Hoofdcategorieën
Topicacties

[Discussie] AMD Athlon 64 & Opteron - Deel 5

Pagina: 1 2 3 4 5 6 7 8 9 10 11 12 ... 18 19 20 21 last

Nieuw Topic
Berichten: 678
Reg. datum: 31 maart 2003

quote:
hully schreef op 16 juni 2004 @ 22:13:
heb een MSI KT8 NEO Mobo gekocht met athlon 3200+ 64 en 2 x 512 mb ddr400 en een asus ax800pro.

het rare doet zich voor dat als ik pc opstart ik supersnel in windows xp zit. Als ik dan in windows ben en ik wil dan bijvoorbeeld direct een programma opstarten of via "deze computer" en dan "eigenschappen" de systeemeigenschappen bekijken, het lang voordat het schermpje te voorschijn komt of het programma opstart, terwijl de hardeschijf geen activiteit vertoont. Het lijkt dus alsof er nog zaken aan het opstarten zijn 'achter de schermen' terwijl er dus geen activiteiten zichtbaar zijn. Het gekke is dat als ik -direct nadat ik windows opstart en het bureaublad zie-internet op ga, internet explorer wel zonder een minuut of 2 vertraging opstart...

iemand bekend hiermee en heeft iemand een oplossing?
Internet explorer wordt zowiezo samen met Windows geladen, het zit 'in' Windows. Moest je bijvoorbeeld firefox op je bureaublad hebben zou die ook een tijdje nodig hebben, omdat die standaard niet meegeladen wordt. Kijk anders eens welke recources het meeste CPU gebruiken bij het opstarten of in het register wat juist gestart wordt ?

Tofu wijzigde dit bericht 17-06-2004 20:43 (7%)

« '-_-' »
Berichten: 10.070
Reg. datum: 06 maart 2001

Abit AV8 Socket 939 Preview B)

Abit AV8 K8T800 Pro bordje heeft AGP/PCI lock

Athlon64 FX multiplier is echter niet instelbaar :/ Is dit een K8T800 Pro issue :?
Zijn er andere K8T800 Pro Socket 939 moederborden (P)Reviews met hetzelfde probleem :?

http://www.tweakers.net/ext/f/35031/full.jpg

The trade of the tools
[ me | specs ] Klipsch Promedia Ultra 5.1 + Sennheiser HD-590

Beth
Berichten: 9.102
Reg. datum: 26 januari 2002

quote:
Athlon64 FX multiplier is echter niet instelbaar Is dit een K8T800 Pro issue
De mp is locked en zover ik weet kan je die niet unlocken.

I don't know who this guy is, but I want him and his tortilla's ... DEAD !

Debian GNU/Linux Sid
Berichten: 6.538
Reg. datum: 10 augustus 2000

quote:
Hacku schreef op 18 juni 2004 @ 07:36:
[...]


De mp is locked en zover ik weet kan je die niet unlocken.
Bij een FX is deze toch gewoon unlocked? Ik meen dat AMD heeft aangegeven dat dat voorlopig ook zo zou blijven en een search met Google het onbreken van een lock te bevestigen :).

Leven is het meervoud van lef | In order to make an apple pie from scratch, you must first create the universe.

Rabbi
Berichten: 368
Reg. datum: 23 maart 2003

quote:
silentsnow schreef op 18 juni 2004 @ 02:59:
Abit AV8 Socket 939 Preview B)

Abit AV8 K8T800 Pro bordje heeft AGP/PCI lock

Athlon64 FX multiplier is echter niet instelbaar :/ Is dit een K8T800 Pro issue :?
Zijn er andere K8T800 Pro Socket 939 moederborden (P)Reviews met hetzelfde probleem :?

[afbeelding]
Ik weet dat de Abit KV8 pro, socket 754 met K8T800 Pro NB, ook last heeft met de multipliers. Ze zijn wel instelbaar maar werken niet. Ik denk eerder dat dit een bios bug is; Abit staat bekend om hun uitgebreide tweak opties.

The only certainty is uncertainty... The only constant thing is change...

