Last message on previous page:
Any sort of software that can help with testing? Something to run the hardware hard and make sure it handles itself? Just something to make it easier to find out if anything was mildly damaged.
Thanks,
Download UBCD and run the burn in tests (ie Memtest86+,etc) from there.
Link:
http://www.ultimatebootcd.com/download.html

So I would need to download the following?
CPU Burn-in
Memtest86+
and then one from each category? or?
Also I suppose I put it all on 1 boot cd?
Thanks,
(sorry for going way off topic from the thread =P)
Unzip with 7Zip. Then burn ISO with DeepBurner. Both are free software, search Google for download websites. UBCD contains all of the tests in 1 CD.

Ouch, that's gotta be a defective motherboard or maybe PSU (Corsair is great, but there are always duds).
Good luck with the rest of your components
I'm gonna try this UBCD, it looks extremely useful.
^Yeah, it's must have tool for any DIY PC builders. Has every thing in 1 CD. Has Memtest86+, SeaTools and a lot of other things.

On a lighter note......
I finally got my E2160 @ 3.2GHz (356x9) with the Memory running the way I want it. All my bios options are pretty standard, for over clocking, except for the memory.
I kept my SMM (SPD) at 2.4 so my DDR 800 is running at 854 (356x2.4) and my timings are 4,4,4,12. I also had to overvolt the RAM. The board does 1.8V, and my Kingston RAM specs at 2.0V. The DDR2 OverVoltage at (+0.2V) just wasn't cutting it so now I'm at (+0.4V).
CPU is at 1.4875V in BIOS
I ran a 12 hour small Prime with no errors and a 4 hour Blend Test. I ran the blend test with the side off so I could monitor RAM temps. They barely got warm to the touch.
This is as far as I can take this chip with this board I think. I also tried to post at 3.3 (367x9) today with zero luck. I had CPU Voltage all the way up to 1.6 and the next step on this board is 1.8V, I don't think I can bring myself to try and post at 1.8V. I loosened my timings and added (+0.2) to the (G)MCH
Let me know if anyone has CPU volt options between 1.6 and 1.8, because I don't and it might be a BIOS version thing.
| Shadow703793 wrote : Yeah, the E2xxx CPUs need very high voltages after 3.25Ghz. Your temps are good. |
I can build on this too, my E2180 runs rock stable at 1.4125V BIOS @ 3Ghz. Yet to get it to just do 6 hours prime stable at 3.2Ghz I needed 1.525V in BIOS. Not only that the heat got quit intense with such a high voltage so I stuck with 3Ghz.
So does anybody have CPU voltage options between 1.6 and 1.8?
Please advise.
Even if there is, it's really not a good idea to go past 1.5V. After all, you do want to keep your cpu for 3years right? Setting it to 1.7V will reduce it from 3years+ to a mere 2years, considering normal use.
I think it will post at 1.65V.
When was the last time you kept a chip for over 3 years?
Stay below 1.5v, unless you plan to upgrade in about 1.5years. I still have a volt modded chip thats still running since 2003. (Check my sig)

