Bij Got Apex? hebben ze een review in elkaar gezet van een koel Kryotech Athlon 800 systeem. Om hem te benchmarken draaide ze Quake II en 3DMark MAX met een TNT2 Ultra kaart. Voor alle details, check de review hier.

No other company offers anything close to the Kryotech at this point in time. Would we recommend this system? Well, that matters. Yes, it is one heck of a system. However, it is $2000 US in bare bones configuration, so it's not exactly cheap. So if it's for a company where labor is incredibly expensive (ie. high end 3d modeling, architecture, etc), then it will more than pay for itself in a matter of days. Also, if you're only satisfied with the latest and greatest, then this is a great way to go. Besides, when faster Athlon processors come out, you can always transplant it into this excellent platform. With the Kryotech Cool Athlon 900 shipping now at $2200 and the SuperG 1Ghz system coming soon in December, it's hard to say this is the one to get though. Also, if you're just a casual user, AMD's Athlon 700 processor is just a hair over $700 at this time, and adding a case and motherboard to that still takes you to just under a grand.

Volgens bronnen van The Register zullen de eerste (ongefixte) i820 modderplanken met twee RIMM slots eind november geïntroduceerd worden. Ook heeft The Register verzonnen dat Intel wellicht een Lite versie van de i840 server chipset gaat bouwen, de i840 staat namelijk wel sympathiek tegenover configuraties met meer dan 2 RIMMetjes:

And an Intel customer, who does not wish to be named, has now told us that he is expecting his two-RIMM mobos to arrive in late November.

Another Intel customer says that the embarrassment is even greater because its Cape Cod mobo -- which uses SDRAM is ready to roll and works perfectly, but the company will not release it until its so-called flagship mobo, the Rambus-based Vancouver, is ready to roll.

Sources close to Intel's plans add that it is under enormous pressure to ship the i820 mobo this year, otherwise it will lose enormous face both with Rambus Ink and the other six members of the Seven Dramurai.

[...] As we reported earlier this week, the i840 chipset has no such Rambus problems, and that has prompted speculation that Intel may create an i840 lite for the desktop market.

According to the slide, entitled Intel 840/820 Status Update W42.1, the i840 remains on target for W44 (that's next week, for us mere mortals on the planet earth).

ATi, fabrikant van half brakke videokaarten, heeft officiële support voor open source Linux driver ontwikkeling aangekondigd, zo lees ik in deze press release:

ATI Technologies Inc. (TSE:ATY, NASDAQ: ATYT) announced today that it is fully supporting the development of drivers for the Linux Operating System and is endorsing the open source movement by releasing 2D, 3D and multimedia programming specifications for its industry-leading RAGE graphics technology.

Recognizing the phenomenal growth and increasing popularity of Linux, ATI is committed to ensuring that the open source development community has access to technical development information on all its key components.

``ATI has made tremendous strides in the past year working with key developers on our support for Linux and we are on target to become the graphics company of choice among Linux users,'' said Henry Quan, vice president of corporate marketing, ATI. ``We will continue to provide the hardware and software support developers need to help ensure the success of Linux.''

``ATI's opening of the programming specification for their graphics chips will allow unhindered development of open source drivers that exploit the full potential of the hardware, and also gives curious developers a look inside the workings of modern 3D accelerators,'' said John Carmack, co-founder and lead programmer of id Software.

Internet2 - het supersnelle research netwerk voor amerikaanse universiteiten - heeft z'n officiële debuut gemaakt, zo bericht ABC News. Ter demonstratie werd een hoge resolutie HDTV filmpje (270Mbps) realtime van Stanford (Californië) naar Seattle geduwd:

Attendees watched a high-definition television stream from Stanford University, over California’s research network, through Internet2, through Washington state’s network and finally to UW. The broadcast was of higher quality than digital wireless transmissions. “These transmissions are production-quality video,” says Amy Philipson, one of the managers of University of Washington’s ResearchTV project. “This is the convergence that everyone’s been talking about, a real intersection of television and the Internet.”

[...] Even Internet2 can only handle 622 megabits per second, making multiple streams nearly impossible — for now. One of the biggest challenges facing researchers, according to Internet2 spokesman Greg Wood, is to increase the bandwidth available on the research network or to figure out a way for multiple users to take advantage of a single stream.

