Go to Home Page GuidesHow to ArticlesReviewsForumsFrequently Asked QuestionsNewsLinksPotpourri

Site Search

 

Abit KX7-333/KX7-333R DDR Athlon Motherboard
Last updated: 10/3/2002

QUALITYMasking, wave soldering, and the trace and ground plane layout are quite good.

The silk-screening on this board is fairly good; however, it is not as good as the EpoX boards we have reviewed which were rated excellent in this category.  Labels are difficult to find and read.

The DDR-SDRAM sockets are of very good quality and have what appear to be gold-plated contacts (they are difficult to see).

I counted 48 105C electrolytic capacitors. The EpoX 8kHA+ with the VIA KT266A chipset had 53 (and audio). That is not a really significant difference and is still a lot of capacitors for a motherboard.

The processor core voltage regulator circuitry consisting primarily of 1) an Intersil HIP6301 Multi-Phase Buck PWM Controller, 2) three Intersil HIP6601A gate drivers, 3) six Fairchild FDB6670 N-channel logic level PowerTrench MOSFETs, the large electrolytic capacitors in the picture, and the four toroid coil inductors.  At a glance, it looks like it came right out of the Intersil spec sheets and app notes and it compares well with the recommendations in the AMD Voltage Regulation Design Application Note (PDF).  The regulation circuitry for the DDR memory, AGP interface and other chips use most of the rest of the Electrolytics and employ Intersil controllers and regulators.

Going back to the picture, I can't say I like the metallic-looking (it's not conductive when measured with an ohmmeter) piece of tape next to the CPU.  I've never seen anything like that before on a motherboard.  Thinking it might be a protective covering for an engineering change, I e-mailed Abit and asked them what it was.  The response was:

"I think you are talking about the Overclocking Strips. They give a little more stability when tweaking your CPU." 

Prodding them did not produce any more information.  Maybe, they can't explain how it works (if it does).  Looking around the Internet with a search engine turned-up more info/rumors.  Apparently, Abit employees used this as one of their tricks to overclock their own computers and Abit merely included them in the manufacturing process.  This makes them engineering changes after the fact and a poor practice.  They certainly do not look like something one should see on a professionally-designed, quality product.  One might presume that the tape increases the capacitance between traces it covers and distributes noise spikes/ringing on one of them to the others, thus weakening them on all of them.  That is not the way I learned to decouple logic circuits.

There are two more pieces of tape on the bottom of the motherboard.  They are clear tape and may be just protecting the battery, etc. from shorts.  They don't look good either--band aids.

I do not overclock processors and do not recommend it, having made that mistake several times in the past and suffering the consequences of the corrupted personal and customer data that resulted.  AMD emphatically does not recommend it either.  Doing it may void the AMD warrantee.

Lately there have been a flurry of problems reported with the heatsink-fans that motherboard manufacturer's have been installing on VIA Northbridge chips. Unscrewing the fan from the heatsink and looking at the bottom reveals that the one used for the KX7-333 is an ADDA ADO412MS-G70.  The specifications for the fan are at: http://www.addausa.com/specifications/dclist.pdf.  The mean-time-between-failures is not specified, but fans with sleeve bearings usually have a specified lifespan of about 20,000 hours and single bearing, ball bearing fans roughly twice that long.  My experience with chipset (see http://duxcw.com/digest/Reviews/MBs/Epox/8kha/4.htm) and CPU fans with sleeve bearings shows that you will be lucky if one doesn't break within a year of average use.  Furthermore, before they do break they often expel bearing residue in the form of a fine black powder, which seems to have an affinity for motherboards and is probably not too good for them. 

Passive cooling would be more reliable and the chip/motherboard should be designed to make it possible.  If fans are used, they should be high quality and have ball bearings.  If the CPU is not overclocked, this motherboard will work without the fan.  If it breaks, simply unscrew the four screws holding the fan to the heatsink, while holding the heatsink so it won't work loose, and unplug and discard it.  The heat sink is barely warm to the touch when the fan is removed and the computer is working hard.

10/6/02 Just three days after this review was originally published, the Northbridge fan started making that screeching noise one typically hears, when first turning on the computer, that indicates the bearing is seizing and starting to fail.  Larry

< Previous | Contents | Top | Next - Performance >

Copyright, Disclaimer, and Trademark Information Copyright © 1996-2006 Larry F. Byard.  All rights reserved. This material or parts thereof may not be copied, published, put on the Internet, rewritten, or redistributed without explicit, written permission from the author.