| Dunkel wrote : I think it will post at 1.65V.
|
My last pc was a AMD Athlon XP 3000+ with a BFG 6800OC. The PC before that was a PIII with a 16MB Voodoo vidcard. The PC before that was a 486. I make huge upgrades at a time. obtw, I have a Q6600 and a 8800GTS 320MB right now.
| Dunkel wrote : I think it will post at 1.65V.
|
I would truly, really, and honestly like to know exactly how much time you really take off for such over voltages...But really how many people have OC their chip using excessive voltage and eventually years later have it finally die? All I have heard people say is it's gonna take X years away and this and that, but how many people have actually tested to see how long it would really last? For all I can really say with what 'proof' I have read the damned thing could last 5 years with what some would say an above average overvoltage. Of course heat is the other limitation, but I've never read and article where someone said X cpu died after Y years with Z overvoltage. Of course we could always make the educated assumption that it will decrease the life, but by exactly how much? And who's not to say a non-OC cpu should last 20/30/40 years? If thats the case lets say OC decreases the cpu's life by 50%. Since we don't even really know the max lifetime the OC cpu could maybe last say 10 years... I dunno, If it wasn't for heat I would probably OC my E2180 to the max.
Could NB/SB temps on my MB be stopping me from booting @3.3GHz or higher?
Has anybody had a problem with a high OC with there respective chips, and found that actively cooling the chipset would add stability?
I have found that increasing the MCH voltage by even .1V will cause my board to fail to POST. I've been wanting to try active cooling, but I don't see an easy way to mount a fan.
Any suggestions?
| homerdog wrote : I have found that increasing the MCH voltage by even .1V will cause my board to fail to POST. I've been wanting to try active cooling, but I don't see an easy way to mount a fan.
|
During my personal experience I have never had a situation where upping the MCH or FSB voltages gave me 'extra' stability. The system would readily crash as though it never changed. Although I never used the extreme FSB of above 400Mhz, but I'm a little skeptical that changing the MCH/FSB voltages will make a difference. Perhaps I should go back to testing my 8x400 setup with my E2180...
I can post with my MCH @ +0.2V, but that shouldn't be my issue. I've read that we shouldn't need to ad voltage there unless we are closer to 400fsb. I'm trying to post @ 367x9 but I'm defiantly going to try a x8 sometime soon.
If you have room for it you can use the Antec Spot Cool. Tom's just used one in there high end build this month. They like it a lot.
| homerdog wrote : I have found that increasing the MCH voltage by even .1V will cause my board to fail to POST. I've been wanting to try active cooling, but I don't see an easy way to mount a fan.
|
I was able to do this on an old computer I rebuilt for my buddy. The NB fan was pretty much shot and made that awful typical grinding noise when a fan is at the end of its life. So I took that off and rubber banded a 80mm case fan to the CPU heatsink Fan. I couldn't believe how well, yet mickey moused the setup worked. The trick is I know the DS3L NB cooler is kind of tall so depending on how tall your cpu fan is you might have to fuss with it. I actually might throw another 80mm fan in my case to see if I could cool it down. When I get home I'll post you a picture if you can't visualize it.
Hmm, I might try that. Thanks.
Still, I'm quite satisfied with my overclock, so I don't have any real need to up the MCH voltage. Then again, I never had any real need to do any of this...
| bildo123 wrote : I would truly, really, and honestly like to know exactly how much time you really take off for such over voltages...But really how many people have OC their chip using excessive voltage and eventually years later have it finally die? All I have heard people say is it's gonna take X years away and this and that, but how many people have actually tested to see how long it would really last? For all I can really say with what 'proof' I have read the damned thing could last 5 years with what some would say an above average overvoltage. Of course heat is the other limitation, but I've never read and article where someone said X cpu died after Y years with Z overvoltage. Of course we could always make the educated assumption that it will decrease the life, but by exactly how much? And who's not to say a non-OC cpu should last 20/30/40 years? If thats the case lets say OC decreases the cpu's life by 50%. Since we don't even really know the max lifetime the OC cpu could maybe last say 10 years... I dunno, If it wasn't for heat I would probably OC my E2180 to the max. |
How many people had their chip die on them because of excessive voltage? Very few. Why? Cause they're pc hardware enthusiasts. Ocing makes you a hardware enthusiasts, which means he or she will upgrade to the latest tech, which is like at least once a year. Thus enthusiast will be on his or her new system, ignoring the older system. And by ignoring the older system, it is not used frequently, so it's life is prolonged.
My friend, a rather hardcore gamer, uses his system 8hours a day. He has not much knowledge on hardware. He had a P4 with 1GB of ram and some ATI vidcard (forgot which one). Within 2years, the vidcard died (the temps were below 60C) and the HD died. Within 3 years, his ram died. Within 4 years, his cpu and motherboard died. He did not have any major heating issues. This is of course all on stock, with no OC.
It has been more than 2 year since, and he has a new mobo, PentiumD, 1GB ram, HD, and a 6600GT. I wonder how long that will last him?
^He should get his PSU checked out.
Heres the 'stylish' setup I did to replace the northbridge fan on that pc I was talking about.
Simply, but effective.
*Edit*
Now that I took a second to really look at it, the damned NB fan is bigger/better than the cpu fan
| bildo123 wrote : Heres the 'stylish' setup I did to replace the northbridge fan on that pc I was talking about.
|
Actually, it's much more effective to use a smaller fan, such as a 60mm. Why? Just look the NB. The majority of the heatsink is under the center of the fan, where there isn't any airflow. Unless you move it, it's better to get a smaller fan imo.
| Evilonigiri wrote : Actually, it's much more effective to use a smaller fan, such as a 60mm. Why? Just look the NB. The majority of the heatsink is under the center of the fan, where there isn't any airflow. Unless you move it, it's better to get a smaller fan imo. |
Ah, but this is were the angle of the shot threw you off. In reality the 80mm is actual right centered so that the NB sinks left border is in line with the left border of the fan so its not stuck in that 'dead space'. Interesting thing though, the heat went down roughly 5C when the fan was sucking air from the sink as opposed to blowing on the sink, so for those of you wondering which way to direct the flow, sucking from the fins and outwards towards panel is the way to go.
| bildo123 wrote : Ah, but this is were the angle of the shot threw you off. In reality the 80mm is actual right centered so that the NB sinks left border is in line with the left border of the fan so its not stuck in that 'dead space'. Interesting thing though, the heat went down roughly 5C when the fan was sucking air from the sink as opposed to blowing on the sink, so for those of you wondering which way to direct the flow, sucking from the fins and outwards towards panel is the way to go. |
Indeed, it did throw me off. I still feel that using a smaller fan would be more efficient. Not cooler, mind you, but more efficient per cfm.
| Evilonigiri wrote : Indeed, it did throw me off. I still feel that using a smaller fan would be more efficient. Not cooler, mind you, but more efficient per cfm. |
Probably, but this is all I could scrap up laying around
Still amazing how this ancient (most parts in here are 6+ years old) and it could play WoW like a champ at average resolutions....which I'm trying to sell to my buddies friend because he can't play for his windows 98 box ain't up to snuff....
Updated 4/14/2008
Added new Prime95 Images. More text edits. Added Disclaimer