Hehe dit is zware humor: een Linux tooltje waarmee je processen kunt killen, verpakt in een Doom interfeesje:

As I was listening to Anil talk about daemons spawning processes and sysadmins killing them, I thought, "What a great user interface!" Imagine running around with a shotgun blowing away your daemons and processes, never needing to type kill -9 again.

[...] Id Software has generously released the source for Doom, which has been ported to Linux. I downloaded one of the many versions and added a few lines of code that would spawn a new soldier for each process, renice the process when it is wounded, and kill the process when it dies. [break]

Cool . Op deze site kun je 'm downloaden. Ook vind je daar een lijstje met voordelen van Doom als admin tool, zoals: [/break] - The machine load is immediately apparent to the player, who can see how crowded a room is. The player can eyeball many machines from a high vantage point and go down to a room that needs maintenance.

- A new sysadmin can be given less power by providing her with a smaller weapon. A rank beginner may not be given a weapon at all and be forced to attack processes with her bare hands. It would take a foolhardy player to attack a room full of monsters, just as a newbie should not kill a bunch of important processes. A more experienced sysadmin would have time to stop a newbie who is trying to kill the wrong process. The real work could be left to those with the big guns. The truly great sysadmins could have BFGs.

- Really crowded systems would regulate their own load because monsters occasionally kill each other. Once the population in a room goes down, the monsters will stop attacking each other.

- Sysadmins could cooperate or compete. Doom is a natural environment for player-to-player interactions. A team of players can cooperate to take care of a heavily-loaded system, or they can even take out rogue sysadmins who are killing the wrong processes.

Sjappie schrijft: "Een revolutionaire techniek om plaatjes te scalen, op deze manier kunnen nummerborden nog op de meest vage foto's herkend worden":

Shortcut Software Development is specialized in advanced graphical software solutions. The past year we have been working on an advanced digital image-scaling technology, called S-Spline. We personally feel this to be a revolutionary step forward in digital image-scaling and we are confident it will become the new standard.

We have released a demonstration program and we would very much appreciate any comments and suggestions regarding this software. Our website address is http://www.shortcut.nl/ A downloadable version of the software is available from http://www.shortcut.nl/Sspline/sspline.htm

We are looking forward to your comments. Any feedback is more than welcome. Thanks for your time,

Vincent Weberink

Hier een paar screenshots (beetje verwazigd door downscaling, check Shotcut.nl voor meer voorbeelden):

Voor alle blije tweakers en tweakerinnen die een poging hebben gewaagd om de Live!Ware 3 drivers van de Creative Site te trekken en gemerkt hebben dat dit onbegonnen werk is (je mag blij zijn met 1K per sec), heb ik een aantal snelle NL mirrors opgezet, waaronder eentje op Cameron's wonder server:

- http://cameron-diaz.student.utwente.nl/tweaker/lw30full_9x.zip

En twee Win98 FTP servers (online totdat ze vanzelf doodgaan):
- ftp://tweaker:downloads@192.87.216.41:2000/d:/downloads/lw30full_9x.zip
- ftp://tweaker:downloads@192.87.216.104:2001/downloads/lw30full_9x.zip

Mocht de force niet meewaaien, dan kun je altijd nog deze snelle FTP in Duitsland proberen.

Deadalien stuurde deze link naar een artikel van Licht.nl (afkomstig uit de Televaag), over de DVD-videorecorder die Philips afgelopen week heeft gepresenteerd:

Afgelopen week kondigde Philips aan de nieuwe dvd-videorecorder veel eerder op de markt te brengen dan tot dusver was aangenomen. Adri Baan, topman van de Philips-divisie voor consumentenelektronica, kwam tijdens de Consumer Electronics Show te Las Vegas nog met een tweede verrassing: opgenomen schijfjes kunnen op gewone dvd-spelers worden afgespeeld. De dvd-rewritables worden compatibel met de huidige spelers.

De nieuwe dvd-videorecorder gaat bij haar introductie ergens tussen 1.700 en 2.000 gulden kosten. Daarmee gaat deze recorder die een zeer hoge beeldkwaliteit koppelt aan gebruikersgemak, een rechtstreekse concurrent worden van de 'high end' vhs-recorders.

