View Full Version : Corsair PC-3000 issues
I've got an EPoX 8K3A+ mobo, and I'm using Corsair XMS PC-3000 RAM. Well, because of stability issues, I'm only running at stock speeds (so getting PC-2700 RAM speeds). Well, unfortunately, in SANDRA, I'm only pumping about 2.0Gb/s of memory, which is what I should be getting were I using PC-2100 RAM. Also, if I bump up the RAM's specs anywhere above normal, I get serious stability issues (BSODs, etc). Now, I traded in the RAM and got a new stick, but the problems still persist. Could this be a faulty mobo?
Originally posted by Phleg
I've got an EPoX 8K3A+ mobo, and I'm using Corsair XMS PC-3000 RAM. Well, because of stability issues, I'm only running at stock speeds (so getting PC-2700 RAM speeds). Well, unfortunately, in SANDRA, I'm only pumping about 2.0Gb/s of memory, which is what I should be getting were I using PC-2100 RAM. Also, if I bump up the RAM's specs anywhere above normal, I get serious stability issues (BSODs, etc). Now, I traded in the RAM and got a new stick, but the problems still persist. Could this be a faulty mobo?
stock speed for you would be equivalent to pc2100. (133mhz)
pc 2700 is much higher. so you are running your pc3000 at pc2100 speeds not 2700.
have you tried messing with the ram timings?
Originally posted by Maxvla
stock speed for you would be equivalent to pc2100. (133mhz)
pc 2700 is much higher. so you are running your pc3000 at pc2100 speeds not 2700.
have you tried messing with the ram timings?
No...in my BIOS it's showing I'm at 166MHz for the RAM, 133 for the FSB. Those are stock speeds for the board. And I can't fiddle with the RAM timings, or else my system won't boot, or will start BSODing every few minutes.
Also, if I overclock my system (even to as little as 140MHz instead of 133), the system refuses to boot. I thought it was the memory until I replaced it with another DIMM...am I correct in assuming that it's probably the mobo?
does your motherboard come set stock with a 3:4 memory divider?
thats weird it would be 133/166 stock.
well if its 166 it could be the ram, but i doubt it.
is this a new board you just bought or is the memory the new part?
All of it is new. And by stock I meant the base for all my components...it's at 333 for the mem, 266 for the CPU, 66 for the AGP, and 33 for PCI cards.
Originally posted by Phleg
All of it is new. And by stock I meant the base for all my components...it's at 333 for the mem, 266 for the CPU, 66 for the AGP, and 33 for PCI cards.
since you said 266 for the cpu this means your fsb is at 133x2 which is your DDR speed its not 333. thats just the specification of what it will run when you set the FSB higher.
But it's showing me that my RAM clock is set to 166MHz. I'm using the KT333CE chipset, and I'm positive that my RAM is running at 333MHz right now.
Originally posted by Phleg
But it's showing me that my RAM clock is set to 166MHz. I'm using the KT333CE chipset, and I'm positive that my RAM is running at 333MHz right now.
unless its got a divider you can't have the cpu at 266 and the ram at 333.
do me a favor. download WCpuID 3.0 (http://cgi2.tky.3web.ne.jp/~nrklv/cgi-bin/softdl.cgi?wcpu30g.exe) and take a screenshot of it running then post it here.
this will straighten a few things out.
It's a KT333, the 333 comes from the 333MHz DDR data rate on the ram... At present there is no 166fsb AMD, they are 133. The ram runs at fsb + 33MHz to achieve the 166MHz base frequency for the memory. This is all bone stock KT333 architecture and in no way mysterious or obscure.
Maxvla: every time I run it, my comp freezes =\ and this was even after a completely fresh install. I'm in linux now, though, so are there any alternatives for that?
And yes, the motherboard has a 4:3 divider...
KnownKiller
08-12-02, 02:01 AM
Under Advance Chipset/Dram Timing you have this?
System Performance- normal
Current FSB Frequency-133Mhz
Current Dram Frenquency-166Mhz
Dram Clock- 166Mhz
Dram Timing- Manual
Sdram Cycle Length-2.5
Bank Interleave- 4 Bank
Precharge to Active-3T
Active to Precharge-6T
Active to CMD-3T
Dram Burst Length-4
Dram Queue Depth-4 Level
Dram Drive Strength-Auto
DRAM Command Rate- 2T command
The rest auto
Fast R-W Turn Around-Disabled
Continuous DRAM Request-Disabled
Write Recovery Time-3T
With the CPU set to:
CPU clock-133
CPU ratio-Auto
Vcore- Auto or 1.75
For the Dimm where it says add voltage at +0.10 or 0.20 V
Does the system boot at that?
