| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Comments:
<0> And memory leaks
<0> :D
<1> SQL SErver easily can use a gb
<1> IIS another 256
<1> OS 512
<1> maybe i'll go with 3gb
<2> I have like 1 GB (less actually... between 800 and 900 MB), and it seems enough. My machine never ever swaps.
<3> SQL Server uses a gig?!
<1> it can
<1> I used to run it with 14gb dedicated (AWE)
<0> 3 GB and you lose dual channel
<1> Ashe: really? 2x 1gb and 2x 512? that can't use dual channel?
<0> Ah I thought you wanted 3x 1 GB
<0> Dunno about 2x 1 + 2x 512, I guess it'd depend of the motherboard but you'd better check first
<1> aight, didn't think about that
<0> That's the kind of thing that brings you 150 BSOD/hour
<1> i'll check
<1> lol
<1> don't need that :)
<3> lol. its sad when your process with peak memory usage is AIM
<3> I just checked my process list heh
<3> Nothing has peak memory usage over 60MB for me.
<0> The other day at work we had a service taking 24 GB of memory
<0> (or 4x 6 GB, technically)
<0> I called the developers, they said
<0> "yes, we have a memory leak, it'll be fixed for the end of the year"
<0> So I restarted their service
<3> thats one helluva memory leak... and by the end of the year?!
<0> It went from 42 MB to 550 MB in 30 seconds
<3> They don't release periodic updates or osmething?
<0> They do
<0> Every week there's unitary patches for every software
<0> And twice a year major release
<3> and they cannot fix a memory leak?
<0> Nope, they ****
<3> thats rather sad.
<0> They tried to blame me too
<0> Cause I write the software that runs "below" their service
<0> Of course that piece of software works fine on like 200 clusters but MY software did put a leak in THEIR service
<0> heheh
<3> For server software, I would consider a memory leak rather high priority
<0> Especially when there's 24 GB of RAM on that particular server
<0> :D
<4> what are the gnu c++ libraries called in linux?
<0> And when the customer has 600 calls per second on that machine
<0> libc++
<3> For workstations, no one really cares cause people are all accustomed to restarting every other day thanks to win98
<0> I restart my laptop once a year
<4> thanks
<1> aight, i'm out of here
<3> I cannot really complain that much about XP... only time I restart is for updates requiring restart
<3> anyone used synctoy for XP?
<4> anyone have any idea what this is? ld is messed up? /tmp/ccAudkbC.o:(.eh_frame+0x11): undefined reference to `__gxx_personality_v0'
<4> collect2: ld returned 1 exit status
<4> that happens when i call popen first time i thought i was using it wrong then i tried with an example from a man page
<4> is it an error with gcc?
<5> Hey, I'm still working on a program to send through another application's socket - I keep gettting WSAENOTSOCK when calling WSADuplicateSocket or send() despite the fact that the socket's ID is certainly correct
<5> Could this be caused by the fact that the target task is a different one than the caller, or am I doing something wrong with the socket identifier?
<6> yes
<5> Yes to which? ;x
<6> you can't send through another process's socket just because you know the socket id
<5> Calling WSADuplicateSocket doesn't work, either
<5> I'm guessing this is way more complicated than it appears
<6> you'll likely need to inject your code into that process
<5> Ah, ok, thanks
<7> I suggest you use some form of IPC to send the data to the process/thread owning the socket
<6> what the heck are you trying to do?
<5> It's just something to send and recieve packets in addition
<6> cheat in a game?
<5> to the ones being sent by another process
<5> Pretty much, but it's just for the fun of getting it right.
<6> figures
<5> Heh
<5> Thank you, Asmodee, I'll look that up
<4> nevermind i got it man i feel like a dumb***... .c not .cpp hah!
<4> java is evil it make me forget C...
<8> "Arrays are like virgins. They are very careful about what they allow in themselves". <-- Best array explanation ever.