Dit keer geen behuizing in zuurstok kleurtjes, maar iets waar je wat aan hebt: dankzij de nieuwe 33MHz Dragonball chipset van Motorola zal Palm Computing begin volgend jaar een handheld met kleurenscherm kunnen releasen:

Palm has been working on a color-screen device for months and will deliver it by the first half of 2000, said Michael Mace, chief competitive officer for Palm, at the PalmSource 99 conference here today. The conference is Palm's annual meeting of hardware and software developers.

Color represents the next big thing for the handheld community. Currently, the vast majority of handhelds come with black-and-white screens, which offer low image quality that can limit viewing options. With color, the devices will move closer to desktop-PC quality.

[...] More robust internal technology is making the shift possible. Motorola announced yesterday that a new 33-MHz Dragonball chip will include support for color displays. Current Palms use a 16-MHz Dragonball chip.

Fiets richting News.com voor het complete verhaaltje (met o.a. info over het falen van WinCE).

News.com heeft een artikeltje gepost over de opening van AMD's 400 miljoen dollar kostende Fab 30 in Dresden (zie deze nieuwsposting voor de press release). Dresden zal 0,18micron koper Athlon's produceren en zal naar verwachting tegen midden 2000 de eerste 1GHz versies afleveren:

But as with all things AMD, the new factory represents a potent opportunity and a significant risk. On one hand, Fab 30 will vastly expand AMD's production capacity, long an Achilles' heel for the company, while giving it a momentary technology edge over Intel. On the other, the Dresden facility has been expensive to build and won't be cheap to run, especially worrying for a company that's lost nearly $400 million over the last three quarters.

[...] Fab 30 also will allow AMD to market a product the industry leader will not have until 2001. All of the Athlons from the Dresden facility will be based around copper chip technology from Motorola.

Over time, copper is expected to enhance processor performance because the metal conducts electricity better than aluminum. Under current 0.18-micron manufacturing techniques, the difference between the metals is fairly negligible, according to some analysts. But copper advocates disagree, arguing that, if anything, moving to copper now will provide AMD a valuable learning experience.

Ace's Hardware heeft professionele 3D benchmarks van de GeForce en een aantal Unix Workstations bij elkaar geharkt:

Ok, take this as a FUD to the unix workstations market. However with the recent introduction of the Geforce256, for the first time in history x86 becomes a platform which doesn't suffer from second rate performance, while still offers its primary advantage, price. From benchmarks such as viewperf and indy3d, the K7 600+ and Geforce combo is challenged by very few unix/risc based workstations. The price for a setup like this due to the recent hike in memory will likely run about $2000 to $2500, counting the Geforce at around $300 and with 256MB+ memory. So here are a look at some geforce indy3d and viewperf scores compare to the highest official scores on sense8 and spec.org.

De simpele Athlon / GeForce combo blijkt in veel gevallen sneller dan de professionele Unix workstations (met een 10 maal zo hoge prijs). Check de frontpage van Ace's Hardware de resultaten.

AMD heeft z'n nieuwe fabriekje in Dresden (0,18micron, koper) officieel geopend. Hier een stukje uit de press release over deze naar eigen zeggen meest geavanceerde fab ter wereld:

AMD today dedicated the world’s most advanced wafer fabrication facility in Dresden, Germany. The grand opening of the company’s Fab 30 marked the successful completion of a project which began with groundbreaking in October 1996. The state-of-the-art fab is being readied for volume production of leading-edge microprocessors for Microsoft Windows computing. Fab 30 activities are currently focused on characterization and qualification of the facility. The company expects to begin production of AMD Athlon™ processors featuring copper interconnect technology later in the current quarter with first revenue shipments planned for the second quarter of next year. Investment in Fab 30 will total $1.9 billion, approximately DM 3.2 billion. AMD’s investment ranks as one of the largest foreign investments in East Germany.

[...] Sanders noted that a pre-production version of the AMD Athlon processor built using the technology being installed in Dresden achieved speeds in excess of 900 megahertz.

The Street, een of andere site waar je voor moet dokken (dat suckt), heeft een nogal onheilspellend artikeltje gepost over de Rambus problemen met de i820 chipset...

