crazyhorse bracht een link richting Overclockers Workbench, waar men een tabelletje met de specs van de VIA Samuel heeft gepost. Deze proc is gebaseerd op de WinChip C5 core en zal geproduceerd worden op een 0,15 of 0,18micron procédé. Naast 128KB L1 cache is er 64KB full-speed on-die L2 cache aanwezig. De release van de VIA Samuel vindt plaats in het derde kwartaal, tegen kloksnelheden van 500 tot 700MHz:

3dfx heeft een setje nieuwe Win9x beta drivers gereleased voor de Voodoo3 2000/3000 videokaartjes. Hieronder de info voor versie 1.04.04:

What's new in 1.04.04 BETA

1. Please download and view the FAQ below for more info about this and future BETA drivers
2. Fixed Midtown Madness crashes out during races
3. Fixed exiting Glide Unreal

Download ze bij 3dfx Gamers.

JeroenB schrijft: "Je kan zeggen wat je wil van Microsoft, maar deze keer zijn ze wel HEEL innovatief bezig ":

Persoonlijk vind ik dit ontwerp toch wel een tikkeltje extremer dan de 'open' designs van Apple .

Japan Terror Corps (aka Akiba's PC Watch) wist een Playstation II te bemachtigen en sloopte het ongelukkige apparaat direkt open om de insides aan een nadere inspectie te onderwerpen. PS2 open en bloot:

Playstation II moederplank

PS2 Rambus chippies

PS2 controller

Sony Emotion Engine

PS2 kast

Verder heeft wazige japanse site ook schema's van de Playstation II architectuur:

ZDNet heeft een interview gepost met Shawn Fanning, de schepper van het door velen zeer gewaardeerde programmaatje Napster. Recentelijk kwam Napster diverse keren in het nieuws toen een aantal universiteiten besloten om het proggie van hun netwerk te bannen. Het enthousiaste genapster van de studenten bleek een nogal extreme hoeveelheid traffic te genereren:

ZDNet: Has bandwidth usage become a top priority? Schools have banned the use of Napster due to the amount of bandwidth it uses.

Shawn Fanning: It's interesting because a long time ago, before there was even a product, another founder, Shawn Parker, who's 19... he and I had been conversing on the Internet. We met through the Internet security community. He said this has the potential to be the first broadband killer application, and it has sort of become the truth because obviously it's so bandwidth intensive. I mean, it has been an issue. It's been all over the press, and schools are deciding to shut it down. We released a public announcement saying that we had some potential solutions that we were working with, and once we had something concrete we would start implementing it and approach schools. So it's definitely an issue, but it's something that we're working on. Hopefully we'll have a solution soon to present to both universities and corporations, because its popularity has just been incredible. It spread through universities and companies very quickly. So the bandwidth issue is definitely a big concern of ours.

Paul DeMone van RealWorld Technologies heeft part I van zijn Intel Willamette artikel online gemikt. Dit eerste deel gaat voornamelijk over de geschiedenis van Willy en zijn welbekende voorganger; de P6 core die gebruikt wordt in de Pentium Pro, Pentium II en Pentium III en Celeron. Verder wordt de architectuur van de Willamette core globaal besproken:

Willamette is a very deeply pipelined processor. It uses 20 pipe stages to execute integer instructions including the 4 pipe stages associated with fetching uops from the trace cache. If you included the pipeline stages associated with fetching x86 code from the L2 cache, decoding it into uops, and loading uops and program mapping/flow information into the trace cache the total number of pipelines stages probably approaches 30 or more.

The branch misprediction penalty appears to be at least 19 clock cycles when the correct path is present in the trace cache. If the trace cache misses, then the branch mispredict penalty is considerably higher. This compares to a minimum branch mispredict penalty of 11 cycles for the P6 core. The P6 uses the two-level Yeh and Patt adaptive branch prediction scheme. Despite the fact that the P6 predicted branches correctly around 90% of the time it still lost about 30% of its potential performance due to branch mispredicts. Although the Willamette will no doubt use more modern branch prediction techniques like gshare and dynamic prediction strategy selection, its huge mispredict penalty will make its performance very sensitive to the efficacy of its branch prediction algorithm(s) on the particular code being run.