If you were to say keep the same settings as above but go 140 in the CPU clock you will be running that ram at 166+33=199FSB=398Mhz...
If you set the CPU clock to anywhere from 140-165 it will add +33 to the total for the ram... so remember that.. and thats only when ya set the DRAM timing above to 166.. If you where to set that at 133FSB and the CPU clock at 133 it would stay at 133 1/1... so with those settings you could run the cpu clock at 140 and it would keep the ram at 140FSB as well.
Best bet if ya want to see some speed is to run the ram and the CPU FSB the same.. So you would want to select a DRAM timing of 166 and the CPU clock at 166.. then it would be 1/1 in sync they call it.. Thats where the speed is.. Ya aint gona see much of a difference only running the ram at 166FSB and the CPU at 133... Though if you where to do that you would have to unlock the CPU so you could use a lower multiplier and higher FSB so to keep the CPU in spec without adding much heat.. If you where to try a CPU clock of 166FSB and a DRAM timing of 166 at that default 11.5 multiplier you would be running that cpu at 166x11.5= 1909Mhz which i doubt it would do anyways but maybe with some water cooling.
Though anyways.. the way it works is at 133-165 you have the 4th divider... At 166+ you use the 5th divider for the PCI bus.
Ya might already know that but just encase and for the other people that use mainly Intel.
Also some have reported and i have noticed that ya cant go into the bios and change everything at once and think its gona boot up afterwards even though ya set the settings up correctly. ya have to start with just settings the DRAM timing to SPD and the CPU at 133Mhz.. Then on the next boot change the DRAM timing to 166... Also sometime it will lock out the other settings in the bios for the DRAM clock.. Like sometimes you will have the choice of 133 and 166 or SPD, other times only SPD or 166..
Also check which bios you have... The newest ones will show you the CPU and FSB clock in the boot up screen and even the multiplier and stuff and even the hardware health voltages at the bottom. But dont go flashing those bios till ya get it stable at the right default settings first.. Cause ya wouldnt want the computer locking up or erroring durring a bios flash.. could spell disaster with your system never tuning on again.
So just try those settings above first and see if everything is good... If it is then you can try tweaking the timings a little better for the ram cause we all know Corsair is the shiznot!
:beer:
KnownKiller
08-12-02, 02:04 AM
Oh yeah.. also make sure to use either the 1st DIMM slot or the 3rd one from the CPU for the memory. Those are the most stable slots.
ALso make usre our using a good Power Supply for that AMD set up. Thats alot of stuff and that power hungry AMD chip and graphics and that stick of ram all love alot of juice.
Yeah its gona be tough to figure out whats the prob if everything is new.
I'm using a PCP&C 425W PS. that *should* be enough. However, one of the lights on my case didn't work once I installed that mobo...so maybe it's not getting enough juice...
And yeah, those are the settings in my mobo, except no bank interleave. And I *would* run 166/166, except for the fact that I can't even OC it to 140 and boot up.
KnownKiller
08-12-02, 02:46 AM
Originally posted by Phleg
I'm using a PCP&C 425W PS. that *should* be enough. However, one of the lights on my case didn't work once I installed that mobo...so maybe it's not getting enough juice...
And yeah, those are the settings in my mobo, except no bank interleave. And I *would* run 166/166, except for the fact that I can't even OC it to 140 and boot up.
Ok well the power supply should be fine.. ya aint making this easy since everything wasnt tested before in a different setup.
Well where you can start... well for one if one of the lights on te case aint working then ya probably have the connector in wrong.. try turning it around.
2. Take the side door off the case and make sure ya got enough ari flow in the case.
3. Head here and download a program called memtest86 v3.0 (http://www.memtest86.com) Run the installer and have a blank formatted floppy in the floppy drive.. then boot up with it in there.. it has its own OS so there shouldnt be a prob there... If ya can loop that test with no errors then your memory is fine. The toughest tests are 6-7 you can select to just run those by dropping the console with the "C" key and select test and then 6 then enter,enter..