Beth
Berichten: 9.102
Reg. datum: 26 januari 2002

quote:
odysseus schreef op 18 juni 2004 @ 09:57:
[...]

Bij een FX is deze toch gewoon unlocked? Ik meen dat AMD heeft aangegeven dat dat voorlopig ook zo zou blijven en een search met Google het onbreken van een lock te bevestigen :).


Idd totaal vergeten.
De FX is idd unlocked, maar of dat voor de s939 ook zo is?

I don't know who this guy is, but I want him and his tortilla's ... DEAD !

Beth
Berichten: 9.102
Reg. datum: 26 januari 2002

2,6GHz op 130nm?
quote:

Advanced Micro Devices may push the release of its high-end microprocessor for desktops produced using 90nm fabrication process into the second quarter of next year, X-bit labs has learnt from sources close to the company. But before the chipmaker unveils its new microprocessor design, there will be a yet another 130nm product targeting enthusiasts.

As reported, AMD plans to introduce its AMD Athlon 64 FX-55 chip in the Q4 2004. The chip will be clocked at 2.60GHz, but is still likely to be made using current 130nm Silicon-on-Insulator process technology. A more powerful processor geared towards hardcore enthusiasts – AMD Athlon 64 FX-57 – made using 90nm SOI fabrication process is expected to emerge in the Q2 2005. Previously high-end desktop parts were referred to as “code-named San Diego”.

AMD Athlon FX microprocessors are positioned as top-of-the-range offerings from Advanced Micro Devices. While the architecture of the Athlon 64 FX is the same compared to the Athlon 64, e.g. the chips can execute the same software, including 64-bit software, the chips feature large 1MB level-two cache as well as dual-channel memory controller, bringing some additional speed bumps for the chips that cost approximately $800 today.

By contrast, typical AMD Athlon 64 processors that cost from around $200 to $750 today, either feature single-channel memory controller or only 512KB of secondary-level cache delivering slightly lower performance compared to the “FX” parts.

By the time the AMD Athlon 64 FX-57 hits the market, the Sunnyvale, California-based chipmaker is likely to launch its Athlon 64 4000+ (Q4 2004), Athlon 64 4200+ (Q1 2005) and Athlon 64 4400+ (Q2 2005), sources said.

All the new AMD Athlon 64-series processors will be drop-in compatible with Socket 939 infrastructure that can supply up to 105W of power to central processing units.

Representatives for Advanced Micro Devices do not typically comment on information from unofficial sources.
http://www.xbitlabs.com/news/cpu/display/20040618024759.html

I don't know who this guy is, but I want him and his tortilla's ... DEAD !

Ek een sterke Jerke !!!

quote:
Hacku schreef op 18 juni 2004 @ 13:32:
[...]


Idd totaal vergeten.
De FX is idd unlocked, maar of dat voor de s939 ook zo is?
FX is FX, dus dan moeten ze gelijk zijn.

T E A M: Grutte Pier [Wa Oars] / Fryslân wer mei Kening \___ R@H#4

Beth
Berichten: 9.102
Reg. datum: 26 januari 2002

Nee want de core is anders (s940: sledgehammer, s939: clawhammer) dus het kan zijn dat de FX s939 locked is, maar dat zullen we wel ondervinden door overclockingen van één of andere gekke jap :p

I don't know who this guy is, but I want him and his tortilla's ... DEAD !

Berichten: 1.104
Reg. datum: 18 februari 2002

Vandaag mijn A64 CPU ontvangen met deze stepping e.d.

CBASC 0416VPMW

ADA3200AEP4AX

Ik zie het verder hier nergens staan?
Iets bekend hierover?

I’m just a digital guy in an analog world.

Gestoorde IRC'er
Berichten: 818
Reg. datum: 21 april 2000

quote:
MiMiC schreef op 18 juni 2004 @ 20:27:
Vandaag mijn A64 CPU ontvangen met deze stepping e.d.

CBASC 0416VPMW

ADA3200AEP4AX

Ik zie het verder hier nergens staan?
Iets bekend hierover?