AMD Zone schrijft in een artikel over de 1GHz Athlon dat Compaq binnenkort een 1GHz Athlon in haar assortiment neemt. Volgens de bron van AMD Zone komt Compaq in verband met beperkte supplies niet met een 1GHz Pentium III:

Compaq will be introducing computers with 1GHz processors and allow consumers to custom build their computers AMD’s new CPU. Other OEMs are expected to announce 1GHz systems as well. Compaq will not be offering a 1GHz Coppermine due to the short supply, the source said [break] In een ander nieuwsbericht meldt Yahoo dat Gateway morgen met een 1GHz Athlon komt. Het systeem moet 2.999 dollar gaan kosten: [/break] Personal computer makers that use AMD Athlon chips are expected to announce systems using the new processors Monday, the Chronicle said. A source told the newspaper that Gateway will sell its 1-GHz systems starting at $2,999.

Nieuwe DPC stats van Marc Heida:

Hier zijn ze dan, de stats van 4 maart. Alle data van 3 maart zijn ook verwerkt , maar daar is geen overzicht van .

De DPC Hitparade Top 10 van 4 maart 2000

1. 47,293 Interior Lulu(26, Totaal:311,341)
2. 8,597 Ferion Power!(18, Totaal:393,936)
3. 6,071 The Vinz(467, Totaal:11,530)
4. 6,019 Bugsie(75, Totaal:140,150)
5. 5,571 DPC Division Asten/Someren(176, Totaal:59,257)
6. 4,646 BRILSMURF(183, Totaal:56,159)
7. 4,139 sienema(2, Totaal:1,069,707)
8. 3,749 Chainsaw(16, Totaal:404,537)
9. 3,688 HarryB(13, Totaal:515,264)
10. 3,679 jgroot(82, Totaal:129,835)

En de nummers 11 tot en met 20:

11. 3,412 Eddict(27, Totaal:308,262)
12. 3,008 Nasty(9, Totaal:625,146)
13. 2,947 Sueno(45, Totaal:209,745)
14. 2,868 Liam_xi(609, Totaal:6,367)
15. 2,656 j.deij(156, Totaal:71,338)
16. 2,580 Sir Soes & Shellshock(74, Totaal:142,124)
17. 2,577 Richard.Oostindie(158, Totaal:70,083)
18. 2,479 2fast4u(33, Totaal:261,415)
19. 2,210 FalconM(58, Totaal:172,724)
20. 2,152 nicolai(35, Totaal:240,187)

De Top 15 teams van 4 maart 2000

1.( 1) 664,130 Slashdot.org
2.( 2) 610,537 #DeathMatch
3.( 0) 596,912 AnandTech
4.( 3) 538,023 Russian Team
5.( 0) 412,706 HackZone Team -- Combined power of Russia and exUSSR countries
6.( 0) 380,755 Team Warped (OS/2)
7.( 0) 369,147 The Amiga RC5 Team effort
8.( 0) 366,911 Team EvangeLista (Macs Rule! :-)
9.( 3) 271,733 Japan FreeBSD Users Group
10.( 1) 258,283 Dutch power cows
11.( 2) 257,055 Libertarians for Privacy (Freedom Works!)
12.( 1) 237,133 Japan Windows NT Users Group
13.( 3) 194,777 L0phT Heavy Industries
14.( 0) 187,684 Team Hungary STB+
15.( 1) 187,496 CCC.de
58.( 12) 55,920 The #koffie Team

Teams om ons heen (total stats):
27.( 1) TeamUFies (All Hail the Dust Puppy!) (Tot:50,956,897, Gis:148,825)
28.( 1) Prof. James Challis' Most Excellent UK Team (Tot:50,921,410, Gis:97,692)
29.( ) AL-Mail Users (Tot:48,851,503, Gis:78,965)
30.( ) Team RimArts UsersGroup (Tot:47,475,040, Gis:71,056)
31.( ) Dutch power cows (Tot:44,315,223, Gis:258,283)
32.( ) NetBSD Users (Tot:42,426,830, Gis:79,046)
33.( ) SwissTeam (Tot:38,998,644, Gis:82,112)
34.( ) QUALCOMM Incorporated (Tot:38,396,928)
35.( ) Michigan State University (MSU Spartans) (Tot:37,943,529, Gis:103,382)
(Teams waarvan de score van gisteren ontbreekt, hadden geen score in de top 100!)

