AMD K8L

http://img98.imageshack.us/img98/6312/k8lcloseupzt4.gif

http://img97.imageshack.us/img97/8839/dedicatedl1pg8.jpg

http://img237.imageshack.us/img237/3934/dedicatedl2yx1.gif

http://img220.imageshack.us/img220/4258/k8lcachel3eb8.gif

http://img213.imageshack.us/img213/6980/htxnt7.jpg

http://img132.imageshack.us/img132/9528/hypertransportwt2.jpg

http://img152.imageshack.us/img152/3893/k8ldicefullsz8.gif

http://img246.imageshack.us/img246/1815/k8ldicehalttx6.gif

http://img137.imageshack.us/img137/7905/k8lfpuul9.gif

http://img218.imageshack.us/img218/903/k8lmoddesignol1.gif

http://img241.imageshack.us/img241/5589/k8lroadmapup6.gif

http://img80.imageshack.us/img80/7215/torrenzajx7.jpg

Read more ->
wikipedia!

x-bit labs

Conclusion in X-bit labs

The future processor from AMD, currently known as K8L, will be the next step in the evolution of the K8 series. The improvements we learned about so far such as increased to 32 bytes instruction sample, improved branch prediction algorithm, introduction of out-of-order reading, will eliminate a few bottlenecks and improve the integer performance. The expansion of SSE instruction sets will help improve the performance significantly in applications dealing with heavy floating-point or integer calculations using SSE instructions, where K8L will be able not only to compete successfully but even to outperform Conroe. The inability to decode and retire 4 commands per clock cycle in some cases may result into tangible performance gaps in integer applications. However, it may not be of that much importance in most cases, because the typical instruction execution pace in real integer applications does not exceed 2-2.5 instructions per cycle because of the data dependence. K8L performance may also be increased in other ways without raising the number of decoding and retirement pipes. Here I am talking about increasing the scheduler queue depth, reducing the number of false dependences (thanks to mechanisms similar to stack engine) and performing out-of-order reading skipping the writes. We do not know today if AMD will be able to implement these features in their new processor. We also do not know if they will modify the caching and prefetch system. However, this is very important for successful competition of the new K8L against Conroe processor in the entire range of tasks.

The successors to K8 architecture require a lot of work. They need not just to compete with Intel solutions, but to outperform them, and that is why their architecture needs to be free from the K8 bottlenecks and has to be able to decode and process up to four instructions per clock. Alas, developing and perfecting a new core is a tremendously laborious and time-consuming task. Even if a new architecture is under development at AMD, it will hardly become market-ready in the next couple of years. It’s hard to tell what this rapidly developing industry will have become like by that time!

Dynamic I/O bus introduced in K8L architecture

AM2 get upgraded to AM2+: AM3 is postponed!
Gud news + Bad news :!

AMD to enter K8L era in 2H 2007

AMD Quad-Core Altair upcoming in 2007 Q3
Here AMD claims more than 40% improvement! Now is it truely possible?


If u have more Info on K8L. Link it here!
213 answers Last reply
More about tomshardware
  1. Clearly, K8L is something of a revolutionary platform. The processors themselves aren't anything new, but HTX and Torrenza... that's some cool stuff. And it might just be enough to kill Conroe/Woodcrest (as far as platforms go).

    Something else I've noticed is enhanced RAS. RAS is sometimes more important than CAS. If they manage to do something cool with RAS, K8L could see low latencies with existing DDR2 memory.
  2. Quote:
    Even if a new architecture is under development at AMD, it will hardly become market-ready in the next couple of years.


    ?
  3. Quote:
    Something else I've noticed is enhanced RAS. RAS is sometimes more important than CAS. If they manage to do something cool with RAS, K8L could see low latencies with existing DDR2 memory.
    _________________



    They mean Remote Access Services.
  4. I knew that! ;-)

    Seriously, I thought they were refering to Row Address Strobe.
  5. Quote:
    I knew that! ;-)

    Seriously, I thought they were refering to Row Address Strobe.


    Both wrong tbh.... RAS = reliability and stability

    http://www.eweek.com/article2/0,1895,1844683,00.asp
  6. It's better than Intel's current products for sure... I would just be wary of what Intel may pull out of the bag, 45nm before the end of 07 is possible from Intel....
  7. Quote:
    I knew that! ;-)

    Seriously, I thought they were refering to Row Address Strobe.


    Both wrong tbh.... RAS = reliability and stability

    http://www.eweek.com/article2/0,1895,1844683,00.asp
    quite no!
    BaronBS is ALWAYS RIGHT, remember: ALWAYS RIGHT!
    mpjesse=noob jesus :)
  8. Don't post heaps of pictures like that you stupid noob.
  9. awww poor 56ker complaining about pics.

    lol
  10. Quote:
    Don't post heaps of pictures like that you stupid noob.


