Taiwanese motherboard makers amateurish

WE'RE A SMALL reseller/assembler/consultancy here in Australia, with a smallish unit count, but it keeps our company busy with higher-value services rather than comodity home retail.

We always try products our disti's suggest. So we tried the Abit LX6, VP6 (VIA 694X dual CPU) and then KG7 (probably the crapiest motherboard ever made).

Without exception the LX6 and VP6 motherboard have failed completely. Other Intel, Tyan and Asus motherboard we used in that period, which far out number the Abit product are still running. The chassis/PSU/HD combinations are pretty well all the same over those years.

Abit quietly removed ECC support for the VP6 in a BIOS upgrade, devaluing these servers for our clients. I sent 4 emails requesting a reason, or re-instatement but received no reply.

The KG7 motherboard has an entire troubleshooting website for it at viahardware.com (or "sudhian"). In a desparate search for reliability for this board we would upgrade to the latest BIOS only to have the machine fail to POST not recognising the high quality Kingston DDR we were just using in the previous BIOS version. All but 1 emails out of 8 to Abit support regarding problems with this motherboard were ignored.

Sudhian link

In the four years we experimented with Abit product I sent 12 emails and received replies to 2. We have removed all KG7s from our clients and replaced them with Intel.

In the seven years we have dealt with Tyan, I have sent 29 emails and had replies/resolutions to all.

The Taiwanese need to learn a huge amount about customer service; if you don't buy a 40-foot container a week of product, they don't give a toss. Their websites are amateurish, frequently overloaded and slow (eg Asus). Another example;

http://www.seventeam.com.tw try the "E-Mail" button, you'll get a non-deliverable, which we alerted them a fortnight ago. They make all the PSUs for Inwin.

AMD/Nvidia need to be very very hard on these low quality, poor service Taiwanese mass manufacturers (VIA included). Client's and dare I say it dealers are smarter than ever and I think increasingly are buying quality, which we associate with Intel/Tyan, and unlikely to include AMD or VIA.

Regards,
Connor

Subject: Sony makes PS2 on a chip

... thus underlining why Microsoft's Xbox was doomed to the loss-making pile before they even started.

Microsoft can't put an Intel processor, an Nvidia graphics card and I/O processor and SDRam and pack them onto 1piece of silicon to reduce manufacturing costs. Manufacturing the XBox will always be much more expensive than manufacturing a PS2.

Peter Lindsey

Subject: Intel and their use of English

As an Englishman I found many of the Phrases you use incomprehensible.Please don't critisise other when your own choice of words is so poor.

carl studt

Mike replies
You misspelt criticise. Also, I'm not an Englishman, that's a deadly insult...

Intel is not the bleeding Oxford English Big Dick

Mike

I was surprised to read an editorial ('opinion', 'commentary') actually condemning ('putting down', 'dissing') people for making use of the English language - words like 'innovative' and 'anomaly' are hardly unusual ('weird'), nor should they be taxing ('hard') for a smart writer such as yourself to decipher ('figure out'). The English language is shrinking quickly enough without the process being artificially accelerated ('speeded up', 'hurried along', 'rushed') by those too lazy to read a couple extra letters per word. Where do you draw the line? Is everything supposed to monosyllabic ('short')? And no, I was not an English major. Just someone who doesn't want us all to have only twenty words to choose from because everything else was just too darn demanding.
Cheers - PKRead

Subject: Opteron quirks

Egads, did you see Tom's memory bandwidth figures?

Seems he did not even enable dual channel.

Does he do that kind of crap on purpose?

Leave it to an amateur to hobble one of the platform's strengths.

Maybe I'm wrong but how else can Opteron with Dual DDR 333 have less bandwidth than The Xeon with dual DDR 266?

Xbit showed the clawhammer coming very close to the theoretical bandwidth limits of DDR 400, in single channel mode of course.

The opteron has dual channel on each of two processors, effectively doubling bandwidth also.

Perhaps it is an asymmetric issue, like using DDR400 with a 333fsb Athlon. The Opteron has a native fsb of 800 and supports DDR400, so perhaps we wait for DDR400 ecc reg before this is solved

Something is not right, and since Tom's makes no mention of it in their article, I'll assume for now that it was an oversight on their part.