4.If the ram is good.. then grab a copy of Prime95 (http://www.mersenne.org/freesoft.htm)
Install that and run it and stop whatever it wants to run and select the tortue test after setting up the CPU settings in teh options menu... If ya get threw the first set of tests or just like maybe 10 of them your CPU should be fine. Though loop it over night to make sure... Also a good program to burn in the CPU with.
5. make sure to install a hardware monitor program to watch the system and CPU temp and voltages... Epox supplied one with the included CDROM though it doesnt seem to work for me... So i would advise to grab a copy of HWdoctor.exe from Abits web site here (ftp://202.168.194.114/pub/download/fae/kr7a-raid_hwm.exe) I'm using the saem one on the same board... its the same winbond chip so it does work.
KnownKiller
08-12-02, 02:49 AM
Oh yeah make sure you have installed the Via 4-in1's drivers...
if ya havent got it yet get them Here (http://www.viahardware.com/download/index.shtm#4in1) .
Make sure to get the latest or the offical non bios ones.. the 4.37 final should do fine.
I've got the 4 in 1's. Memtest can run for hours (the windows one, where you specify how much RAM to use) without a problem, unless I'm overclocking in the slightest bit. Any benchmark/burn-in can run perfectly fine...until I overclock. Then memtest picks up errors, Sandra and 3D Mark BSOD, and I usually can't even boot. The light was plugged in fine. I even cut off the LED and attached another one to see if the LED was faulty.
KnownKiller
08-12-02, 04:26 AM
Welp i guess you should look into returning the M/B. Dont buy another Epox board. Buy the Abit KX7Raid next time. Better overclocking and better memory bandwidth.
Either that or your system is overheating...reason ya cant overclock at all.
That memory will do 200FSB easy.. That CPU should overclock atleast some. SHould run up to 1700Mhz no probs, if not 1800Mhz.
KnownKiller
08-12-02, 04:34 AM
Last bit of advice is to check and make sure that the jumper on the board is at 133FSB and not 100FSB.
Also feed the ram a bit more volts.. go up to 2.75 total.. So give it +0.20/.
Corsair will warrenty the stick up to 2.80 volts.
If the stick will run stable at 166 with no errors i dont see the ram being at fault... I know the thing is rated for 180FSB but you cant try it that high cause of the system probs. If the stick was a bad stick it would error at whatever setting.
Oh yeah, I keep forgetting to mention this...
My CPU runs at about 76 degrees celsius. However, I'm using a Delta super high and a ThermalRight AX-7, so I should be getting much lower temps. I asked about this a couple months ago, and someone said it was because the sensor was closer to the die. Is this correct?
KnownKiller
08-12-02, 03:22 PM
No its not closer really.. Its really lower then any other board i've seen.. Its small.. Also its centered right below the die.. Onlike other boards That usually a little off center.
The board has an internal die reading as well though you will have to use some other program to from from that senser then the hardware doc. Not sure the name though.
heheh but hell 76 degrees C??? WOW.. And that thing is still running eh? I mean crist.. mines running at like 48-50C if i let it and i dont like that. So much that i'm not running it at defualt speed and voltage just to keep the heat down.
edit... oh found the name its motherboard monitor 5... Might want to look on fileplanit or something cuase i'm not sure where to get it.
I have it =\ And so yeah, I probably have a faulty mobo? I know it's not the processor or RAM (or atleast, I'm pretty sure), because I've switched out the RAM, and I've switched out the CPU.
Also, I got WCPUID to work, here's the readout:
And this is the MBM readout:
To note, I'd had an 1800+ AGOGA in the computer. I bought a processor for someone else, and noticed it was AGOIA, so I asked them if they would mind switching. The AGOGA one ran at 60 degrees celsius, and this one runs at 75 degrees =\
Edit: and during Prime95, it hit 92 deg celsius. Daaahamn.
KnownKiller
08-12-02, 04:48 PM
heheh yeah those temps sound crazy..
Yo i'm sure your already using this but anyways just encase.
Use Artic Silver 3 thermo goop..
Heres (http://www.arcticsilver.com/arctic_silver_instructions.htm) There instructions on how to apply it correctly just encase ya dont know... I mean hell i've been using the stuff for years now and didnt know the way that they tell ya how to do it... Much better results now.
