Tomshardware do a bit of undervolting

The forum for non-component-related silent pc discussions.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Patron of SPCR
Posts: 216
Joined: Fri Aug 29, 2003 10:58 am
Location: Sheffield, UK

Tomshardware do a bit of undervolting

Post by the_smell » Wed Sep 29, 2004 6:39 am

Just found out about this article, I haven't looked through it in detail, but if you feal like modding a normal althon XP to be recognised as a mobile it has a bit of info there...

I still think a proper mobile athlon is the best way to go!

*Lifetime Patron*
Posts: 3214
Joined: Wed Nov 12, 2003 10:06 am
Location: Pittsburgh, PA

Post by sthayashi » Wed Sep 29, 2004 7:07 am

How can THG be so f***ing clueless
THG wrote:The only limitation is the CPU core voltage that cannot be varied under Windows - AMD keeps the corresponding utility under lock and key for security reasons. Thus a compromise between maximum and minimum speed is called for
And there was no reference made to 8rdavcore. They seemed to focus more on underclocking rather than undervolting.


SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ

Post by Rusty075 » Wed Sep 29, 2004 7:42 am

I don't even know why I bother reading THG's "techie" articles anymore, but here's my input anyway:

1. Stumbled through 4 typo's/grammar/logic errors on the first page alone...then I just stopped counting. ("Intel, for example, dreams of the "digital home," yet only delivers components that would last long without active cooling controls." :roll: )

2. "This involves opening the multipliers, which only these processors allow." Wrong.

3. As sthayashi mentioned, they obviously don't know that there are several windows/linux apps and mobo-utilities that allow on-the-fly vcore adjustments. More quality research from the boys at THG. Hint to THG: Try Google. :roll:

4. No discussion of the ways in which actual XP-M's are different from regular XP's.

They pretty clearly miss the whole concept that reducing vcore is much more useful than reducing the Mhz. In my opinion, the only reason to underclock is to get a better undervolt. They don't even appear to have thought about the idea of undervolting at stock speeds.

"Zalman's CPNS 700A-CU" I'm sure Zalman loves the mis-identification of their product. More quality editing from THG.

I like the total lack of information on where there "Power consumption" numbers come from. You cannot use an ammeter to meaure power consumption in an XP or a P4. (power is drawn from multiple lines, in unknown proportions, depending on the mobo)Useless.

Lol....I bet I won't get a christmas card from Tom this year.

Posts: 26
Joined: Fri Jul 09, 2004 2:35 am
Location: Sydney, Australia

Post by stmok » Fri Oct 01, 2004 8:42 am

Agreed...Their attempt to calculate the thermal output of either P4 or Athlon is totally wrong.

I honestly would avoid their regular hardware site. It has become a pityful excuse of amature crap that even high school students can do a better job of.

One minute the article talks about the AthlonXP and how to do "this and that" to it, and then all of a sudden, they throw in a P4. WTF? Stick to the subject that is mentioned in the title...Its about underclocking/undervolting an AthlonXP CPU, why the flying f**k do you need a P4 in it?

Their methods and approaches seem a bit "inconsistent" in areas.

They mention mobos that don't work with this mod process...How about a list of mobos that do and do not work? (That would actually be useful)

What about Socket-A Semprons? (I know they don't do to well in overclocking, but what about underclocking/volting?)

Their is far better. (Since its articles are written by someone who DOES have some clue).

Posts: 257
Joined: Sun Mar 28, 2004 8:30 pm
Location: nanaimo BC Canada

Post by vortex222 » Fri Oct 01, 2004 9:07 am

its the most useless article to date on there site. besides the article they did many years ago on 34 different Gforce4 TI4200 videocards, the benchmarks indicated they were all the same speed.

OHHH and dont forget the 50 articles they did doing the Duron and Celeron Race.

Patron of SPCR
Posts: 216
Joined: Fri Aug 29, 2003 10:58 am
Location: Sheffield, UK

Post by the_smell » Thu Nov 04, 2004 4:58 am

Tom's done it again... :)

When looking at the intel stock LGA775 cooler...
'During our stress test the cooler ran at full speed, reaching a noise level of 46.3 dB(A), which still allowed work in a relaxed manner.'

Sorry I couldn't resist

Post Reply