Thats not 6 months ago. You got here 7 months ago. That would make you wrong. =PQuote:
Originally Posted by brentpresley
Printable View
Thats not 6 months ago. You got here 7 months ago. That would make you wrong. =PQuote:
Originally Posted by brentpresley
Ha ha lol XDD Ah man got to love it here at the forums. I was being sarcastic, not trying to make any kind of comeback. Atleast Intel finally caught up after 3/4 years agenst a 3 or 4 year old arc. What do you expect magic ferry dust? The earth was not made in a day, nore are companies or new cpu arc's for that matter.Quote:
Originally Posted by brentpresley
Next thing I'll be saying is in 6 months, boy didn't AMD make a comeback in 2007. I bet in 2008 Intel will come back as king after that and I'll switch back and forth all day and night every year just to have the better system. Cool how it works like that. One min AMD is the best then Intel then AMD like it is with DAMMIT and Nvidia. The War goes on forever or it would be no fun.
i have to agree with brent here... throughout all of k8's history, the percentage of chips hitting 3ghz on air is very low. in their recent history, its decent with single core opties hitting 3ghz probably half the time if not more, and dual core opties hitting 3ghz about 20% of the time. thats still far from c2 duo, and its amd's fault they dont have an answer to conroe, theres no excuse for it, they canceled two projects...
I have no doubt that many people have 3GHz K8s but I also have no doubt that the average K8 OC (1C or 2C) is far less than 3GHz. I'm no expert on K8 overclocking but I know plenty of people who couldn't get more than ~2.7GHz from a 1C or ~2.5GHz from a 2C K8 with good air-cooling. I'm talking 24/7 stable of course, not suicide screen shots.Quote:
Originally Posted by xenolith
That was my point. AMD never claimed that Brisbane would clock 50% higher than Windsor, that it would feature 4MB of cache, RHTT or any other performance improvements. All those rumors came from AMD fanboys who, IMO, do AMD more harm than good with their unrealistic expectations and fantasies that AMD can't deliver on.Quote:
Originally Posted by Serge84
I didn't mean to imply that you are an AMD fanboy. I was merely commenting on them in general. It's not like you were making, spreading or defending any of the aforementioned fanboy predictions and rumors, right?Quote:
Originally Posted by Serge84
Maybe we should do an AMD myths of the year. I got the little idea from theinq, since Charlie is debunking all the AMD ideas they never materialized for the work of fanboys and others. So I would like to make alittle stand here for all the AMD people that was mislead by rumours and false information and putting the guildt back to where it came from.
http://www.theinquirer.net/default.aspx?article=36580
We all know the synthetic thing dont hold water, since games are heavily affected. And we all know how much on time the 1year delayed 65nm parts are...Quote:
That brings us to the last big introduction of the year, AMD's 65nm parts. Contrary to fanboy beliefs, it released them when it promised months ago, to the day. They had better power usage than their predecessors but little if any speed difference. The longer latency caches have basically dropped the performance by tiny bits on several synthetic benchmarks, blink and you won't notice the difference
http://www.firingsquad.com/hardware/...view/page5.asp
Anyway, lets start:
Myth 1, lots of cache for K8 is the thing! 4MB FX62 and Opterons!
http://www.theinquirer.net/default.aspx?article=29738
http://www.theinquirer.net/default.aspx?article=28639
Charlie seems to personally owe us one on that one.
Myth 2, 65nm K8s will be 40-50% faster!
http://www.theinquirer.net/default.aspx?article=32322
Didnīt hold water....
Myth 3, Reverse Hyperthreading!
http://www.theinquirer.net/default.aspx?article=31022
http://www.theinquirer.net/default.aspx?article=32589
http://www.theinquirer.net/default.aspx?article=32790
Myth 4, K8L in Q1 2007!
http://www.theinquirer.net/default.aspx?article=32948
There are a few more myths ofcourse. But I find it funny that the same site making them calls everyone else fanboys. So no RHT, No large cache and no superclocked K8/K8Ls to be expected.
He usually answers with quite a temper when he is put to a corner. While calling you all kinds of things while making you remember that he is the god. And gods can never be wrong and is always right.Quote:
Originally Posted by brentpresley
maybe we should include nn into this list to
You are right.Quote:
Originally Posted by Fred_Pohl
lolQuote:
Originally Posted by brentpresley
AMD reported 20 percent dual-core K8L performance boost in comparison to dual-core K8 (obviously at equal freq)
http://www.watch.impress.co.jp/akiba...ibax2006b.html
Here is an excerpt from a recent email exchange between myself and Charlie concerning his recently expressed disdain for hardware reviewers who expected too much from Brisbane.
My email:
Charlie's reply:Quote:
Perhaps the reviewers put too much faith in The
Inq's bull:banana::banana::banana::banana: rumors and were therefore
disappointed that AMD didn't deliver on any of them?
Perhaps they expected the 50% higher clocking
Brisbanes with 4MB of ZRAM L2 cache and the 25%
single threaded performance boost from RHTT that
your rag wrote about? Perhaps they even expected
them a lot earlier as you hinted at back in March?
You AMD PR whore hacks never fail to amuse me. :)
http://www.theinquirer.net/default.aspx?article=32322
http://www.theinquirer.net/default.aspx?article=29738
http://www.theinquirer.net/default.aspx?article=30638
Quote:
--- Charlie Demerjian <charlie@stonearch.net> wrote:
Once again, you prove your ignorance.
