wolfR4, in a clean way, and sometime I getting tired of the time I spend on this tool... I hope I will finish this rewritting part soon.

Keep up the great work.
This will be so cool when you get it dun.
And thank you for your work.
Talk to you soon.

This will be awesome! arsetea will soon die...

I have both the tuxabug and the Go Go Bug

And Dual Core Processor

If you need a tester, I'm your man

e-mail me: thalesbelchior@hotmail.com

A great news for advanced track makers (and custom tracks users consequently).

I think the custom models support is on a good way... It's still incomplete and the inside code is quite dirty as I have reshaped it several times, but the first tests seems to be working good. (eg. I can have automatically the good skymap on "skyscrapers").


@Aylown

Aylown wrote: I have both the tuxabug and the Go Go Bug

Sounds good. I'm going to send you the test file soon...
EDIT: That's done, I didn't use your email, but check your PM.

Cool!
Can you tell us what custom elements we will
be able to use in the custom tracks?
Like sound models gfx?
This is just what we need.
Thanks!

enabling MP3 would be appreciable. The carfix (TMMCC) included with this would be a good idea

rickyd wrote:Cool!
Can you tell us what custom elements we will be able to use in the custom tracks?

Well, the exhaustive list would be too long, but I think all .m objects from 'models' directory, the skymaps, sun.bmp, dragon.bmp and water.bmp textures, all .wav files from one stock track chosen as a template (but any stock track can be chosen), and maybe the generic sfx from the wavs directory... Hope it's enough

But there will be some restriction on the file path length. When patching the value in revolt, changing the lenght is practically impossible. It means that custom models will probably have to be placed in another directory than the track itself (a directory with a very short name, like 3 or 4 letters only, it's a bit long to explain for now).

cat wrote:enabling MP3 would be appreciable. The carfix (TMMCC) included with this would be a good idea

The MP3 stuff has already been discuss here (urnemanden (and kay?) asked for it), and it is stated in the TODO list in the first post: I won't do it for now, I'm not very interested by it. But if a lot of ones asked for it, maybe one day I'll try.

For the carfix, I always though there is no need to have so much cars at the same time in revolt, and that the use of an external tool (not sure of the name, but for example carorganizer) to enable/disable used/unused cars would be better. But I will look at it, I think it won't be too difficult.

if your looking for TMMCC (carfix) its on themeandme's site.
http://themeandme.de
will we be able to use the .ace files? they are for sounds and appear only in stock tracks. they do not work at all in custom tracks.

and also, forgive me if im wrong but i always thought that the gogo bug was called the tuxabug not the bugatux?
awesome work

I took a look at the carfix (or rather at the carbug) yesterday. Well, it was less easy that I would have thought. It seems that there is a kind of stack overflow, for the little who understand what it means (there is a place in revolt where it don't know which intruction to execute because this information has been overwritten by cars informations). I don't know how the TMMCC works, but I think it uses another trick than correcting this directly.
EDIT: I maybe have an idea about how TMMCC could work. Does anybody has ever noticed that when using it, some custom cars are never used as opponent?

Able to use .ace file? Could be a good idea. In fact, I didn't know they wasn't usable in custom tracks. But anyhow, if the inside format is unknow, it won't help much as you won't be able to create your own file. And we have nothing about this (or any other thing related to 1207 patch) in the source code. Before I work on this, I have to be sure that we know what this file does, and how.


About the bugatux/tuxabug name: It's really bugatux...
It came from a french player whose name was Tux or something like that, who was the first to report this bug. And in a somewhat bad french, "bug à tux" means "Tux's bug", or "bug of Tux". And "bug à tux" was contracted in "bugatux".

You may find this insteresting:

http://spreadsheets.google.com/ccc?key= ... AI5t-SbIKw

Feel free to apply changes on the sheet.

Regards,
ADX

Zipperrulez wrote: they are for sounds and appear only in stock tracks. they do not work at all in custom tracks.

For Sounds? I always thought that the ACE files only were included in the 1207 & 0916 patch, not comming from the original Re-Volt CD.

urnemanden @ Feb 18 2010, 04:38 PM wrote:

Zipperrulez wrote: they are for sounds and appear only in stock tracks. they do not work at all in custom tracks.

For Sounds? I always thought that the ACE files only were included in the 1207 & 0916 patch, not comming from the original Re-Volt CD.

3D sounds, so only for 1207, not 0916.

-

@ADX
I don't know if you were answering for something specific we are talking about here, or if it's just because you think that this link can help in a general way.
But it's true that it can help, I will bookmark this link.

Ah, that makes sence. Does the 3D Sound feature work well, btw? It could be a good reason for me to change to 1207 if it made Re-Volt compatible with 5.1 systems.

I don't know for 3D sounds, I just own a stereo system. Moreover, I've heard that hardware acceleration for DirectSound3D has been dropped in Vista, but I don't know if there is a working emulation layer or something like that.
Having 3D sounds in revolt could be fun, I don't know if someone has already tried and see the difference...


