INQUIRER "like a soap opera"

It just struck me why I have now seemingly become addicted to the Inquirer.

While you do seem to report things first time and time again, I don't think that's it. It hit me when I was reading yet another story about how you've pissed off some exec, this time some guy from Morgan Stanley who, like so many, has told you to go and do some more research. Clearly these people don't understand the appeal of your little webzine.

It's not the news that I find so captivating, but the outlandish attitude you have when talking about just about anyone. Every article spews forth opinion after (often unqualified) opinion, and this is no doubt why people have said time and time again that you show an immense amount of bias in your reporting. We see people time after time fuming and "flaming" over the criticisms that you hand out like a confectioner might hand out penny sweets. It's pure drama and I love it.

I mean god, IT is a vibrant business and everything, but technology articles don't half get boring after a while. It's a relief to read articles that offer a healthy dose of narrative pizazz while actually getting the important points across, and these quaint tales of "he said, she said" tiffs with people like nVidia, HP and Sun offer welcome relief from what might otherwise be a very dry piece of writing.

Well done for pushing the boundaries of IT journalism and carving a new niche for yourselves in the market. Tabloid IT journalism with an injection of soap opera. Whatever will happen next? I can't wait to see.

sincerely

Alex Swabey

CRT monitors are dead, monitor firm says

"So what about the 46% of CRT sales? Hollmann claims these will mostly be sold into Eastern and Central Europe, where, he claims, price is "still the main driver".

The professional graphics market (pre-press, digital photography, etc) will also be a continuing market for CRTs for quite sometime, because you cannot *accurately* calibrate an LCD display for true colour rendition. And, there is no new LCD technology on the horizon that can allow display colour calibration on par with a CRT (such as the incomparable Mitsubishi SpectraView series).

http://pro.necmitsubishi.com/

Hiya News Desk,

I had to laugh @ this bit....

So, while you can, buy the biggest CRT screen you can for the cheapest price. But that's what we're saying, not him.

Gazing as I am, on my Samsung SynchMaster 1100p+, in all of its 21' glory.

Reports of its death, are exaggerated....

*BFN*

Ten good reasons not to buy AMD chips

Dear Sir / Madam / Non-Gendered Being

I read the article entitled "Ten good reasons not to buy AMD chips" and couldn't help but think you weren't really putting enough effort into it. Though I have an Athlon XP myself (damn 2200 Thoroughbred A furnace!), and my next chip will be a 2500+ Barton AQXEA (yes, I'm very choosy now on stepping) which I intend to overclock to the high heavens, I am very annoyed by AMD's PR team.

One of the main reasons I hate PR is that they seem so split apart from reality that they cannot remain in tune with the real people in the market. I know that AMD has to use the dreaded xxxx+ system to compare it with Intel, but WHY do they have to do it so unfairly and dishonestly? (Before you say it, yes, we KNOW they claim it's based on the Thunderbird megahertz scale, but nobody believes them!)

When the first T-Bred A Athlon XPs came out (argh, too many As close together there), AMD decided that it would be unfair to release a 1.7GHz chip to compete against Intel's 1.7GHz chip (was it the Williamette?), so they decreased the speed to give consumers a fair comparison. For a time it was good, right up until about the time the Barton came out and Intel released chips with the 800MHz Quad-Pumped Front Side Bus. Can the AMD PR team say, "Fudge!"?

So, the 2500+ has a stock speed of 1.83GHz??? True, it has a FSB of 166 (333DDR) and an extra 256K L2 cache, giving it 512K total, but if you've compared the 3200+, for example, against the P4 3.2C, it simply doesn't compare. (Spontaneous outburst of "Nothing compares, to you!") The only good thing about the new Barton range is the 2500+, and only because you can easily overclock it to the stock speeds of the 3200+, which has a speed of 2.2Ghz and a standard FSB of 200 (400DDR.) I have every hope of running mine with PC4000 ram on a 250MHz FSB and getting it to over 2.5GHz with a Prometeia Mach II.

For anyone who's not used to using these figures on a daily basis, it's confusing and annoying. AMD are essentially lying to consumers. The chips do not provide the performance they have said, and I would only recommend an AMD system to someone over an Intel system IF they intend to overclock or if they need a system on a budget. If they have no intention of overclocking (stupid fools), then I would have to recommend an Intel based system, simply because it will give you more out of the box. I may not like Intel, but you cannot dispute the benchmarks (unless it's 3DMark03 ;) ) If I post on overclocking forums, and I do, I generally find that most people will go for AMD because they can squeeze a lot more out of AMD chips that they give by default, but what about your average consumer who would dispute that "overclocking" is even a real word???

So, why not ask some computer enthusiasts on forums before writing an article for their opinions on the AMD-Intel war beforehand? Some of us know our stuff. (Not all of us though, as you soon learn by posting on overclocking forums for a while!!!)

Thanks for reading this far.

(Please feel free to correct any grammatical and/or spelling errors that I have missed in my haste. [No, Ed.])

Re: Rick Reroy's Writen:

Wherein it was wroten;

"And if you Unix users out there are feeling smug then let's hope you've remembered to update your sendmail and unzip software within the last few weeks."

