Learn how to a build a cloud-first strategyRegister Now

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 317
  • Last Modified:

SetTimer

Hi everyone...
I have a program that needs two timers running simultaneously... is that possible with Visual C++ version 5?
Thanks
0
rose337
Asked:
rose337
  • 3
  • 2
  • 2
  • +1
1 Solution
 
alexoCommented:
Sure!  You can see that SetTimer() accepts an 'nIDEvent' argument which is an integer that acts as the timer ID.

Note however that timers are not a C++ feature.  Such questions are better asked in the MFC or the windows programming areas.
0
 
warmcatCommented:
Bah, alexo 'lightning fingers' beat me to it.

Yes indeedie.  Just use SetTimer() twice with different nIdEvent numbers.  On a WM_TIMER message arriving, examine wParam against the nIdEvent numbers used in SetTimer() to find out which of your timers fired.  In MFC, OnTimer() has a UINT nIDEvent parameter that does the same job.

Regards,

-Andy
0
 
alexoCommented:
On the other hand, you can use two TIMERPROC functions and forget about WM_TIMER messages.  Not my preferred solution but it has its merits.
0
Technology Partners: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
tsauyCommented:
Win'95/98 has a system timer of low acuracy. It seems to me that only the timer slower than
20 Hz is acceptable on a 200MHz Pentium PC. On the other hand, NT has much better results.
0
 
stiffCommented:
Regarding tsauy's comment concerning the accuracy of the Win95 timer routines. There are two more significant problems which cannot be avoided even if the timing routine is hand crafted. First, there is no way in the win95 environment (that I'm aware of) to disable the non-maskable interrupts. This means that any system level function which is driven by these interrupts (such as disk access) will affect the accuracy of the timer. Second, there is an inherent execution latency in the win95 operating system of up to 150 milliseconds. This means that even though your code is clean, if there is another process running which the scheduler deems of high enough priority, it will execute and any event which terminates during that interval will not register as finished until control is returned to your program. The only solution which we have found which enabled us to circumvent this problem was to use a non-multitasking environment such as MS-DOS (yes it is still of value). The latency problem will affect any preemtive multitasking OS.
0
 
warmcatCommented:
stiffy -

In fact there is no upper limit to the '95 interrupt latency, and it's nothing to do with the scheduler.  This behaviour is caused by Ring 0 VxDs turning off Ring 0 interrupts while they get on with some long process.  It is these interrupts which are the backbone of the preemprive multitasking, and if they're locked out the VxD has, uh, uninterrupted access to the CPU.  The most common culprits here are video drivers, the Matrox Mystique one having been observered by myself to provoke latencies > 300mS.  Intensive disk access was observed to provoke latencies longer than 1 second!  This was on a 300MHz P2 using a 'scope to measure the time between the IRQ and the action of the service routine.

Also, I rather suspect that the question was posed by someone wanting a long, maybe multisecond, period regular event where the jitter is not important.

Regards,

-Andy
0
 
warmcatCommented:
... oh, and:

>The latency problem will affect any preemtive multitasking OS.
 
Is not true, 'hard real time' OSs such as VxWorks offer guaranteed interrupt latencies which can be quite reasonable.  It's the implementation of '95's guts which allow outrages like 1 second latencies.
0
 
stiffCommented:
warmcat

I appreciate your comments. I was simply thinking in terms of
Visual C++ (as stated in the question).

Regards Stiff (stiffy? ...oh brother!)
0

Featured Post

Important Lessons on Recovering from Petya

In their most recent webinar, Skyport Systems explores ways to isolate and protect critical databases to keep the core of your company safe from harm.

  • 3
  • 2
  • 2
  • +1
Tackle projects and never again get stuck behind a technical roadblock.
Join Now