And a hard drive, is gonna fail

I see that my article was posted with many replies by readers to debunk. I just wanted to point out to other readers that when we sent this information for review, we made it a point to state that we wanted someone else to verify what was happening. If you think this does not work, can someone actually test the theory? Our testing showed that the partition tables on both partitions are intact and that writing data to one does not corrupt or reduce the size of the other partition.

The comment that I thought had the most merit was from Jeff Garzik explaining the existence of free space on the drive. That at least lets me know how there came to be unused space on the drive. However, for the comments from Matt, we are not accessing the virtual partition from Ghost. It is in fact being deleted once the process occurs. After that partition is created, we are booting into a different drive with the original drive as a slave. That slave drive contains the virtual partition and the unused partiton space. The virtual partition is then deleted (VPSGHOST I believe is the name) and the new partition is formatted and designated a drive letter.

I do not think these are overlapping partitions because file space is not affected when you write to one or the other. I think that Jeff may have this right in that we are accessing portions of the disk that were reserved for bad sector rewrites and space unallocated due to inspection failure. However, to extend Jeff's idea regarding binning. Yes, processors are binned to weed out lower performers and sell them into the market as lower performance parts.

However, we all know that when demand is high for lower parts, processor manus will take higher quality parts and limit them to lower speeds. We all saw this in many of the 2.4c P4 parts that have been on the market and would overclock using ramped up FSB speeds. In essence, just because a part is the lower performance part (or in this case a lower amount of storage on disk) does not mean it was faulty or unable to perform at a higher speed (or larger capacity).

The idea that if HDD manus could up space with no penalty, they would is not valid either. They are still slaves to the market place and while 200GB+ drives are at a premium, they will need smaller drives to sell into the mid-price market since few people dash out and throw down hundreds of dollars for the largest disk available.

Most people will select what is reasonable to their pocketbook and need. So all the drives at 120GB cannot just be drives that didn't test out, there must be some in the market that have good platters and are limited for marketing and sales reasons. Unless of course they are intentionally sabotaging their own platter yields just to get lower quality parts to HAVE to sell into lower density HDDs. Somehow that seems unlikely since they constantly work hard to up density and yield.

That being said, if Jeff Garzik's explanation of what that extra space is really used for is correct (and it sounds logical to me), then the next question is how will this affect overall drive stability? Not well, I would imagine as Jeff points out.

I will be the first to admit I am going on a lot of supposition and deduction but there are still some interesting issues that I would like to see someone address by doing this process first hand. It takes about 15 minutes so surely someone can find the time and prove or disprove that this process is working. The Editor is correct in stating this probably invalidates your warranty so do what I did. Use some old drives you have laying around. I don't foresee any real use for our old 20GB drives in the bin in the back room (lol) and the warranty is non-existant. Warranty will not be an issue and then we can see if this works or not. My original message to the INQ was such that I expected someone out there to try this and verify what we have seen as valid or invalid but at least try it first It was not to make wild claims that this works definitively or is even a good idea. But is is really interesting none the less...

Thanks,

Wiley

Noisy Cisterns

Mike,

I know how you feel about the noise level of typical systems. However, it is possible to set up a very quiet Athlon XP based system (and presumably P4 as well) You have to spend a bit more here and there, but surely you have to do so for that mini ITX system your article describes, not to mention the fact you have very limited and expensive upgrade options with the "mini AGP" slot.

To give an example of how much I prefer quiet, my current system is a Celeron 533A running with its CPU fan unplugged -- it reaches just under 50C at load, and has been running fine that way for several years. The only fan in that system is a PC Power & Cooling power supply that's supposed to be noiseless. It is pretty quiet, and I've been happy with it. One of the reasons I hadn't upgraded was that I didn't think I could keep a system that quiet with today's CPUs.

Recently I decided to give it a shot however, and started with an Athlon XP-M 2600+ mobile mentioned in one of your articles a couple weeks ago. For cooling I got a Zalman CNPS7000-AlCu heatsink/fan. Weighs nearly a pound, and costs $35, but its compatible with Athlon XP, Athlon 64, and Pentium 4 so I figure it'll have a long life. I also got a couple of Zalman's system fans and their 300w power supply -- there are other power supplies that are supposed to be even quieter, but they are harder to find and I figured I could always use my PC Power & Cooling one if Zalman's wasn't up to snuff.

This exceeded my wildest expectations. I am testing it now, but I can run it on default 1.45v at 2.2GHz and CPU temperatures of 43C, at the desktop XP default of 1.65v at 2.4GHz and CPU temperatures of 51C, both measured under load and stable, with all fans in "quiet mode". I probably won't go that high since stability is more important to me that speed, but its nice to know the upper limits. Overall I can't detect any difference in the sound level of the new system and my old one. To say I'm pleased is an understatement.

I figure I spent over $100 on the Zalman stuff (and I have no affiliation with them, BTW) while I could have used the el cheapo power supply the new case came with and some $5 fans and heat sinks, but for $100 or so extra to have minimal noise is well worth it to me.

This is more just to let you know rather than something for the site, but you can certainly print it if you want, just don't use my name/email, thanks.

Doug S

India

Just to add a note to the India's time is here.

To coin a phrase 'outsaucing' surely took some intelligence, why not use some of that intelligence to make this world a better place. Surely Mike, you can try linking a UNICEF banner or some charity organisation logo on Inquirer than pretty women in an expo, an old man or some oldie goldie images and help this world. You might end up saving someone's life - his or her's good wishes outsourced to you may make you a better man.

Good Day

Gill

More on Hard Drives

Sir,

This article, while very true to OEM systems such as Compaq and Dell, etc., is true, but as far as a raw drive purchased from a computer store, it would be false. There is no hidden partitions on any raw drive purchased separately, read as not purchased in an OEM system, and should be corrected in your publication.

Scott S

Martha Stewart and Esther Tigre

RE: "HP insisted on having 32-bit AMD 64-bit chips" 9 Mar 04

Dear Ms Tigre:

You can't be serious in expecting your readers to believe that story. Even Martha Stewart had a better alibis. I still plan to file a report with the Securities & Exchange Commission concerning AMD being forced to cripple its 64-bit chip, as yet another restraint of trade on the part of Intel

The explanation you pose flies in the face of all logic. Why would HP give up the best tool to gain market share against Dell if they were not forced to by contractual obligation with Intel?

As pointed out in Forbes, a journal which Fiorina shurely reads, the I386 chip was on the market for years before Microsoft brought out a 32 bit Windows. Not a single article appeared in the press about customers suffering "disorientation and depression" because they could only run 16 bits. It is much more likely that Intel wanted to prevent the buildup of a pool of 64-bit capable computer for exactly the reason specified in Forbes.

Regards,
Dick Griest