Wanneer halen we ze in:
TeamUFies (All Hail the Dust Puppy!): 61 dagen
Prof. James Challis' Most Excellent UK Team: 42 dagen
AL-Mail Users: 26 dagen
Team RimArts UsersGroup: 17 dagen

Dutch power cows is vandaag 10e en heeft een kans van een op 130 om de sleutel te vinden!

Abdul attendeerde mij in dit nieuwsartikel over de Sony Emotion Engine op dit plaatje van de Sony PS2 moederplaat:

De mannen van iXBT-Hardware hebben een uitgebreide review online gemikt van Asus' nieuwe mammaplank gebaseerd op de VIA Apollo Pro 133A chipset, de P3V4X. Tijdens de benchmarks werd duidelijk dat de Asus P3V4X totaal niet overweg kon met nieuwere VIA drivers, dit resulteerde namelijk in een performance loss van 30% in Quake III. Pas nadat er werd geswitcht naar VIA Service Pack 4.17 werd dit euvel verholpen. Het bordje bleek echter wel zeer stabiel en biedt ook veel mogelijkheden voor overklokkers:

In terms of overclockability, a really significant advantage of ASUS P3V4X compared to all other mainboards on VIA Apollo Pro133A is the possibility to change the AGP frequency divider manually. However, you can switch between 1, 1/2 and 2/3 with dips, which is active only when you configure your CPU in Jumper mode. There is simply no setting in BIOS Setup equivalent to this. That is why if you use your ASUS P3V4X in jumperfree mode, the AGP divider is locked and equals 1 in the interval between 66-83MHz, 2/3 - between 85-120MHz and 1/2 - between 124-166MHz FSB. So, if you wish to be able to change the AGP divider - use dip-switches.

And now a few words about the practical results. For overclocking we used the same poor Intel Pentium III 500E FC-PGA, which we overclocked on all the previously tested boards. Unfortunately, ASUS P3V4X proved not very nice here. As you probably remember, we managed to get the maximum of 775MHz (5x155MHz) on Chaintech 6ATA4. As for P3V4X, it showed comparatively poor results: we failed to surpass 135MHz FSB frequency (the CPU reached only 675MHz). That is why we don't have the heart to call it an overclocker's choice.

Elliott Lee Hazen van AnandTech heeft een review geschreven van de Shuttle AV64 Apollo Pro 133A moederplank met 1x AGP, 5x PCI en 2x ISA (geen AMR). De 8 verschillende bussnelheden kunnen middels het Award 6.00 PG bios ingesteld worden.

While not an ideal for the extreme overclocker, Shuttle has succeeded in making a competitive VIA 133A motherboard. There are not as many FSB speeds as competing boards offer but by allowing voltage core adjustment and an adequate number of FSB speeds for 133 MHz chips, this board should appease the average overclocker. The stability of the board was excellent meaning you can rely on this board to run smoothly. The AV64 provides a quality Apollo Pro 133A solution for almost anyone's motherboard needs.

Tenchusatsu heeft op de Silicon Investor forums een errug handige vergelijking met de voor- en nadelen van Direct Rambus DRAM en DDR-SDRAM in mekaar gemikt:

RDRAM

+ Low number of pins: Excellent for chipsets, less expensive north bridge design, also excellent for having two or four RDRAM controllers in parallel
+ Finer granularity: As memory density increases, fewer chips can be used.
+ Smaller packaging: (Or so I'm told; I can't be sure.)
+ Efficient in bandwidth utilization.
+ Great for systems which don't need high capacities (gigabytes) of memory.
- High prices: Hopefully, this should come down as RIMM production ramps up.
- Huge transition hump from SDRAM to RDRAM production.
- Lower max bandwidth compared to DDR, but efficiency makes up for it.
- Slightly longer latency: For desktops, this can have a noticeable impact on performance, though I believe the chipset makes a much bigger difference on latency than RDRAM itself.

