The Register heeft in dit nieuwsbericht wat info losgelaten over de prijzen van Intel's Coppermine processor die in de week van 25 oktober gereleased wordt. Hieronder heb je een flinke hap uit het bericht:

We can now reveal the dealer prices for the upcoming Coppermine parts, which will be introduced in the week starting October 25. Distributors worldwide are alerting their dealers early, in order to capitalise on the opportunity. Our information comes from two separate dealers (sorry, resellers), both of which would prefer to stay anonymous.

The top of the range 733MHz part, a Pentium III with 133MHz front side (FSB) bus which includes 256K of cache and active cooling, will cost $810 when bought in boxes of 10, and $790 if dealers splash out for 100 pieces. It has the exciting designation BX80526U733256E.

A 700MHz Pentium III but only using the 100MHz FSB but also with 256K cache and active cooling will cost $785/10, or $770/100.

The 667MHz/133 will cost $630/10, or $620/100, while the 650MHz/100 will cost $610/10 or $595/100. These have the same amount of cache and active cooling.

Intel's next step will be to introduce Coppermine parts with larger caches, as it gradually moves to the .18 micron model. At the spring Intel Developer Forum, Pat Gelsinger, a senior VP at the corporation, revealed that by February 2000, his company would have several fabs bashing out the Coppermine technology.

The Socket 370 parts, which are 500MHz/100FSB and 550MHz/100FSB units coming with 256K cache and spelling the beginning of the end of slot one, will cost $255/10 and $385/10.

Intel will also introduce fast Xeon parts on the same day. The 733MHz/256 will cost $865 for 10, while the 667MHz will cost $688 for 10. It should be carefully noted that Intel is beginning to push Xeon and the desktop processors far closer together, as it readies future processor debuts.

We have been shown copies of the current Intel price list, which is enough to make the most hardened marketing manager at a PC manufacturer, or a distributor, or a dealer, quiver. At the same time, it will be harder than ever for people buying PCs to figure out the starting price.

Lees hier meer.

Bij nV News hebben ze een review gebakken van de Microsoft SideWinder Game Pad Pro. Hieronder een hap uit de test, en een screenie.

One feature that instantly became apparent was the Directional Pad (D-Pad) and the Shift key to its lower-right. The D-Pad sports two software-controlled modes; Standard and Proportional modes. In standard mode, the D-Pad performs much like an 8-way Hat-switch (like on advanced flight-sim joysticks). The D-Pad only relays movements of up, down, left, right, and each of the four diagonals - much like how the original SideWinder Game Pad performed. Standard mode is ideal for sports games, or top-down games that are centered around 8-way movement directions.

However, in Proportional mode, the D-Pad is 'transformed' into a free-floating infinite direction recticle. The beauty of this mode is that it behaves like a pressure (or velocity) sensitive joystick. The more pressure you exert on the D-Pad in any direction, the greater response in your game. The Shift button enables the ABC, XYZ, and Trigger buttons to act as a second set of different functions, in effect giving you sixteen programmable buttons. These can either be in Game Presets or user-bound and saved in the SideWinder Game Controller Software.

Digital Clips heeft het tweede deel van hun Creative Labs Annihilator review online gemikt. Ze hebben nu o.a. benchmarks van de GeForce kaart op een K6-III 450 gepost. De resultaten zijn alleen niet echt om vrolijk van te worden...

Although we all know the K6-III 450 has a poor FPU core (essentially the same as a K6-2), I'd never imagine it to be, well, so bad. After seeing the K6-2 and 3Dnow! running Quake 2 a few years ago, I'd expect at least a decent performance. As you can clearly see from the graph, the K6-III 450 lags even behind the slowly aging Celeron 300A. Rather than fill rate limited, the graph above shows the CPU being the bottleneck. Argh, not the performance one will or should expect from a 450 Mhz CPU.

No 3Dnow!?

There can only be one reason. The 3Dnow! support in the Creative/nVidia Open GL drivers for the GeForce is extremely lacking. No where in the Blaster control was a 3Dnow! optimization tab or checkbox, and the numbers above show. Strangely, I believe this problem does not only extend to the GeForce, TNT and TNT2 implementation of 3Dnow! has, up till this date been relatively meagre. nVidia, if you are reading, please, don't forget the user base here.