So give it a read for the hell of it.
Originally posted by KnownKiller
heheh yeah those temps sound crazy..
Yo i'm sure your already using this but anyways just encase.
Use Artic Silver 3 thermo goop..
Heres (http://www.arcticsilver.com/arctic_silver_instructions.htm) There instructions on how to apply it correctly just encase ya dont know... I mean hell i've been using the stuff for years now and didnt know the way that they tell ya how to do it... Much better results now.
So give it a read for the hell of it.
Yup, been there done that ;)
KnownKiller
08-12-02, 08:49 PM
Originally posted by Phleg
Yup, been there done that ;)
hahahaha... well i guess theres nothing left to do but to take that M/B out back and SHOOT IT! Put it out of its missery.
Buy a abit KX7 if ya want to do some overclocking.
I have both and am tring to get the women to let me buy another Abit KX7 to replace the Epox in the second computer since it cant stand the board with its overheating and crappy overclocking.. Plus the fact that, that board desroyed 2 new sticks of Kingmax 3200 i bought a few weeks before on the first boot with that board.. didnt even last like more then 5 secs in the bios for the first time.. I was like cool its working then bam! The computer shut off and it took me all night to come to find out that the 2 256 sticks were dead.. The board still worked though.
Last time i took the epox board out of the case to add a different heatsink i come to find that all 4 of the mosfets (2 on front and 2 on back were burning up.. I could see that it was getting hot or something.. Cause where the mosfets were soldered to the board it was a brown yellowish color... So i dont know what the hell is going on with that board... Seems fine but those mosfetts are makin me worried that it could happen again.
Tamasha_Strife
08-12-02, 09:05 PM
With 92C load temps it's no wonder your system gets unstable. What are the ambient temps? Also before you blame the board and what-not, you should get some kind of accurate reading as to what your CPU temps really is at idle/load. No CPU will OC any with temps like the what you're reporting
Also you said you have a PCP&C yet your voltages aren't super good, I thought they said all their PSU were made with 1% error tolerance max...
KnownKiller
08-12-02, 09:19 PM
Originally posted by Tamasha_Strife
With 92C load temps it's no wonder your system gets unstable. What are the ambient temps? Also before you blame the board and what-not, you should get some kind of accurate reading as to what your CPU temps really is at idle/load. No CPU will OC any with temps like the what you're reporting
Also you said you have a PCP&C yet your voltages aren't super good, I thought they said all their PSU were made with 1% error tolerance max...
Well yeah thats true about the P/S... though it dont really matter how great the P/S is.. it depends on how far the power has to travel threw the board to its destonation. So if it travels far then it will weaking before. The length of the ATX connector also plays a roll in that as well.
i just check what you were talkin about with his lines... they look fine to me... The 3.3+ is the same as mine... low ofcourse at only 3.26... that bothered me right off the bat cause i thought forsure that i couldnt overclock cuase of that... but it wasnt that.. Just the ram couldnt do over 170 Max.. That was with Samsung Original DTL chips.. I long since returned those sticks after trying 8 total of them... Now with the Corsair i can climb up to 200 with the board... Though its stable i have to lower the CPU quite abit cause its showing that its cookin at 48-50 idle with 1.75 volts. Also there artifacts all over the screen cause the chipset cant handle the speed.. I had to drop down to 190 just to get rid of that.
Plus with even at 200FSB max timings 2T i had a crappy sandra bench of only 2600+/2200+... that blows! On the Abit KX7 i get around 2900+/2700+ at 200.. At 205FSB max timings 2T i get 3055/2833+... Lot better if you ask me.
So again... no overheating with the Abit and also way faster bandwidth on the abit plus higher overclock as well.. cant beat it.. 102 bucks t www.newegg.com
Originally posted by Tamasha_Strife
With 92C load temps it's no wonder your system gets unstable. What are the ambient temps? Also before you blame the board and what-not, you should get some kind of accurate reading as to what your CPU temps really is at idle/load. No CPU will OC any with temps like the what you're reporting
Also you said you have a PCP&C yet your voltages aren't super good, I thought they said all their PSU were made with 1% error tolerance max...
Ambient temps are (off the top of my head) 25 celsius. And I had wondered a bit about my power supply, too, since the voltages aren't dead-on. However, I've heard you can also attribute this to a ****ed mobo.
