Saturday, February 23rd 2008

Intel Planning Six-Core Processor, Will Call it 'Dunnington'

Intel is planning on serving a heaping pile of pain to AMD's revenue/stock figures again in a few months, by developing a six-core juggernaut. While AMD is still tweaking on a way to merely get four cores to work in tandem, Intel is hard at work shoving two more cores on one die. This six-core monstrosity will be succeeded by the even beefier Nehalem micro-architecture, which could have up to eight cores on one die. Most of the Dunnington project is still top-secret, but some say that Intel already has most of the hard work done.
Intel has already put together a die, the size of a postage stamp, with three dual-core 45nm Penryn chips on it sharing a 16MB L3 cache. Allegedly, we'll see the Dunnington in either Q2 or Q3, this year
.Source: Gizmodo
Add your own comment

110 Comments on Intel Planning Six-Core Processor, Will Call it 'Dunnington'

#1
trt740
AMD better pull out whatever big guns they have now or ding dong the witch is dead.
Posted on Reply
#2
trt740
Tatty_One said:
Lol, based on what? the fact that as of today ATi has one of the top 5 fastest cards around and probably in about 6 weeks time one in the top 7 fastest cards...........

HD3870x2
8800GTS G92
8800GT G92
8800 Ultra G80
8800GTX G80

Then youu go mid/lowmid and you get......

HD3850
8800GS
9600GT

where in most things most of the time the HD3850 gets beat by the other two..............

http://forums.anandtech.com/messageview.aspx?catid=31&threadid=2154749&enterthread=y

And whats left?................................. Couple that lot with the fact that NVidia is outselling ATi 4 cards to every ATi card! Now please dont get me wrong.......the HD3870x2 is swwweeetttttt and I shall be getting one so genuinly my comments above are not fanboi......just dont quite see the logic in your statement is all :confused:
should be 8800gs
9600 gt
hd 3850
with AA on and AF this card cannot match either of these cards not even a 3870 can unless you turn them down or off


Look at the benches the eye candy is turned off or down. The ATi cards cannot do AA well at all and A/F is not good either until you get to a 3870x2
Posted on Reply
#3
trt740
fitseries3 said:
yeah, but i think they have nvidia cornered this yea in the graphics department.
I agree with this the 3870x2 is unreal and the new chip are reported to run double their speed. That was a statement from ATI to its stock holders so I would think it is true.
Posted on Reply
#4
hat
Enthusiast
trt740 said:
AMD better pull out whatever big guns they have now or ding dong the witch is dead.
Yeah, a super soaker :laugh:
Posted on Reply
#5
Tatty_One
Senior Moderator
trt740 said:
should be 8800gs
9600 gt
hd 3850
with AA on and AF this card cannot match either of these cards not even a 3870 can unless you turn them down or off


Look at the benches the eye candy is turned off or down. The ATi cards cannot do AA well at all and A/F is not good either until you get to a 3870x2
Yeah but I did not list in any particular order so I was not saying the first on the list was better than the 2nd or 3rd.
Posted on Reply
#6
trt740
Tatty_One said:
Yeah but I did not list in any particular order so I was not saying the first on the list was better than the 2nd or 3rd.
still AMD hit it out of the park with the 3870x2 as Nvidia did with the 8800gt, 9600gt and 8800gs
Posted on Reply
#7
Tatty_One
Senior Moderator
trt740 said:
still AMD hit it out of the park with the 3870x2 as Nvidia did with the 8800gt, 9600gt and 8800gs
Agreed but one out of the top 5 isnt very good although I am seriously considering getting a x2 but TBH am going through (amazingly) a period of extreme boredom with overclocking/benching.......having said that, I am looking at the Q9550 on it's release :D
Posted on Reply
#8
Wile E
Power User
Tatty_One said:
Agreed but one out of the top 5 isnt very good although I am seriously considering getting a x2 but TBH am going through (amazingly) a period of extreme boredom with overclocking/benching.......having said that, I am looking at the Q9550 on it's release :D
Me too Tatty (on the boredom). I think it's because we both hit a lull, where we already know our hardware's max. Getting the Vmod going should perk us back up. lol.
Posted on Reply
#9
tiys
blah blah...I bet after this comes out, you'll see a 6 core AMD.

Intel goes dual, next up is AMD going dual
Intel goes quad, next up is AMD going quad
Intel is going hex, next up is AMD going hex.

It's quite funny to think about it.
Posted on Reply
#10
Wile E
Power User
tiys said:
blah blah...I bet after this comes out, you'll see a 6 core AMD.

Intel goes dual, next up is AMD going dual
Intel goes quad, next up is AMD going quad
Intel is going hex, next up is AMD going hex.

