Intel secret X86-64 project still on the go

Advertisement (Why?)

THE END OF A cunning Intel plan to develop an X86-64 microprocessor is not yet nigh, according to sources a micron away from the chip giant. The secret project – first revealed here in the INQUIRER and codenamed "Yamhill" – has however, metamorphosed since the early days, with some Intel politics muddying the water.

Those particular politics led to Intel positioning its future "Prescott" processor as a sort of consumery desktop Clawhammer killer at the Developer Forum held in San Jose in February. The project is now, according to sources, codenamed *T – that's right, with an asterisk, with a secret chip developed straight from the X86-64 specification.

This is kind of an insurance policy on Intel's behalf. AMD's current troubles which can be summed up in the phrase "no cash", have given Chipzilla a breathing space, and it hopes that Hector and his gang won't be able to pull the proverbial Clawhammer-Rabbit out of its magic hat. Intel's insurance policy remains firmly in place, however, but the one thing it definitely wants is to distance any project like this from its beloved Itanium platform, which will chug on regardless.

News source: The Inq

Performance

• 10% improvement in golem mesh rendering performance.
• Fixed unnecessary particle lighting overhead problem.
• For performance reasons, most carcasses are cleaned up more quickly now.
• For performance reasons (mainly in marsh) dropped energy rifles are cleaned up after 10 seconds.
• M08B: improved performance by making sure generator rotating movers only rotate when they are visible.
• MM_Marsh: added field generator and turret to PreCacheList to avoid hitches when these appear.
• M12: fixed particle overhead.
• Added new code/changes to limit overhead due to ragdoll deaths (see “Advanced Tips for Improving Performance” in the readme for details).

Misc Fixes

• Sound fixes. Should reduce occurrences of sound “stuttering” that some people have had due to sound overload / thrashing.
• Fixed player dodge speed and height. Can be modified via user.ini settings (see DodgeXYVelocityScale and DodgeJumpZScale in the [U2Pawn] section).
• M08B: lowered carcasses, blood to floor.
• M09E: fix for 2 fems staying "inert" if the fem with the RL is killed before being triggered.
• Fixed damage warnings (accessed nones) along with fix for damage effects
(blood) being spawned so far inside character's mesh that it doesn't show up
(currently only affects mukhoggs).
• Sunlight fix (flickering in waterfront and potentially other places).
• MM_Waterfront: fall out of world fix.
• M12: fixed sun disappearing in opening cutscene with ParticleDensity=0. Also CS_Outro, PA_Acheron, PD_Acheron.
• Display gamma/brightness/contrast values when changed via F10/F11/F12 keys.
• Fix for inert skaarj lying prone on bank in M08B (skaarj will now react to hearing the player even if player bypasses trigger).
• Fix for “Joe’s” script breaking in M10_Avalon in some cases.
• Fixed changing actor detail level in-game could result in invisible NPCs (in TutA).
• Made key bindings delete button more obvious (click on this to delete key bindings).
• Fix from Jason Yu for bots sometimes not respawning after being killed in botmatches.
• Added files needed to fix umod support. The umod registry settings will also need to be modified for Unreal 2 umods to work. We’re working on making a separate downloadable installer for this but in the meantime, people who installed the game in the default folder can click on the .reg file in Unreal2Help.
• Increased crawling spider damage by 3x.
• Fixed rammers not attacking properly.

Misc Changes

• Added carcass cleanup particle effect.
• Added support for modifying player movement speed via unreal2.ini setting (for the duration of the game). For example, to increase the player's movement speed by 50% set PlayerSpeedRatio=1.5 in the [U2GameInfo] section.
• M08B: fixed trigger that player could avoid.
• Added 70 and 72 Hz to refresh rates options.
• Support for scaling the hud / ui to better support multiple monitor setups. This can now be controlled via the ScaleX, ScaleY, OffsetX, OffsetY properties in the [UI.UIConsole] section in Unreal2.ini. For example, use ScaleX=0.333 and
OffsetX=0.333 for three screen support.
• Added “mouse over” context information for the EAX check box. Warns against enabling EAX on non-Audigy / Audigy 2 cards and warns about possible performance issues.