DDR SDRAM

+ Higher capacity limit per channel: RDRAM only allows for 32 chips per channel, but with double-sided, double-stacked DIMMs, SDRAM allows for 128 chips per channel. This leads to 4x the max capacity.
+ New density technologies (256 Mbit coming up, 512 Mbit after that) will probably show up on SDRAM before RDRAM.
+ Easier to implement ECC and chipkill features than on RDRAM. That's critical for high RAS in servers.
+ Higher max bandwidth than RDRAM.
+ Slightly lower latency than RDRAM.
+ Likely cheaper to manufacture.
- Pin count requirements are significant. I think it's 2 to 3 times that of RDRAM.
- Because of pin count, it's harder to group two or more DDR channels in parallel.
- Possible electrical concerns: Despite the difficulties implementing RDRAM, implementing DDR isn't going to be easy.
- Coarser granularity than RDRAM. You can have fewer than eight chips on a DIMM, but the packaging will be larger and more expensive per chip.
- Less efficient at utilizing bandwidth, which negates DDR's max bandwidth advantage over RDRAM.

GamePC heeft een 21" monitor roundup neer geknalt met een vergelijking van de Sony G500, Sony F500R, Hitachi 814 en Mitsubishi 2040u. Laatstgenoemde werd als beste bevonden:

So after all that, we are leaning towards Mitsubishi's 2040u as being the best of the group. Considering what you get with this monitor and its suggested retail price, it is basically a steal. The 2040u is totally backed by Mitsubishi, meaning any kind of support will be provided within the three years of purchasing. And let me tell you, their support for their products is no joke. We couldn't have asked for a better monitor 21-inch then the 2040u. Even though the other three monitors are basically top of the line, best you can buy, etc… there are just some little things that need to be adjusted or changed to make a better impression to the consumer.

Things like the F500R having that displaying problem. Sony has been known for its quality and just shocks us that their quality control let a problem like that slip by. We are sure this is not a normal thing for Sony, but the price tag on the F500R is a little steep. It's probably not going finding its way in a home environment any time soon. The G500 on the other hand was perfect.

De popi Iiyama Vision Master Pro 510, die ik zelf graag in een roundup zou willen zien, werd helaas niet getest.

Stomped.com heeft een interview gepost met een manneke van Dell Computer. Twee week geleden maakte Dell bekend dat het een exclusieve GeForce versie met 64MB DDR geheugen bij haar high-end game computers gaat leveren. Uit benchmarks van MaximumPC blijkt dat deze 64Meg kaarten een behoorlijk stukkie sneller zijn dan normale GeForce DDR-en waar wij arme tweakers het mee moeten doen. De Dell GeForce is overigens niet in de losse verkoop te krijgen dus je zult er een systeem omheen moeten kopen (of je moet op het foute pad gaan ):

JCal: How much input did Dell have in the creation of this new video card? Did the company engineers help nVidia with designing the card?

Zucker: We actually worked with NVidia from the start. We have a large staff of multimedia engineers that work on various audio, graphics, and video projects. The original GeForce board was a co-development effort between NVidia and Dell. We were involved with the schematics and layout and helped NVidia create what is considered the reference design for the board. This board is then sent out to other parties who then use it to create their own designs. We actually wind up assisting the entire industry with development but feel it is the right thing to do for quality and reliability reasons. We take this reference design and have it manufactured with our supervision with manufacturing testing being performed on actual Dell machines.

Een dag na de release van CloneCD 2.0.1.1 brengt Amphora ons het nieuws dat versie 2.0.2.1 van dit handige bakprogramma al weer uit is:

Version 2.0.2.1, 04Mar2000 Compatibility fixes:

Fixed hang / crash on startup of Windows 95/98 under certain configurations (more than one other CD-Writing Software installed together with Toshiba DVD-ROM drive)

Blocked incompatible Firmware Versions 1.01 to 1.03 of Plextor PX-W124TS, so people can't create coasters

CloneCD (1,3MB) kun je hier naar beneden hengelen.