It's quite funny to think about it.
Actually, AMD was developing Dual well before Intel. Intel just rushed pD out the door right before X2's release, to try and steal a little of AMD's thunder.
Posted on Reply
#11
MagnusEgallo
true dat, true dat.

tiys: your facts you must get strait, amd the first true dual core did make, count does not the pentium-d, it was/is but 2 p4's with duct tape attaching.

octacore could amd go if follow intels duct taping method they used, but poor lazy design does amd not favor, FYI intel quads are not native, duct tape 2 core2duos they did togather as 1.

WileE, thunder did they try and steal, fail did they, poor design=poor performance=netburst core that they used.

tiys next will say amd steals and copys 64bit x86 extentions they use, when invented them amd did.......
Posted on Reply
#12
DanTheBanjoman
Señor Moderator
MagnusEgallo said:
true dat, true dat.

tiys: your facts you must get strait, amd the first true dual core did make, count does not the pentium-d, it was/is but 2 p4's with duct tape attaching.

octacore could amd go if follow intels duct taping method they used, but poor lazy design does amd not favor, FYI intel quads are not native, duct tape 2 core2duos they did togather as 1.

WileE, thunder did they try and steal, fail did they, poor design=poor performance=netburst core that they used.

tiys next will say amd steals and copys 64bit x86 extentions they use, when invented them amd did.......
The discussion is old, the way 2 (or more) cores are bundled is completely irrelevant. A Pentium D is as much a dual core as an X2 or C2D. Saying things like "duct tape" is just popular/fanboy talk. As you know C2Q still outperforms Phenom X4, and since you say "duct taping" is a poor lazy design per se and poor design = poor performance you're obviously wrong somewhere.

AMD already said the cheaper and faster method Intel uses is the better one months ago. Making a native quad core didn't do AMD any good, they got behind and it didn't perform as planned.

Additionally Netburst isn't a poor design, on paper it was good. Problem was heat, Intel planned to clock them far higher which would do them good. It was more a failed design than a poor one.

I think you don't have your facts straight either. Then again yours are mostly based on fanboyish arguments while I have no clue where tiys got his.
Posted on Reply
#13
MagnusEgallo
DanTheBanjoman said:
The discussion is old, the way 2 (or more) cores are bundled is completely irrelevant. A Pentium D is as much a dual core as an X2 or C2D. Saying things like "duct tape" is just popular/fanboy talk. As you know C2Q still outperforms Phenom X4, and since you say "duct taping" is a poor lazy design per se and poor design = poor performance you're obviously wrong somewhere.

AMD already said the cheaper and faster method Intel uses is the better one months ago. Making a native quad core didn't do AMD any good, they got behind and it didn't perform as planned.

Additionally Netburst isn't a poor design, on paper it was good. Problem was heat, Intel planned to clock them far higher which would do them good. It was more a failed design than a poor one.

I think you don't have your facts straight either. Then again yours are mostly based on fanboyish arguments while I have no clue where tiys got his.
if a poor design netburst was not, then why fail did it?

good on paper and in practice are to very diffrent things, yes......

effectivly admited netburst was a bad move has intel.

phenom a bad move for enthusist market was, shines in the server market does it, price for performance much more cost effective is it, performs higher in spicific server related tasks it does when compared with like clocked xeons it does.

links i would like to amd wrong saying they where.

only a fanboi would say a good idea/design was netburst when prooven epic fail it was compared to older designs, when having admited a wrong move was it have its own creators.

no point in argueing is there, can see i that the dark path you have taken.....
Posted on Reply
#14
[I.R.A]_FBi
4*1=4
2*2=4
1*4=4

same thing, i dont care how it works intel makes it work.

why yall so conerned bout "duct taping" its the duct tape going to fall off and you have to go service it?

or we log onto tpu on day and look in the news section

"Core2Duo's recalled as duct tape falls off before end of 3 year warranty"?

i guess not, stop being silly and harping on things that dont matter.
Posted on Reply
#15
btarunr
Editor & Senior Moderator
MagnusEgallo said:
if a poor design netburst was not, then why fail did it?.....
Fail it didn't because it be made to compete with AMD K7 not K8, and compete it did and did well.

MagnusEgallo said:
good on paper and in practice are to very diffrent things, yes......

effectivly admited netburst was a bad move has intel.

phenom a bad move for enthusist market was, shines in the server market does it, price for performance much more cost effective is it, performs higher in spicific server related tasks it does when compared with like clocked xeons it does.
Bad move Netburst was not. Successful it was against the generation it belonged to and the competitors' architecture it was made to compete with.... succeed it did. Phenom a bad move is true. Opteron quad an equally bad move is true too. Sway it has none of over Intel's designs, thermal envelope it had but so did Intel's Low-power Xeon lineup. Performance edge it had none.