About the carfix stuff, I'm creating a specific topic, it's better.

jigebren @ Feb 18 2010, 04:44 PM wrote:

urnemanden @ Feb 18 2010, 04:38 PM wrote:

Zipperrulez wrote: they are for sounds and appear only in stock tracks. they do not work at all in custom tracks.

For Sounds? I always thought that the ACE files only were included in the 1207 & 0916 patch, not comming from the original Re-Volt CD.

3D sounds, so only for 1207, not 0916.

-

@ADX
I don't know if you were answering for something specific we are talking about here, or if it's just because you think that this link can help in a general way.
But it's true that it can help, I will bookmark this link.

Just chit-chat

To celebrate the return of antimorph here, I will try to release a alpha/beta version of WolfR4 quite soon (maybe this WE).

I'm waiting for Aylown's anwer about bugatux, and I have some (~ a lot of) things to check before I can release it.
Of course it will be in beta/unfinished stage, without custom models support (not yet finished), but you could benefit from bugfixes, tweaks, etc. Check first post for current state (look at the [!]).

Great, no?

jigebren @ Feb 19 2010, 10:53 PM wrote: To celebrate the return of antimorph here, I will try to release a alpha/beta version of WolfR4 quite soon (maybe this WE).

I'm waiting for Aylown's anwer about bugatux, and I have some (~ a lot of) things to check before I can release it.
Of course it will be in beta/unfinished stage, without custom models support (not yet finished), but you could benefit from bugfixes, tweaks, etc. Check first post for current state (look at the [!]).

Great, no?

Great yes But celebrate me not.

Currently I do ponder.

https://developer.mozilla.org/en/JavaSc ... ctypes.jsm
https://developer.mozilla.org/en/WebGL

antimorph wrote:Great yes smile.gif But celebrate me not. unsure.gif

Yes, celebrating the return of one of the fisrt (or maybe the first) revolt patcher... I think it's a good opportunity.
Even if it's just to have a look, it's good to see that you still feel concerned.


But, well, how can I say?
Maybe I was expecting that my announcement above would have created a bit more enthusiasm here... Or was it not clear that it would be the first available release of my WofR4 tool?
I'm a bit disappointed by this lack of reaction. As for the carfix topic: a lot of work, and not a simple thanks.
The real fact is that I had no much time today, but don't expect some release in the coming days, I don't really fell like working for such a thanklessness.

If it is the satisfaction of people's gratitude you seek, you will fall flat on your face.
I still enjoy people's gratitude, but I gave up on prioritizing it. Personally, I care about this project, just not as much as I should... because I'm not a big Re-Volt fanboy right now. I wish you luck on this and please finish it. Don't let people's ignorance get in the way of you. People in this community tend to be lazy leechers- they'll take but they won't say a word. Even I am sometimes.

Well I for one was ignorant but then I'm still catching up on these accomplishments (it will take many weeks if not months I think for me to get my head around all the work that's been done), I think it's incredible what you have managed to achieve. I would look forward to discovering even more and would be happy to collaborate with you. I worked in (intellectual) isolation on this stuff for a while before it got the better of me, but it surely did...

I'm certain that there's a great deal we can learn from each other, I've not been idle in the years I've been away.

Well, antimorph, I don't know if you're going back here at the wrong or at the right time:
- at the wrong, because I don't know why, but this last days, the lack of support here was actually beginning to annoy me, and the interest I had in sharing all my revolt labors was really about to fly away...
- at the good, because it would be a pleasure to collaborate with you, and I agree we could benefit from each other (as I have no doubt I'll be the one who would benefit the most ).

jigebren @ Feb 21 2010, 10:09 PM wrote: ... but this last days, the lack of support here was actually beginning to annoy me, and the interest I had in sharing all my revolt labors was really about to fly away...

If others are like me, there is very little input we can actually offer, but yet a great deal of interest in the overall project.
Antimorph is most deffinately the one on a similar interlectual level that can provide an alternate or like minded perspective of all your work acheived so far.
It is just a real shame that we can not get RST's input on the work he has done for the Pheonix R3, that would make a solid base for a great think tank.

Your work is and will be fully appreciated by all, I am sure of that.

You know, what is sometime boring me is the feeling that nobody take the time to answer even simple question: look at the tuxabug topic, and count the number of guys who tried the arto's tool and sent back the result... and it wasn't very complicated.

manmountain wrote:If others are like me, there is very little input we can actually offer, but yet a great deal of interest in the overall project.

I understand very well, and it good to avoit useless chit-chat, but sometime a little bit of supportive words can't do no harm.
And I'm afraid ones don't want to post here because they fear they don't fit the 'similar interlectual level'.
Let me tell you there is nothing here that you can't understand, just things that you're not accustomed to. Sometime discussion can be related to a specific issue (eg. with arto or antimorph), but keep in mind that even I wouldn't have understand anything to this a few days/months before...

It is just a real shame that we can not get RST's input on the work he has done for the Pheonix R3, that would make a solid base for a great think tank.

Yes, I have been contacted by him in the beginning of the 512 patch, but as our opinions were too different, I think the disscussion between us has come to an end. But sometime, I'm afraid I begin to think he was not totally wrong.