Cedar Mill based Celeron is an OC king

Slightly off topic here....

Yeah, but look at the voltages. FYI: high-voltage = short-lifespan. @ 1.456v is safe to say you can have it running 24/7. Remember 1.5v+ is the killer
Click to expand...

For chips that default to 1.35, you're starting to ask for trouble at 1.55+.
For chips that start at 1.40, 1.5v+ isn't going to kill much of anything if you have decent cooling. Sure when you're in the area of 1.60+ on air, then you're playing with fire.

Back on topic...

It appears Intel will have a very small lead when they finally get 65nm products out the door. That very small lead may disappear when AMD releases an A64 with a DDR2 controller onboard. Also, when AMD starts to use it's own 65nm process, I think (IMHO) that they'll hurt Intel again with performance/watt.

It will be nice to see Intel have something competitive on both the performance and power fronts, even if it is only briefly.
 
freeloader1969 said:
Slightly off topic here....
It appears Intel will have a very small lead when they finally get 65nm products out the door. That very small lead may disappear when AMD releases an A64 with a DDR2 controller onboard. Also, when AMD starts to use it's own 65nm process, I think (IMHO) that they'll hurt Intel again with performance/watt.

It will be nice to see Intel have something competitive on both the performance and power fronts, even if it is only briefly.
Click to expand...
I think AMD and IBM were going to use the 65nm during the second half of 2006. Also, when M2 comes out, it will be good, but Intel has been behind AMD for more then a few months, and four months after AMD releases M2, Intel will have Conroe. Intel spent 2 years developing the Conroe processor so, it will definitly beat AMD in performance/watt. Merom is going to have twice the performance/watt as Yonah which is already higher the P4.
 
Duke3d87 said:
I think AMD and IBM were going to use the 65nm during the second half of 2006. Also, when M2 comes out, it will be good, but Intel has been behind AMD for more then a few months, and four months after AMD releases M2, Intel will have Conroe. Intel spent 2 years developing the Conroe processor so, it will definitly beat AMD in performance/watt. Merom is going to have twice the performance/watt as Yonah which is already higher the P4.
Click to expand...

Make that 6 years.

Btw , do you know that Yonah's design started concurrently with Dothan in 1999 ?

The oldest architecture in Intel labs seems to be Nehalem.Work on it started around 1999-2000 and is supossed to be out by 2008.That is 8 years.

The interesting stuff is wheter Conroe 2 or Nehalem will be the first with an IMC.Initially , Merom microarch was suposed to be the 1st with IMC , but that was cut out because they needed the new arch faster for obvious reasons.
 
Duke3d87 said:
I think AMD and IBM were going to use the 65nm during the second half of 2006. Also, when M2 comes out, it will be good, but Intel has been behind AMD for more then a few months, and four months after AMD releases M2, Intel will have Conroe. Intel spent 2 years developing the Conroe processor so, it will definitly beat AMD in performance/watt. Merom is going to have twice the performance/watt as Yonah which is already higher the P4.
Click to expand...

The News is, AMD and IBM are at least a year away from a viable 65nm, even the AMD Fans sites point that out. From Dec 27 to the second week of Jan 06, Intel will launch Yonah, Presler and Cedar Mill (a Celeron version) and the Chipsets to support them. Now that's a Fact. AMD and IBM have to work the kinks out of SSOI at 65nm. Again, this is IBM doing most of the legwork, NOT AMD. Add to that, the folks who buy the first AMD 65nm Processors will be nothing more than beta testers.

http://www.pcworld.com/resource/article/0,aid,123850,pg,1,RSS,RSS,00.asp

From the time it takes for first stampout to shipping for revenue can be about 1 year. This use to be 18 months. There's always a risks if companies jump the gun before fully testing something like this. As these guys point out, AMD will test 4 new technologies at 90nm prior to moving to 65nm. I'd bet the AMD 65nm Processors will ship in April or May of 2007 August of 2007 wouldn't surprise me. Hell, Intel will have moved from Conroe A to Conroe B by then. Not to mention DDR3, ICH8 and SATA 4.

Not disrespecting the thread, the plain old Cedar Mill would make a better pick than the Celeron version.