Very nice. Those images are HUGE!
Hopefully soon I will be upgrading to a DS3L from my Abit IP-95, which for the life of me I can't seen to overclock my CPU with. I haven't tried very hard, and am not concerned with doing so until I get a new board. Anyway, I was wondering, everyone seems to use a e2180 or e2160 to overclock with. I have an e2200. Does everyone use the cheaper ones because you get similar overclocks for cheaper, or do 200's not overclock as well for some reason? Would a higher multiplier allow it to run with less heat (266x11 vs 333x9 or similar, I am looking at 3ghz or less)? For cooling I plan on using an Intel stock P4 cooler (all aluminum, twice the height of the e2xxx coolers) until I get something better. If need be, I'll lower the overclock to around 2.5-2.6 when I use the stock cooler. Now, I was also wondering if there was anyway to do something like Speed-step. I know not to use it for overclocking, but my computer is on for 8 hours a day and would like to save energy and put as little wear on the CPU as possible. Or would my only real option be Standby when I'm not using it?
^Yes. The E2x00 chips are all limited to a max OC of about 3.2Ghz. After that voltage increases are huge for small FSB gain, which is not worth it.

Where do I disable the speedstep? Or the setting that keeps the MHZ from changing. CPU-Z clocks it at 2670.16MHz to 4005.21Mhz. I just want it to stay at 4.0mhz constant.
^ Simple.
Go to "Advanced BIOS Features"
Set "CPU Enhanced Halt (C1E)" and " CPU EIST" (aka SpeedStep) to "Disabled"
Note: DON'T EVER Disable "CPU Thermal Monitor 2 (TM2)"

Thanks shadow!
Dang Shadow, I could never do the ^ thing as much as you do. It takes me too damn long to make a post; someone else has almost always chimed in by the time I click "Submit your reply."
In the text you say up the FSB voltage at >310MHz, but the BIOS picture shows 400. Which is it?
Also, I have the same board and I'm wondering what the "System Temperature" is ("Temp1" in SpeedFan). It's reading ~50C under load and I thought it might be the Northbridge temperature. However, I put an aftermarket cooler on it and the base of the heatsink is cool to the touch, so I can't figure out what it is. It varies from 45 to 50C, so it's not the Tjunction or Tcase. Any thoughts?
| Quovatis wrote : In the text you say up the FSB voltage at >310MHz, but the BIOS picture shows 400. Which is it? |
I run the FSB at 400MHz and I've never had to increase the voltage. Actually, my system becomes less stable when I increase the voltages...
| Quovatis wrote : In the text you say up the FSB voltage at >310MHz, but the BIOS picture shows 400. Which is it?
|
Its either CPU or GPU(?).

It can't be the CPU as it's nearly constant around 45-50C from idle to load and it takes ~15 minutes to get to that value unlike the CPU that shoots up. Somehow I doubt the BIOS would be reading the GPU temperature, but perhaps. It could also be the southbridge, as it was warm to the touch, but not hot.
What does others' "System Temperature" read who have the board? I'm just concerned because something is hot that's not the CPU, or else it's giving an erroneously high reading.
The northbridge on P35 boards can get extremely hot if not properly cooled. This is one of my only complaints about the DS3L; that heatsink is kinda weak
| Quovatis wrote : It can't be the CPU as it's nearly constant around 45-50C from idle to load and it takes ~15 minutes to get to that value unlike the CPU that shoots up. Somehow I doubt the BIOS would be reading the GPU temperature, but perhaps. It could also be the southbridge, as it was warm to the touch, but not hot.
|
Can you post a screen shot?

| homerdog wrote : The northbridge on P35 boards can get extremely hot if not properly cooled. This is one of my only complaints about the DS3L; that heatsink is kinda weak |
Agreed. That's why I use a 40mm fan on it.

| Shadow703793 wrote : Can you post a screen shot? |
I guess I can when I get home, but what for? In the BIOS (F7) go to PC Health, and then I'm talking about the temperature that says "System Temperature". This same temperature is read as "temp1" in SpeedFan.
I just want to make sure what it is before I keep overclocking on this board. Thanks for the help
^For me Temp1 is Northbrige temps. It never goes above 37C (under 9hrs of Prime95 continuous loading) with fan. Never goes above 39C without fan. If it is motherboard temp it's too high. See and make sure the heat sink is not loose,etc.

What's the max 'safe' temp for the NB? Rather, what's the general rule of thumb for NB temps of P35's?
^I say not to let Northbridge temps go beyond 45C for X38/P35. No more than 50C for X48.

I thought it's higher than that. I know before it was 40C, but it's quite different now.
With no fan in the case, my P35 NB temps goes to 48C. This is with an FSB of 266MHz and default voltages, which is rather hot, so it led me to believe that the max safe temp. is like 50C or so.
^Really? I have never gone over 45C on any of my PCs (Including ones I built and sold)

I reset the NB heatsink and temps are still the same even at stock...48C. Maybe just a bad sensor because the heatsink is cool.
There are 1167 identified and unidentified users. To see the list of identified users, Click here.
