raju Posted November 26, 2006 Posted November 26, 2006 holy ****, the new bios on read delay 7 boots up over 450fsb, stability testing is in progress as we speak, on an e6600 no less, vnb is 1.77... I will amend the guide soon as I have some results... what a board... regards Raja Share this post Link to post Share on other sites More sharing options...
raju Posted November 26, 2006 Posted November 26, 2006 this is where I just ran into the trc issue... memory settings on auto, i dunno if it's a misreport, or the bios just sets the bank cycle time wrong over certain fsb's.. here's a screenshot of what I mean.. Orthos is running fine so far, so I guess it may just be a misreport... Share this post Link to post Share on other sites More sharing options...
raju Posted November 26, 2006 Posted November 26, 2006 I just set them manually and the trc is back again to normal... Share this post Link to post Share on other sites More sharing options...
IvanAndreevich Posted December 23, 2006 Posted December 23, 2006 How did you get TRC to be 16? On Auto I get it set to 5. On manual settings, I get it set to 1. How the heck can I get it to some reasonable value? The above happens at stock and with an overclock. I think I am getting instability because of this. I know memory is usually quite sensitive to this timing (at least DDR1, in the past). PS: only got the board yesterday Share this post Link to post Share on other sites More sharing options...
raju Posted December 23, 2006 Posted December 23, 2006 depends on your fsb and divider Ivan... regards Raja Share this post Link to post Share on other sites More sharing options...
IvanAndreevich Posted December 25, 2006 Posted December 25, 2006 Divider: 400 MHz FSB: any, as far as I can tell What's the big problem with adding the TRC option to the bios? Share this post Link to post Share on other sites More sharing options...
raju Posted December 25, 2006 Posted December 25, 2006 also depends on the spd timings of your ram.. Share this post Link to post Share on other sites More sharing options...
raju Posted December 25, 2006 Posted December 25, 2006 without a signature means I can't see what you've got... Share this post Link to post Share on other sites More sharing options...
WintersFury Posted December 28, 2006 Posted December 28, 2006 After reading all this I just have a few questions. What does the voltage adjustment that the bios has in MV really do? I get the impression that execute disable bit and c1e don't have to be turned off but may impact stability? So experiment and turn off if needed? I'm mainly interested in voltages as I was having a game patcher crash running at 300fsb on the 800 divider and figure either my NB or CPU needs a bit more in the way of voltage. I didn't bother Orthosing it as it's just till I get my PSU back (blew after 8 hours running stock) but I'd imagine it's not stable there for some reason or another... Ran all my benchies fine so I figured it was stable enough to game with *shrug* Thanks for any help Share this post Link to post Share on other sites More sharing options...
Thraxz Posted December 28, 2006 Posted December 28, 2006 Raju, you're a machine. I wish I could hang with you but I can't do half of what you do. Though I'm pulling out primes at 3.8GHz. Ugh, I just can't believe how hot this chip is. I know if I could keep it cool, it'd do so much better. Share this post Link to post Share on other sites More sharing options...