Donnie27
 
Donnie27 said:
The News is, AMD and IBM are at least a year away from a viable 65nm, even the AMD Fans sites point that out. From Dec 27 to the second week of Jan 06, Intel will launch Yonah, Presler and Cedar Mill (a Celeron version) and the Chipsets to support them. Now that's a Fact. AMD and IBM have to work the kinks out of SSOI at 65nm. Again, this is IBM doing most of the legwork, NOT AMD. Add to that, the folks who buy the first AMD 65nm Processors will be nothing more than beta testers.

http://www.pcworld.com/resource/article/0,aid,123850,pg,1,RSS,RSS,00.asp

From the time it takes for first stampout to shipping for revenue can be about 1 year. This use to be 18 months. There's always a risks if companies jump the gun before fully testing something like this. As these guys point out, AMD will test 4 new technologies at 90nm prior to moving to 65nm. I'd bet the AMD 65nm Processors will ship in April or May of 2007 August of 2007 wouldn't surprise me. Hell, Intel will have moved from Conroe A to Conroe B by then. Not to mention DDR3, ICH8 and SATA 4.

Not disrespecting the thread, the plain old Cedar Mill would make a better pick than the Celeron version.

Donnie27
Click to expand...

link: "We are qualifying equipment as we speak. Fab36 was designed to be a 65-nm from the start. We are making great strides with IBM and our SRAM yields are hitting all our milestones," Sonderman told EE Times in a telephone interview. "The whole idea is to begin to process 65-nm in the middle of 2005 and bring it into production in 2006. We're certainly on schedule." Sonderman said wafers first ran in Fab36 in March 2005 and that first "wafer outs" would occur in the second-half of 2005."

from the article you linked to: "AMD plans to start making 65nm chips with all four strain methods in the second half of 2006 at its new fabrication plant in Dresden, Germany, Kepler said."

I don't know where you get that guesses from, but judging by the article you yourself posted as well as any other article i read so far on AMD/IBMs 65nm process, really, i think you are way off. If your guess with a release date as late as aug2007 would be correct that would be more than one/one and a half years late, which would be quite a drama. don't get me wrong, i don't think AMD couldn't bork it that badly - i just don't think IBM will :D

Why not just be happy that intel's stuff is coming along quite nicely, really, no need to predict doomsday for AMD ;)
 
wizzackr said:
link: "We are qualifying equipment as we speak. Fab36 was designed to be a 65-nm from the start. We are making great strides with IBM and our SRAM yields are hitting all our milestones," Sonderman told EE Times in a telephone interview. "The whole idea is to begin to process 65-nm in the middle of 2005 and bring it into production in 2006. We're certainly on schedule." Sonderman said wafers first ran in Fab36 in March 2005 and that first "wafer outs" would occur in the second-half of 2005."

from the article you linked to: "AMD plans to start making 65nm chips with all four strain methods in the second half of 2006 at its new fabrication plant in Dresden, Germany, Kepler said."

I don't know where you get that guesses from, but judging by the article you yourself posted as well as any other article i read so far on AMD/IBMs 65nm process, really, i think you are way off. If your guess with a release date as late as aug2007 would be correct that would be more than one/one and a half years late, which would be quite a drama. don't get me wrong, i don't think AMD couldn't bork it that badly - i just don't think IBM will :D

Why not just be happy that intel's stuff is coming along quite nicely, really, no need to predict doomsday for AMD ;)
Click to expand...
I agree, AMD's 65nm process is on track for introduction in Late 2006, which is most likely September, from my guess. The transistion will continue though into 2007, where they most likely will be done by Mid 2007. By the time AMD's 65nm stuff is out though, from the current looks of things Intel with have it's products based on NGMA out so there won't ever be a need to compare 65nm Dulal Core Turions vs Yonah as the processor it will be engaging is Merom.

I am most interested in Merom/Conroe, Yonah, Presler, Cedar Mill are all simple 65nm to just test the 65nm process out in mass scale production levels. You can tell, Cedar Mill is a simple optical shrink of Prescott-2M, while Yonah transistor density isn't that much higher then Dothan, they aren't really pushing the 65nm super hard heere like they did with the Prescott launch in 2004.
 
wizzackr said:
link: "We are qualifying equipment as we speak. Fab36 was designed to be a 65-nm from the start. We are making great strides with IBM and our SRAM yields are hitting all our milestones," Sonderman told EE Times in a telephone interview. "The whole idea is to begin to process 65-nm in the middle of 2005 and bring it into production in 2006. We're certainly on schedule." Sonderman said wafers first ran in Fab36 in March 2005 and that first "wafer outs" would occur in the second-half of 2005."

from the article you linked to: "AMD plans to start making 65nm chips with all four strain methods in the second half of 2006 at its new fabrication plant in Dresden, Germany, Kepler said."
Click to expand...

