[Bug 27896] New: Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

Summary: Wine is too fast ;-)
           Product: Wine
           Version: 1.3.25
          Platform: x86
        OS/Version: Linux
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: kernel32
        AssignedTo: [e-mail]
        ReportedBy: [e-mail]

When I launch Borland C++ 5.02, the "about dialog" contain a count timer in aim
to control the speed of some printing in screen.

But with my CPU, (final time - start time)= 0 then this result is use to divide
a constant  => "divide Error"

My question, Is it possible to make wine running at a slow speed click time to
avoid this problem? (I presume it should be used for some older games)

In other word, Is it possible to virtualise the speed of processor in purpose
to use older program?

thanks.
Stephan

It would be cool to develop/maintain under wine rather windows...

PS: I know is a older framework but...

[Bug 27896] Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

Nikolay Sivov  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|kernel32                    |-unknown

--- Comment #1 from Nikolay Sivov  2011-07-25 09:30:08 CDT ---
How does it compute this difference, how timestamps are gathered?

[Bug 27896] Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

Alex Bradbury  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #2 from Alex Bradbury  2011-07-25 11:29:04 CDT ---
(In reply to comment #0)
> In other word, Is it possible to virtualise the speed of processor in purpose
> to use older program?

I've had success using cpulimit for this purpose.

What behaviour does the program have when running under Windows on the same
hardware?

[Bug 27896] Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

--- Comment #3 from www.recover.eu  2011-07-26 01:55:09 CDT ---
> I've had success using cpulimit for this purpose.

How does it work? where can I find this option? I hope this solve my problem
;-)
> What behaviour does the program have when running under Windows on the same
> hardware?

Under windows, it is not a problem because this "O.S." is so heavy and slowly
that the difference between timestamp is different to 0, but, I think it could
be a problem later also (but the microsoft team will find something to slow
their OS. ;-) ).

I think there is no behaviour under windows then, at term, if it work under
wine if not we'll migrate the program C++ under a alternate framework

thanks.

Special thank to all developper of wine. You make a good job!

[Bug 27896] Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

--- Comment #4 from Nikolay Sivov  2011-07-26 02:05:37 CDT ---
(In reply to comment #3)
> Under windows, it is not a problem because this "O.S." is so heavy and slowly
> that the difference between timestamp is different to 0

That explains nothing, please try to figure out how it works to get timestamps
(API calls for example). This could be still a bug in Wine.

[Bug 27896] Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

--- Comment #5 from Alex Bradbury  2011-07-26 02:09:31 CDT ---
(In reply to comment #3)
> > I've had success using cpulimit for this purpose.
>
> How does it work? where can I find this option? I hope this solve my problem
> ;-)

I should have been clearer, it's an external tool
(http://cpulimit.sourceforge.net). It's probably packaged for your
distribution, it definitely is for Debian and Ubuntu and the man page describes
how to use it. It may not work in your case, but I've had success with a game
where sound plays far too quickly. It may not work in your case because the
program works by constantly sending SIGSTOP and SIGCONST to the process, but to
prevent your divide by zero error the process I believe it would need to be
suspended between acquiring final_time and start_time, which might not happen.

I agree with Nikolay, you shouldn't dismiss the possibility of this being a
wine bug.

[Bug 27896] Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

www.recover.eu  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID

--- Comment #6 from www.recover.eu  2011-07-26 03:50:26 CDT ---
Due to my precedent test, I assume the bug is RESOLVED. 
The problem where in other place than I thought. 
Not in timestamps but in loading the police 'Courrier New','Arial','Time New
Roman',... present in every windows OS and not by default in wine.

When you install polices, IT works.

cfr. AppDB RAD C++ version 5.02

Thank and sorry for this

[Bug 27896] Wine is too fast ;-)

http://bugs.winehq.org/show_bug.cgi?id=27896

Austin English  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #7 from Austin English  2011-08-02 11:02:34 CDT ---
Closing.