• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

More Bad News

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
gamefoo21 .............. If you notice , I said that things at OCZ have changed alot , in fact they now have RAM which is often in the top three .Their customer support actually responds to email and answers the phone :eek: , this is a vast improvement from earlier days in fact having RAM which runs at its rated speed consistently is a great improvement . So please don't mistake the 'old' OCZ with the new one .

I must however admit that the whole fake silver thermal paste issue 1-2 months ago does look somewhat fishy and reminds a bit of old times . For now I'll give them the benefit of the doubt .

http://overclockers.com/articles938/
http://www.ocztechnology.com/displaypage.php?name=recall
http://overclockers.com/articles938/index02.asp
 
While the mainboard issues are bad news, I think the situation should be considered in more detail than "prescott sucks" or "IC7 rules".

It cannot be denied that Prescott chips draw loads of power and produce massive heat. This has two consequences: CPU cooling is a 'challenge' and current day mainboards are running on their last legs when powering them.
This thread is about the mainboard problem. So to me it seems that the logical thing is to either...

1) wait for the new boards due soon ( http://www.xbitlabs.com/news/mainboards/display/20040221080602.html ), which will eliminate the problem altogether*, or

2) make the best of it now by optimizing mainboard cooling and monitoring the BIOS detunes that manufacturers like MSI have put out.

What seems illogical to me is to deny that there is a problem and claim that current mainboards will handle things 'just fine', or claiming the opposite that all is lost when you plug in a Prescott chip.

* Regrettably, the xbit link also states that the new chipsets i915 and 925 put out much more heat by themselves than earlier designs, requiring a heatsink on the southbridge too. A disappointing turn of events IMO, but separate from what is happening to current boards (think mosfets etc).
 
I agree that the Prescott is for the more advanced overclocker and that people who are unwilling to address the extra power consumpution/extra heat should think twice. Certainly choice of motherboard is important. At this point I've only used the Abit IC7 and Abit AI7. I can now recommend that the AI7 is probably not the best choice since the PWM gets too hot. The IC7 series seems to be working out better and the Asus P4C800 appears to be doing ok too. Looks like we should totally avoid the MSI boards for the Prescott. Warning people of the potential hazards of overclocking the Prescott is one thing... running around telling everyone to stay away from them is a different story.
 
I agree batboy, its muich to early to tell anyway. Intel has too many resources and has been at this too long. But I guess the amd fanboys have nothing better to do. Of course their only intent is to help us poor little Intel foks out. The Prescott has some issues, but also has shown a lot of potential. We all knew that it was going to have problems with some mobos. It will perform a lot better when we have mobos that are designed for it and I suspect that in the next few months we're going to see much more of the potential. In the mean time if you're going to overclock one, then do what batboy says and the first step is to pick the right mobo. As for this thread, thanks for pointing out that I should stay away from the MSI boards when I buy a prescott, but it appears that the max3 that I have and the P4C800 that I have will work just fine.
 
Last edited:
at this time I would have to agree with everything cowboy x said....

after carfully rereading his (cowboy's) statments, I feal that even batboy may have jumped the gun when stating ...
"" Are you on AMD's payroll to discourage people from buying them? ""

man, cowboy x, I know how you feal...
I'm quite sure that you know all the grief and name calling I've had when I've repeatedly pointed out nVidia's poor IQ and cheating ways. (and to think, I'm a big nVidia fan)

anyway, be forwarned guys, OCing with current prescotts could be leathal to your mobo down the road....even with cooling on your mosfits.

just a reminder for batboy on the status of one IC7 type mobo:

sbud said:
Alright, last night, I pushed my 2.8 Prescott to 3.850 with 1.52v and was pushing upwards of around 55C under load. Damn, that's fricken hot !!! on water too. I've never seen over 42C load on any Northwood that I've had. I tried to touch the ramsinks on my mosfets and couldn't touch em for more than two seconds, damn hot. About ten minutes of Prime, my rig shuts down down, in the end, my Prescott killed my IC7-G. I've just returned the 2.8E, and will just stick with my good 'ol 2.4c. I just can't see running that Prescott overclocked for a long period of time.