raju Posted December 28, 2006 Posted December 28, 2006 After reading all this I just have a few questions. What does the voltage adjustment that the bios has in MV really do? Depends which voltage you are talking about? If you mean overvolting in general then the response can be quite technical and is better settled by an ee. In laymans terms look at voltage as potential/storage of energy, and current as flow. Higher voltage is more potential to release energy and supply current (depending on how much current the circuit and psu can put out at max and heat cooling ability). Everything has an optimal range, when you go over safe specs you have a margin on most cpu's and nb's to make them run faster when more voltage is applied. The counter balance is heat, and overall voltage handling capacity over sustained lengths of time. That's why we monitor temps. Though we must remember that temps alone will not guarantee a long component life, especially when overvolting. Overvolting creates more strain on a motherboard circuitry and produces more heat, which has to be cooled. Typicaly DFI boards are setup to overclock well at the component level. For a board to be a good overclocker you need a good on board power circuit, a good psu, a reasonable cpu and good ram. Last is bios settings ability that gets covered here quite extensivley in the OCDB. That's why DFI-Street has a OCDB so you can check ballpark voltages for overclocks in general. Most of the entries are on the safe side, so they're usually good to go. When there is more voltage potential available it provides the cpu with more of an energy source to tap into upon load, giving the ability to run at higher speed. It also strains the internal traces more as they are only designed for specified currents and voltage handling, going over spec means faster deterioration over equal time. Though we stick within close range of absolute max spec, unless we have a crazy streak and can afford a potential loss. We counter deterioration to some degree by increasing cooling capacity. I have no idea if this is the question you were asking as it was not quite specific. I get the impression that execute disable bit and c1e don't have to be turned off but may impact stability? So experiment and turn off if needed? C1e is a power saving feature that keeps a constant sense of the cpu, if it's idling the c1e function reduces the voltage to save power, and re-applies your bios set overvolt when it senses the cpu is under load. Disble bit I don't know anything about, but I have heard it's ok to keep off as it's not used. For outright overclocks that are on the brink of the boards capabilities people turn off c1e as a precaution, just in case it does interfere with stability. Otherwise one is always free to experiment with what works for you. If you can get away with using it and your rig is stable, then why not save the power? I'm mainly interested in voltages as I was having a game patcher crash running at 300fsb on the 800 divider and figure either my NB or CPU needs a bit more in the way of voltage. I didn't bother Orthosing it as it's just till I get my PSU back (blew after 8 hours running stock) but I'd imagine it's not stable there for some reason or another... Ran all my benchies fine so I figured it was stable enough to game with *shrug* Thanks for any help There can be more to it than voltage alone, you need to be sure the ram timings are within the ram capabilities at the mhz that divider and fsb gives you first. It's always wise to run varieties of tests to see where the error lies, orthos blend picks up cpu and ram issues, whole 3dmark2001 is good for giving the cpu/gpu and ram a workout together, then use 3d mark05 too. If it's Orthos stable but fails at 3dmark2001, check ram timings first, gpu second, and lastly give the vcore a little boost. On conroe boards this sometimes happens that you can run Orthos at far lower volatges than you can some 3d and games (I speak from experience), use all of them to test. Sometimes a 0.03 vcore increase cures the issue. Dunno if I covered it all, only got limited time, plenty of knowledgeable people here that can chime in if you have more questions... regards Raja Share this post Link to post Share on other sites More sharing options...
raju Posted December 28, 2006 Posted December 28, 2006 Raju, you're a machine. I wish I could hang with you but I can't do half of what you do. Though I'm pulling out primes at 3.8GHz. Ugh, I just can't believe how hot this chip is. I know if I could keep it cool, it'd do so much better. Dunno 'bout me being anything special, I wing it just like you guys do. When I hit those 3.825 prime stable clocks on water I was using a strom and pa120.3, with 2 X ek 1900 water blocks with 90degree elbow joints and a D5 pump, in short a restrictive loop. Temps at around 1.58 vcore were into the mid 70's on orthos load. Secondly it was on my es, which is an absolute king cpu. I have a little theory atm which I have no real evidence or stats to back up, but the b steppings seem to mimmick the es's more in terms of heat. The es could boot everytime on my machine at 455fsb, the a stepping I have can boot out of bios only at that speed, it'll get into xp run orthos blend and anything else I throw at it all day, but it won't reboot from reset or cold power at 455. The only way I can get the 455 back is to reset cmos and boot out of bios again at 455. I know these conroe's have a reference drive voltage, I'm just left wondering if the es's and b steppings actually have a higher reference hence the higher boot up capacity. Stability wise the oc's on my es and retail a are the same so far but board restriction prevents me from full clarifiaction and comparison. Now what could be the difference is the question? 1) are intel putting the temp diodes in different places on a's and b's? - I doubt it 2) Ar the B's different internally - I doubt that too,,, 3) Do the b's have a different default drive voltage - possible, but needs someone with technical insight to clarify... I have not got a retail b so I can't confirm. I may be wrong both techniaclly and in practice, but I know there's some form of difference in boot ability and behaviour, but I do have temp sensors on my evap that confirm the a running a lot cooler volt for volt. Which means that es's being hotter is no myth, reports pit the same conclusion with the b vs a retail debate, which raises the question that there must be more voltage by default in the es's and b's (I don't mean vcore).. regards Raja Share this post Link to post Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now