    Did he step on any Intel fanboy's nerve?! :lol:
    Your intel couldn't give us not even these diagrams for 3 years so go and learn how to talk to people; good manners I mean, you really need it.
  11. Quote:
    Don't post heaps of pictures like that you stupid noob.


    I appreciated the pictures and found them informative.......noob!

    Very exciting this kay eight el. I don't think it will be a Conroe killer, but AMD has a lot of cool things cooking up. I found the two most interesting to be:

    DICE
    Dedicated L1, L2, shared L3
  12. Quote:
    Something else I've noticed is enhanced RAS. RAS is sometimes more important than CAS. If they manage to do something cool with RAS, K8L could see low latencies with existing DDR2 memory.
    _________________



    They mean Remote Access Services.

    Fail

    nelahem would beat K8L by price to performance ratio
  13. Quote:
    Hey, NetBurst was suppose to turn your copper wire into optic fibre and speed up your Internet!

    Yes...But it just turned the PC into a toaster. P3/K6 sucessors were the only evolutive winners.
  14. Cool. Too bad I can't understand most of it :oops:
  15. certainly K8L will beat intel's conroe,but thats not point of discussion becasue main question is when AMD will come out with it? and Will intel be able to switch to 45mn from current 65mn by then and what about new architecture in first half of 2008 and intel is hiring few professional with expertise in graphics development, is it just for ati alternative or something else. lets not forgets that intel can put memory controller onchip, atleast they had sample of it, in 2005 spring IDF.
  16. By the time K8L comes out, or even paper-launched, Intel's products would be based on Penryn; not Conroe.

    But still, I think the K8L would be better; as it would be released about a year after the Core architecture.

    But then, Core is supposed to be replaced with Nehalem in 2008...
  17. Quote:
    certainly K8L will beat intel's conroe,but thats not point of discussion becasue main question is when AMD will come out with it? and Will intel be able to switch to 45mn from current 65mn by then and what about new architecture in first half of 2008 and intel is hiring few professional with expertise in graphics development, is it just for ati alternative or something else. lets not forgets that intel can put memory controller onchip, atleast they had sample of it, in 2005 spring IDF.


    To me, K8L is more of a platform than it is a processor. And to my knowledge, Intel doesn't have anything CLOSE to what the K8L platform will be.

    So to me, what Intel will have in 2008 to compete w/ K8L is irrelevant if they keep using the same chipset platform which hasn't really changed since the original P4.
  18. Quote:
    Fail

    nelahem would beat K8L by price to performance ratio
    So you're saying both Intel and AMD have sent you engineering samples of their new chips, and have already told you the pricing for them once they're available?
  19. Quote:
    Don't post heaps of pictures like that you stupid noob.


    Why not?

    Atleast those were informative... unlike ur stupid post!
  20. No you stupid noob. I'm quite certain its against the tos and its poor netiquette.
  21. Quote:
    Did he step on any Intel fanboy's nerve?!


    And you're pulling this assumption from your ass? Moron. Don't you mean stepping on toes rather then nerves? Dipsh!t.

    Quote:
    Your intel couldn't give us not even these diagrams for 3 years


    WTF are you talking about moron? Did you miss all those IDF shows? :roll:

    Quote:
    so go and learn how to talk to people; good manners I mean, you really need it.


    Go learn to not be such a dipsh!t, you really really need it fanboy.
  22. No noob the slides are old and its against the tos.
  23. wusy:

    Seriously.
    Where can I find info on the VIA Cyrix ?
    I remember years back, the Cyrix benchmarks gave people with weak hearts - complications!
    How are things different now? I need to care about this if VIA will start using lots of it's chips in UMPCs..

    Thanks.
  24. Careful action man ur starting to sound like MMM.
  25. You might wana tone it down a bit AM, as of late Ive only been seeing harsh insults. Lets keep a healthy mix of disscusion and wity insults, eh?
  26. Blah, they shouldn't allow kidies to talk here.
  27. Quote:
    Careful action man ur starting to sound like MMM.


    Yeah, after conroe took the performance crown he wants to take the FORUM_JOLLY crown.
  28. Quote:
    Blah, they shouldn't allow kidies to talk here.


    What are you doing here then dipsh!t?
  29. Quote:
    Blah, they shouldn't allow kidies to talk here.

    What are you doing here then dipsh!t?

    Just being insulted by an angry teenager :lol:
    Go ahead, some more efforts and the crown is yours... :lol: :lol:
  30. Quote:
    No you stupid noob. I'm quite certain its against the tos and its poor netiquette

    You should read the TOS yourself.

