Print 19 comment(s) - last by Martimus.. on Aug 16 at 1:20 PM

A recent correspondence between DailyTech and Digitimes leaves everyone "scratching their heads"

I don't usually post emails from other editors, but this one was just too good to pass up.  The following is a correspondance between DailyTech and Digitimes after DailyTech published a rebuttal to a Digitimes article claiming tens of thousands of Radeon HD 2600 and 2400s had been recalled.
Hi Chris

You seem to have gotten your wires crossed with this article.

The reports Digitimes made had nothing to do with the UVD (that was an unrelated article), the issue (as it said in the reports) was that the cards were unstable and needed a BIOS update to correct them.

It's no wonder your anonymous sources were left scratching their heads if you were asking them about an issue that no one is reporting as existing. Perhaps if you go back and ask them about the correct problem you will have more luck.

I hope you or your editors will revisit this article soon, we stand by our original report.


Ricky Morris

To this email I replied with the following:
Hi Ricky,

I believe you writers have taken several liberties in the article in question.

AMD’s most recent errata calls for two bugs in the most recent A13 silicon (A14 was shipped to manufacturers): UVD failure and a reported hang during 3D Mark. These two errata are intertwined and fixed with the same BIOS update.  If the defect that Ms. Chen and Mr. Tsai documented is completely different, then it’s the first anyone at AMD or any of the AIB partners have heard about it. 

According to the Digitimes article:
“The issue was not only encountered by small makers, Asustek Computer, Micro-Star International (MSI) and Gigabyte Technology all experienced the problem too, noted the sources.”
All three of these manufacturers confirmed that they have not found defective ASICs in their inventories.  In addition, all three confirmed that no Digitimes employee ever inquired them about any recalls – at least not through any of the employees they probed, which I understood was a lot.
“One first-tier maker has recalled over 20,000-30,000 units already, noted the sources.”
In addition to the fact that those 3 top-tier guys denying that any recall is in place, an AMD rep reached out to me to mention that this statement was also completely untrue.  I believe the word “fraudulent” was used.
“It is believed that most of the defective cards are still in the channel or on their way back to the makers with only a small number of having already been purchased by consumers, noted the vendors.”
When I inquired about this point, specifically, my contact mentioned the Acer / UVD point demonstrated in Mr. Ram’s article.

A memo forwarded from AMD to all of its APAC add-in partners follows, with an attachment of the Digitimes article cited on DailyTech:

Last week an article appeared claiming that AMD shipped UVD-disabled Radeon HD 2400 to some of our customers. Please review the statement below and feel free to use this with your customers to assure them that *all* Radeon HD 2400 and 2600 products they receive are 100% UVD enabled.
“We can confirm that all shipping ATI Radeon HD 2400s and ATI Radeon HD 2600s have working UVD functionality.  We are shipping a number of HD 2400 A13s and their UVD functionality is enabled in the August Catalyst driver drop.  All of the remaining HD 2400s, and all of the HD 2600s, have UVD support in the current driver.”
It’s certainly a possibility that AMD and these manufacturers have all colluded to spread disinformation about the topic at hand, but given the inaccuracies of other reports by Ms. Chen and Mr. Tsai I’m going to stick with the advice of Hanlon for this one.

Kristopher Kubicki

Scraching their heads, indeed.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

I don't think this is a good idea
By Martimus on 8/11/2007 2:28:47 AM , Rating: 3
While this blog was interesting, I am not sure posting a private e-mail for everyone to see is such a wise idea. That is usually a good way to lose sources as people lose trust in being able to talk to you without having you repeat what they say to you to everyone else. I would remove this before too many of the people actually read it, since it is still early and not many people have seen it yet. Of course you are free to do whatever you want, this is just my experience based on similar mistakes that I have made.

By Martimus on 8/11/2007 2:30:27 AM , Rating: 2
To elaborate, you could write a report that has all of the meat, but doesn't name names or show exact context of the comments made.

By Rampage on 8/11/2007 3:20:24 AM , Rating: 2
This isnt that big of a deal to remove. Its damaging to no one really, its merely interesting. DTech intends to report the news, and theres nothing even embarassing for DT or DT here.

Regardless, due to YOUR comment I already have a copy of this page saved on my PC. :)

By James Holden on 8/11/2007 3:39:35 AM , Rating: 2
I didn't see anything particularly damaging to anyone here. I was a little surprised that the digitimes editor claimed his report was a separate bug though: AMD publicly stated that the error digitimes found was for UVD.

So what I want to know, is AMD blowing smoke or Digitimes?

By leidegre on 8/11/2007 5:44:27 AM , Rating: 2
The Internet was invented for flaming! Get used to it, or burn like everyone else.

RE: I don't think this is a good idea
By SandmanWN on 8/14/2007 10:16:51 AM , Rating: 2
Not much of a source to lose when they make such a bad judgment in reasoning and have nothing to back it up with yet still support the article. Bleh!

