The problem is hyperthreading is typically pretty good to have on,
depending on instances of different servers running on it, as a loose
rule of thumb I'd personally have it on if you use more than 1
dedicated server per "real" cpu. So if you use > 2 game servers on a
dual, I'd try it on personally and see.

I think the problem is there's no "proper" way to my knowledge (and
again I don't know windows that well) to properly figure out how much
"spare" there is on a virtual cpu, but that doesn't mean its not
useful to have there. So I would guess valve can't do any more than
the o.s will report.

It wouldnt surprise me even if hyperthreading is turned off, it still
shows as double what taskmanager has (assuming you have 2 real cpu's),
which is actually "sort of" correct. I.e if it shows as 95% of one
cpu, thats actually correct as it can never use the 2nd cpu once it
hits 100%, only other processes can. Thats typically the problem with
large servers, as huge slot count server would likely better run on an
64bit Ath4000 than a dual 3gig xeon, and the xeon naturally comes into
its own once running multiple instances.

Basically though, don't read too much into stats or the taskbar in my
book. Watch what individual cpu's are doing.

On 5/12/05, Whisper <[EMAIL PROTECTED]> wrote:
> --
> [ Picked text/plain from multipart/alternative ]
> May try then with HT turned off and see what happens.
> On 5/13/05, Steven Hartland <[EMAIL PROTECTED]> wrote:
> >
> > That is the case Ian is surprising the amount of people who don't
> > know how to interpret taskman results, ive had many an argument
> > about CPU usage due to this :P
> >
> > Steve / K
> > ----- Original Message -----
> > From: "Ian mu" <[EMAIL PROTECTED]>
> >
> > > Just wondering as a 40 player server will naturally take a fair bit of
> > > juice, could it be that stats shows a usage on a single hyperthreaded
> > > cpu (i.e 1/4) and task manager shows the total which isn't quite
> > > correct? (ie there isnt really 4 full cpu's).
> > >
> > > Don't really know windows well enough in how it reports cpu usage
> > > compared to Linux though, just wondered if it might be the "cause" of
> > > the different reporting.
> >
> > ================================================
> > This e.mail is private and confidential between Multiplay (UK) Ltd. and
> > the person or entity to whom it is addressed. In the event of misdirection,
> > the recipient is prohibited from using, copying, printing or otherwise
> > disseminating it or any information contained in it.
> >
> > In the event of misdirection, illegible or incomplete transmission please
> > telephone (023) 8024 3137
> > or return the E.mail to [EMAIL PROTECTED]
> >
> >
> > _______________________________________________
> > To unsubscribe, edit your list preferences, or view the list archives,
> > please visit:
> > http://list.valvesoftware.com/mailman/listinfo/hlds
> >
> --
>
> _______________________________________________
> To unsubscribe, edit your list preferences, or view the list archives, please 
> visit:
> http://list.valvesoftware.com/mailman/listinfo/hlds
>

_______________________________________________
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds

Reply via email to