No pinhead, I just quoted a silly word and didn't think about changing it. lol. NOT a bad word per se but NOT excepted in some circles and shouldn't be used in a forum such as this.. hope that makes some sense. lol. ....gm
Anybody experiencing instability with NeroVision 5 of the Nero Ultra 8 Suite (v8.3.2.1) while on an overclocked or close to stock system? I'm not sure, it may just be the app cause everything else runs fine. I will compare in adobe Premiere 3.0 later. Just wondering if some-one here knows Nero a bit better as far as what kind of sub system stress it taxes on a OC'd system if any. NeroVision 5 of the Ultra 8 suite uses these 3D animated menues and it does tend to be more a resorce hog already. Early v8.1.1.0 was worse.
abuzar1, Kids! LOL!! You slow it down and see if the temps go down. If they don't then you have to go over your case cooling and see where it's lacking. I mean why burn it up? That's just crazy! We know what you've accomplished already, so what's the point of risking your machine? Sometimes you have to be a bit of an engineer to get things right! I ran a 3.0/800 Prescot at 3.84GHz in a very poorly ventilated case. I had to try different fans and different directions with the airflow of those fans to get it to run cool enough. Where I was living had no AC, and with summer temps of 115F, it was a challenge. Even thought the P4 Prescot was a known heater, I managed to keep it under 60C while encoding. It just takes time, and that dirty word, Patience to figure it out! LOL!! Best Regards, Russ
OK! I've been doing some adjusting and testing. Some processors are great overclockers, some are good and some just don't overclock very well. My E6750 is a good overclocker, but not great! 3.55GHz is it! I sort of figured considering I have to raise my CPU Voltage more than I see other people have to do to get the same results. First off, It won't even post at 1.40v. If I set it on automatic at stock 2.66GHz, the auto sets it to 1.44v. Since it should run at a much lower voltage stock and the Prime95 CPU test, which stresses almost no Ram fails at lower than 1.49v, I have to believe it's the processor. Memtest passes with no errors, at 1066GHz, so to me that clinches it. At 3.55GHz it runs good, so I'm not sure yet whether I want to RMA it or not, but I think I might. The benchmarks below are good, but it bothers me that it won't post at 1.40v and that the Auto sets the CPU voltage to 1.44v. It takes 1.49v to be stable at 3.55GHz, and I've seen many instances of the E6750 getting to 3.6GHz with the default voltage to ignore the problem. I can get another 2000-3000MB/s at least, of memory bandwidth by messaging the memory subtimings! Still that high a voltage just to post bothers me. Here's the latest benchmarks! Not too shabby! I think I will take a few days and think it over and then decide whether to RMA it back to Intel. Best Regards, Russ
If it won't post at normal voltage then it's probably buggered. What if you had a board that can't raise the voltage?
if thats what it autos to its the chip telling the bios thats what it wants, it would do this with a board that gives the user no vcore option too.
Really? Fair enough. I assumed that some boards wouldn't have the hardware required to adjust the voltage in software, whether or not that adjustment was visible to the user or not.
sammorris, It's been that way since day one. I just didn't snap to it. I just started out on automatic and didn't pay any real attention to the voltage until I started overclocking it. I was going to see how high it would go on Auto. Next time I'll check by turning the auto off! The numbers are good, and I can get the memory bandwidth up to about the 8200-8300MB/s range, but that takes a lot of time to do, as the changes are small and many with a lot of testing in between. I don't think I want to have to do it twice with all the time it takes. BTW when you were surprised with my 3DMark 06 score a while back with my 7600GT those memory sub-timings were what did it. I didn't change anything except fatten the memory bandwidth about 5000MB/s. Maybe I hurt it taking it up to 4.0 briefly! Taking it to 1.56v, just to feed my Ego was stupid. My box that the CPU came in says right on it, "Max Voltage 1.350v"! I was just so damn obsessed with seeing 4.0, and Nuck had done it at that voltage and his CPU is still all right, I decided what the hell! I also had the Intel Tech tell me 1.5v was safe and I figured that a short foray to 1.56v wouldn't hurt. Mine was also from the early batch of G0 stepping E6750s, so maybe there were changes made after I bought mine. I know that they stopped shipping them for a while just after I bought mine! I had the memory set to 2.3v after checking with DocTY first, and since the memory used to run at 932MHz with the timings at 5-4-4-12 with the E4300 at 3352MHz and then it wouldn't run over 888 with this CPU, I should have been suspicious then. The only thing difference in booting up with the Dominator memory is It will run at 3.6MHz, and run benches. With the G.Skill 800MHz memory, it wouldn't clock over 888MHz also at 3552GHz (notice a pattern here)! I mean what's the likelyhood of that with two different bus speeds because of the different mulltiplier. I was blaming the memory when in reality it's very likely the CPU! I only just realized it now that the CPU was at the exact same speed, because I backed the E4300 down to 3.3GHz. I was getting the figures from the calculator based on the known memory speed and came up with 3552 that way for both the 800 and the 1066 memory! I guess you're never too old to learn! LOL!! Best Regards, Russ
See, this why I never did an extreme overclocking test. It might not make a difference, but by the time you find out it does, it's too late... I find it funny how I'm using a lower voltage than your CPU requires at stock to make 3.15Ghz!
Russ, sucks about your CPU, I would RMA it. Also I use a Thermaltake Armor case(the one without the 250mm sidefan).
@theonejrs, It may very also in part be the mob and chipset. 2.3v is just as stressfull on an MCH to do daily as 1.6v is on a CPU and mobo. If you didnt cool the mosfets, I'd suspect that first, or as well. Do you have another mobo to test the E6750 in? I find many different mobos auto select at various readings. My 2 E6750's have VID of 1.255 & 1.23000 respectively. My Q6600 has VID of 1.225, oddly enough, it's lower. Unless they're supposed to be. bothe E6750's auto to 1.35 but droop to 3.2v and lower on the GA-P35C-DS3R. On my old IP-35 Pro, it held closer to 1.32v & since C+P state worked it lowerred to 1.10?v I guess it wouldn't hurt to just RMA it if you have nothing else to test it on. it sort of sounds like my experiences with 2 older Intels. One P4 3.4E was at 4.06GHz for so long it acted weird when I needed to boot as stock. It freaked out untill I OC'd it a tad. THe same for a P4 660 where the lowest vcore in BIOS was no longer available. Not sure if it was an ASUS or Abit thing or the CPU. (it happened in two, an Abit Fatality and ASUS P5AD2-E. I also saw this with my recent, but still in daily service, Pentium D-950 (C1) to 4.250GHz. It takes a bit of tweaking to get it to POST as stock when I had to last for a BIOS flash. I never heard of this either, so I'm sure it's just a fluke. But to me, it's almost like they got used to running at nominal spds for so long, they thing it is stock. Or at least thats how stable they are. For my Current E6750, I install and service as 500x7 as stock cause thats what the mobo and CPU seam to love and run rock solid. All features of ACPI mode work if RAM is left to 1:1 for DDR1000. If I dump the 5:6 for DDR1200, mobo now needs more volts all around. Underclocking, or more so undervolting is cool too, I did that at times for the fun of it. It's easy for lots since better CPU's are set Auto for more then they may need. I can undervolt my E6750 @ stock clok to 1.254v or use rated 1.346v to 3.2GHz on 500x7. Not bad since tomorrows CPU's are aiming at 2000FSB and we're already on 1600FSB normal.