Consequently, they showed the opteron performing miserably in any memory intensive application, while absolutely dominating in the pure CPU tasks.

bloody shame.

It would have been nice to see someone slap a PCI GF4 mx 440 or Radeon 7500 in an opteron server and similarly hobble the competition to give a better Idea of the platform's performance. Throwing an 8 mb trident against a 9700pro is a bit ludicrous.

Jason

Thank you for an in depth look at the launch. I spent an hour and a half (gasp) checking your site, Bennet's and Tom's. It's humourous, for the past 3 hours AMDZone has had the headline "45 minutes to go".

Good to see the little knee-nicker going for chipzilla's nads, that is until they decide to do something or Madison 2. Amazing what oodles of what real men have and cash will do.

I hope Satan's enthusiasm will rub off on his minions in industry, and if it does and AMd starts selling servers in the double digit percentiles, I'm sure AMD stock will be back down to $3.12 in no time !

Thanks again for your usual breath of fresh air amongst the flatus.

Peter Valiunas

http://www.theinquirer.net/?article=9048

excellent. well written and extremely educational. thanks for all the work that went into this. i thoroughly enjoyed it.

now if only you could get it read by the mainstream media. 8^)

good job!

andrew

And I though I was the only one who noticed things didn't add up. At the same time I think their numbers for music CD's being burned seem grossly exaggerated. I have burned several hundred CDR's, but less than a dozen of them were of music, most of those were copies of favorite CD's I already owned to use in the car, the rest where compilations, and this is after having a CDR for over seven years. Besides that even if I didn't have the ability to burn music CD's I can honestly say I wouldn't have bought more CD 's over the past seven years, but I can also say if they had a better product I would have bought more albums.

Email address supplied

I am not privy to the nuances of the case, as regards the RIAA vs. Verizon, that is for the courts and it should be interesting. Also, I am not a supporter of Mr. Ashcroft or his vision of US law, but I do fail to see how traffic in copyrighted material is generally protected under the First Amendment.

Article I (known as the First Amendment) says:

"Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances."

I am also amazed by the apparent stance of both The Inquirer and some of its reporters. You have often stated your opposition to others copying your work. And I do note that you have a copyright notice at the bottom of your page. Are not musicians entitled to the same protections? The ability to steal and the right to steal are not one and the same.

This stealing of copyrighted material is not good for the future of an online community or the Internet. It is too bad that probably we will all be negatively affected by the actions of the copyright thief.

James

Subject: J2ME vs. Symbian games

You wrote in your article about mobile games ("Symbian neglected at quiet mobile gaming show") the following:

"Developers appear to be concentrating very much on Java/J2ME at the expense of rivals such as Brew and MoFun. More surprising was the general reluctance to develop games for Symbian given that two of the most popular games handsets - 7210 and the 7650 - are from Nokia."

First of all, the 7210 is not a Symbian phone, because it is a Series 40 model with the standard Nokia operating device. The 7650 has Symbian OS and Series 60, such as the 3650 and forth-coming N-Gage.

I think it's still quite obvious why developers are concentrating on J2ME games instead of Symbian products: there are a lot more J2ME supporting devices than what there are Symbian handsets.

Nokia has estimated that it will sell about 10 million Series 60 (Symbian) phones this year, which is only about 10 % of all the phones they have estimated to sell this year, supposing that their market share will remain at about 37%. Then again, most today's mid-range and even lower range models support J2ME.

I have also worked in a J2ME or Symbian project where we have considered this issue whether to build a J2ME (or Java MIDP in our case, to be exact) application or just build a Series 60 / Symbian application. We acknowledge that a native Symbian (C++) application would give us a better advantage of hardware and software of the phone, but a Java application gives us instantly a larger target audience of consumers.

Best regards,
Tero Lehto

Subject: chairman Jerry Sanders III, said, contained 100 million transistors and described the die as smaller than your smallest fingernail

At 193 sq. mm.., Cherry Sanders wasn't hooting about AMD's small die size advantage over Intel's 1130 sq. mm. Xeon - was he?

193 sq. mm die size - 89 watts power consumption.

AMD must be cooking a LARGE TURKEY !@!!

Paul R. Engel

µ