^^^
Charlie gets owned. :banana: :clap: :woot:
obviously I'm missing something. sure we all know charlie but what does he have to do with this topic? care to fill me in?
Because he helps the rumour mill and even fuels them personally about things that never exsisted.Quote:
Originally Posted by biohead
In other words, he is one of the people responsible for all the imaginary crap.
Would that be compared to the lower performing 65nm K8s or 90nm? And in what and where. Sythetic bench, or real world average. And if 20%, then there is along way to the goal when K8L is basicly just C2Ds at 2.5Ghz at 125W TDp.Quote:
Originally Posted by MAS
well i dont care if you think i am an AMD fanboy,but you are always talking negative about AMD,why?Quote:
Originally Posted by Fred_Pohl
if intel is faster than AMD next year,i will buy an intel quadcore rather than an AMD one,i really dont care about the brand, as long as my pc is fast
my point was, if you would have know something about Z-ram,you wouldt have send that mail, because the first generation of Z-ram only worked on 100MHz,while the second (recent) generation works on 400MHz,still a bit too slow to implement into modern cpu's,maybe next generation Zram wil find its way to cpu's
btw,i just read some articles from Charlie,and i have to agree with him,you really prove your ignorance
AMD and IBM have stated publicly that this technology allows for a 40% faster switching transistor than from a standard 65 nm design without all the three letter acronyms (TLA’s). In a complex design like a CPU this could mean a theoretical 50% overall clockspeed increase going from AMD’s 90 nm process to AMD/IBM’s 65 nm process all the while staying within the same power envelope.
does he say amd will produce a CPU with a 50% increase in clockspeed?? NO
The next gen of AMD FX chips will have the full 'big cache' Opteron level of L3 cache, most likely 4MB.
He doesnt say a word about ZRAM,and he is only 2MB off,K8L will use 2MB L3 cache
AMD IS GOING to announce 65nm parts next week at a gala event. We are not supposed to know about it, so don't tell anyone that we are aware of it. About #*$*(% time.
Well he may be a bit wrong here,but if he meant giving info or something about 65nm part he might be right,but this article isnt of any significance
I think i proved my point
NOW CAN WE PLEASE GET BACK TO K8L RELATED INFO INSTEAD OF THIS PURE BS
How Brisbane related to K8L? Have you mixed up Brisbane and Barselona, haven't you??? :confused:Quote:
Originally Posted by Fred_Pohl
What you and Charlie expected from Brisbane other than a shrink, reduced power consumption and potentially higher OC ?
AMD reported 40% per average increase compared to K8 under certain conditions, and this was relatively recent, at Dec Analysts Day.Quote:
Originally Posted by MAS
But probably those chineese know more than AMD :P
amd never said what it compared to k8. it could very well be that they compared quad core star to k8. as a sidenote, please refrain from calling star k8l, because that is NOT its codename.
40% is probably for FP codes .
Are the K8L Es here yet, the thread to long to read through.
Thanks
Quote:
Originally Posted by GoThr3k
If my skepticism of ludicrous fanboy fantasies, FUD rumors and overly optimistic vaporware predictions equates to "always talking negatively about AMD", I am guilty as charged. :D It's not that I have anything against AMD, it's just overly optimistic fanboy rumors that annoy me. I think that AMD can and will improve on both their 65nm process and the venerable old K8 core but it won't happen over night and the ideas that AMD can ever produce 50% higher clocking K8s or improve IPC by 40-50% over K8 with their stars CPU lineup are ridiculously optimistic IMHO.
AMD using ZRAM as CPU cache wasn't my idea, dude. I know that ZRAM is not yet ready for that and maybe never will be. The only point you proved is that The Inq didn't directly state that 50% faster Brisbanes would launch in June 2006 with 4MB cache and RHTT. I never said that they did but The Inq did start the rumors which were then exagerated and parroted by AMD fanboys all over the WWW. It isn't hard to see how overly optimistic fanboys would interpret what The Inq wrote and unless The Inq's writers are incredibly stoopit, they knew what they were implying and how it would be interpreted when they wrote it.
There are a lot of other bogus Inq rumors that support what I'm saying but I don't have the links handy or hours to spend searching for them. If you choose to ignorantly believe their FUD, be my guest.
Now let's get back to talking about those vapor K8L ES chips that didn't show up in December... :p:
The Akiba article said that (my understanding at least): AMD representatives told the press that the level 3 cache of dual core Stars family(Kuma) would bring roughly 20% increase in performance(probably average figure,dependant on the apps) over previous K8 generation X2(dual core,ie Windsor or Brisbane)!Quote:
Originally Posted by brentpresley
Note that they emphasised cache in this information.
They were clearly talking about the addition of the L3 cache and the effects of it on the New Core,compared to the older K8 architecture.We can assume that the New Core benefits more from the shared level 3 victim cache due to it's largerly improved execution units and imroved design,overall.
They didn't say that Stars will be 20%(overall ) faster than K8!At least we can't conclude they said that only because of this one article(written in Chinese btw).
Although some people here believe that they have been around for several months under super-duper-top-secret NDA, IMO we won't see them until sometime in Q1-07 with a paper or possibly even a hard launch sometime in Q3-07.Quote:
Originally Posted by day187