Slowly, the toll of the Intel-Rambus fiasco is becoming clear. Executives in the computer-memory industry told TheStreet.com they now believe the Camino problem is so intractable that Rambus technology may never work, at least not in any way that would make Rambus chips marketable or profitable. The problem goes beyond dollars and cents; many in the semiconductor industry are critical of how Intel has effectively forced Rambus technology on the marketplace. The cost of a Rambus failure to the computer industry, they say, could reach billions of dollars in potentially wasted inventory and development and lost revenues.

[...] Industry sources say the quick fixes for Rambus that Intel is likely looking at will either result in a slower system or limit the amount of memory in a PC because they will reduce the number of memory slots. And while Rambus chips might be usable for these chipsets, many say the motherboards made might still have to be scrapped.

Logistically, it would be impossible to produce new DRAMs this quarter even if Intel can launch its chipset now, executives at four memory chipmakers say. Outside of Intel, all Rambus DRAM production has stopped, they say. "A DRAM takes four months to make," says Jim Sogas, director of the DRAM division at Hitachi (HIT:NYSE ADR). "At the very, very minimum it would be 90 days."

Om het artikel te kunnen lezen heb je een trial login nodig: Your username is: blaat / Your password is: demand184 .

EETimes heeft een artikeltje neer gemikt met info over de plannen van chipset bakkers VIA, SiS en ALi. VIA, de belangrijkste van deze drie, verwacht tegen 2001 de grootste x86 chipset supplier te zijn. Verder ziet VIA opperhoofd Chen meer heil in Double Data Rate SDRAM dan in Rambus:

Via also is expanding beyond PC133 chipsets, and plans to support double-data-rate SDRAM “whenever the market wants it,” Chen said. Via will not support Direct Rambus DRAM, however. “DDR is going to be a far bigger selling product than Rambus,” Chen said.

Intel has said it is considering producing its own chipset to support DDR for servers, and once again Via believes the two companies can complement each other in the market.

“We have already helped Intel in the 440BX [chipset] shortage,” Chen said, claiming that without Via's compat- ible chipset line as an alternate source, Intel's microprocessor sales could well have suffered.

iXBT Labs heeft een review gepost van de PIII-600E (100MHz FSB) en PIII-600EB (133MHz FSB), de eerste met betrouwbare Coppermine benchmarks. Uit de tests (op een VIA plank met PC133 SDRAM) blijkt dat de Coppermine in Winstone net iets sneller is dan een gelijk geklokte Athlon. In situaties waar FPU performance belangrijk is, zoals Quake2/3, blijft de Athlon ver voor lopen, zolang de videokaart niet tegen z'n fillrate limit zit. De conclusie van iXBT:

Well, with the help of its Coppermine Intel managed to catch up with AMD Athlon almost bloodlessly. It had just to shift to a new 0.18 micron technology, which helped to limit the manufacturing expenses and to obtain higher working frequencies. No special architectural changes were needed. However, now we all know that Intel will hardly create anything faster and newer for Slot 1. The next mainstream processor based on Willamette core will be designed for an absolutely different processor socket. At the same time AMD's Athlon still has room for improvement. But even manufactured with 0.25 micron technology Athlon proves not a little worse than 0.18 micron Intel Pentium III. So, the last word belongs to AMD, which can easily leave Intel behind only with its higher clock frequencies.

Alles wat AMD nu nodig heeft is de switch naar 133MHz SDRAM (of VC DDR, nog leuker). De Coppermine's (die volgende week officieel gereleased worden) zijn inmiddels ook onze Pricewatch binnen gewandeld (en ze zijn niet goedkoop).

JC heeft een interessante posting over het Willamette nieuwsartikel van The Register op z'n site geplakt:

Register put up a very interesting bit here. It's about a surprise Willamette introduction in February of 2000 ("paper launch" in December, chip actually appearing two months later, according to the article). I passed this by despite the fact that a good ten percent (slight exaggeration, but you get the idea) of y'all emailed the URL to me. It just doesn't seem likely, considering the design, to our collective knowledge, hasn't taped out (and if it did, it was likely recently). Takes about a year from tapeout to production. You do the math. However, as I said, I wasn't going to put up a link to it, but I just realized something (thanks to Jocelyn Fournier, I think, for nudging me in this direction). The specint95 score of the P7-1100 shown at that register article is utter crap. If it is really the case that it is that slow, then Willamette will be pretty pathetic for servers, especially if you consider the 1MB on-die L2. The quoted score is 43 at 1100MHz.

