> You can also use TDateTime, but if the clock on your computer is
changed
> during time-measuring, you would get the wrong values.
Which is a very good reason for not using it! And all the floating
point arithmetic and conversion from system time is much slower.
QueryPerformanceCounter can b
Angus Robertson - Magenta Systems Ltd wrote:
>> I am unfamiliar with GetTickCount. Is that an API or something?
>
> it also wraps after 49 days, so
> some care is needed for applications that run longer without a reboot.
For calculating ticks elapsed I use this function:
function CalcTicksAppar
> GetTickCount returns milliseconds since Windows booted. It's
> a core Windows API used for most application timing, although
> it only has a resolution of about 20ms (I think), it also
> wraps after 49 days, so
> some care is needed for applications that run longer without
> a reboot.
>
> I am unfamiliar with GetTickCount. Is that an API or something?
GetTickCount returns milliseconds since Windows booted. It's a core
Windows API used for most application timing, although it only has a
resolution of about 20ms (I think), it also wraps after 49 days, so
some care is needed fo
Thank you for your reply.
I am unfamiliar with GetTickCount. Is that an API or something? It
doesn't seem to be associated with THttpCli. Can you give me a small
example of how to use GetTickCount.
Thank you.
> I'm using a ThttpCli to download a 500kb file from my server. I could
> go bigg
> I have seen but had not time yet to examine the problem.
Thanks.
You will understand the problem in a minute or so when examining. There is a
FHtmlCharSet wich is used for the html-part, but there is no way of setting
the value unless you derive the component. I just made a set-er an get-er
for
> > IMO it is better design to have all access to a given
> > component from only one thread. And in that case, you can
> > define NO_ADV_MT to have better performances because you
> > avoid using a critical section.
>
> Thank you, this answers my questions.
>
> I use 2 servers and some clients in
I have seen but had not time yet to examine the problem.
--
Contribute to the SSL Effort. Visit
http://www.overbyte.be/eng/ssl.html
--
[EMAIL PROTECTED]
Author of ICS (Internet Component Suite, freeware)
Author of MidWare (Multi-tier framework, freeware)
http://www.overbyte.be
- Original Mes
> > And what was the cause? In TMimeDecEx you set destination stream
> > after the part was decoded. You lose your data, because you don't
> > save it:
> > Hope this helps... :)
>
> Not really, you say there's bug in TMimeDecEx, but you don't say what
> you changed to get your 'result'.
No r
Did anyone see my message below from a week ago?
I guess the property I made should be moved from published to proteced.
Regards Bjørnar
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of Bjørnar Nielsen
> Sent: 26. august 2005 12:59
> To: 'ICS supp
> IMO it is better design to have all access to a given
> component from only one thread. And in that case, you can
> define NO_ADV_MT to have better performances because you
> avoid using a critical section.
Thank you, this answers my questions.
I use 2 servers and some clients in same exe, b
> I wonder when it is safe to define NO_ADV_MT.
>
> Must I not define NO_ADV_MT at all when I use threads?
>
> What if I use a socketserver in one thread and another socketserver in
> another thread and there is no interaction between the threads? The client
> connections always stays within the t
I wonder when it is safe to define NO_ADV_MT.
Must I not define NO_ADV_MT at all when I use threads?
What if I use a socketserver in one thread and another socketserver in
another thread and there is no interaction between the threads? The client
connections always stays within the threads.
13 matches
Mail list logo