Actually, because(*) the sendmail issue is not a UNIX issue because sendmail is not a part of UNIX and the unzip issue is also not a part of UNIX. I would suggest that you should have written it thusly:

And if you Red Hat distro users of unzip or sendmail out there are feeling smug, then let's hope you've remembered to update your packages. Those of you who have signed up for the $60 Network Basic program would have been in a position to have your systems automatically updated months ago when these issues were first discovered, of course.

(*) I was taught not to begin sentences with the word 'because' because the learned ones have declared that such a thyng is wrong. My blynd cousin, however, whose name is Hacktuel and whom we nicknamed Hactuely, told us that we would shurely be able to get away with it if we would just say his nickname first, then went ahead and started any sentence in the world with 'because'. I have changed the spelling of Hactualy to Actually to please the authorities. And, yes, I do often have too much time on my hands and I'm trying to prevent my otherwise idle mynd from becoming the Devil's Playground, as the godly ones are wont to grimly remynd me of.

Gene Mosher
ViewTouch

It would seem that once again AMD is trying to make itself known amongst the common people, and have benchmarks that reflect its processor in the best light. Oh wait, this is the first time they are doing this! We know AMD for a long time has had chips that outperform Intel chips and are cheaper than Intel chips, they tried a few ad campaigns more recently the AMD ME campaign. Unfortunately, it was useless. Its time for AMD to come out with an advertisement that basically says that they make CPU's just like Intel, and that their CPU's are “cheaper” and better. Everyone knows what Intel makes, not everyone knows what AMD makes. Given some information they could increase their market share very well. Yet they don't do it, and they come up with ingenious plans to tie a motorcycle to their bike and then claim the bike can go this fast. Its time for AMD to hit the in-duh-viduals to borrow Dilbert's phrase. The people who read reviews on websites make informed decisions and polls have shown that they pick a good CPU. Check for instance the latest poll on HardOCP see Next CPU-HardOCP. There are a substantial number of people who are going to make their next CPU an AMD one. Those who are choosing Intel processors have already been exposed to reviews on AMD processors and motherboards and have seen the difference. Either they want to blow the money on a P4 or are hardcore Intel fans(or have some other feasible reason for choosing Intel, but as an AMD fan boy I must say that there are none :P ). AMD needs to hit those that don't dabble enough to read reviews or visit tech websites on a daily basis. They posses the knowledge to put together a computer but aren't fanatics. They will pick Intel just because it has familiarity and “seems” to be the best based on funny looking aliens on TV. AMD marketing is generally a failure, its idiotic, incompetent, and probably full of people that don't even know exactly what the company makes. Rigging reviews is not marketing, its cheating, and its deceitful and dishonest to those that have followed it throughout the years.

Ajay Desai

Currently, comparing windows to it's closest (and arguably superior) competitor OSX, the biggest glaring difference is the GUIs. XP isn't anywhere near as elegant as OSX's openGL-based Aqua. However, because the hardware is seperate, OSX in many ways is not directly competing with XP. On the other hand, 4 years is a long time for the KDE and Gnome people to get their act together on a cohesive GUI. Linux's lack of a common interface is well documented, and work has already begun to fix the problem. The weight of Appearance/Interface should not be underestimated. Perhaps a graphical overhaul as a windows interim release would not be such a big mistake, especially if Gnome3 or KDE4 comes out and blows XP's current interface out of the water.

...and which compiler.

When I did my PhD, I did some analysis with whole-system hardware instruction traces, and with SPEC95. The traces were huge, but great: they showed a machine in both system and user state, doing file IO, getting network interrupts, the works.

The SPEC95 programs did not statistically resemble the traces. Each was unusual in its own way, and "small" by many definitions of small. The one exception was "176.gcc". It was big enough, and spread out enough, that it was actually a very good proxy for a trace.

When I want to know how well a new chip will run big bloated general purpose system code, I just look at "176.gcc", and throw the rest over my left shoulder.

Which is why this page is interesting. AMD uses the Intel C compiler in its benchmarks, because that's been tuned to a faretheewell to give good SPECmarks. But notice the charts at the URL above. Intel's compiler gives 745 on "176.gcc", but on the same (opteron 240) chip, GCC 3.3.1 gives 993, which is 33% faster.

So when the AMD64 comes out, what I wanna be told is the GCC "176.gcc", and skip the rest. Preferably a number bigger than the Itanic's 1554.

Oh come on, PC OEM's have been doing this for years. I used to work in the R&D lab of a major PC OEM back in '98, and you can bet when we sent a system out for review we cherry-picked the best components we could that regardless of the expense, whereas the components used for actual production systems for any large OEM are picked on price, compatibility, reliability and availability first and performance last.

We wouldn't try and switch for something with an obviously different spec, like the Dell example of 5400 vs 7200rpm on the hard drive, but you'd pick a drive with really fast seek times or a large buffer (say 512kb instead of 128kb which most people did not think to ask about back then) or just a drive from a better manufacturer, like an high-end IBM (This was back when IBM was good) instead of a low-end Maxtor model.

I always told friends if they were buying a PC on the strength of a review to make sure to ask the manufacturer for a system just like the one in the review and check all the components matched up or demand a refund. You could sometimes get a really good deal that way if you knew enough about the cost of the components involved.

Aunty Dan