Printable View
We never seen a K8 65nm ES (Only one from some guy in japan about 8 months ago) yet the production chips are out in dells and hard to get. They kept them secret. We will never see a K8L until production but ES's of them have been made since it was taped out in the end of 2005 so ES's are out for chosen few that will not say a word to anybody. Prob been out for a few months.
there were very little 65nm k8 es for a reason, in that there was no reason for alot of es as it was just a die shrink. like i said in a thread before, there were plenty of am2 es around, as it required new mobo logic, as would k8ls since new am2+ mobos are going to be released, so if there were actual es samples out in any meaningful quantities we would have at least heard of them. also, k8l was taped out this year, i dont know where you heard 2005...
The 90nm is 20W TDP, even if the transisition is better than AMD's poor 90nm-65nm transition, it's not going to go lower than 10W.Quote:
Originally Posted by nn_step
It's a terrible value since it has only 4 DIMM slots, reducing its usefulness for any significant 2S applications.Quote:
3) Don't be so quick to judge. It is still a wonderful value for a good 2P system.
I dont think there are K8L out since 2005 or i should at least heard about them or know someone that has one.Quote:
Originally Posted by Serge84
Don't you mean 2006?? ;)Quote:
Originally Posted by metro.cl
i did a quote he said 2005 :confused:Quote:
Originally Posted by Lightman
Considering that AMD just demoed the first working K8L silicon this month and all it could do was run task manager, I think it's pretty unlikely that AMD has been sending out ES chips for months.Quote:
Originally Posted by Serge84
Why is it that AMD fans seem to always expect soooo much more than AMD promises or delivers? Cut AMD some slack guys. It's hard enough for them to meet their own goals without their fanboys adding greatly exagerated performance expectations and cutting months off their timelines.
Just because you don't hear about them doesn't mean they aren't thereQuote:
Originally Posted by Fred_Pohl
Just because something is possible doesn't mean it is likely.Quote:
Originally Posted by nn_step
true, kind of like how the average Conroe overclock isn't ever going to hit 5GhzQuote:
Originally Posted by Fred_Pohl
...or how the average K8 overclock is likely never to hit 3GHz. :p:Quote:
Originally Posted by nn_step
actually it is the average for Single cores, not quite the dual cores but then again you never specified that. :p:Quote:
Originally Posted by Fred_Pohl
right in case of 90nmQuote:
Originally Posted by brentpresley
not right for 65nm
Yes because 65nm donīt even reach 3Ghz :p:Quote:
Originally Posted by MAS
i know 3100 (xtremesystems) and 3180MHz (Belgium) results with HT below 1000 (right) and several 2800-2900MHz results with HT above 1000 (suspicious) :)
http://www.matbe.com/articles/lire/3...90nm/page3.php
http://www.xtremesystems.org/forums/...d.php?t=126558
Quote:
Originally Posted by MAS
It's just too bad a 3GHz 90nm K8 would probably match or even outperform these 3.1 - 3.2GHz 65nm K8s. Even your own X2 @ 2.7GHz would come pretty close. ;)
http://www.firingsquad.com/hardware/...view/page5.asp
90nm 4600+ > 65nm 4800+. :slapass:
L2 latency increase = :nono:
I highly doubt that. I belong to another forum where the members aren't as 'Xtreme' as this place and the avg 1C K8 OC there is only ~2.7GHz. Only one member ever hit 3GHz with an Opty 148 and that was with the case side panel removed and the air-cooled PC sitting in an open window in mid-winter. He refused to run most BMs but he did manage a few and some CPU-Z screen shots. The highest benching water-cooled 1C K8 OC is a Opty 146 @2.8GHz but he runs it at 2.6GHz for daily use. The avg 2C K8 OC there is ~2.6GHz and the avg C2D OC is ~3.2GHz for a performance delta of ~2000+ on the AMD PR scale. :p:Quote:
Originally Posted by nn_step
From what I could tell, the FX-74 @ 3GHz was 11% faster than a 2.66GHz Kentsfield in Cinebench 9.5. While in 3Ds Max 9, the Kentsfield was 15% faster than the FX-74.Quote:
Originally Posted by nn_step
I find it strange therefore that you would use the word "soaring" while describing the Quad-FX's lead, while choosing not to use that word to describe the Kentsfield's lead (even though it was bigger). Based on the numbers in the benchmark, a Core 2 Quad @ 3.0GHz should get 1620 points in Cinebench, beating the FX-74 at the same clockspeed.
Not exactly soaring, and this is in a 64bit operating system using a benchmark where the K8 traditionally does well.