By my guesstimations (with the help of idiot from Ace's), an Athlon at 1100MHz would score between 50 and 55 (perhaps subtract a point or two for dropoff from linearity), depending on whether or not you optimize for prefetching. This means that Athlon pastes these alleged Willamette scores in specint. Actually, from the look of it, given Intel's Coppermine presentation at µPF, it seems that Coppermine is also faster than Willamette in specint. I didn't check at all with the Winstone score, but as you can see, if Register's data is true, then it isn't really great news for Intel. I don't know about you, but I'll prefer to believe the more reasonable assumption that Willamette will come out in 2000 Q4 (or 2001 Q1) but will be totally rippin' in performance. [break] En later: [/break] Registerama: If you look hard, you'll see that Register has (as of this writing) unpublished that "Willamette 1100MHz" story that caused such a ruckus yesterday (the file is still there, but it is no longer linked to on their page).

Het artikeltje (van 18 oktober) staat nog wel in de archieven, maar is van de frontpage (tot 17 oktober) verwijdert.

Hmm, Steve staat strak van de coke? . Nu heeft-ie voor de vierde keer binnen een week z'n policies aangaande de G4 orders aangepast. Zoals het er op dit moment voor staat wordt de hele downgrade-operatie teruggedraaid voor openstaande (US) orders:

And now we have Jobs saying, yes, we can honour existing orders after all. Go figure... So, all orders submitted before 13 October "will be honoured at the originally quoted prices, including those placed with the company's resellers", Apple said.

What brought about the change? According to reports, not only further complaints from understandably indignant customers, but the threat of a number of class action lawsuits.

"We apologise to our customers for upsetting and disappointing them during this past week," said Jobs. "Our actions today will hopefully set things right."

Meanwhile, British buyers appear to have been left out in the cold. An employee of one London-based reseller told Web site MacFixIt: "As of close of day on Friday, it would seem the policy on old orders has only been changed in the US. We could get no information out of Apple regarding if the change would affect the UK. We certainly weren't phoned to say reinstate your old orders."

Check The Register voor het volledige nieuwsbericht.

insane schrijft: "Automatiserings Gids bericht ons dat M$ dit jaar forse winst heeft gemaakt. Ondanks al die tegenslagen(?) maakt onze oom Bill nog steeds veel winst":

Als gevolg van een "gigantische vraag naar PC's", zoals Microsoft dat omschrijft, is de kwartaalwinst van de onderneming gestegen met 30 procent tot een bedrag van 2,19 miljard dollar. De cijfers hebben betrekking op het eerste kwartaal van het lopende boekjaar 2000.

De omzet in het kwartaal steeg met 28 procent tot 5,38 miljard dollar. De resultaten komen als een verrassing, vooral gezien de sombere berichten die Microsoft zelf in het afgelopen jaar heeft verspreid. De firma was bang dat de aanschaf van nieuwe apparatuur en bijbehorende software zou worden uitgesteld tot na de eeuwwisseling. Dit blijkt dus niet zo te zijn.

Bernie schrijft: "eindelijk heeft Asus op de site de V6600 aangekondigd, ik heb alleen de beloofde 64Mb versie nog niet kunnen vinden":

Dat is 'm dus. Een overzicht met de features, specs en andere dingen die we al lang weten vind je hier.

Neelix schrijft: "Eindelijk is er een link verschenen om Live!Ware 3 te kunnen downloaden. Alleen is die server nu naar de klote en op de FTP heb ik het nog niet gevonden. Afwachten dus maar....".