And how would I get accurate readouts on my die temp if my mobo's sensor is giving that for a reading? Wouldn't my processor die in a heartbeat if it were *actually* that temperature?
Nope, sorry, it's around 30-35 celsius. That normal or so?
KnownKiller
08-13-02, 04:19 AM
ya about on the mark... Though usually with a good heatsink you would see like temps in the 10+ range from the system temp idle.
I'm sure something is reporting wrong even though the internal die temps are suppose to be accurate. So if that was the case and those temps from the die are right your CPU would of been toast long ago. So like i said.. something aint right on that board.
i believe something like 60C is the max before something wrong happends like lock ups or a dead CPU.
I've got a ThermalRight AX-7 heatsink and a Delta High Output fan, so I don't think I'm suffering in the cooling department.
Speaking of which, does anyone know if that will fit in the KX7? It's a rather large HSF...
KnownKiller
08-13-02, 04:40 AM
Originally posted by Phleg
I've got a ThermalRight AX-7 heatsink and a Delta High Output fan, so I don't think I'm suffering in the cooling department.
Speaking of which, does anyone know if that will fit in the KX7? It's a rather large HSF...
Best to ask about that in the Abit forum.. head here (http://forum.oc-forums.com/vb/forumdisplay.php?s=&forumid=43)
I dont see why not anyways.. though i dont know cause i'm using a Switech MCX462.
Yeah, I did that too to cover my bases ;p
And umm...by the way, sorry for posting this in the memory forum...I thought it might be the RAM at first, but as I typed the post, I wavered towards thinking it was the mobo... =P Indecisiveness sucks, I think.
badgers
08-13-02, 12:40 PM
this is in response to the original post.
Your score is correct.
Your FSB is 133 and your mem is 166
Your FSB is limiting the performance.
There is no point in running PC2700 unless you increase your FSB to 166 as well.
Either unlock your processor or see if the CPU will just take the jump.
there have been several reviews that have noted that 133FSB just sucks when running PC2700 memory.
http://www.deviantpc.com/reviews/cxms3200/p3.shtml
Hrm...got the new mobo, but not having much more luck overcloking. Stability is much better, with no BSODs at fastest timings/settings at 133MHz FSB 166MHz DRAM. However, I can't get anything higher than 137MHz for an overclock, even with all other settings to their most stable. /sigh...
badgers
08-13-02, 08:36 PM
unlock your CPU and up the Vcore to the CPU-try 1.9Vcore
also, 2.6V on the memory may help
Check your heatsink and your temps.
KnownKiller
08-14-02, 02:39 AM
Originally posted by Phleg
Hrm...got the new mobo, but not having much more luck overcloking. Stability is much better, with no BSODs at fastest timings/settings at 133MHz FSB 166MHz DRAM. However, I can't get anything higher than 137MHz for an overclock, even with all other settings to their most stable. /sigh...
Whats the stepping code of that CPU?
AGOIA.
Gaaa...this is getting annoying. I've now replaced the CPU, mobo, and RAM atleast once. I'm getting enough power to everything, and I've got more than sufficient cooling on the computer (temperatures are now well within normal range, at 40C cpu, 30C case. All my PCI/AGP cards are high quality (Gainward GeForce 3 Ti-200, SoundBlaster Audigy Gamer, Intel Pro/100+ Management ethernet), and I tried doing everything without either PCI card anyways. The best I can figure at this point is that my graphics card doesn't like the higher AGP bus. I guess what I'll try to do next is unlock the CPU, run at 166/166 on the 5:2:1 divider, and use a 9.5 CPU multiplier, so everything is running at its normal spec except the FSB itself.
Edit: Hrm...even overclocking as little as 140 renders my system unbootable. My video card won't display any output, even the BIOS. At 136, it's bootable, but freezes during SANDRA's multimedia benchmark, with everything else at its most relaxed settings.
/grumble
KnownKiller
08-14-02, 04:34 AM
um... dude, I hate to tell you this..
You are using the wrong divider at that 133-141 FSB you trying to use.
Look..
133/4=33
133/5=26.6
140/4=35
140/5=28
So lets see here... 140/5=28x2=56... Your trying to not unly run your PCI buss at under specs at 28, your also trying to run your AGP under spec at 56.
ya need to keep the PCI at 33 or higher.. Dont try lower.. Prob why most of the time it wont boot.