The Opteron 100 series o/c's are highly dependent on steppings, as in most cases. Many, many people here @ XS were able to get to 3ghz stable w/air in the early runs (i.e., CABNE and CABYE). I haven't been paying much attention to what people are getting with current steppings though.Quote:
Originally Posted by Fred_Pohl
Edit: I just checked the 207 page opty o/c thread. They're still not doing too bad. 3107mhz @ 1.4v on air
Just like what AMD said about 65nm, they did exactly what they said they would. Bring a cooler less energy cpu out. They never said it would perform any better. The slower ns in the cache is due to them thinking of releasing current chips with more cache if needed. Said by AMD.Quote:
Originally Posted by Fred_Pohl
I am clearly not a AMD or Intel fan anymore. I am going to get a mATX board as soon as I find one I like or just get a NF680 775 atx and see what my E6400 can really do. I don't expect anything from AMD or INTEL it comes out when it comes out. Don't get the wrong ideas about what I think of the situation because frankly you know nothing about how I think from a performance prespective.
I am all about performance now no matter the cost. Getting ready to build a extreme liquid cooling system for my comp and would love to use it on my C2D.
Oh really? Then I must have a rare chip or maybe its just my skill of getting 50% OC's from a limited chip like AM2's are. 939's got past 3ghz all the time back in the day before conroe but how many times has that been said. You where here you should know and so was I.Quote:
Originally Posted by brentpresley
I reached 2.96ghz on a A64 3000+ 754 before I even had a AM2. And I also reached 2.5ghz on a 2500+ sempron but the memory controler on it was a dud so it was only a 75% OC.
Do we remember the world record data base? Also the new one. Thats a lot of rare chips ya know. Most people can't reach the cpus limit because they are afrade of frying it at a high voltage or it gets too hot. The people that can afford to push there chips are more likely with the right cooling to reach much higher then 3ghz but there are not that many people that can buy or build phase coolers. Thats a more reasonable anser.
Unless your comparing AM2 and leaving out the 939. We all knew that was the king of OCing before conroe but conroe is king now. That doesn't mean you can say this and that to something that wasn't true. 939's are better then AM2 before AM2 came amd made kick ass 939's now the AM2's are just crappy. I don't know what they did but the 939's where made superiorly compared with AM2 chips.
We can show plunty of AMD's that run past 3ghz up to 4.2ghz exactly. But that was way too long ago and forgoten in time because of C2D. C2D is the best but before that K8's where king for 3 years for a reason. No P-D could compare period even OCed to 7.2ghz. It could not touch a 3.8ghz FX-60 back in the day because that is the equivilint to a 8ghz to 10ghz P-D. Now AMD is just slower at the moment, no big deal to me.
http://enthusiast.hardocp.com/articl...VudGh1c2lhc3Q=Quote:
Originally Posted by Theli
According to this it was only 1 to 3% faster. But all I will hear is. Man no it wasn't it was 20% of 50% faster. Really? Show me where. Realisticly C2D and C2Q is faster but not by such rediculessly high numbers. Its only 17% faster in 32-bit mode and about 10% faster in 64-bit. But still C2D has much of its advantages in OC is why C2D is much better then AMD when AMD can't get much more then a 50% OC C2D's do higher then 100% OC's on low end models.
ummm wow that is really ironic you know.Quote:
Originally Posted by brentpresley
Considering you posted in a few threads about a year ago on how people's K8's should Hit 3Ghz easy. :stick:
Get your story straight
I'm rounding here, so accept the tolerances of being human :slap:Quote:
Originally Posted by brentpresley
http://valid.x86-secret.com/show_oc?id=132953Quote:
Originally Posted by brentpresley
http://valid.x86-secret.com/show_oc?id=132951
:l Theres your 3ghz from me. You forget fast don't you? Short memory span. Alsimers getting to ya old timer?
http://valid.x86-secret.com/show_oc?id=108727
4ghz validated.
http://valid.x86-secret.com/show_oc?id=94127
4.17ghz validated.
http://www.hwbot.org/
http://www.ripping.org/database.php?act=records
Don't hold your breath too long it might kill ya when I already made my point. Besides that much mhz is nothing when its that much under 3ghz come on dude it was a 754. But now I just showed you a 3ghz validation for me. Your point means nothing since I proved otherwise. And with liquid cooling in oil with nitrogen pellets I beleave my chip will go higher. I stopped because that is very hot for air with the voltage thats required I need atleast water to reach 1.8v/1.9v at a decent temp.