Hot Hardware hoopt ons blij te kunnen maken met deze review van de Thermaltake Golden Orb, een heatsink in de vorm van de HP Polarlogic. In vergelijking met de standaard Intel Celeron cooler ging de CPU temperatuur met de Golden Orb ongeveer 11 graden Fahrenheit omlaag:

As we all know, there is only one thing that matters when is comes to what really makes a good Heat Sink, performance. Looks, style and design all take a back seat to how cool a Heat Sink will keep your processor. We don't have a lot of data in terms of how the Orb does vrs. the rest of the world's Heat Sinks that are available on the market. What we decided to do was compare the Golden Orb against Intel's original factory stock heat sink that comes with the new Coppermine processors. The Intel Sinks are no slouches either. They have decent fans, good size but not so good retaining clips.

[...] Whether you are a hard core overclocker or not, the Thermaltake Golden Orb packs an unbelievable amount of cooling into an already great design. This shows a 12% improvement at idle and 10% improvement at load, with the Golden Orb.

BDW: deze vergelijking in percentages is uiteraard nogal onzinnig bij temperaturen.

CDRW Central schreef een review van de 4-speed Teac CD-W54E IDE rewriter. Hier een stukkie:

D to CD copy at 4x and 2x was a dream, we burnt over 10 CD-R's on the Teac, all being a mixture of audio and data CD's as the source, even as image files from the hard drive. During these tests, we ran other apps to see if we could get the Teac CD-W54E to buffer underrun, we even toured the Internet while it was burning data from the CD-ROM source and the hard drive, not once did it fail. Now 16 - 18 months ago, we would have had trouble to do this with any CD Recorder, but today...... well !! When burning a full CD, it can take up to 20 minutes to burn at 4x speed, so you will have to be a little patient while this goes on, and twice as long at 2x speed.

PS2K.net wist een Playstation II te bemachtigen en sloopte het apparaat meteen open om de ingewanden te fotograferen:

Below we have photos of what is hidden inside the PlayStation 2 hardware, including the DualShock2 and the 8MB Memory Card.

The motherboard is nice and tidy, but notice tha size of the main chips! The 'Emotion Engine' is said to be the biggest processor ever mass-produced. It seems like it's the truth! Enjoy these photos from a Core Magazine reporter in Japan:

Mocht je geïnteresseerd zijn in screenshots van PS2 spellen, dan kun je de PS2 coverage van Gaming Age checken. Onderstaande screenies zijn afkomstig van Ridge Racer V en Tekken Tag Tournament:

Thanks denhis voor de link.

Gamespot heeft een review gepost van de Adaptec 29160 Ultra 160 SCSI controller. De 29160 kan gebruik maken van 64-bit PCI slots, hoewel de adapter ook in normale 32-bit slots werkt (64-bit PCI is echter wel vereist om de maximale bandbreedte van de 29160 te benutten, deze is bij 32-bit 33MHz PCI namelijk beperkt tot 132MB/s). In vergelijkende WinMark benchmarks met een Quantum Atlas 10K bleek de 29160 sneller te zijn dan de 2940U2W:

The Business Disk Winmark's score was about 6.8 percent better using the 29160. This is statistically significant, as it's above the 3 percent margin of error. The High End Disk Winmark came in at a whopping 15 percent higher, which was pretty startling. Remember, we're using the same drive, CPU, and memory, only varying the SCSI host adapter. Also interesting was the fact that the CPU utilization dropped a bit, which indicates that the Adaptec controller chip is more efficient. When you check out the actual, measured transfer rates, these don't change at all. This reinforces the conclusion that there's an overall efficiency increase and that the Atlas 10K is already moving data as fast as the controller will accept it.

De Playstation 2 heeft gisteren zijn debuut gemaakt in Japan, met als gevolg dat winkels aldaar met duizenden PS2-seekers geflood werden. PS2K.net en PS2.IGN.com maakten een impressie van de hectische introductiedag:

After some waiting, stores in Akihabara opened their doors as early as 7AM this morning in order to accommodate tens of thousands of awaiting customers. The media presences were enormous, as everyone swarmed around the first few PlayStation2 adopters. The lucky few were whisked in and out of stores, having been waiting outside select stores for nearly two days.

By 9AM, some lines in Akihabara were literally thousands of people deep, with no signs of letting up. While some major chains were distributed relatively large quantities of PlayStation2 systems, it's veritably impossible to satiate the swelling demand in the streets. Here are a few screens from early this morning outside a select few Akihabara retailers:

Thanks denhis voor de tip.