By Martimus on 8/16/2007 1:20:42 PM , Rating: 2
When you make it public like this, they aren't the only source you lose. You lose others who read this and think, "What if he shows the same disrespect to me?" They begin to question whether they can trust giving this person any information that might reflect badly on them if it came out in public. What was written here was pretty benign, but the fact that it was written in the way it was written raises questions about whether it is safe to trust the person with sensitive information. That was why I wrote the warning. I couldn't really care less about this feud they have, or the actual information this article contains, because it has nothing to do with me.

lots of mystery motives..
By hellokeith on 8/11/2007 12:19:03 AM , Rating: 2
I don't usually post emails from other editors

So why did you? If this is a he said she said situation, posting an uncorroboratable email gains you no more credibility than you had before posting it. I trust DT more than other sites already, so what is to gain here?

RE: lots of mystery motives..
By KristopherKubicki on 8/11/2007 12:23:25 AM , Rating: 3
Well, I thought there was relevant information in the emails that was not revealed in either article. I have nothing but respect for Mr. Morris and his team, I just think they were way off the mark here.

RE: lots of mystery motives..
By omnicronx on 8/11/2007 2:53:49 PM , Rating: 2
I am siding with you on this one, if another site is willing to post wrong information, and on top of that are willing to defend it even after they are proven wrong, someone has to call them out. It sounds like they contacted DT not the other way around. Good job for not making too much out of it either, you played your cards well.

Scraching their heads, indeed.
By crystal clear on 8/12/2007 8:45:56 AM , Rating: 1
Yes M.S. & AMD will be Scraching their heads, indeed
with this news-

ATI driver flaw exposes Vista kernel !

An unpatched flaw in drivers from ATI creates a means to smuggle malware past improved security defences in the latest version of Windows and into the Vista kernel.

Microsoft is working with ATI on an update which security watchers warn might be far from straightforward to roll-out.

The existence of the security flaw in ATI's driver came to light after developer Alex Ionescu released a proof-of-concept tool called Purple Pill that created an easy way to load and unload unsigned (potentially malicious) drivers on Vista. The utility circumvented new anti-rootkit defences built into Vista by turning off checks for signed drivers.

Ionescu pulled the utility hours after its release after realising that the ATI driver flaw Purple Pill uses, which he learned about in a presentation by Vista kernel security expert Joanna Rutkowska at Black Hat last week, is yet to be patched.

The functionality of Purple Pill is similar to Atsiv, a tool designed by Australian developer Linchpin Labs, as part of a research project into driver signing. Microsoft responded to the creation of Atsiv by revoking its certificate and classifying the utility as malware

RE: Scraching their heads, indeed.
By 265586888 on 8/13/2007 12:45:02 AM , Rating: 2
And already fixed :)

So please stop the shouting and whinning, thank you.

RE: Scraching their heads, indeed.
By crystal clear on 8/13/07, Rating: 0
By KristopherKubicki on 8/13/2007 8:40:25 AM , Rating: 2
Easy tiger :)

Sept 10
By crystal clear on 8/13/2007 8:30:31 AM , Rating: 1
I make this post as it related to AMD & Digitimes-

AMD to launch two Barcelona-based processors in September

AMD schedules to launch two 2-way Opteron (Barcelona) server processors on September 10 this year, according to sources at server makers.

The Opteron 2348 and 2350 will have core frequencies of 1.9GHz and 2.0GHz and will be priced at US$320 and US$390 in 1000-unit tray quantities, respectively. In October, AMD will launch the 2.2GHz Opteron 2354 with 95W TDP and a price of US$610 in 1000-unit tray quantities, however this processor should actually appear in the channel later this month, noted the sources.

The Opteron 2300 series will include Opteron 2340, 2350, 2352, 2354, 2356, 2358 and 2360. Models at frequencies higher than 2.0GHz are expected to launch in the fourth quarter, according to an earlier announcement by AMD.

AMD also has plans to introduce the 8-way Opteron 8300 series which will include Opteron 8348, 8350, 8352, 8354 and 8356 models, added the sources.

AMD declined the opportunity to respond to this report, saying it cannot comment on unannounced products.

Source: Server makers, compiled by Digitimes, August 2007

RE: Sept 10
By crystal clear on 8/13/07, Rating: -1
RE: Sept 10
By CyborgTMT on 8/14/07, Rating: -1
RE: Sept 10
By crystal clear on 8/14/07, Rating: -1
RE: Sept 10
By CyborgTMT on 8/14/07, Rating: -1
"Intel is investing heavily (think gazillions of dollars and bazillions of engineering man hours) in resources to create an Intel host controllers spec in order to speed time to market of the USB 3.0 technology." -- Intel blogger Nick Knupffer
Related Articles

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki