groovetek

Yeah - I'm surprised there has been no reply yet - let's hope they have indeed read this thread and are currently doing some testing... For benchmarking/overclocking, this board is second-to-none, because of the features if has, but the tendency to kill hardware is too high... from CPUs to memory to graphic cards, and killing itself too... I think the new version of DFI's NF4 series should combat all these plaguing issues that have cost us enthusiasts thousands of dollars... RIP my venice (damn 704 bios bug)...

I totally respect DFI and Oskar Wu, even AG, so that's not in question... but nonetheless I am disappointed that, as an OVERCLOCKER, without having personally made any mistake, my hardware died due to a BIOS fault, yet at that end, it is not an official BIOS and thus I have accepted the consequences, of course, as you can see.. I'm not disputing anything, but I'm just pointing out that there could be some improvements to the BIOS testing/board debugging...

Well, here we have 3 sides... 1 is arguing that more official bios's should be released more often so that it's not ALWAYS at the user's own discretion to give it a shot to see if it fixes their troubles with hardwar compatibility. Another side is arguing that BIOS's being released shouldn't have fatal bugs like this, be it BETA or NOT (I'm in this category). And the third side is arguing that you shouldn't even overclock in the first place, and that if anything happens (EVEN not as a direct result of overclocking), you are the one to blame. Well shoot me if I'm wrong, but what's the point of manufacturers even having PC4000/4800 RAM (hell even OCZ have some DDR600+ RAM made specifically for DFI NF4 series), all these weird and wonderful heatsinks and other cooling gear? They literally encourage users to overclock - On the boxes of many motherboards they have big advertisements about how good at overclocking they are, and what safety features they have. Overclocking has come to a point in the last 2-3 years where if you were even half-interested, you would probably be doing some overclocking. HECK, many manufacturers even overclock the CPU for you without you even knowing (just to name a few names - MSI, Gigabyte). Then what, does AMD say, "hey no, the board had set up the CPU speed by 100mhz and increased voltage by 0.05v, sorry, bad luck, you overclocked it"? No they don't, let me explain why. FIRST OFF, CPUs and other components don't just die unless you're running a voltage higher than the tolerance limit of its manufacturing process. Not to mention, with many lines of CPUs in the past, certain higher-end CPUs were simply overclocked/overvolted versions of the slower ones. In the case of this BIOS - the motherboard, NOT the user, pumped in stupidly high volts into the CPU (in my case, I calculated it to be 2.02volts into my venice, killing it instantly). If you look on xtremesystems forums - the only CPUs/RAM dying are the ones being pumped at very high voltages, used with extreme cooling, and quite literally taken beyond reasonable speeds. I'm not saying DFI is WRONG in this case - hell - I admit that I AM RESPONSIBLE for the death of my CPU because I had made the choice to even overclock it in the first place. HOWEVER, I feel that as an overclocker - I DID NOT do anything that was ill-advised. I simply downloaded the BETA BIOS to see if it made my G.Skill BH5 work properly - and voila it fixed my memory compatibility problems, and I wanted to get a voltage abit above 1.55 (which is the max you can set without ABOVE VID % control), and so I set 1.20 * 133%, which, mind you, has been fine in every BIOS I've tried so far, giving me a nice 1.56 vcore.) And then, whaddya know, some new BIOS bug that fries ur CPU. As any right-minded person can see, I'm not condemning DFI at all, I'm simply trying to emphasize that such fatal bugs really should not exist in the first place, be it a beta or alpha or whatever. DFI is not obligated to do anything about my case, which is fine, but if these BIOS's keep introducing such costly and fatal problems in BETA BIOS's, then one must question who and how these BIOS's are being written/tested. AG, you used an example that if you drive a car at 150mph and if the wheel falls off, bla bla bla, etc... You are comparing apples to oranges. It simply does not apply. If you crashed the car, at ANY speed, they will not replace anything - because you, as the user are responsible for accident. If you wanted to compare cars to CPUs, then using your perspective, it means that if I didn't mount the cooler properly (user mistake), but I didn't overclock the chip at all, and it died - they will replace it for me. EVIDENTLY this isn't the case. Don't take this as an offence - I'm not trying to prove you wrong or anything - and in fact the majority of the points you have listed are VALID. However, there is a difference between VALID and REASONABLE. You, of all people should understand that DFI makes boards for overclockers, probably over 90% of boards sold are to people who want to use the overclocking features implemented on the board (otherwise why put it there in the first place?), Now, if the board starts killing hardware when the user hasn't made any fatal error (ie, no cooling on ram that's overvolted to 3.4 volts, or, setting 2 volts for the CPU), then you cannot deny the fact that there will be alot of unhappy people, many of which are experts in overclocking, and had they chosen a different board, the same BIOS/manufacturer fault would not have existed. Then again, OF COURSE, different manufacturers would have their different problems as well. However, sadly, it's no secret that currently, DFI boards are killing the most hardware compared to other boards, for example the 4V jumper issue (which again, was a BIOS problem, as a fix was released), and now this. I really like the vast amount of options available in DFI boards, and who knows, I may buy one again in the future. But it's evident that between the NF2/NF3 754 boards to the NF4 series, alot more danger has arised.