<3> yes, except people don't go out of their way to use arrays.
<6> heh
<4> hah
<0> That's not even close to funny El_Barto
<9> it's kinda perversion
<9> aaron_: i don't find correct that to praise a language against another.
<0> He's not praising one
<0> He's saying java ****s
<0> And everybody knows that
<10> ...
<10> i argue that java rocks
<0> Yes, but you ****, and everybody knows that
<11> ROTFL
<10> valid point.
<4> as far as ease is concerned i think java is good but if efficientcy is a concern then java dosn't work too well
<9> dunno, java has a lot benefits too, hard to say it's ****in..
<10> aaron_ : ?!
<0> I think he's trying to say it's slow
<10> first of all, it's really not slow.
<0> Yup, it's not slow
<9> hm probably
<10> i was under the impression that that myth was debunked a few years back.
<4> well the java vm is slow
<0> You just don't understand JVMs emulate a 100 MHz CPU
<10> really.
<10> even the one on the dallas TINI which runs at 40mhz.
<2> Why is the Athlon 64 FX-57 faster than the Athlon FX-60 in this chart? http://www23.tomshardware.com/cpu.html?modelx=33&model1=238&model2=328&chart=59
<10> a 40mhz cpu emulating a 100mhz cpu. amazing.
<4> i have to port a entire research project to c++ cause java couldn't cut it...
<4> or a good example is run limewire
<4> play some music files
<4> while using limewire...
<10> and blame java.
<0> It's 2x 2600 MHz
<0> And the other one is 2800 MHz
<4> unless they improperly implimented threading in limewire
<12> Hi, I need to do something in C and I don't know how to proceed... I have to write a small program that uses socket that will run a command on a remote server and parse the output to the client, command is something like ls or even a find, can anyone help ?
<0> So for applications that really need the frequency
<4> its due to the java system
<0> It's faster
<9> don't java accuse wrongly :)
<0> And the second core is not used
<2> Ashe`: Ah, thanks :*)
<2> Ashe`: Can you help me on that page? Which benchmark would be best for me to look at? (since I want C++ compilation to be fast)
<12> I have done my socket... it's only the part to run the command and send the output to the client
<0> CPU Drystone
<11> garf1: #c, popen
<2> ok thanks
<12> thanks
<9> Ashe`: i am not sure, but the Intel cpus has a intruction set that such as
<0> meyrnn: athlons support SSE
<9> when you code int a=blah int b=blah cpu can knows it will possibly processed as a+b blah
<9> than can cache it and works faster isn't it?
<0> You mean branch prediction?
<9> yup kinda
<0> All CPUs have that
<9> i mean more advanced instruction sets are implemented.
<0> Then I'm not sure what you're trying to say
<0> The only chances in the instruction set is SSE1, 2, 3 and x86-64/IA64
<9> i mean which one has more advanced sets athlon or intel?
<0> changes even
<10> all cpu's have that?!
<0> meyrnn: they both have the same
<10> as far as i know, branch prediction is relatively new on the scene.
<9> hm
<9> k
<0> First google hit: http://www.x86.org/articles/branch/branchprediction.htm
<0> The first microprocessor in the Pentium family introduced a simple one-level branch prediction mechanism with many ludicrous quirks. The later versions, Pentium MMX, Pentium Pro, Pentium II, etc. have longer pipelines and therefore a higher need for effective branch prediction.
<0> Now not sure what you call relatively new
<0> If the scale is vawjr's life
<0> Then it's very recent
<10> the scale is the cpu's life.
<0> Well, 1993 is not exactly recent
Return to
#c++
or
Go to some related
logs:
#squid
p0seidon movie
isotonic inurl:undernet
#AllNiteCafe
#AllNiteCafe
#MissKitten
#AllNiteCafe
namel
#chatzone
12WHAT ARE THE RESIDENTS OF THE ISLAND OF LESBOS CALLED?
|
|