AI

• Mod authors can once again enable script warnings / errors by setting
bDisableErrorMessages=false (by default all script warnings / errors are
silent). These were completely disabled for RC7 with the intent of
re-enabling these for the patch. Note that you have to enable the Component=Console.MessageArea line in UIScriptsUI.ui (remove the “;”) to see debug messages in-game.

Editor

• Fix for not being able to paint on terrain in editor.
• Fix for terrain build crash if no terrain map in level.
• Fixed editor "first run" (UnrealEd.ini added to system folder).
• Fixed WBrowserStaticMesh::UpdateMenu crash.
• Fixed crash if you clicked on delete in the decolayer terrain
tool, without a decolayer actually being selected.

Testing / Stats

• Improved in-game profiling support / logging.
• Improved in-game stats. Added “stat lock”.

Other

• Added Mike Lambert to in-game credits.

(5 replies) #1 leebobs on 26 Mar 2003 - 18:31

Hmmm.... Interesting, maybe, just maybe; Intel are scared of AMD?!? "People will use 32bit till the end of the decade" - Don't think so. "Who needs more that 640k or memory" - Good Question Bill.

#1.1 tntomek on 26 Mar 2003 - 19:03

Although I am an AMD supporter, I don't think Intel is worried at all. AMD is in serious financial trouble; how long do you think they can keep reporting net loses for?

#1.2 Tobbe on 26 Mar 2003 - 19:30

[neoquote=#1.1 by tntomek]Although I am an AMD supporter, I don't think Intel is worried at all. AMD is in serious financial trouble; how long do you think they can keep reporting net loses for? [/neoquote] You're correct, Intel won't invest in x86-64 if AMD will fail in their own project. Intel already have 64-bit CPUs and they want to use that architecture instead due to the bottlenecks that x86 has.

#1.3 kairon on 26 Mar 2003 - 20:29

[neoquote=#1.2 by Tobbe]You're correct, Intel won't invest in x86-64 if AMD will fail in their own project. Intel already have 64-bit CPUs and they want to use that architecture instead due to the bottlenecks that x86 has.[/neoquote] x86 is 64 bit too, is it not? the 16 bit pc's had chip names like 80186,80286,etc

#1.4 vacs on 27 Mar 2003 - 01:15

no, x86 implies a 32-bit CPU. The "x" can substitued by any number 086,186,286,386,486,586 and so on. All 32-bit CPUs...

#1.5 MZatko55 on 27 Mar 2003 - 13:15

Don't mean to be a dick, but you are wrong. The Intel 8086, 8088, 286 were 16-bit cpus. Also, as a sidenote, x86 refers to a set of hardware specific functions used in low level programming languages such as Assembler. One limitation of the x86 set of instructions besides being based off of a complicated instruction set (CISC) which means it issues more functions to do the same amount of work. Also, I know that it doesn't support hardware division like an equivelent RISC chip such as sparc, power, and alpha. It is said that AMD's new 64 bit instruction set is a superset of the x86 instruction set, which means it will add more hardware level functions available which may mean things like hardware division may be available and other bells and whistles while maintaining compatibility of the base x86 functions. One of the reasons Intel probably wanted to abandon the x86 instruction sets with the Itanic is because the x86 architecture is a nightmare, to be blunt. To all veterans of assembly on an Intel machine, you know what I mean.

Last edited by 23592 on 27 Mar 2003 - 17:01

(9 replies) #2 vacs on 26 Mar 2003 - 20:15

What would I laugh if Intel really releases a x86-64 CPU which is faster then AMD's Hammer. It's quite plausible that Intel really wants to do that. Itanium are just for high-end servers and not for the consumer domain which this unannounced CPU would be aimed at. That way AMD would do all the work for Intel (make sure that software is available for their 64-bit extensions) and Intel would have a CPU in direct competition to the clawhammer

#2.1 MZatko55 on 26 Mar 2003 - 20:43

I don't know what you would laugh, but I would laugh the theme to Airwolf. I hope Intel does release this chip and puts AMD out of business because I'm starting to get sick of all this competition crap.

#2.2 buroja on 26 Mar 2003 - 21:00

Yeah, it totally sucks to have more than one option. Die, AMD, die! I don't know about the rest of you, but I can't WAIT for the return of the $1000 flagship CPU.