3DMark Max Pro

What? No results? The reason is simple. I can confirm the lack of 3Dnow! support in the drivers simply because the benchmark will crash every time after 4 - 5 seconds when 3Dnow! optimizations are used. When I switched to Intel MMX optimizations, the benchmark would run flawlessly, albeit, at a slideshow.

Voor de rest vind je er ook een hele hoop Unreal Tournament & Quake 3 screenies. Begin hier met part deux.

Agossie, het ziet er naar uit dat de 500MHz van de PowerMac G4 niet deze maand uitkomt. Door errata problemen zal hij pas in december worden gereleased, aldus dit nieuwsbericht.

Sources said that the problem -- which only arises when the G4 is run at speeds of 500 MHz or higher -- can result in some corruption in the processor's data cache. Motorola's recommended workaround is to enable the "GlobalWaitR" register in the processor, which, while preventing the problem, slows timing throughout the chip.

Motorola's Technical Support Hotline confirmed both the existence of the errata and the workaround, which it acknowledged entails a "speed hit."

Although the G4 processors in shipping Power Mac G4s contain the errata, their sub-500-MHz speeds keep them from encountering the corruption problem, sources said.

Indeed, another source said, this issue might never evince itself in Macs, since the OS doesn't manipulate data rapidly enough to cause the problem -- the glitch would more likely effect more-efficient embedded operating systems. Even if data corruption should occur, a source said, the result would be nothing more than a system freeze, easily fixed with a restart.

Jason van AnandTech heeft een review gepost van de Cambridge SoundWorks Desktop Theater DTT2500 Digital Speakers. Dit heftige setje wordt geleverd met een Dolby Digital decoder:

Yes, that's right. You can actually watch a DVD movie on your PC now and watch it the way it was intended to be watched. In Dolby Digital AC-3 sound. Dolby Digital is essentially an advanced form of digital audio coding, allowing us to enjoy movies in a whole new way. Dolby Digital was first experienced in 1992 when movie audiences got their first taste of real sound quality. Dolby Digital AC-3 (5.1) consists of five discrete full range channels: left, center, right, left-surround, right-surround and of course number six (or the ".1" of "5.1"), the LFE (low frequency effects) channel. If you are unsure what the LFE channel is, it is commonly known as the subwoofer. Cambridge had introduced Dolby Digital to the PC desktop with their PCWorks Desktop Theater. The new 2500's are a MUCH better unit. To listen to your DVD's in Dolby Digital format, you will require a DVD Decoder card in your PC, something like a Dxr3 or Real Magic Hollywood Plus. I used the Real Magic for my tests with the unit. I tested the sound quality with various movies and one of my favorites, Top Gun. After inserting the DVD, off go the lights, and kick back and relax. The movie sounded just fantastic. Having a home DVD system already, I am fairly use to DVD and its mind-blowing sound quality. To hear this kind of quality on your PC is what will blow you away. When the gaming world starts taking advantage of DVD and Dolby Digital, things are really going to change in the PC speaker community. With these speakers you are ready for what the future has to offer.

Check de rest van de review hier.

Entech heeft een nieuwe versie van de videocard/monitor utility Powerstrip in elkaar geschroefd. Versie 2.52 heeft nu preliminary support for the GeForce256, improved control over the Permedia3 and Oxygen/VX1, and some other improvements all around with respect to custom CRTC timings, in-game gamma adjustment, etc." Haal hem hier naar bendeden. (667kB)

Ik was net wat aan het rond fietsen in de .plan tracker van Bluesnews, blijkt dat daar de afgelopen dagen een behoorlijk discussie is los gebarsten over de gebrekkige support die game developers tot op heden van Apple hebben ontvangen (zie ook dit niewsartikel van eerder deze week). Rick Johnson, Soldier of Fortune programmer bij Ravensoft, vat het hele gebeuren treffend samen (nouja, deze .plan update is best wel heftig ):

Just to clear up some confusion about this whole Apple topic, as I’ve been seeing a lot of emails with people reading into what I said.