OK actually now that we're talking about official BIOS's, here is my next question - what if an official BIOS had the same bug as this 704 BIOS? It seems all too possible that such a problem could be overseen during testing. And so lets say this 704 BIOS was released as an OFFICIAL, and people were having their CPUs fried, is DFI obligated to replace their CPUs, if AMD's RMA system refuses to replace the CPU, since it had died because of overvoltage (which in this case is what the BIOS can do)? I understand AMD do not cover CPUs that have been subjected to increased voltages - if there is any way for them to diagnose and come to that conclusion.

We understand this completely - nobody is disputing the state of the BIOS's... but as far as I can say... the 310 official BIOS is quite poor especially when it comes to memory compatibility, and this forces people to update to a BETA BIOS as required. It has come to my attention that, like me, alot of people are (or in my case, were, until my CPU got destroyed by the VID bug), having alot of success with the 704 BIOS. Not only has this BIOS fixed so many memory issues for both UTT/TCCD-based modules, it has support for X2 chips, and consequently I feel it is a good base BIOS for DFI to work on an official release. I firmly beleive less than half of overclockers who use the DFI board are using the last official BIOS... Not to mention that it seems other manufacturers don't have problems with a wide range of memory without the need to tweak up timings/play with BIOS's. Don't get me wrong - I understand the DFI is a tweakers' board, which is why I bought it, and have enjoyed tweaking it, until my CPU got fried because of the BIOS.

OK now that i know the problem, I cancelled my order for the Abit AN8. However, what I want to know is, will there be a fixed version of this BIOS released? Because it allowed my G.Skill GH BH5 UTT kit to clock very nicely asynchronously...

Gosh that's EXACTLY what I did , except it was 1.525 before I changed it... sigh... my venice instantly died... Hard to get RMA in Australia... AMD RMA procedure is very tight... and I have to go through my retailer first, who is interstate and I have to provide explanations...

The frustration you guys are feeling is understandable, but I don't think it is justifiable to act aggressively towards DFI and/or Oskar Wu. I have lost precious hardware too because of this, but hey - that's life, or more accurately, hey- that's computers. But most importantly, I think it's unfair to DFI/Oskar Wu to associate deaths of other hardware components to this BIOS, at least, not YET. The problem has been explained almost thoroughly enough - the BIOS has introduced VID control errors with the 123,126,133, & 136% settings. If anything, maybe the situations under which using those settings would cause voltage spikes could be explained, otherwise it's detailed enough and we just have to live with it.

The VID above % was never supposed to be accurate in terms of how it controls the voltages, so trust the BIOS reading. CPU-Z and other software is known to report slightly lower values. BEFORE YOU CONTINUE, please read this though: http://www.dfi-street.com/forum/showthread...00&page=1&pp=15

this may or may not apply to you - but is important that you take note of it, as it was why my 3ghz on air venice 3000+ was killed: http://www.dfi-street.com/forum/showthread...00&page=1&pp=15

i just emailed them... lets see how long it takes for them to respond...

Oskar Wu - how did you find out? Was it from my email to DFITS? Well at least now the answer to death of my Venice 3000+ has been explained. I'm very depressed about it... will never find a chip like that again... so then my motherboard is probably still alive? If so, I maybe be willing to give it a second chance... unfortunately all BIOS's before it gave me async problems when running G.Skill BH5 UTT GH kit... will an updated BIOS that removes the CPU-killing voltages but with the same memory settings be released? If so, I will give the DFI a second chance. Any other problems with the BIOS that could kill my new venice 3200+'s coming in? I have thoroughly resetted the CMOS - will putting in my new chips get them killed as well?

I don't know where you got this information from, but I'm almost positive that BH5 will almost 100% clock higher than CH5 at the same given voltage... Which explains why OCZ's curren PC3200 Gold EL BH5 UTT, and G.Skill 1gb-GH kit BH5 UTT both are rated at 2.7-2.9v 2-2-2-5 @ PC3200, and have proven themselves to be high performers at lower voltages than the VX series, which are CH-UTT based.

i think the point he wanted to make was that, there should be a benchmark utility that can stress out the system as much as an OS+StressTest together can do. Unfortunately, MemTest86+ is still pretty much the only one, and hardly stresses out the memory, let alone the system enough to gauge a point of perfect stability. I DO find however, that it is generally quite good for finding a point where at least the system will be DESKTOP STABLE, that is, can boot up, and at least do basic tasks, and usually the max memory overclock is no more than 5-10mhz lower than the speed there MemTest first begins to report errors.

whoah lol... then u've missed out on a year's worth of technology.