sound card(SB16/220/5/1) is wrong in dosbox
dosbox is: sb16/220/7/1

It is understandable that a lot of people get confused by this. This default is historically inaccurate. All Creative cards until SB2Pro shipped with 220/7/1 as default but most SB16 cards shipped with 220/5/1. I used to have SB2Pro years ago and I still remember that the manual had some explanation about using IRQ7, something like: "All documentation attributes IRQ7 to printer but we checked all printers here in our labs and did not find a single one using it so it must be safe." The wording may not be exact but the meaning is. Looks like they upgraded their printers while working on SB16... Said that, all cards since the first SB had jumpers to switch between IRQ2/3/5/7 (IIRC), DMA1/3, and ports 200/220/240 (again, IIRC).

sound card(SB16/220/5/1) is wrong in dosbox
dosbox is: sb16/220/7/1

I also tried that of course 😀

Didn't work either, same behaviour...

Thanks anyway!

Can you choose Gravis Ultrasound (GUS) for Xtracker instead of SoundBlaster?

You need either
sbpro 220/7/1 or
sb16 220/7/5
whereas sb16 has problems with the volume,
so use sb or sbpro

Can you choose Gravis Ultrasound (GUS) for Xtracker instead of SoundBlaster?

Had'nt tried that before, though it's possible. I quickly set up the GUS via config file for 220/5/11/3/5 and installed the drivers refered by the Wiki.

The demo and the standalone player are working now. 😀

However, the full version still quits instantly 😢

I'll just wait to see whether further development of Dosbox will get this running.

Thanks!

Hallo,

i´ve try the same. I want to start the xtracker v1.05. I´ve do all the things that do have done too. Also my v1.05 dosent run and i´ve got the same "fileintegrität"-error when i try to start the program.

At last i thought that my version of xtracker is damaged, and i wanted to try to get an old version from a person ho worked also with the tracker. But i didnt do that. I am also didnt try to start the Tracker on an old realy Dos-Maschine (because, before i read your post i thought that is a problem with my personal copy of the tracker).

But now i am very happy, that i am not alone. Maybe we have two chance:

1st: On www.xtracker32.org there are two people there worked on a Win32-Version of the tracker. Maybe one of them are crazy enough for helping us.
2nd: Matthias Ksoll aka Cosmic an old Radical Rythms Member and one of the 100%-User of the Tracker is still connectable under monolab.de, ksoll2000.de and so one. Maybe he knows something about the problem.

Greetings, Mario

1st: On www.xtracker32.org there are two people there worked on a Win32-Version of the tracker. Maybe one of them are crazy enough for helping us.

I already knew xtracker32.org. Luckily I was able to get this version 😀 If you want to know more just pm me.

However, it's discontinued, I've got the latest working beta. Unfortunately, there are some things missing compared to the dos version, and some of them are rather important(don't know them by heart). That's why I put all my effort into getting it working with DosBox...

I doubt that the developers at xtracker32.org will be able to help with the dos version, because they aren't D-Lusion and certainly don't have access to the source code to tell what's wrong. They programmed xt32 from the scratch trying to build it to be as close as possible to the original one.

2nd: Matthias Ksoll aka Cosmic an old Radical Rythms Member and one of the 100%-User of the Tracker is still connectable under monolab.de, ksoll2000.de and so one. Maybe he knows something about the problem.

I didn't ask Cosmic yet, however I'm not sure he's the right person for this problem. As you stated, he was a user, not a programmer.

Hopefully DosBox will run XT in later versions, time will show. Meanwhile I'll try VMWare and Qemu, maybe they will be fast enough some day 😀

As only the demo version is available for download, there
is no way to tell what's wrong with the full version.....

Hi wd,

I want to try to send you a privat message with the full version off the App via vogons, but a technical mistake "Ihr Dokument enthält keine Daten" by submit the mailform prevented to send the message.

Please send me your mailadress.

Mario

Hi all,

here the latest news:
wd found out that the problem with version1.05 is supply back to an harddisk-check via Interrupt 13 from the programm. Inquiries over the int13 are not supported however by DOS box. It does not become clear, for what this inquiry happens, because something like this isnt in the Shareware v1.03.

Fellaw found out that d-lusion produced a version1.07. This was only a beta-version with a betatest-screen in the beginning, but without any startmistakes (it is possible, that this version havent the int13-Checker). The problem ist, that the version 1.07 is not a regular test-release.