“You guys just want free stuff to take it home…”
Game developers (or at least Raven) don’t ask for equipment from manufacturers to “take it home so we can have some cool things.” This equipment is used to develop on, to test our games on, and upon occasion, to add special features into the games to support that equipment. Examples of this pertaining to SoF would be when we received a Savage S4 video card, and added texture compression; Razer mice and adding support for their 5 buttons; Creative Labs SBLive! Card and 2.0 support; Aureal’s A3D and adding 2.0 support; etc.

“Your anti-apple / anti-mac attitudes are going to bring you down…”
Up until recently, Apple has been “ignoring” the needs of the gaming community. Mac has typically been a graphics / publishing computer, educational computer. “Games and education don’t always go in the same sentence” one might argue. The market has never been very large to support hard-core games. Units were typically substantially lower in sales, which when you add in marketing, packaging, distribution, and other promotional expenses, it can be a very difficult market to break even. We do have a few people at Raven who own Macs, including the Art Lead of SoF and the Product Director.

“Why do you expect Apple to give you free stuff? Just buy it yourself…”
Back in early May at E3, Apple was very eager to work with us. They promised that they would get us equipment, support, etc. Star Trek: Elite Force would seem like the perfect title to make a big hit on the Mac. Time went by, nothing showed up. We started to ask Apple what’s up through emails and phone calls, but we never received any word back. A few weeks ago, we received the Mac version of Heretic 2. Having nothing to test it on, we placed an order for a G4 a few weeks ago. But at this point, we also realized that if Apple wasn’t going to return our messages, that maybe they really weren’t going to be sincere about “wanting games” and being “developer friendly.” Which prompted my small .plan update two days ago.

Generally, we get equipment from manufacturers because it is either not available to the public yet (as in the case of AMD sending us K7s) or for us to investigate the product to see if we can support it. We simply can’t go out and buy every piece of equipment a gamer might have. Most of the time, the equipment we receive is unsolicited. Because of my .plan update a few days ago, we even got an email from a Cyrix representative to send us some computers with their CPU on it to test our games on.

I received many inquires from magazines about this topic. Typically, they would ask “is it Apple that you aren’t getting support from or the Apple community?” I’ve been around Apple community for a long time, way back to the Commodore 64 days, so I know how strong the Apple community is, and their support was never questioned.

The night of my .plan update, Jim Black of Apple made a phone call here. He told us that our “seed equipment” was approved of at Apple, but wasn’t sure of the status, since he had changed positions internally. Yesterday, we received a few more phone calls from Apple. They weren’t too pleased with my update (though we decided as a company to have me do this update). They had indeed sent us 2 Macs mid to late August (3 months after E3), but not to Raven, instead delivered to Activision. Since we were in between producers at the time, and that the Macs didn’t appear to be labeled for Raven, they got absorbed into Activision.

Still though, this doesn’t explain why our messages weren’t being returned (inquiring about the equipment), nor did anyone at Apple really know what was happening with the situation until yesterday. Whenever we get “big” equipment from people, they are always immediately following up with us if we received it, works properly, etc. AMD has always been very good to us, sending a bunch of machines (most recently a few K7s) and they are always on the phone the next day, making sure we got them, and working through any problems we are experiencing. My .plan update resulted from us exhausting every possible avenue we had to contact Apple. Normally these situations are handled behind the scenes, but in this case, we felt we had no choice, and had to go to a more public forum.

Apple has now gotten back to us; we are being placed into their development program, and we should be getting those Macs shortly. Once we get them, there’ll be an update about how Heretic 2 is looking, plus James will make a .plan update on the status of Star Trek: Elite Force (hopefully getting it up and running will only be minor work). We appreciate Apple getting back to us so quickly now, and providing us with additional equipment.

Just a quick follow-up on Heretic 2. We’ve since run the port of Heretic2 that LogicWare has provided us. Runs remarkably well, and you wouldn’t be able to tell the difference between [break] Inmiddels heeft Raven twee G3's van Apple gekregen, waarvan er eentje dood arriveerde... .plan update van James Monroe:[/break] We got some macs finally, and they're pretty spiffy. Two g3s, one of which seems to have been involved in an impact test. Tt didn't come out alive. Tt has these clever handles on all corners which are great for moving the machine around. The front top handle had been smashed and cracked. This resulted in the corner of the case to which it was attached to pop its rivets and bend outward, slighly opening the case. Also, internally, the heat sink had come off the proc. (very nice lever action way to open the case, btw.) I found a metal strap floating around in there and reattached it. Now it powers up and actually runs.