Staat wel in de topicstart, maargoed.
Athlon 64 3200+, Newcastle core (512KB L2 Cache), CG stepping. Geproduceerd in week 16 van 2004.
 
Ek een sterke Jerke !!!

quote:
Hacku schreef op 18 juni 2004 @ 18:32:
Nee want de core is anders (s940: sledgehammer, s939: clawhammer) dus het kan zijn dat de FX s939 locked is, maar dat zullen we wel ondervinden door overclockingen van één of andere gekke jap :p
Nee want dan moeten ze het niet meer FX noemen, ze verkopen ze unlocked en met 1 MB anders is het geen FX.

T E A M: Grutte Pier [Wa Oars] / Fryslân wer mei Kening \___ R@H#4

Berichten: 994
Reg. datum: 02 december 2001

Ik kwam dit tegen :

AMD Socket 939 Memory Testing

http://www.legitreviews.com/reviews/939mem/

Iemand all gezien :?

DFI Lanparty Rev.B ( A+ ) * Waretcooling CPU, NB * AMD XP2800+ AQZFA 0338 VPMW 10x252 * Iiyama Pro454 * KINGSTON HyperX BH-5 3200/512 2-2-2-11* WD 800JB 2x * ASUS CD 50x * Windows XP prof. * nForce sound

« '-_-' »
Berichten: 10.070
Reg. datum: 06 maart 2001

quote:
oc4ever schreef op 18 juni 2004 @ 10:41:
[...]


Ik weet dat de Abit KV8 pro, socket 754 met K8T800 Pro NB, ook last heeft met de multipliers. Ze zijn wel instelbaar maar werken niet. Ik denk eerder dat dit een bios bug is; Abit staat bekend om hun uitgebreide tweak opties.


Bedoel je niet toevallig de AGP/PCI lock?

Overigens is Bios 15 al uit, en de AGP/PCI lock werkt in deze versie.
quote:



Dan krijgen we waarschijnijk ook een Socket 939 Athlon64 4000+ 2600MHz 512KB Newcastle. 90nm in H2 2005 is zoiezo tragisch, dat zou namelijk betekenen dat het lange tijd stil gaat blijven qua nieuwe processors.

Een Socket 754 2600MHz processor gaat zeer lastig worden qua PR. Een 2600MHz Newcastle zou dan boven de 3400+ uitkomen, aangezien de 2400MHz newcastle al op 3400+ zit. Waarschijnlijk zou het dan 3600+ worden. De 2600MHz ClawHammer zou dan een PR van 3900+ krijgen, aangezien de 2400MHz ClawHammer ene PR van 3700+ heeft.

In dit scenario heb je dus een 3400+, 3500+, 3600+, 3700+, 3800+, 3900+ en 4000+ :/

The trade of the tools
[ me | specs ] Klipsch Promedia Ultra 5.1 + Sennheiser HD-590

Rabbi
Berichten: 368
Reg. datum: 23 maart 2003

quote:
silentsnow schreef op 19 juni 2004 @ 07:18:
[...]


Bedoel je niet toevallig de AGP/PCI lock?

Overigens is Bios 15 al uit, en de AGP/PCI lock werkt in deze versie.


Het is echt een probleem met de multipliers: Soms boot de cpu op een lagere mulitplier, soms boot hij niet als je een lagere multiplier instel. Het vreemde is dat niet alle borden dit hebben.

Op Got heeft Milanza dit bordje al getest:

quote:
Met bios 15 als ik iets boven de 2,8 volt selecteer in uGuru loopt ie vast. Met 14 had ik dit niet en kon ik nog 1:1 draaien tot ongeveer 235mhz op 3,2 volt. Met 15 kwam ik niet verder dan 220 1:1 ofzo. Het gekke is dat als het de htt en het geheugen asynchroon lopen ik het geheugen makkelijk tot 228 kan krijgen 8)7. Ook kan ik de latencies niet veranderen in de bios, want dan boot ie gewoon niet. En als ik hem te hoog heb geklokt en ik zet hem weer normaal, dan boot ie ook niet.
Hopelijk komt Abit snel met een nieuwe revisie of bios update.