Read it a little more then. I said from first stamp out to shipping for cash, takes one year. If this guy says AMD will tap-out in the summer of 2006. If everything goes well, that means you'll not be able to buy one until the summer of 2007 or August being more realistic. Those Technologies will show up on 90nm Processors first. AMD's PR person is from the School of USS Grant. Remember that saying? "If you can't dazzel them with footwork, baffle them with BS!"

wizzackr said:
[I don't know where you get that guesses from, but judging by the article you yourself posted as well as any other article i read so far on AMD/IBMs 65nm process, really, i think you are way off. If your guess with a release date as late as aug2007 would be correct that would be more than one/one and a half years late, which would be quite a drama. don't get me wrong, i don't think AMD couldn't bork it that badly - i just don't think IBM will.

Why not just be happy that intel's stuff is coming along quite nicely, really, no need to predict doomsday for AMD ;)
Click to expand...

Way off of what? AMD said they were on time and if they are, they'll start this summer read June/July 06. Then June/July 07 is on time. August 07 is 2 or 3 months late, not 18 months late :rolleyes: I didn't guess anything LOL! Intel stampped out 65nm almost a year ago. Anand says AMD is a year behind Intel on 65nm. I think you're getting confused by two things here. You're mixing up how far AMD is behind Intel on Process and when AMD will SHIP for sales (NOT meerly tap out) launch it's own 65nm. Surely you don't think they go straight from first Tap Out to stampping and on sale at the same time? Please tell me you're not that new to this?

http://www.digitimes.com/news/a20051221A1001.html

AMD didn't bork anything. I didn't predict any friggin Doomsday either. Please don't confuse this with some guy at AMD putting a needless spin out to the press. AMD is a year behind Intel and that doesn't mean they're late at all. Now if AMD says they'll SHIP for revenue in July-06 and don't ship until July-07, then that makes them a year Late.

Anand said:
We won't be revisiting the issue of power consumption, as we already did that at the end of our last article, but needless to say, Yonah is the most efficient dual core processor that we've tested to date. Granted that it does have the advantage of being on Intel's 65nm process whereas the Athlon 64 X2 is still based on AMD's 90nm process, but given that AMD is around a year away from transitioning to 65nm, it is an advantage that Intel has the right to enjoy.
Click to expand...

PentiumM at 90nm is cooler than Athlon64 and a Dual Core 90nm would still be cooler than X2=P That's why some say they should change that site's name to AMDtech. Even with their Cheerleading, they say AMD is a year behind. Again, it's NOT when they tap-out but when they ship and or you can buy one. If that's not the case, then even if AMD were shipping the first tap-out, they're still a year later than Intel tapped out last year, sheesh!

My AMD based rig is two weeks old today.

Donnie27
 
Donnie27 said:
I think you're getting confused by two things here. You're mixing up how far AMD is behind Intel on Process and when AMD will SHIP for sales (NOT meerly tap out) launch it's own 65nm. Surely you don't think they go straight from first Tap Out to stampping and on sale at the same time? Please tell me you're not that new to this?
Click to expand...

Hey Donnie - I thought AMD taped out the first 65nm processors in mid 2005, aiming for revenue production in summer 2006 (probably only one line of processors and not the entire lineup?), which is why I thought your predicted availability in august 2007 was too far behind for my guessing - especially taking into account in how far IBM is actually involved into their transition. (EETimes: "Sonderman told EE Times in a telephone interview. "The whole idea is to begin to process 65-nm in the middle of 2005 and bring it into production in 2006. We're certainly on schedule.")

No problem, though, as I really couldn't judge it with my admittedly little knowledge on the topic. However, no need to get all personal, either ("Please tell me you're not that new to this?" etc.) ;) wel'll see what happens...

marry christmas,
j

PS: How old is the anandtech article you quoted from? I couldn't find it...
 
wizzackr said:
Hey Donnie - I thought AMD taped out the first 65nm processors in mid 2005, aiming for revenue production in summer 2006 (probably only one line of processors and not the entire lineup?), which is why I thought your predicted availability in august 2007 was too far behind for my guessing - especially taking into account in how far IBM is actually involved into their transition. (EETimes: "Sonderman told EE Times in a telephone interview. "The whole idea is to begin to process 65-nm in the middle of 2005 and bring it into production in 2006. We're certainly on schedule.")

