DerekT Posted January 5, 2007 Share Posted January 5, 2007 I personally go by the general rule of thumb 0.2v added to the VID. If your VID is 1.325 then I would advise no higher than 1.525. However, I am a bit conservative and 1.575 with low thermals is very likely fine. I would use Core Temp though to test your temps. http://www.thecoolest.zerobrains.com/CoreTemp/CoreTemp.zip Use Orthos and run it for a minimum of 30minutes, then take a copy of the Core temperatures. http://sp2004.fre3.com/beta/orthos_exe_20060420.cab What are your Orthos loaded temperatures? Link to comment Share on other sites More sharing options...
Yoxxy Posted January 5, 2007 Share Posted January 5, 2007 VID is not the best indicator for Intel cpu's. Just go by the default voltage of 1.35. Also try to make sure that your core temperatures are not over .26x + 43.2 where x is watts of usage. That is where your computer will in the end have serious problems. I have an X6800 retail with a VID of 1.0000 (lowest I have seen on a C2D). Now it is near impossible to say I should be running 24/7 @ only 1.2 volts or even 1.275 so you need to just stick to the default voltage of 1.35. I have also run an X6800 ES that has a VID of 1.1250 at 1.45v for the last 5 months 24/7 with no problems. I would howerver say that you should not have Core Temps over 70 degrees C prolonged, or your system in the long run will be in danger. Link to comment Share on other sites More sharing options...
DerekT Posted January 5, 2007 Share Posted January 5, 2007 I personally find VID to be the best general indicator (for one to start their overclock at) although one needs to find their own plateau regardless of using the VID or 1.35v. I have worked with a Core2 E6600 L628A with a vid of 1.325 and another Core2 E6600 L630B with a VID of 1.235. The same +2 rule seems to apply for my general purposes and as I said above, one needs to find their own rule of thumb with regards to plateau. Anyone who follows another's system plateaus is not an Enthusiast and needs to learn their own system requirements. My general rule of thumb is not anothers although it can not hurt as my general rule of thumb is a very conservative one. The thermal output of the 1.235 VID is quite a bit higher than the thermal output of the 1.325 when both are set to 1.55V. This extends the view that the higher end of thermal displacement will show greater. Keep in mind that I said "Indicator". There is NO hard and fast rule for either. Of course, as an Enthusiast, one should always test and validate their findings on their systems. I would howerver say that you should not have Core Temps over 70 degrees C prolonged, or your system in the long run will be in danger. I agree with this statement. I would personally not go over 60C long term. Link to comment Share on other sites More sharing options...
Yoxxy Posted January 5, 2007 Share Posted January 5, 2007 Correlation in this case unforunately does not equal causation. Link to comment Share on other sites More sharing options...
DerekT Posted January 5, 2007 Share Posted January 5, 2007 Proof will stand by testbed until there are greater proofs. The statement "Correlation does not suggest causation" is not fundamentally so if one can demonstrate a consistent correlated relationship and then this often will suggest a greater possibility or probability of some causal relationship. For you to use that term in this situation shows that you are not grasping the concept. If you're suggesting that the VID is not provable rather than the static 1.325v, then you are disputing the accuracy of the record of an event rather than the relationship between these events. The accuracy is not in dispute and thus your use of this term is in error although such a term is useful on grounds that prove a disputed relationship. You however, dispute without proof and offer this cliched statement as the "Proof". :D: OK, if you say so. Link to comment Share on other sites More sharing options...
deuspit Posted January 10, 2007 Author Share Posted January 10, 2007 and 48.6ns... is it a good value... Can i have more? Why labs don't arrive ta this value with a bigger overclocking?? Thanks for all men... Link to comment Share on other sites More sharing options...
DerekT Posted January 11, 2007 Share Posted January 11, 2007 and 48.6ns... is it a good value... Can i have more? Why labs don't arrive ta this value with a bigger overclocking?? Thanks for all men... I don't understand what you mean. Can you explain this? Link to comment Share on other sites More sharing options...
deuspit Posted January 11, 2007 Author Share Posted January 11, 2007 When i tested my memory latence, i have 48.6ns... When i look at the best overclocking in the corsair lab, they have 54ns of latence... So i don't know if my value is real or wrong... because i think that corsair are certainly (:roll: ) better than me to overclock (really better :D:, i'am sur)!!! so i'm sur that 54ns is the best real value that we can have... Can you explain me that? Really thanks... Link to comment Share on other sites More sharing options...
DerekT Posted January 11, 2007 Share Posted January 11, 2007 What latencies do you speak of? Can you post a screenshot of the program and the latencies you speak of so that I can explain i? As well, if you have a latency of 48 and the Corsair Labs have a latency of 54 that means that your have the better latency. Also keep in mind that many variables can effect this and your system might well be different in version and revisions. Even two identical motherboards can show a difference due to fabrication issues that might show a difference in months of fabrication. Link to comment Share on other sites More sharing options...
deuspit Posted January 11, 2007 Author Share Posted January 11, 2007 all are here: ..i hope.... ask if you need other things http://www.hiboox.com/image.php?img=ef142079.jpg#gd Link to comment Share on other sites More sharing options...
DerekT Posted January 11, 2007 Share Posted January 11, 2007 Your need to update your Everest software. Either that or something is wrong. There is no possible way that a Core2 Northbridge Memory Controller can beat an AMD OnChip Memory controller with latencies. Not at your DRAM and CPU/DRAM speeds. http://i41.photobucket.com/albums/e269/Ropey123/ev.jpg Link to comment Share on other sites More sharing options...
deuspit Posted January 11, 2007 Author Share Posted January 11, 2007 Last version of Everest Ultimated edition 3.50.761 http://www.hiboox.com/image.php?img=25545c7.jpg and... now? Could you help me? Is it an answer anywhere Link to comment Share on other sites More sharing options...
Yoxxy Posted January 11, 2007 Share Posted January 11, 2007 That seems good for those timings. Best I can muster is mid 40's with 3-3-3-5-1T @ 880 mhz. That is nowhere near stable for 24/7 though, so I would be happy with what you have. Link to comment Share on other sites More sharing options...
deuspit Posted January 11, 2007 Author Share Posted January 11, 2007 Strange results for 4-4-4-12 timings ... May we put less timing to this ram...? like 3-3-3...etc? Isn't it so dangerous? Link to comment Share on other sites More sharing options...
Yoxxy Posted January 11, 2007 Share Posted January 11, 2007 Not dangerous. There is just a very low probability that it will work stable. You can try to boot whatever you can and as long as you don't raise the voltage over warrantied voltage you will be fine. Link to comment Share on other sites More sharing options...
deuspit Posted January 12, 2007 Author Share Posted January 12, 2007 I am at 2.25V... it isn't the warrantied voltage i thinks... but it is the best for my O/Ced PC!! if not it doesn't work... :p Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.