Ik krijg inderdaad ook zo'n lekkere Microsoft ODBC error. Ondertussen mogen we ons vermaken met de tekst op de Creative Live!Ware site:

Live!Ware 3.0 CD is now available for purchase online. The CD version is onsale in several countries (soon available in more) and includes extras that will not be available for download for the Web version. Here is the list of software content that is exclusive to the Live!Ware 3.0 CD :

- Creative Digital Audio Center (Lite Version)
- Creative Wavestudio 4
- Creative Recorder
- Soft Karaoke
- Sound Blaster Live! Experience
- MiniDisc Center
- EMU General MIDI banks
- Demo MIDI clips
- LAVA! MV3 samples
- MP3 clips
- Standard 30 SoundFont banks
- 25-50 SoundFont banks (>1000 instruments)

DaNose schrijft: "Ook Tresh & gang hebben een poging gedaan om zoveel mogelijk performance uit de Geforce te persen. Ik vind de resultaten eerlijk gezegd een beetje tegen vallen, alleen bij extreem hoge resoluties is er een verbetering te zien":

We saw that most of the overclocking only yielded small performance gains. In Quake 3, gains averaged around %10. The gains were higher in the high resolution, high quality settings, but the low frame rates at those high settings make the gains almost useless. We did see the overclocked SDRAM board hold its own against the DDR board in Quake 2, but we have to wonder how much frame rate improvement we would see if we could get the engine core higher than 130MHz. We'll probably see cards that can handle higher speeds as NVIDIA improves GeForce 256 yields, but for now we can only wait.

We originally thought the boards were stable at 135MHz, but extended use proved us wrong. Sounds like we need to operate on the GeForce's heatsink and fan solution, but we'll have to make up a plausible explanation for Mr.Perez first. "Umm.. Freak accident. We accidentally dropped it. The heatsink flew off and the card slid across the concrete, finely grinding down the top of the chip, exposing the copper plate. We used thermal compound to cover the wounds."

In het meest ideale geval (Quake III met high quality settings) is het verschil tussen de reference DDR kaart en de overgeklokte kaart op 135/340MHz 22%, da's zeker niet slecht. Check de resultaten bij Firingsquad. Lees voor meer info ook het GeForce overclocking verhaaltje van Commander Tom.

mikato schrijft: "Door The Register al eerder in de wereld geholpen, maar nu ook op Computable in plain ol dutch":

Intel komt naar verluidt in december al met zijn 32-bit processorreeks Willamette, die een kloksnelheid van 1100 MHz moet halen. Het vervroegen van de lanceringsdatum met maar liefst negen maanden is een reactie op de MHz-voorsprong die concurrent AMD momenteel heeft met zijn Athlon-processor (K7). Volgens berichten in Amerikaanse media zou Intel

Willamette in december al aankondigen, om de processor twee maanden later in PC's te hebben. Die tussenperiode moet dienen om computerfabrikanten de tijd te geven hun voorraad systemen met Coppermine-processoren weg te werken. De Coppermine wordt volgende week officieel onthuld, maar gebruikt dezelfde P6-kern die in de Pentium Pro-chip al zijn debuut maakte.

Hmm, blijkbaar toch iemand die iets met dat wazige nieuwsberichtje van The Register doet. De 'serieuze' nieuws sites zoals News.com, Techweb en EETimes hebben nog niet bericht over een mogelijk vervroegde release van de Willamette. Op de AMD Roadshow gisteren sprak men nog steeds over een delay naar 2001.

hjs schrijft: "kwam bij Voodoo Extreme dit lijntje tegen :

Ok, so you won't need to grab those unofficial v3.52 Detonator based drivers after all, Ed Downing sends word that nVidia has posted new Detonator v3.53 on their site. "

Detonator driver voor Win9x, klik hier. Voor NT gebruikers, download ze hier. BDW, naast de GeForce 256 kun je deze drivers ook gewoon voor TNT/TNT2, Vanta en M64 kaarten gebruiken.

hjs schrijft: "Onze vriend tom heeft het nu getest in hoevere de nVidia GeForce is over te klokken":

Het blijkt dat de "gewone" GeForce met SDRam met standaard core/memory clock van 120/166 Mhz is over te klokken naar 145/183.

The findings in this overclocking article have more of a scientific than a practical value to most of us. The gains achieved by overclocking GeForce256 are rather little, mainly due to the fact that the possible increase of the clock rates is rather small as well. It doesn't come as a surprise that overclocking the memory of a SDR-GeForce will help in many cases, especially at high resolutions. Cranking up the core clock makes most sense on DDR-boards.