MagnusEgallo said:
links i would like to amd wrong saying they where.

only a fanboi would say a good idea/design was netburst when prooven epic fail it was compared to older designs, when having admited a wrong move was it have its own creators.

no point in argueing is there, can see i that the dark path you have taken.....
Netburst was bad say fanboys. It wasn't bad at its time say people with rational thinking.
Posted on Reply
#16
DanTheBanjoman
Señor Moderator
MagnusEgallo said:
if a poor design netburst was not, then why fail did it?

good on paper and in practice are to very diffrent things, yes......

effectivly admited netburst was a bad move has intel.

phenom a bad move for enthusist market was, shines in the server market does it, price for performance much more cost effective is it, performs higher in spicific server related tasks it does when compared with like clocked xeons it does.

links i would like to amd wrong saying they where.

only a fanboi would say a good idea/design was netburst when prooven epic fail it was compared to older designs, when having admited a wrong move was it have its own creators.

no point in argueing is there, can see i that the dark path you have taken.....
Like I said, unexpected amount of heat. Besides typing really annoying you don't read either it seems. Please type normally, I doubt many people will have a serious discussion with you otherwise. I from this point won't, until you bother to type normally.
Posted on Reply
#17
MagnusEgallo
bad was netburst for its day, p3 tualatin faster was, based off of p3 tualatin is core2, netburst one point had, chips selling to the dumb who think/thought that clock=performance, complain intel fanboi's do about athlon running hot in those days, but forget presHOT they do, 90c common was with preshot, athlon classic and socket A those temps never did reach.

a success only was netburst selling to noobs was because "2.8 ghz must be faster then 1.8ghz" yet failed to realise did noobs and fanboi's alike, 9ICP(athlon) greater then 6ICP (netburst) so lower clocks mattered did not.

One good use did netburst have, encoding........

tryed have you to run 64bit windows on netburst cores? unbaribly slow it is, turtles stampeeding thru peanut butter it reminds me of......slower then even the worst semperon 64 it is......
Posted on Reply
#18
[I.R.A]_FBi
actually my socket A runs far hotter than my 478 ...
Posted on Reply
#19
MagnusEgallo
DanTheBanjoman said:
Like I said, unexpected amount of heat. Besides typing really annoying you don't read either it seems. Please type normally, I doubt many people will have a serious discussion with you otherwise. I from this point won't, until you bother to type normally.
then talk to me do not, my feelings it will not hurt, the darkside i avoid, darth banjoman you are.
Posted on Reply
#20
MagnusEgallo
[I.R.A]_FBi said:
actually my socket A runs far hotter than my 478 ...
depends on cooling and core revision this does, amd retail cooling form back then epic fail was....EPIC!!!

a preshot i had, retail intel cooling i used, 90c temps it would reach, my palimino those temps never have reached, even on epic fail retail amd coolers.
Posted on Reply
#21
[I.R.A]_FBi
MagnusEgallo said:
depends on cooling and core revision this does, amd retail cooling form back then epic fail was....EPIC!!!

a preshot i had, retail intel cooling i used, 90c temps it would reach, my palimino those temps never have reached, even on epic fail retail amd coolers.
my intel stayed a 45 degrees all the time and never moved .. currently my athlon serer is warming my room @ 50-63 degrees depending on load

also my intel was very slow, like cold mollasses on a winter morn ...

For an entertaining conversation force-speak makes.
Posted on Reply
#22
MagnusEgallo
[I.R.A]_FBi said:
my intel stayed a 45 degrees all the time and never moved .. currently my athlon serer is warming my room @ 50-63 degrees depending on load

also my intel was very slow, like cold mollasses on a winter morn ...

For an entertaining conversation force-speak makes.
agree do i, :D


as stated above, epic fail in those days was amd retail cooling, large sinks then intel did use, with high speed fans many times, and slow they where, have had a few netburst rigs have i, slower then mollasses on a winter morn where they all, 3.4gz with HT did not help......glad cheaply i got it, EPIC fail it was......(had ddr2 800 did it and slower it was then 754 3700+ clawhammer, sad was i.....made good $ selling a friend to i did, intel fanboi was he, payed more then i spent on it did he :)
Posted on Reply
#23
Morgoth
hey starwars fanboy type normal !
Netbrust is still good for what i do!
cant wait for my first native quad core from intel ^^ named bloomfield :D

You Underestimate My Power!!
Posted on Reply
#24
Azazel
no need for it at this moment in time...2-4 cores are enough for this moment in time
Posted on Reply
#25
candle_86
who gives a crap, for one the P4 was not a gaming chip, but for media encoding ect it creamed AMD, go back and look. For another who gives a rats ass about all this, we all know when the day is over we buy what are budget will allow.


And star wars boy, turn 13 and stop taking like a moron please, it got old after the first post
Posted on Reply
Add your own comment