So i tryed to get in contact with the old d-lusion-crew. You can see all programmers under www.d-lusion.com. But i´ve got no request.
The d-lusion-crew

D-Lusion released recently on its homepage the two commercial programs Rubberduck and Drumstation to the free Download. Possibly i can speak with one of them to give the really last version of x-tracker likewise officially freely so that it comes to no copyright difficulties.

Greetings, Mario

Hi,

i've got an answer from one of d-lusion. I ask him about the version-history, and i inform about the int13h-Problem in connection with dosbox. First in the origin language, german:
---
Hallo Mario,

interessantes Projekt, danke für die Information.

Wir haben damals noch eine Version 1.07 herausgebracht, diese benutzt
jedoch als Kopierschutz die gleiche Technik über den int13h (wenn ich
mich noch richtig erinnere, ist ja schon etwas länger her 😉, so dass
dies nichts helfen dürfte.

Hast Du noch Ideen, wie das Problem ohne Neuübersetzen des Quellcodes
zu lösen wäre? Wir wären diesbezüglich für alles zu haben...

Grüße,
Thomas
---
and here in the translation:
---
Hello Mario,

interesting project, thanks for the information.

We at that time brought still another version out 1.07, these used however as copy protection the same technology over the int13h (if I remember still correctly, am already somewhat longer ago;), so that this nothing might help.

Do you have still ideas, how the problem without new translating of the source code would be to be solved? We would be in this connection for everything to have...

Greetings, Thomas
---
What should we do know? Maybe the dosbox-crew can tell us something about the problematic of implemeting the int13h-Routines.

Greetings, Mario

Int13 is implemented, but you have to create disk images
to use it (works much like bochs then). There is no way
around this.

Sounds like they want to keep their protection code, if
not it would be possible to remove that part and repackage
the program.

@Mario:

Thanks for your effort. Maybe I'm able to figure out a workaround with the DosBox Debugger(or within native DOS using Softice). Shouldn't be too difficult, as Int13h is a suitable breakpoint...

@wd:

Hm, sounds interesting. I'll give that a try(not tomorrow but somewhen on thursday or friday).

I believe that D-Lusion don't have access to the code anymore, maybe they don't have the rights anymore or simply deleted it. That would explain why Thomas asked to modify the binary instead of the code.

Well according to the mail, it isn't clear if the dlusion people
really like to have their binaries modified, especially as this
means removing a part of their copy protection in this case.
And it affects the registered, copyrighted version only, not
the demo.

To get around the checks: in the dosbox debugger, set a breakpoint
at int13 ("BPINT 13 *"), when reaching the breakpoint trace until
outside of the function. There should be a mov, a xor, and a cmp.
This and all the following JNE have to be NOPed out, the JE has
to be taken (so modifying the first JNE to unconditionally jump
to where the JE points to works best).

hth

I've got some news about that:

Meanwhile I've tested Beta 1.07 for a while, and it works quite well.

I set Dosbox to

output=opengl
frameskip=1
default scaler
core=dynamic
cycles=35,000
type=sbpro2
sbrate=44100

to have almost 100% correct output. The cpu utilization is about 10-20% with these settings on my system(see below). As soon as I lower the frameskip to 0, the output begins to wobble and the cpu goes up to 60%.

Maybe that's a general Dosbox prob, because it's the same in cubic player.
Still haven't been able to debug the 1.05 version successfully yet 🙁

Last but not least my system specs to compare the performance:

WinXP Pro SP2
Celeron D 3,06@3,61
512 MB DDR333, 32M for Dosbox
Dosbox CVS from 02/18/06 without any patches

Hi

I have the same problem. Does anyone have any news for this? Where did you get the Version 1.07 from?

Finally some good news after years of trying etc.

Xtracker 1.05 runs under the recent Dosbox SVN version using imgmount and boot for MS-DOS 5.0.
MS-DOS 5.0 had been installed with Bochs since I couldn't install to work with Ctrl+F4.

XTracker/X-Play work like charm, though SB16 output has some crackling. SBPro output is next to perfect.

Side note:

During the last few years I've also been trying to run DOS + XTracker on i686 hardware natively.
Intel 845 chipset + Northwood P4 /VIA KT266 + AMD Thunderbird 2400+ were the most recent parts I succeeded with.
Details of my current setup:
Shuttle SK41G
Thunderbird 2400+
512MB RAM
40GB HDD
Win98SE in DOS mode
ALS4000 soundcard