    Quote:
    RULES OF CONDUCT

    User agrees not to post material that is knowingly false and/or defamatory, misleading, inaccurate, abusive, vulgar, hateful, harassing, obscene, profane, sexually oriented, threatening, invasive of a person's privacy, that otherwise violates any law, or that encourages conduct constituting a criminal offense.
  31. Ah the good old labeling based on assumptions by the old fella. Good work gramps. Heres some useful info for you, just because someone is older doesn't mean they're better in any way. Are you over 50? Sure sound like it.
  32. Against stupid people who wont post here again its quite acceptable.
  33. Quote:


    RULES OF CONDUCT

    User agrees not to post material that is knowingly false and/or defamatory, misleading, inaccurate, abusive, vulgar, hateful, harassing, obscene, profane, sexually oriented, threatening, invasive of a person's privacy, that otherwise violates any law, or that encourages conduct constituting a criminal offense.
    Pity....That rules out about 98% of posts. :wink:
  34. Quote:
    just because someone is older doesn't mean they're better in any way.


    Right, but if someone insults people who bring info about K8L, that means he's definitely out of his brains.
    OOOOO You source of all knowledge, after taking back all the BS that came out of your mouth, why don't you tell us this and put a stone over it: Do you think that K8L will outperform Conroe or not?! That's the only point here, otherwise go cheating with your neighbours, this is not the place.
  35. Quote:
    Right, but if someone insults people who bring info about K8L,


    Sigh again you totally miss the point. Its about posting all the pictures not bringing rehashed K8L info.

    Quote:
    after taking back all the BS that came out of your mouth,


    The irony.

    Quote:
    Do you think that K8L will outperform Conroe or not?!


    How should I know? What a fanboy, thats not even the issue at hand and you have to bring it back to that. True AMD fanboy. They shouldn't let people in homes near comps with the internet.
  36. c'mon, let's stop with the "compliments" esspecially action_man. no need for that.

    by the way, i think K8L could be a succes if AMD will succeed in implementing the processors with the new chipsets provided by the newley bought ATI. the core power management, is something close to what intel introduced on conroe... but i like the ideea of dedicated L1 and 2 and a big, shared L3 cache.

    by the way, i'm not a fanboy so don't call me that way.. i'm more of a "performance whore" migrating from company to company :))
  37. It's not out yet but you surely can understand by the diagrams how that compares with Conroe. Would you say that even the people @ X-Bit labs are Stupid Noobs?! :lol:
    Beople here talk about Nelahem (or what it's called) and BearLake-X, plans of 2010 while this thing is boming out next year. C'mon, be so kind to give an opinion :twisted:
  38. Diagrams are good and all but benchmarks is what counts at the end of the day. Yes it'll be fast but fast I have no idea.

    Quote:
    Would you say that even the people @ X-Bit labs are Stupid Noobs?!


    Again with the assumptions from your ass?
  39. Ya, it's still much cleaner than your mouth!
  40. Wise words.
  41. Quote:
    Do even know the history of how the first A64 and Conroe proved themselves to be worthy in their respective era?
    Not through some stupid flow diagram, but by credible benchmarks from the first ES.

    No benchmarks from ES, no game.


    :lol: If we had the benchmarks we wouldn't need to share opinions (Do I have to explain the difference between facts and opinions?) The game will be over when we have the benchmarks!
  42. welll... then we'll just have to wait for the benchmarks. damn it's gonna be a long year:D but interesting:P but anyway, diagrams give you a huntch of what's to come, on that you can agree!
  43. Theres no helping you gramps. BTW if you're so much older and wiser then us mortals how about you share your age.
  44. Stating my opinion does not make me a fanboy. I see wider busses and executin units, add the 65nm shrinking and I THINK K8L will be better than Conroe. Anything ba about this?! :roll:
    This does not mean that AM can pour his BS on me allover this thread!
  45. Quote:
    Theres no helping you gramps. BTW if you're so much older and wiser then us mortals how about you share your age.


    Of course, I am 61.


    :lol: Just kidding, I am just 22, but it means nothing. Here the problem is not age; it's your arrogance.

    P.S: Who the heck is giving me a single vote for each comment?!
  46. Only 1 year older then me, thats rich. Your problem is your stupidity, cant really offer any advice expect do the opposite of everything you currently do. If it doesnt help it'll at least be amusing.
  47. Quote:
    Only 1 year older then me, thats rich. Your problem is your stupidity, cant really offer any advice expect do the opposite of everything you currently do. If it doesnt help it'll at least be amusing.


    I don't mean to offend, but read again what you just wrote...It's one of the most stupid things I have ever read.

    Btw, no need to offend to get a point shown (you did that when you said that only diagrams won't show how a system will perform).
  48. Quote:
    ..Your problem is your stupidity


    I know. I've been chatting with you for hours; a person that at the age of 21 does not know what is an opinion and how to respect it...
    I HAVE to be stupid :(
  49. No you're stupid and talking crap to someone smarter then you.
Ask a new question

Read More

CPUs Performance AMD Integer