[break]En een stukje verderop:[/break]

Especially in Quake3 at the high quality setting the increase in memory clock translated into an increase in frame rate at the same percentage. This shows that even DDR-memory cannot quite compensate the shortcomings of GeForce's memory interface. I was told by NVIDIA that a 256-bit wide memory interface was close to unrealizable, but I am sure that this is what it will take for future 3D-chips. The only alternative could be the much beloved RDRAM.

Volg deze link voor het hele artikel.

Ik kreeg van roadrunner deze link naar Deeztech, waar men een aantal gruwelijke experimenten op een K6-2 400 heeft los gelaten .

Ever wonder what an AMD K6-2 looked like without that little metal cover over the center of the CPU? Me too, and that means one of a couple of things. Either we all have way too much time on our hands, or we are some sick twisted individuals. Now here is the kicker, I haven't seen this done on any other site to date. I could be blind, and just have passed it over but I think I would have remembered something like that. I mean why hasn't anyone done it before now? Everyone is ripping puny heatsinks off Riva TNT boards and such with wild abandon.

Na operatie draaide dit slachtoffer stabiel op 504MHz, waar de arme stakker eerst niet verder kwam dan 448MHz.

Diederik is een beetje off-topic: "Heerlijk deze nieuwe wereld . De thedigitalbits.com heeft screenshots van de cover en disk. Merkwaardig dat er NTSC op de DVD staat, dat geldt namelijk niet meer...mmmmm.":

De 'eerste Star Wars, The Phantom Menace'-bootlegs op DVD zijn verschenen in Azië. Bij The Digital Bits kregen ze een coverscan toegestuurd.

Epic heeft een nieuwe versie van de kont schoppende Unreal Tournament Demo gereleased:

This version will be the last demo version avaiable before UT appears on store shelves. This version is NOT network compatible with previous versions, so we encourage everyone to upgrade. It doesn't contain any new content - only performance improvements and bug fixes. ngWorldStats will also be reset for this demo release.

- Download links are available here. More mirror sites will be added soon!
- There is a list of changes since the last demo.
- Be sure to check out the list of Known Issues as well. [break] Hier een paar punten uit die 25 kilometer lange changelog: [/break] Performance
- Networking bandwidth and server CPU use improved
- A.I. performance improvements and tweaks
- Replicate all player skins in use to clients when joining to avoid hitch when first see new player
- Performance improvements and cleanup of HUD, messaging, and console
- Precache left hand weapon meshes if using left hand.
- Improved dynamic detail dropping to maintain minframerate
- store list BSP nodes for decals to greatly improve decal clipping speed
- sort decals by texture to improve performance
- increased characters per font page for default font to increase performance

AI
- Improved bot AI for impact hammer
- Added variety of playing styles to default practice bots
- Improved auto adjust bot skills, and skills now remembered from level to level
- Bots use rocketlauncher locking on more
- Bots have adjustable strafing ability

Gameplay
- Game time now unaffected by timedilation
- Kill tossed weapons faster
- Improved/louder gib sounds
- Fixed switch to best weapon twice and then its always pending bug
- Improved reduced HUD (with just frags, ammo, health, and armor) - shows name of weapon switched to if weapon is hidden [break] Voodoo Extreme heeft een lijstje met download mirrors: [/break]
Mirrors:
Sharky Extreme
MN Gamers
Download the Demo from AVault
Download the Patch from AVault
Download at Stomped.com
Download the demo from FilePlanet
Download the Patch from FilePlanet
Download from Game Rush
Download from 3DFiles

De files staan inmiddels ook op de waanzinnig (vooral voor Surfnetters) snelle FTP van Niels (Cameron Diaz) bij de UTwente.
http://cameron-diaz.student.utwente.nl/tweaker/utdemo348.exe
http://cameron-diaz.student.utwente.nl/tweaker/UtDemoPatch348.exe

Gert schrijft: "Bij Riva3d gejat. 3dfx probeert nu op alle manieren winst te maken. ":

Hmm, oude fossiel uit de STB tijd, met prehistorische Riva128ZX chipset. "Opens up a virtually unlimited pathway to visual realism and high performance", spreekt van grote bewondering voor de concurrent .