| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Comments:
<0> Send all complaints to /dell/null/
<0> Thanks pb.
<0> pb How goes the 1 gig cpu?
<1> 1 gig cpu? that's ancient
<2> doriz: If it does the job, who cares how old it is?
<0> SHUDDAP.
<2> I've got a 733MHz P3 machine in production.
<2> With an 18GB drive.
<0> You go fdiv_cpu.
<2> Thanks, XTR.
<3> is there a way to view my serial key using the recovery disk
<3> or is it not on there
<0> gcc invinitjig.c 2>error-msg >/dev/null
<4> i ran memtest for 10hours and only got 3errors, does that necessarily mean anything is wrong?
<5> fdiv_bug; i've got a quad 250Mhz server in production :D it has a dozen or so 18GB scsi drives, plus 2 external raids of 14 drives scsi each
<5> bobbuilde; ONE error indicates something is wrong.
<4> shi
<4> +t
<2> peerce: Quad 250? What is that, something Sun?
<4> hmm that really ****s
<5> yeah, a old Sun Ultra Enterprise 3000 aka e3000
<2> Nice.
<5> boat anchor
<5> thing is the size of a kegerator
<6> Mmmmm... beer
<5> also got a pair of e/450 suns, dual 400MHz. and a Dell dual 600Mhz xeon win2000 server (that one has 6 * 145GB SCSI raid so its a good file server)
<4> lol old stuff yet it stil is powerful :)
<4> i got a dual 400Mhz compaq and it runs great for linux
<5> my home linux box is a p3-450 :D
<4> hehe
<4> my firewall linux pc is 66Mhz with 32mb of ram
<4> works fine
<5> the home linux box used to be a p1-100, but the spam filtering on my email started to really bog it down
<4> hehe
<4> if you do a ramtest and get errors, is it necessarily something wrong with ram, or could it be mobo as well?
<5> yes
<4> k
<5> or just that the BIOS speed settings are a bit too aggressive.
<4> i put 1 stick in, ran it, no errors, took that out, put other stick in, ran it, no errors, put them in together, get errors
<5> if your BIOS has an advanced section on memory timings, try slowing down the CAS a tick or something.
<7> peerce, so, it couldn't be the northbridge causing memtest errors?
<5> TedSki; it could be. could be almost anything.
<4> well i set it to what corsair told me to set it to over phone
<4> CAS 2 @ 1.85V
<5> ok, try cas 2.5
<5> see if the errors go away.
<4> hmm well for CAS2 ram im not even gonna run CAS 2.5/3 because thats what i paid for lol
<5> are you overclocking anything?
<4> nope
<5> you realize that the actual system wide performance difference between CAS2 and CAS3 is barely measurable with a microscope and tweezers?
<6> I *********e, so I just happen to have a microscope and tweezers
<4> well my old ram was CAS3 and this ram is CAS2 both PC3200, and i see a big difference so far
<4> except for the random crashes which blow ;)
<6> it's not about if it blows, it's about who it blows :)
<5> so you're running at or under 400Mhz FSB ?
<4> 400Mhz
<4> yes
<4> its actually PC3500 ram, (418Mhz i think), but i run it at pc3200 to test stability first
<5> is this system dual channel ?
<4> yup
<5> so in theory, the two DIMMs are on two seperate channels. THAT will make FAR more of a difference than cas2/2.5/3
<5> try CAS 2.5
<4> its weird that they run fine as singles but have errors together
<4> doesnt make any since to me
<4> and i tried different slots
<5> not at all wierd. capacitive loading of a 400Mhz memory bus.
<4> i was using 1 and 3, then i used 2 and 4, i get less errors, but still errors :(
<5> light can travel about 2.5 feet in the time of one cycle. and there's two transfers per cycle, because its DDR.
<5> wait, shouldn't you use 1 and 2 ?
<4> not on this motherboard
<4> its 1 and 3 or 2 and 4
<4> 1 and 2, 3 and 4 are not dual channel
<5> so 1/2 is one channel and 3/4 is the other channel ?
<4> i think its 1/3 2/4?
<5> what chipset on this motherboard ?
