Moore is a Moore-on and Nforce 2 problems

This is a good little writeup. It is true that boeing etc have been trying to build super sonic and/or huge assed modern jets, but why pay 10x the price to get somewhere a few hours faster?

Our business is growing pretty well (considering the economy) and just the other day we were thinking about getting a bunch of new machines. A print server box, a few more work stations etc and my boss came up with 5 DELL PIII 866 mhz machines (used) *WITH* win2k licences & software and monitors for around $200 a pop. These are like DC-10s or 737s -- not big, not fast, but they'll get you where you are going. They'll run open office or with a RAID card and memory turn into nice linux print/data servers. I bet this is going on all over the place. A few years ago we'd have shelled out $1000 a box.. I guess we aren't doing the economy any favours but ... i'm not losing any sleep over it :) The licences are key -- it's getting very hard to order DELL or whatever with 2k installed and if you walk into staples they want $299 for 2k (we are not large enough for site licences or other fancy M$ deals). So it's almost worth it to buy crap old machines just for the damn software.

So although the 'fighter jets' of the world (gamer machines, uber high end servers etc) will continue to be produced -- and might follow moores law, the average business/production commuter can limp along with 1-2 ghz machine no probs.

keep up the good work
-z

---

wow -- this has to be the lamest article i've seen from the Inquirer. predicting the end of moore's law is about as interesting as predicting that airplanes can't carry 13000 people...

Email address supplied

---

The reason why the comparison doesn't work is that 13,000 people don't all want to travel from Heathrow to New York at the same time. There are limitless uses for nigh-infinite amounts of processing power, however.

Also, the price/performance ratio of planes _has_ still been going up. What's changed is that flights have been getting progressively cheaper rather than larger/faster.

10 years ago I couldn't travel London/Edinburgh for £30. I certainly couldn't cross the atlantic for the prices they now quote.

I'm sure that eventually processors will eb as powerful as they can possibly be. But they can still carry on getting cheaper.

Andy

---

Economically unfeasible? Please, commercial aircrafts work the same way the shuttle program works. Keep the design and force people to use and re-use the plane. If there is a problem, you ask the same manufacturer to build the part and sell it at a very high markup. That is capitalism.

Look at HP. They are only making money on the consumables of the printers, not the printer itself. Instead of scrapping the whole plane and build one that costs less, is faster, higher payload (or people), pollutes less and can be easily fixed, innovation has become to who can put the internet in first class! But that is like asking for a printer that doesn't run out of ink (cartridges).

Parallel lines can be drawn with the design of the CPU too. Innovation is using smaller pipelines so you can put more next to each other and increase the crunching power. Personally i don't believe that that is the only way that we can calculate data.

Well, there is no necessity to innovate and come up with new things. As most know, necessity is the mother of all inventions. Then again, we, the paying public, are not even asking for a plane that can carry 1000 people from England to Australia in 4 hours without polluting the air we have to breath.

Regards Sven

---

It is indeed NOT a law. Moore was/is a moore-on.

Peter

---

Not a flame a comment.

I think the combat pilot didn't quite understand why Moores law is holding true.

The reason it probably will hold true is because "it's the technology it creates thats creates tomorrows technology" so anything that gets in it's way gets is fixable, it will only be our ability to apply it fast enough that will slow / stop it or for comercial or security reasnos, but lets not go there :O)

P.S. If the creation of planes had helped the creation of planes then they probably would be carrying 13000 people.

Email address supplied

---

What surprise such a comment would come from someone of Tower Semi; they still have serious trouble making their .5 micron process work properly :/

Name and email address supplied

Nvidia can at least be happy that it has plenty of followers even if there has been a problem with the Nforce 2. Here's a tiny selection of what arrived in our inboxes...

Thanks for your timely and in-depth response concerning the nForce2 BIOS issues lately. I see that most of the problems stemmed from users o/c their nForce2 boards by large amounts, causing the board to "hang" and be damn near impossible to reset with a 200Mhz FSB Duron to plug into board to "reset" it afterwards... like you mention in your article, should nVidia have stressed the use of that jumper more than they did, or were mobo makers simply lazy, cheap, or negligent by not including the jumper that was included in the reference design? Probably a combo of both, but might never really know the answer....

I have only mildly overclocked my Asus A7N8X deluxe board, thus never hit "the wall" like other users have, and thus have never had to worry about "resetting" my board... but good to know for the future...

The important thing is that nVidia responded immediately to the problem, and very soon this whole BIOS thing will become a "non-issue" as board makers implement the jumper or configure the BIOS so it doesn't "hang" and is able to reset itself properly in the future....

So , it looks like there was indeed a BIOS/jumper problem , but mainly affecting extreme overclockers ( but not always ). It did not affect me personally, but the fact you reported the bug in the fashion you did, caused a lot of attention and forced nVidia to jump on the problem immediately and resolve it... had you not posted your article, they might have taken longer to acknowledge and resolve this issue.

So I take back my initial disdain and criticism of your first article, as the problem never affected my board nor the h/w reviewers online, causing me to initially get my back up thinking you were blowing a few user's experiences out of proportion....but in fact you were right all along, there was a definite problem, and you did the right thing, helping the whole process along until nVidia finally admitted the truth and resolved it... so kudos to you sir!!! Keep up the great work there!!

Regards,
Jim

---

The Asus A7N8X Deluxe motherboard doesn't just suffer from overclocking the CPU bus, but if you push the memory too hard. The on-board audio reports that the system can't boot due to a memory problem, but then doesn't allow you to do anything about it.

Luckily, you CAN restore order by using the jumper on the motherboard to reset the CMOS settings. It's annoying, but it's not TOO difficult.

Dave

---

You mentioned that the MSI nForce 2 motherboard was the only motherboard with the 100MHz FSB jumper. However, I have an Asus A7N8X that has a jumper that sets the board to operate at either 100MHz or 133/166MHz. Would this not be the same type of jumper?

As for the problem with the nForce 2, I too believe that my board suffers from this same problem. Whenever I try to overclock the FSB out of spec, the board refuses to post. I have tried everything but as soon as I move the FSB only 2 or 3MHz out of spec the board refuses to post. But it runs fine at 100, 133 or 166MHz….. very strange. However, when the board refuses to post I haven't had to set the FSB to 100MHz. All I've had to do is reset the CMOS via either the Clear CMOS jumper or by simply removing the battery.

I thought I would share my experience with the nForce 2.

Sincerely,
Dean

---

Just to let you know, I experienced the same difficulties as reported in your articles. In addition, I received some very odd messages and BIOS behavior that appeared to make no sense. For instance, sometimes the board would boot up with the wrong multiplier and the board would refuse to POST with the error message that the board was improperly OVERCLOCKED!! Often, if the boot process was interrupted before passing the process from the BIOS to Windows, the USB ports would fail. And of course, even the slightest attempt at FSB overclocking rendered the board DOA. Luckily, I discovered a way around all of this, although it is cumbersome and time consuming. It is possible to recover the board by removing the battery and then moving the jumper that adjusts the FBS bus from 200 MHZ to 233/266 MHZ. This resulted in the board booting up in factory reset. Naturally, all the settings were lost, but the board could be restored. Rather disgruntled, I gave up on trying any aggressive settings for fear of the long process of reset and BIOS tweaking that is certainly a pain. I am glad to hear that fixes will be available soon. In the meantime, perhaps others night profit from hearing of my experiences with the ASUS A7N8X.

Cheers,

Raymond

µ