Now, down to business. I've heard code warrior is the way to go for mac programming... hen i'll get to find out how well our STL and CPP code changes behave on the mac. The PC and the Mac version. I believe this is still a beta version.

Chip ontwerpers gaan in hun zoektocht naar steeds hogere performance het pad in van heavy parallism, zo bericht EETimes. Intel kiest hierbij voor het sterk compiler-afhankelijke instruction-level parallelism (ILP), terwijl IBM daar nog een flinke schep tegenaan gooit door twee cores in één processor te verenigen, het zogenaamde chip multiprocessing. Compaq / API pakken het weer anders aan: middels thread-level parallelism (TLP) werkt één enkele processor als een virtueel quad processor systeem:

Slater and other analysts said that while Intel's Itanium is expected to fare well when it debuts next year, the use of instruction-level parallelism may be coming to an end for new architectures as designers begin to see the advances in chip- and thread-level parallelism that IBM, Compaq and others are leveraging.

Indeed, keynoter John Hennessy, co-developer of the first commercial RISC chip and professor of electrical engineering and computer science at Stanford University, cited a looming transition away from ILP.

"These techniques are getting ever more complicated. I don't see any performance wall, but there are steeper slopes ahead," he said, noting the complexity of using techniques such as trace caching and value speculation. Bigger advances will come as designers embrace parallelism through multithreading, but that requires a significant transition, he added.

"We are entering a domain where designers need to employ multiple threads, and that requires software support," Hennessy said. "That means we have to help software guys think of new ways to deal with parallelism. It's time we get started on the process of moving to multithreaded software models."

Acer Labs heeft nieuwe IDE drivers gepost, waaronder ook een exemplaar voor Linux. Verder heeft MSI een nieuw BIOS voor de MSI 6167 plank klaar gebakken en vond Gigabyte het nodig om hetzelfde te doen voor z'n GA-71X bord.

AMD Zone heeft Quake3 benchmarks van een GeForce 256 + Athlon 600 systeempje bij elkaar weten te rapen:

System: Athlon 600mhz, 256megs pc133 ram.
Settings: High Quality, no sound.
Game: Q3Test v1.08
Creative TNT2Ultra @ 800x600 = 60.3fps
NVidia GeForce256 @ 800x600 = 80.4fps

De jongens en meisjes van Firingsquad hebben een artikeltje in elkaar geschroefd over de Dual Head feature van de Matrox G400:

Summing it up The bottom line is that DualHead is like a lot of the things you take for granted in life. Sometimes you don't know you need it until you try it. But once you've tried it, there's no going back. The day I'm forced to trade out my G400 for a card that doesn't support dual monitors is the day I will probably break down in tears. The thought makes me wonder one of two things - when is Matrox going to announce and ship their next video card (G600? Or G800?), and when will NVIDIA, 3dfx, and ATI come out and implement something like this?

Ennuh, ik heb 't ff nagevraagd bij Reinder - onze lokale dual head held (jaja, 22" + 19") - en die bevestigd de verslavingsfaktor van een 2600 pixels brede desktop .

FPS3D.com heeft een review van de HighPoint XStore Pro drivers neer gemikt. Uit de opmerkelijke benchmarks blijkt dat de drivers in eerste instantie een (flink) stuk trager zijn dan de standaard Windoos busmaster drivers, maar sneller worden naarmate de data in de cache hangt (wat in de praktijk weinig voorkomt):

Ok, time for my opinion. Don't use this driver! So what if the app will launch 10 times faster if you run it again right after you close it, that's not all the time! I'll pay the punishment for the one or two times I do that per day to get my everyday performance back. Really, no one launches the same application over and over and over. Do me a favor, if you come across a disk with HPT XStore Pro written on it, destroy it, burn it, just don't let it fall into the hands of some unlucky amateur tweaker!

Ik kwam via HardOCP bij deze wazige site, waar we worden geconfronteerd met een nieuw fenomeen in overklokkers land: de koelmuur...

...hoe je plaatselijk een lage druk gebied creeërt...