<4> nforce 4
<4> if you put them in 1 + 2, or 3 + 4, they run single
<5> k, so thats two independent channels rather than dual bank interleaved ala intel's stuff.
<4> my problem before was all my games were crashing, but since i put them in 2 and 4, i get no crashing *SO FAR*, but i get memtest errors none the less which bother me
<5> seriously, try both banks with CAS 2.5
<4> hmm say i get no errors, what will that prove
<4> mobo cant handle it?
<5> that corsair is blowing smoke.
<4> ah
<4> "We run 2.0 CAS" == bull**** then?
<5> a single chip probably runs cas2 just fine, but with more than one, you need a little more time
<4> ah
<4> well if 2.5 does work, ill just call them up tomorrow and bitch at them for cas2 not working
<5> some of my servers, if you populate all 4 rows in each of the 8 ram banks (yes, thats 32 dimms), you can't run 400Mhz, you have to slow down to 333Mhz
<4> ah
<5> but the server company quite clearly specifies this.
<5> as they test and characterize that ****.
<8> keyboard flatly refuses to work after the POST screen
<5> you're undoubtably using OEM stuff you homebrewed, so YOU are the 'engineer', and YOU get the characterize it.
<4> hmm well corsair says they are supposed to run in CAS2 on dual channel together @ 418Mhz properly
<4> well its not OEM its retail..
<5> bobbuilde; could be a motherboard limitation too. ask the mobo people (***uming you speak chinese)
<5> this is a retail system?
<5> dell or hp or something?
<4> no parts are retail
<5> huh?
<5> I'm talking about the system.
<4> oh yeah
<4> it is ;)
<4> OEM :P
<5> its up to the system integrator to characterize SYSTEM WIDE performance
<8> so
<8> my keyboard isn't working
<8> i dont give a **** about hp and dell and compaq and crap
<8> i had even worse problems with those systems
<8> i have a perfectly good system that I know works well... just need to get that keyboard working <:)
<4> ha
<4> i just got off phone with tech support
<4> he said, in a non-server system, 3 hours in a 10hour span is normal
<4> whats the truth to this?
<4> 3 errors** not hours
<4> he said its very minimal...what bitches, in my server downstairs (ecc ram though), i ran it for 32hours on 4GB of ram, with 0 errors
<4> maybe because its ecc?
<5> thats pure bull****.
<4> hmm
<5> NO ERRORS are acceptible. anything above that means SOMETHING IS WRONG.
<5> would you mind if your bicycle wheel only fell off once or twice a week ?
<4> haha
<4> he said to call mobo company, unfornately, DFI doesnt give tech support from what i see
<5> DFI is cheap junk.
<5> slapped together with very little engineering other than cost reduction of the sub components
<4> well it sure does smoke the **** out of my asus board
<5> it probably has crappy byp*** capacitors on it
<4> what would you recommend?
<5> the Asus board uses the same chipset?
<4> yup
<4> $240 for asus, $200 for dfi
<5> and the DFI is faster? in what manner?
<4> performance wise, this pc is much faster
<5> measured how?
<5> all other components (disk, video) identical ?
<4> just overall notice of performance in games
<4> mostly cpu/ram
<4> not disk/video so much
<5> asus actually tests their **** and sets things to conservative reliable settings.
<5> by all means if fast-n-flakey is what you like, stick with that then.
<5> you'll be happy.
<4> hmm true
<4> actually ill just try this ram in the asus pc :)thats a good test
<5> so you have a asus nf4 and a dfi nf4, both have the same exact CPU, and same exact video card, and the DFI is 'noticably faster' in games ?
<4> i like my tyan mobo for my server though, little pricey, but i guess you get what you pay for there
<4> yeah
<4> my pc is stable with my other ram init
<9> hello
<5> you might fire up SiSoft Sandra on those two systems, and see if they are running different memory timings.
<4> same, i used cpu-z to check
Return to
#windows
or
Go to some related
logs:
#winvista
#gentoo
#online
#politics
static-fxfeeds.
co-op intarnet
#computers
#gentoo
Cannot redeclare class db_error
#online
|
|