The AGP needs 66Mhz to run.. Again dont try lower only higher hehh.
I'm sure ya knew this.. I know how it is to pull hair over a computer and i'm sure ya just didnt think of it as you were thinking of other settings holding ya down.Plus on the Epox board it set the divider for you...which i really dont like..I like to know what the divider is.
So encap...
PCI needs 33
AGP needs 66
133-165 use the 4th divider
166+ use the 5th divider
:)
Happy overclocking
KnownKiller
08-14-02, 04:38 AM
Sorry double post!
2nd time thats happened to me! argh!
KnownKiller
08-14-02, 04:55 AM
Well i may have been why off on my last post.. I think it was just seeing you say the 5th divider and the 140FSB in the same paragraph that lead me to that idea since you said your graphics card wouldnt even show the BSOD.
Ya know i got the same card as you do.. The Gainward Ti200 128MB. Its fine right now running at 82Mhz on the AGP.
Well ideas... If that card is bad or the settings are it could be causing all the trouble..
1.Got another card to try out?
2.Your not overclocking the card right?
3.Ever inspect the card before for damage or something that looks wrong?... Is the card brand new to where ya never used it in another system before now?
4.Can you try the card in a freinds computer to see if its all good?
5.Ever take the GPU H/S off and inspec the coverage of the goop they used on it? I mean sometimes the damn heatsink aint even touching the GPU... If your gona try that i'd advise that as a last resort cause if ya can still return the card it would be best to do that cause ya might damage the card removing the heatsink. Though if your gona do that ya need to warm up that goop first so it will come off easy.. Though it might not be the same goop i'm thinking of hehe.. Might come off easy.
I can think of more though your not even touching that high of a FSB to even think of those other things.
Sorry, was a typo...correct divider was 4:2:1
Okay, I have...
Removed all PCI devices, removed the floppy drive, removed my DVD-ROM drive, removed my CD-RW drive, removed my spare hard drive, unplugged all auxiliary fans (case fans, etc.), so that NOTHING is attached to my computer except the bare minimum. I went into the BIOS, set all settings to their safest, set RAM FSB to 133, and made sure all BIOS settings were correct. I overclocked to 140FSB, and freeze on startup (same place as usual at 140FSB). So I know it has to be one of the following, since these were the only items that were connected.
A) Motherboard
B) RAM
C) Processor
D) Video card
E) Primary HDD
F) Power supply
Now, I can rule out motherboard, because I just fsckin' replaced it, and while things are a bit better (the old board was incredibly quirky), I still have oc'ing issues. It's possible that I got two bad boards in a row, but highly unlikely.
I can rule out the RAM, because I have had it replaced when I thought it might be causing the problems when I first got my upgraded hardware. Also, I can run perfectly fine on aggressive timings at 166MHz, so at 133MHz on relaxed settings, it should be perfectly fine. It's theoretically possible to be the RAM, however, it's unlikely.
I can not rule out the processor. I just swapped it out for a nearly identical one a few days ago, however, I cannot verify to within reasonable doubts that it is not causing the problems. I will see if I can get a friend to loan me his, and I will investigate further. While unlikely, this is one of the better candidates for the problem that I can see as of right now.
I can not rule out the video card. This is, in my opinion, probably the most likely culprit. It could just be a fluke, and for whatever reason not tolerate higher AGP bus speeds--I don't know. However, the fact that it won't give me ANY display, not even the BIOS, when I'm at higher than 70MHz AGP makes me suspect it. I'll try to borrow a friend's video card to run a few tests.
I can rule out the hard drive. First off, unless there's something horribly, horribly, seriously wrong, I have no clue as to how it could affect this. It's theoretically possible, but once-in-a-billion-lifetimes unlikely.
I can't rule out the power supply. While it hasn't been giving the dead-on voltages I had expected, they aren't far enough off to be causing major problems. However, I could be worng about this. Even so, ramping up the voltages hasn't worked in making the computer stable at higher speeds. This also points towards the video card, as I am able to give enough voltage to the RAM and CPU where they theoretically should be quite stable. However, this makes me wonder if maybe the AGP slot itself simply isn't getting enough voltage. That seems like it may possibly be causing me problems, so I'm going to install Abits mhardware monitor now and check what the AGP voltage is. Speaking of which, what should it be?