#2.3 Krux on 26 Mar 2003 - 21:18

[neoquote=#2.1 by MZatko55]What kind of stupid comment is that? If it wasn't for AMD you would still be paying 1000 bucks for an 800 MHz CPU........... Idiot competition drives the market. and AMD's new CPU is backwards compatible which means they won't be testing software for anyone it can run 32bit apps just as well as 64bit apps.... soon as the P4 dies Intel will go back to being a respectable company not one that’s just trying to sell to the stupid masses that don't know clock speed isn't everything.[/neoquote] What kind of stupid comment is that? If it wasn't for amd you would still be paying 1000 bucks for an 800 MHz CPU........... Idiot competition drives the market. and AMD's new CPU is backwards compatible whichs means they won't be testing software for anyone it can run 32bit apps just as well as 64bit apps.... soon as the P4 dies Intel will go back to being a respectible company not one thats just trying to sell to the stupid masses that don't know clock speed isn't everything.

#2.4 MZatko55 on 26 Mar 2003 - 21:56

It's called sarcasm. I'm an AMD stockholder.

#2.5 tntomek on 26 Mar 2003 - 21:59

LOL you have to work on your sacasm cause I also thought you were serious

#2.6 vacs on 27 Mar 2003 - 01:21

competition is bad for us customers. In the past, Intel developped a new CPU all 2-3 years. Now, due to all this competition we have, Barton, Hammer, P4 Prescott, Itanium 3, Centrino and VIA's C3 CPUs are all coming out in the same year! As soon as you have bought a new one, it's already outdated and you need to consider buying a faster one. Hopefully AMD goes out of business so I can stop buying a new PC every 5-6 months

#2.7 Octol on 27 Mar 2003 - 09:19

[neoquote=#2.6 by vacs]competition is bad for us customers. In the past, Intel developed a new CPU every 2-3 years. Now, due to all this competition we have, Barton, Hammer, P4 Prescott, Itanium 3, Centrino and VIA's C3 CPUs are all coming out in the same year! As soon as you have bought a new one, it's already outdated and you need to consider buying a faster one. Hopefully AMD goes out of business so I can stop buying a new PC every 5-6 months

[/neoquote] If you can [I]afford[/I] a new PC every 5-6 months, count your blessings. In any case, what you want is not possible, whether or not AMD goes out of business. The people at Intel--or any other tech company for that matter--can't afford to sit on their hands or [I]they'll[/I] be out of business. That's what Apple did way back when, and you can see where it got them. I understand how painful it is to have to keep reaching in your pocket just to try and maintain a reasonable level of technological concurrency, but you can't stop progress. The fact is that a PIII 600 on a 66MHz MoBo (my oldest system) still does an admirable job of performing its required tasks more than satisfactorily. Having the latest and greatest isn't an absolute necessity. What's more, you don't have to settle for something that slow; because of the intense price competition you can, right this minute, buy a P4 2.4GHz, 533MHz FSB CPU at New Egg for $162.00--a processor that wasn't even available a year ago! I'm not a gamer, so I can't address a gamer's needs, but I can tell you that my work computer--a P4 2.2GHz (400MHz FSB) box with 1GB PC2100 DDR can run Word [I]and[/I] Outlook 2002, Photshop 7, and multiple instances of IE6 all at the same time without a hiccup; and it can do that while simultaneously recording the output of 16 high-resolution color security cameras. I'd say it's reasonable to conclude that this is more than enough power for most people! Just because a computer isn't the newest available doesn't make it outdated. And then there's the alternative to this rapid-fire development: we could all still be computing on 286 machines running at 12MHz with 2MB RAM, a 20MB hard drive, and a 1.2kbps modem--the top of the line box in 1985.

#2.8 vacs on 27 Mar 2003 - 10:50

funny, someone took my last post seriously, unbelieveable

#2.9 Octol on 27 Mar 2003 - 13:53

Unfortunately, I've heard enough people express more or less that same concern, and they [I]were[/I] serious. I thought I'd reply to that.

Commenting has either been disabled on this article or you are not logged in. Click here to login or register, its free!

Note: Anonymous commenting is disabled in order to keep the quality of responses to a high standard.

Neowin.net