Nice attitude... just what im talking about. :rolleyes:Quote:
Originally Posted by JCornell
Printable View
Nice attitude... just what im talking about. :rolleyes:Quote:
Originally Posted by JCornell
Simple question. Ur building a machine and got 2 alternatives. They cost the same but one offers abit more performance than the other. Wich would u take?
Yes you're right!Quote:
Originally Posted by informal
http://www.pcworld.com/businesscente...hip_plans.html
Quote:
AMD's 12-core chip will include two 6-core processors on separate chips in a single processor package, McCarron said. That is a more realistic goal than including 12 cores on a single chip, which can be expensive to manufacture, McCarron said.
Ah ... enough with that kind of expression ... I don't give a dxxx, fact is a fact ..Quote:
Originally Posted by eXa
Won't you agree more for what I meant above about "Numbers" ? ...
Wow... When are those MCM to be released?
These are just the start. I'm sure if HP and Dell ask for different configs AMD will comply. OEM's have made such requests in the past. If they want a cheaper MCM with 2 Quads, AMD is not going to say NO.Quote:
Originally Posted by bowman
Right now i dont care about numbers nor performance. My point is that u can compare the 2 systems. Heck u can compare it to MMs rig, the pentium D rig or whatever. As long as they can do the same job u can compare it. It is really as simple as that.Quote:
Originally Posted by JCornell
http://pc.watch.impress.co.jp/docs/2.../kaigai_1l.gifQuote:
Originally Posted by Seraphiel
Hopefully these rehashes will keep them afloat in the server/HPC space until they can put a mirror shine to Bulldozer, and then they might get interesting for desktops again.
Bit more performance in how many apps? Surely you're not talking about the limited tests done here, right? Again, does the TLB patch mean anything? Sure prices for processors with known bugs costs less. Too much is unknown to draw a conclusions from pro or con for either of the rigs talked about.Quote:
Originally Posted by eXa
Oh yeah, sure, nothing wrong about it ... Already compare one to another right ? ... :clap: You can see all the complete charts ...Quote:
Originally Posted by eXa
But at the end, who mentioned about numbers of threads compete to real cores ? This is just about what we all discuss about ...
I think a lot of people are getting caught up in arguing the semantics instead of looking at the values for what they are. If LIKMARK wants to see how a 4 Physical core/ 8 Thread Nehalem UP system compares to an 8 core/ 8 thread Barcelona DP system. Whatever, the numbers fall where they do and we all move on. Most people may not find this comparison very valuable due to significant hardware setup differences, but I can see where He finds it important. Sure it isn't a fair comparison, but it doesn't have to be as long as the details of the comparison are on the table and you know what you are getting. The only thing that LIKMARK has done that I might find fault with is that he expresses disappointment with the 8 Thread numbers from a 4 core Nehalem, but doesn't ever explicitly state that he understands why an SMT system would be expected to be lower than an 8 core system (Gainestown). Either this was oversight on his part or he simply didn't understand the architectural differences of SMT vs. physical core.
U dont get the point. All im saying is that u can compare the 2 systems.Quote:
Originally Posted by Donnie27
Im not saying u should buy the "old" amd system, id go with nehalem myself.
Shure the amd system may beat the nehalem in 8 threaded apps, but nehalem crushes it in singel, double and even quad id guess.
But that is not my point.
We all get the point, but some of it just won't listen, thats all ...Quote:
Originally Posted by eXa
If you want to compare Bloomfield with 2*Barcelona, its ok ... you do whatever you want ... But if you loudly mentioned about why Bloomfield result is sucked while compare to 2*Barcelona just isn't right ...
Still it related to "Numbers" right ? ...
PS : No one stop you to compare between system, all the charts is being made by Hornet311 ... so what do you want now ?
Ur attitude aint very good either. What u could have said instead of bashing is something like...
Nice to see ur tests, good to see how well the 4core nehalem actually holds up againts the 8core barcelonas.
If he aint personally impressed, so what?
Impressed to what ? 2*Barcelona ? ...Quote:
Originally Posted by eXa
Oh ... how is this gonna end ... Help plz ...
I think my attitude is fine though, From the first place I don't know it could be compare to Bloomfield, so I provided Gainestown 2.4G(hey just the same to Barcelona, 8 threads vs 8threads) ...
Meaning, why don't pick your own rank to compete ? ...
At the end, just isn't right, how about compete it to Harpertown ? Reasonable right ? ...
PS : I'm suspected this is just a friend of that guy or just about the same guy with different ID ? :rofl:
Yeah nice attitude idd. I dont know Likmark at all.Quote:
Originally Posted by JCornell
Rank..? are u a moron?
If ur building a server is "rank" what ur after? no what ur after is performance, price, price to run and so on. Who cares how many cores it is? its irelevant.
JC i think you should calm down a little bit, you did an awesome job with this nehalem benches and really contributed to this community by taking requests of what to bench and how to bench.
But your recent posts sound quite hostile to me, no disrespect but there are better ways to reply. I think it is pretty obviouse now that some guys want to see different numbers or comparrisons but they can do this on there own and can take the material you provided as a base. Dont get yourself in useless bickerings.
I hop you dont missunderstand this post, its nothing personal against you, i just try to view this in a neutral way.
OK, I'm a moron but I'm :with: :rofl:Quote:
Originally Posted by eXa
Bloomfield is not a server ... And Hey, Nehalem isn't in the market yet ... why you bother to compare the price ... And other things, some believes AMD has nothing else to compete with except prices :up:
See "numbers" again, thus he just denied who cares about how many cores ... :rofl:
:dammit: ... Why I'm here reply to you ... still it's just same brain damage but a different guy ...
Thanks ... I'm cool ...Quote:
Originally Posted by Hornet331
:)
I guess the new tactic to justify for inferiour performance is to do the 2vs1. Or if AMD ends fabless it will be good! Not even to think about those K10s uses twice the power to do...less? But its ACP! We invented that because its good for everyone. nobody runs full loads!
Who the hell educate these AMD diehard fanboys? The former Iraqi information minister? Not even Jihad people are this dedicated.
And all those praising themselves holy. Ohh, Intel fanboys always threadcrap and blabla. What is this?
try new version of winrar 3.71, this is old, not corecctly for dual and more CPUs...Quote:
Originally Posted by JCornell
to download trial version here: http://www.slunecnice.cz/sw/winrar-3/stahnout/
Wow, that is a very interesting pic. It shows Bloomfield running at 2.7GHz which I don't believe is one of the three stock speeds. I suppose that means there already exist board bioses that allow for changing the multi or QPI frequency. I know this was already known, but it's the first actual hard evidence I've seen of it.Quote:
Originally Posted by Chri$ch
EDIT: Nevermind, it is 2.66GHz. I was seeing things.
Was about to comment that, but im not going down to ur lvl.Quote:
Originally Posted by JCornell
XS is idd going down the drain, well maybe not but the quality just aint here anymore.
Im going back to the WC section, its a little better there than the news section anyways.
Maybe I was not meant to carried on from AndreYang posted redirect from XFastest.Quote:
Originally Posted by grunge100
Some Taiwanese posted here will just make things get worst ... This incident ever occured right ... If some know what I meant ...
I think I'm missing something...
I thought Bloomfield line up would take QX9770, QX9650 and Q9550 places on the market, that would mean I'd have one in my case, not in a server rack. Wouldn't be a fair comparison against "the old" QX9650? And maybe 9950BE x Bloom@2.66?
But I don't blame peolple for doing all kinds of comparison, there's lots of perpectives on the matter, we could just try to be polite on comments or keep mouth shut.
PS. Thank you JC for the benches, I hope you keep us posted