So now, I have three possible culrpits:
A) Video card
B) Processor
C) Power supply
I'm about to check out the voltages, so I may be able to check off the power supply in a few minutes. Also, I'm fairly sure I'll be able to borrow a spare video card and processor from a friend, so I *should* be able to figure this out.
If I verify those and it still doen't work? I'll just say God hates me.
Well, I fired up their "Hardware Doctor" program, and, if I am correct in assuming that the center of their settings is optimal, my AGP voltage appears to be a little lower. However, it doesn't seem low enough that it would be causing problems. I don't really know voltages off the top of my head, so tell me if you spot anything that seems too abnormal.
badgers
08-14-02, 06:49 AM
I read that to indicate your Vcore is low and your AGP voltage is low.
stock is 1.75 cpu and 1.5 AGP
I run my CPU at 1.85ish and AGP at 1.7ish(ish means that it is never exact to the 0.01 place :))
KnownKiller
08-14-02, 07:04 AM
Heres what i get with a 100% load on it.. using SETI@home.
The AGP is suposed to get 1.50v Mine is a steady 1.50v.. never moves. The vio is not for the AGP like it was with teh older boards. Now they say that the AGP and the memory share the same line.
Some other lines seem low as well... It might be the P/S after all.. maybe who knows.. Is that an older model or something? is it AMD aproved? I know its like a top brand and pretty damn expensive... though not all P/S's run good with every board out there.
How about another post wtih a load on it.. if ya can.
I really dont see the CPU being the prob cause ya said that ya traded a friend for it right? and the old one did the same thing right?
Hell ya might wana rip your friends P/S out of his case and take it with ya for a night and see... Better hook him up with some beer or something hehe..
KnownKiller
08-14-02, 07:17 AM
The vcore is prob just at default which is 1.75.. prob only seeing it flucuating... Wouldnt worry too much about that.. The real thing i see wrong is the AGP and the mem voltage.. hehe.. Up the mem to 2.7 atleast for now. I know its defult and stuff but still.. i dont know a stick of ram that will run fine at 2.5v.
One last tip here.. Just for the hell of it i'd be like testing this thing outside of your case for now till ya get things figured out.
I mean ya could have something touching the backside of the motherboard that aint supposed to that could cause huge probs if it is. Mostly caused by a standoff thats not being used on the board that is touching somewhere theres no hole on the board.
So if at all possable take the thing out and lay it on a flat piece of cardboard or a M/B box better even.. easier to install the vid card that way. If ya cant get the Power on wire to reach the board from the case durring this just use a flat head screw driver and cross the 2 pins where the cable is nornally connected to power on the board. Or ya could use a key. Kinda like a test bed if ya will.
I mean hell i just had a guy that had probs he couldnt explain with his hard drives not working with them pluged into the board but would if he unpluged them from the board... He was happy after i told him to power up the board outside of the case.
What's Vsb? And, well, most of the others. I don't really know what they're specifically for =\ I really need to take a course on electronics, or read a few books, or something...
Btw, as far as I can tell, the Vcore has stayed solidly on 1.73, with the occasional half-second switch to 1.74 every few minutes or so...so I don't think it's fluctuation.
Searching for alien life:
Originally posted by KnownKiller
One last tip here.. Just for the hell of it i'd be like testing this thing outside of your case for now till ya get things figured out.
I mean ya could have something touching the backside of the motherboard that aint supposed to that could cause huge probs if it is. Mostly caused by a standoff thats not being used on the board that is touching somewhere theres no hole on the board.
So if at all possable take the thing out and lay it on a flat piece of cardboard or a M/B box better even.. easier to install the vid card that way. If ya cant get the Power on wire to reach the board from the case durring this just use a flat head screw driver and cross the 2 pins where the cable is nornally connected to power on the board. Or ya could use a key. Kinda like a test bed if ya will.
Yeah, I've considered it a few times, but it's still kind of a hassle. I've heard of things like that causing problems, so I made sure every riser was *exactly* where it should be. I'll try ripping the mobo out of the case in a bit. And yeah, I know the trick with a screwdriver, too...been messing around with hardware for 5 years or so ;)
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.