No problem, though, as I really couldn't judge it with my admittedly little knowledge on the topic. However, no need to get all personal, either ("Please tell me you're not that new to this?" etc.) ;) wel'll see what happens...

marry christmas,
j
PS: How old is the anandtech article you quoted from? I couldn't find it...
Click to expand...

Merry Christmas to you as well. I didn't take it personal at all.

I'd go with what IBM said before AMD. IBM and AMD have talked about having parts of 65nm ready, they were nowhere stampping out 65nm this past summer. Sure they had layouts but hell, that's about it. Many in the computer industry don't take to kindly to companies who "overstate" their positon constantly.

I quoted from this review. It's at the end of the first page.

Donnie27
 
needmorecarnitine said:
you claim that you trust what IBM says over AMD. The Anandtech link you posted doesn't attribute it to anyone.

Here is a fairly recent article on amd's 65 nm status: http://www.theregister.co.uk/2005/12/07/amd_ibm_stressed_silicon/

and

http://www.infoworld.com/article/05/12/06/HNibmamd_1.html
Click to expand...

First of all, I don't claim shit! I trust what IBM is saying because AMD is using what the hell IBM comes up with=P They can call it partnering as much as they like. If IBM fails, AMD is fucked!

The best estimates show AMD and IBM is at least 1 year and at worst 2 years behind Intel on 65nm. Having come up with parts of the layout or transistors is NOT the same as stampping out. Even many Fans sites don't think AMD will ship 65nm this year=P That's nothing more that wishful thinking. Hell, Intel might be closer to 45nm than AMD and IBM might be to 65nm.

Donnie27
 
http://www.hardwareanalysis.com/content/article/1820/

http://www.theregister.co.uk/2004/08/31/intel_65nm/

Intel bases this claim on the 70Mb SRAM chips it has been fabbing at 65nm at its Hillsboro, Oregon 300mm-wafer development fab. It's not the first 65nm part Intel has produced - in November 2003, it turned out a working 4Mb SRAM chip.
Click to expand...

http://www.theregister.co.uk/2003/11/24/intel_talks_up_65nm_process/

65nm SRAM "Published Monday 24th November 2003"

http://www.cpu-central.com/wwwboard/msg72.asp?id=72459

4.5GHz without breaking a sweat :)

http://www.xbitlabs.com/news/cpu/display/20050823194429.html

by Anton Shilov

[ 08/23/2005 | 07:45 PM ]

The head of Intel showed the first public demonstration of Merom, Conroe and Woodcrest – processors for notebook, desktop and server platforms designed on Intel’s 65nm technology manufacturing process. He also said Intel has more than 10 processor projects that contain four (quad-core) or more processor cores per chip.
Click to expand...

I could post 100's links showing the same kinds of things. The last link shows Intel showing off Conroe and etc... in August. Presler and Cedar Mill was already finished by then and started shipping a couple of months later. That's a big time difference than what IBM and AMD have done and they've not demoed any 65nm processor=P

If there's any more doubt, here's an EE Times writer doing a story for the ChannelWeb. Both respected.

http://www.channelweb.com/sections/allnews/article.jhtml?articleId=174901579

Andy Wei, a member of the technical staff based at AMD Dresden, described a new process that AMD will first retrofit into its 90-nm microprocessors, and then use in its 65-nm designs going into production in the second half of next year at AMD’s new 300-mm fab in Dresden.
Click to expand...

You see, that's an AMD person saying NEXT YEAR (2006), not last summer. Again, AMD will beta test this on 90nm prior to going to 65nm.

Donnie27
 
JetUsafMech said:
Actually, Intel has been working on 45nm since at least 2003...

http://www.reed-electronics.com/electronicnews/article/CA333899.html
Click to expand...

Again, that's just small parts of the process. I'm talking about having Demo Processors like the First 800MHz Hammer for example. Intel showed off both Conroe and Memrom this past August. If AMD shows something this (06) June or July, it will be a year before that product ships for profits. That still puts AMD at 65nm in the summer of 2007. They both weren't supposed to have 65nm (except Presler/Cedar Mill) ready in 2006. That doesn't make AMD late, it just makes Intel early.

Donnie27
 
You must log in or register to reply here.
Back
Top