64bit is good after all

ONE USER WROTE in your article "64bit processors a waste of time" the following (and I can't help myself, i just have to respond..)

The question is WHY do we need to move 64 bit anytime soon? Even most power users are barely using 1GB of RAM in their systems now. It surely isn't the 4GB of addressing space. It can't be because we need 64 bit versions of Word and Netscape sigh.. rm *

ok.. This is a home user. lets light the lights for 'm: In the real world 4GB is not enough. the X86 architecture CPU's are nowadays used more widely than ever. not only at home, not only for dtp, and not only for cad. In small, medium and large bussinesses they are used as (no, really) servers. 4GB here, is not enough. in my serverpark there are several Terminal server boxes for which I just love to have more than 4 GB. These boxes are running almost 100 users each with big app's and memory is my only constraint. Same goes for SQL. Same goes for almost everything. (ever used Notes for 2500+ users ??) more is and always will be..better. and AMD is providing it at lower cost (as to sun, itantium etc) and in the X86 arena, so that real applications like this (Terminal server and SQL, etc) can finally benefit from it...

bring it on. thanks for reading. Met vriendelijke groet (With kind regards),
Niels-Paul Kuppens

Subject: Not much wi-fi action - that's none, at Kennedy, Heathrow

You should come to Vancouver's International airport. They have so much coverage it's silly, you can get it almost to the parking lots. ;) I know the guy who installed it (Radian communications) and they are getting complaints from their 'security experts' as there is "too much coverage".

tim

Subject: Has Intel spannered AMD Fujitsu deal

Yip I'm an AMD Fan boy before I start my flame on Paul R. Engel >D

Think the two things I mainly didn't like about this guys rant was the claim of Poor Performance & It's relative Low Clock Speed, other than the fact he's obviously a Intel Fan Boy in hiding, I'll let that slide...

Poor Performance based on what exactly??? From what I've been reading the world+dog has accepted that the Opteron CPU is great little performer and from your own previous article even Micro$haft says so (anything to dig at Intel) and what is performance based on Megahurts? Or as most enthusiasts know it's based on L1, L2 & L3 Cache, IPC, Onboard Memory Controller (Smite Intel Smite).

Sounds like this guy is still stuck in the past, praying Intel can do something about AMD64, which I hope they will fail again at doing by releasing yet another shite processor just as they did with the P4, what a pile of pi$$.

Before I receive any flames back, I'm only an Enthusiast, so what do I know :D

Plz let the flames begin... Thanks for listening to my rant...See Ya!
Paul-Robert Archibald

Subject: Response to Paul R. Engel's comments on the opteron

Paul claimed the opteron had a "HUMUNGOUS 85 WATTS POWER" dissipation at 1.8ghz?

Opteron is being rated at a thermal design power, or thermal tolerance, of 80 watts, according to sources. But OEMs familiar with the design say AMD is being generous and allowing quite a bit of future thermal headroom for faster Opterons. "That's the spec," said Phil Hester, chief executive of Newisys, a Texas server startup wholly focused on the Opteron, and in which AMD owns a minority stake. "But the actual is more like 40 watts." Part of the Newisys sales pitch is to allow customers to touch the working Opteron chip package, which is "slightly warm to the touch," Hester said.


Paul insinuated a 184 mm square die is outrageously large.....
Itanium 2 madison 421 mm square.

Poor performance..........
www.anandtech.com
well at least it appears unrivaled as a server. Workstation performance doesn't look the best but i'm witholding judgement until nforce3 arrives soon (an actual workstation chipset)

Aiden Morrison

Subject: 64bit processors are a step into the future

Which came first; The Chicken or The Egg, as the saying goes. If no 64bit applications for PC workstations exist, why do we need a 64bit CPU?

1. CAD and 3d Animation , a 64bit x86 machine takes away the need of having to use an expensive SGI machine. Being a Windows or even Linux machine it will play alot nicer with others, being a breeze to setup on a Windows network. Applications that would benefit: Pro/E, Maya, Studio, Solidworks, to name a few. Anyone ever try to network and SGI machine on mixed platform network? It's not pretty. It will bring down the cost and complexity of doing CAD.

2. Video Games. greatly improved 3D rendering.

3. Video editing. Video Editing ecounters the same memory intensive processes still image editing does, but to a greater extent. Having more than 4 GBs of RAM may be helpful when editing Hi-res video.

It was only a few years ago, when people were skepical about the need for a 32-bit OS, let alone a GUI interfaced OS.

I believe the time is right for a 64bit x86 compatible desktop CPU. It may be quite a while before we need it, but I think it will be far sooner when people start wanting it.

Jeff Dranetz

Subject: AMD x86-64 Branding

It seems pretty obvious to me that this edict from mount AMD is an attempt to keep intel out of the x86-64 biz. If Intel decides to try and make up some lost time in the sector, they will be coming in late, and you can't try and instate your CPU architechture later in the game, so if they do come in and use AMD's instruction set, they will be forced to call it AMD64, not the wholly less embarrassing x86-64. I think it's pretty brilliant.

Mike

Subject: Net.wars "I ain't never seen no horses download it" Note

Well, it's true. Small record labels *are* having trouble with all this file sharing going around. One of Belgium's (once most known) labels (Lightning Records) went belly-up last month. Sales were down dramatically over an extended period, even though their music wasn't really less popular.

Everyone is, or was, grabbing it for free..

Regards,
Hilke

Subject: RIAA article 23-04-2003

Jack concludes the RIAA won't sue the consumer. In the US, the RIAA has already sue four University of Maryland students for copyright violation (I don't know if it was for file swapping or distributing CD copies). Additionally, the RIAA issued a subpoena to compel an ISP (Verizon) to reveal the name and contact information one of Verizon's subscribers the RIAA contends is regularly engaging in music swapping.

It appears they are lining up the examples....

Chuck

Subject: Opera

I think the Inquirer should be commended for admitting to publishing an erroneous article and coming clean. In all fairness, the original headline did contain a question mark in the punctuation, and I took this to mean that the Inq staff wasn't sure of its technical merits. By publishing the article and quickly publishing letters critiquing the analysis the Inq actually did a service to the IT community by more or less proving that Opera is in fact not spyware at all. No one can expect IT journalists to have the technical knowledge to dissect every viewpoint that comes their way.

Most mainstream journals in the U.S. (IT related or not) NEVER come clean when they publish misleading or false information. If they do it is buried as an obscure correction in the publication.

Mistakes are part of life, but what separates the ethical from the rest is how those mistakes are handled. This one was handled in an exemplary fashion.

Best regards,
Sean Fowler

Subject: INQ fed duff intelspinmeister info on 875P

Hi Mike,

I had a look at the Intel link on the subject page and noticed a strange thing! The link you gave was:

http://developer.intel.com/design/motherbd/bz/bz_bios.htm

and at the bottom of the page is a note that says:

Updated: Wedmesday, April 25, 2003

Not only is the spelling wrong, but the date is too. Veeeeeery starnge, I would say:-)

Andy Woudstra
Toronto

P.S. SARS - what SARS??? Come on over!

Subject: INTEL TOLD US on Tuesday that it would be a week or so before it would be able to post the BIOS download for the "anomaly"

Of COURSE it will take Intel a week to post this BIOS UPDATE - heck - Intel's people were busy last week - making all those threatening phone calls to AMD Slowpteron vendors - who apparently ignored Intel and showed up at the AMD vapour launch anyway.

Paul R. Engel

µ