The only certainty is uncertainty... The only constant thing is change...

« '-_-' »
Berichten: 10.070
Reg. datum: 06 maart 2001

quote:
oc4ever schreef op 19 juni 2004 @ 11:56:
[...]


Het is echt een probleem met de multipliers: Soms boot de cpu op een lagere mulitplier, soms boot hij niet als je een lagere multiplier instel. Het vreemde is dat niet alle borden dit hebben.

Op Got heeft Milanza dit bordje al getest:


[...]


Hopelijk komt Abit snel met een nieuwe revisie of bios update.


Heb je een linkie naar z'n post? Ik kan het nergens vinden.

quote:


Kleine uitbreiding op de huidige roadmap:

(bablevis vertaling + leesbaar gemaakt)

"According to the AMD official disclosure, the AMD dual core processor is compatible with present Opteron processor socket [Socket 940], the dual core processor core size is 210 square millimeters, slightly bigger than the 0.09 micron [90nm] Athlon64 processor the core size (193 square millimeters). AMD first will promote the Opteron dual core processor. When AMD changes to 0.065 micron [65nm] regulation, AMD will promote the dual core desktop processor. With the AMD dual core processor, each core has an independent memory controller, and uses NUMA (Non Unified Memory Architecture). Between the dual core will use compared to the existing HyperTransport main line quicker connection way."
bron

Ik twijfel aan "each core has an independent memory controller" omdat eerder al duidelijk werd dat er slechts één memory controller aanwezig zal zijn.

Ook is van belang dat de dual core desktop processor op de markt komt als 65nm. Dat kan nog erg lang gaan duren, want 65nm voor de desktop komt volgens geruchten pas in 2006 (bron)

The trade of the tools
[ me | specs ] Klipsch Promedia Ultra 5.1 + Sennheiser HD-590

Beth
Berichten: 9.102
Reg. datum: 26 januari 2002

AMD Opteron bug can cause incorrect results
quote:

GERMAN SITE 3D Chips has spotted a bug report on the AMD site that could cause Opteron systems to hang or incorrect results to be produced.

The bug is related to REP MOVS instructions which could result in succeeding instructions being skipped or an incorrect address size or data size to be used.

Opteron users are being advised to contact their AMD reps for information on a BIOS update, while the firm also says it plans a fix.

The report from 3D Chips is here, and that page also contains a link to the other bug info AMD has published in June.
http://www.theinquirer.net/?article=16683

I don't know who this guy is, but I want him and his tortilla's ... DEAD !

« '-_-' »
Berichten: 10.070
Reg. datum: 06 maart 2001

quote:
Hacku schreef op 19 juni 2004 @ 20:03:
AMD Opteron bug can cause incorrect results


[...]


http://www.theinquirer.net/?article=16683
BIOS update wekt de verkeerde indruk. De BIOS update is een workaround voor het probleem, dus geen echte fix. Een microcode update van de control unit (uPROM om precies te zijn) moet door AMD zelf worden gedaan, en alleen dan is het probleem echt verholpen.

silentsnow wijzigde dit bericht 19-06-2004 22:03 (18%)

The trade of the tools
[ me | specs ] Klipsch Promedia Ultra 5.1 + Sennheiser HD-590

Berichten: 3.543
Reg. datum: 24 juli 2000

quote:
silentsnow schreef op 19 juni 2004 @ 19:16:
Ik twijfel aan "each core has an independent memory controller" omdat eerder al duidelijk werd dat er slechts één memory controller aanwezig zal zijn.
(bron)

Wat goed mogelijk is is dat beide cores zullen beschikken over een eigen 64-bit controller (i.p.v. 128-bit bij de huidige Opteron).

Hiermee is een lage latency voor beide cores gegarandeerd via het "eigen" geheugenreepje en toch is er een hoge bandbreedte beschikbaar door NUMA.
 
Berichten: 3.543
Reg. datum: 24 juli 2000

Iemand dit trouwens gelezen ('t is alweer van een paar dagen geleden) ?
quote:

Sales of AMD Opteron microprocessor – the world’s first and yet the only server chip that can run 32-bit and 64-bit code – for the first quarter of this year beat the aggregated supplies of the chip during the whole 2003, says a report.