I'd advise anyone to pay close attention to motherboard cooling when overclocking a Prescott.

mica
 
Yeah, I saw that post about the IC7 dying. I'm not too worried about one failure. If we start seeing a pattern, then I'll be concerned.

My point is this, Mr. X has been posting in several sections telling everyone to avoid the Prescott like the plague. To my knowledge, he's never even owned a Prescott. People jumping to conclusions is what's what really bugs me. Chicken Little ran around telling everyone the sky was falling and some people actually believed that silly little bird. I think we need much more data before we know the whole story about the Prescotts.

I have one and I'm still testing it. I posted my results in a thread and just left it up to the people reading my posts to decide whether they wanted one or not. Mica, you were following my posts pretty closely, and I believe because of that you decided to hold off buying one. That's fine, you weighed the facts and decided what was the right path for you. Others read my posts and decided to try a Prescott for themselves. They sort of knew what to expect.

I say check out the facts and read some reviews from people that have actually been there and done that before deciding what to do. Like I posted earlier, the Prescott is not for everyone. In fact, my personal opinion is to wait for the next stepping to be released later this spring. I'll try another Prescott then. On the other hand, if you want to buy one now, then go ahead and good luck. I'm not going to tell anyone to run away from the Prescott though. I've seen several people get 4 gig overclocks from them.
 
A wise man learns from his mistakes , but a wiser man heeds advice and learns from the mistakes of others . I don't have to get a Prescott to see what people are saying or the issues that they are facing or having to overcome .

I find people sometimes misunderstand what I am saying . Batboy , I fully respect your oc'ing abilities and I have no problem with you pioneering with Prescott overclocking . It is even better that you and others come in here and let us know what is going on and how to do it ourselves . But agin I say that Prescott as it currently exists is a hit and miss choice at best.

Again I will ask , what is the benefit besides the challenge alone , to overclocking a Prescott as opposed to a Northwood C ? Also what is the benefit to the average Joe who buys one to run at stock speeds , compared to a Northwood at stock ?

To those who want a challenge right now .............. by all means go ahead . I am yet however to see any good reason to get one .
 
You have a point. What are the benefits over the Northwoods? Not much until you can get the Prescott above 3.8 gig, then the Prescott begins to shine. But, of course you have to deal with power consumption and heat issues at those speeds too. Even in the 3.4 to 3.8 gig range, the Prescotts do more SETI work than the Northwood and some of the memory benchmarks are better. Super Pi 1M calculates faster with the Prescott too. Anyway, we're cool Cowboy, I just like a good debate.
 
Who said that it begins to shine anymore than a Northwood at the same 3.8+ ghz ? I mean that is good for the SETI guys ,but so far the other improvements are synthetics .

When the Prescott gets a new stepping and some better boards ( yes they are to blame as well ) I suspect we will then see the benefits of that extra long pipeline and 90nm process with rediculous 4Ghz+ stock speeds ( I hope ) that is where the Prescott is supposed to shine , by running at speeds out of the reach of Northwoods and avoiding any clock for clock comparisons or problems with the older chips . It is also supposed to show great benefit to those programs needing more cache and eventually SS3 apps ( don't hold your breath ) . The problem is that that is not where Prescott is right now . I was very honestly disappointed in Prescott's performance and lack of current benefits . But as I said before I expect improvements to occur and people like you Batboy to be there to tell us firsthand .
 
I haven't said to stay away from Prescott. I'm only saying current Prescotts aren't really shining yet, and Prescott can kill the motherboard if the user isn't careful when overclocking.

Synthetic performance shouldn't be interpreted as reflecting real world performance. You need benchmarks from real applications to have any valid point of comparison. Sure more SETI work is done on Prescott clock for clock, but most people don't buy a computer just for running SETI, although some might. My point is, there are better choices than current Prescott, unless you just want to toy with one.
 
Back