A report over InformationWeek web-site cites analyst Dean McCarron for Mercury Research who claims that AMD supplied about 70 000 of AMD Opteron microprocessors in the first quarter this year. By contrast, the Sunnyvale, California-based microprocessor maker supplied about 65 000 of its server microprocessors in 2003. According to some other estimates, AMD only sold 40 000 of AMD Opteron products last year.

....

After the market leader Intel Corporation earlier this year adopted technology fully-similar with AMD64 that powers the Opteron and Athlon 64 chips customers and vendors gained additional confidence in products from AMD and that could catalyze the success of the Opteron.


Bron: http://www.xbitlabs.com/news/cpu/display/20040616151224.html

Trouwens, 70.000 processoren, dat lijkt best weinig. Maar AMD heeft hier naar verwachting tussen de 50 en 70 miljoen euro voor ontvangen. De Opteron-omzet zal de andere 3 kwartalen van 2004 ongetwijfeld oplopen. Uitgaande van een verkoop van 400.000 tot 500.000 exemplaren over 2004 kom je toch op een kleine 400 miljoen euro (dus bijna $ 500 miljoen) omzet. Ondanks de relatief lage aantallen gaat Opteron dus een substantiele bijdrage aan de omzet van AMD leveren.

grizzlybeer wijzigde dit bericht 20-06-2004 00:47 (19%)

 
Rabbi
Berichten: 368
Reg. datum: 23 maart 2003

quote:
silentsnow schreef op 19 juni 2004 @ 19:16:
Heb je een linkie naar z'n post? Ik kan het nergens vinden.
[OC & INFO] AMD Athlon64 OC resultaten deel 4

Hier nog een tread op Xtremesystems.org over de nieuwe bios 15: http://www.xtremesystems....ght=abit&pagenumber=1

The only certainty is uncertainty... The only constant thing is change...

Berichten: 65
Reg. datum: 28 februari 2002

quote:
W4rlock schreef op 25 mei 2004 @ 16:09:
[...]


Idd, ik denk ook dat het aan aan de BIOS ligt, want het systeem draait verder toppie. De enige BIOS die te downloaden is op de EPoX Europe site is echter ouder dan mijn BIOS. De EPoX USA site heeft nog helemaal geen BIOS om te downloaden :S

Als ik een oplossing vind zal ik 'm iig hier posten :)
Het is een beetje late update, maargoed. Mijn geheugen draait inmiddels op 200 Mhz... na een BIOS flash (versie 05/14/2004 van de EPoX USA site) de waren de problemen opgelost :)
 
Beth
Berichten: 9.102
Reg. datum: 26 januari 2002

De Pentium 4 3,6GHz aan de tand gevoeld

Besluit: de Athlon 64 3800+ kan met moeite mee met de P4 3,6Ghz, ik ben bang dat hij zal verliezen van een P4 3,8.
Vind persoonlijk dat AMD iets mag doen aan hun rating, ze hebben een goede cpu op de markt maar verneuken het een deel met de rating. (voor de s939 dan) 2,6Ghz voor de 3800+ zou veel beter zijn.

I don't know who this guy is, but I want him and his tortilla's ... DEAD !

Heb je mijn replies al gelezen en de Anandtech review gezien?

Deze review op T.net is echt slecht. De benchmark selectie is echt niet goed.
 
Beth
Berichten: 9.102
Reg. datum: 26 januari 2002

http://www.anandtech.com/chipsets/showdoc.html?i=2088
Ik neem mijn woorden terug :)
In games klopt de FX53 (= ~3800+) de 3,6 er overal uit.

I don't know who this guy is, but I want him and his tortilla's ... DEAD !

Pagina: 1 2 3 4 5 6 7 8 9 10 11 12 ... 18 19 20 21 last


Dit topic is gesloten.


VNU Media logo Powered by True

© 1998 - 2008 Tweakers.net - Alle rechten voorbehouden

Uitgever van: