CPU-Z AMD 65nm Brisbane

[metro.cl]

Hello guys as i promised here is a SS of a final silicon of Brisbane, everything default.

http://img213.imageshack.us/img213/1241/am265nmgv5.jpg (http://imageshack.us)


this is a ES but is same as future retail chips, improvements are 0.5 HT setting and also lower TDP, you only need bois update to use this chips

Launch date is tomorow

More info about oc margin will be posted soon :)

Source (http://www.chilehardware.com/foro/am2-brisbane-x2-t50115.html?p=659258#post659258)

[dinos22]

that's great man :toast:

[msimax]

overclock that sucker :)

[Fred_Pohl]

1.32Vcore? I thought it was supposed to be lower than that.

[Nanometer]

It could be that CPU-Z is having problems reading the correct V-core. Not somthing that is uncommon with new CPUs. I also hope the stock V-core is much lower. 1.25v would be great. I seriously hope these chips clock better. 2.5Ghz to 3.5GHz on air would be nice ;)

[bachus_anonym]

Nice one, metro :up: Can't wait for your testing, man!

... hopefully not-wishfully-thinking, this thing will do solid 3.2GHz at normal ambient... :)

[el rolio]

yea, what is actual stock vcore? bios wise?

[dinos22]

Nice one, metro :up: Can't wait for your testing, man!

... hopefully not-wishfully-thinking, this thing will do solid 3.2GHz at normal ambient... :)
hopefully they perform similar clock for clock......amd can't let Intel run away with the lead otherwise prices are going sky high

[Lightman]

You're THE ONE metro.cl! :D :D

I suspect that OC session tomorrow??
Remember, we are waiting ;) .

[nn_step]

ooh half multi so sexy, how do the dividers look?

[MaxxxRacer]

very cool. cant wait to see how they clock. hopefully 65nm is as helpful for AMD as it was for Intel. (other than the obvious cost saving benefits)

[HKPolice]

Everyone is predicting crappy OCs with AMD's 65nm because they're not releasing any chips faster than ~2.8Ghz whereas 90nm 3Ghz chips are everywhere. Hell, even the new 4x4 FX74 is only 90nm.

[Afterburner]

Ahhh.....finally, I was getting bored inspecting 65nm dies :D

[informal]

And you tease us with those possible quad core Barca dies coming from Chartered :D

[zakelwe]

I wonder how these will respond to voltages ?

Regards

Andy

[cirthix]

Everyone is predicting crappy OCs with AMD's 65nm because they're not releasing any chips faster than ~2.8Ghz whereas 90nm 3Ghz chips are everywhere. Hell, even the new 4x4 FX74 is only 90nm.

depends on architecture. i'd like to remind you that conroes are 65nm and the current highest stock clocked chip is a 3ghz xeon. also, with 1.325v stock.

I'm expecting that these chips clock slightly better than current conroes, but not much more.

it would be nice to have an early hint :P *poke*

[pawl]

please sir,

all we want is one superpi run at stock speeds so we can see if this is just going to be a die shrink or if its the new improved slightly( hopefully much ) faster core.

p

PS. heck any bench will do. just give us some hard data so we can stop pontificating.

[Varsos]

I guess tomorrow HE will show us what this baby can do.

[dinos22]

I guess tomorrow HE will show us what this baby can do.
yeah seems from his post :)

[kemo]

Cant wait for updates

[HKPolice]

depends on architecture. i'd like to remind you that conroes are 65nm and the current highest stock clocked chip is a 3ghz xeon. also, with 1.325v stock.

I'm expecting that these chips clock slightly better than current conroes, but not much more.

it would be nice to have an early hint :P *poke*

That's not possible because if AMD's 65nm process was such a breakthrough, why are they not releasing a 3Ghz 65nm part? Look at this roadmap:

http://resources.vr-zone.com//newspics/Nov06/16/AMDroadmap.png

Even in Q3 2007 there will not be a 3Ghz 65nm AMD cpu, max is 2.9Ghz, whereas Conroe is already at 2.93ghz with the X6800.

Most people forgot the exponential current leakage problems that occurred when Intel first switched to 90nm, remember Prescott? They learned from that mistake and the 65nm transition has been smooth for them.

AMD's 90nm transition was smooth because they paid IBM an arm and a leg for SOI tweaks, but it's not cutting it anymore for 65nm.

It's not like the good old days anymore when a die shrink was always a blessing. You have to be more critical and analytical these days. Why isn't AMD releasing a faster 65nm part? Because they can't, the process is no better than 90nm when it comes to clockspeeds, it only allows them to fit more cpus into every silicon wafer therefore cutting costs.

[cupra]

overclock that sucker :)

cant wait to see it :)

[metro.cl]

Ok mates i got some bad news for you:

This will most likely be a softlaunch and if they release some 65nm parts its gonna be really hard to get them.

no performance gain here this is just 65nm am2 cpu same as 90nm am2 cpus

[Starscream]

That's not possible because if AMD's 65nm process was such a breakthrough, why are they not releasing a 3Ghz 65nm part?


part could be the financial point.

they cant move the entire production range to 65nm at the same time.
The biggest chunk of CPUs that AMD sells are the slower ones.
So by moving them first to 65nm you will cut cost at the CPUs that you sell the most.

[gOJDO]

@Starscream
Transition from one proces to another costs a lot of money. Also, their matured 90nm SOI2 is very optimized. Maybe it offers better yields than their young 65nm process.
AMD need less dissipated heat and smaller die to make more powerfull chips with more transistors on squared mm. Thats why they are developing their 65nm SOI3 process.

[LOE]

Starscream:

many people fail to understand it

high end is too small to be a priority

[lapdog]

So I guess that 40% faster switching speed is really just zero. :fact:

[Thorry]

Indeed, the budget market is where the money is (strange enough). Most OEM systems sold use 'budget' parts (which are very powerfull these days, a Sempron 3800+ is budget these days)

[savantu]

So I guess that 40% faster switching speed is really just zero. :fact:

What did you expect ? 40% greater freq ?

My bet is Brisbane isn't freq optimized at all , dumb shrink to get it out fast while the bulk of the resources were concentrated on K8L.

[madcho]

Everyone is predicting crappy OCs with AMD's 65nm because they're not releasing any chips faster than ~2.8Ghz whereas 90nm 3Ghz chips are everywhere. Hell, even the new 4x4 FX74 is only 90nm.

Brisbane is not 2*1Mo, If AMD use Brisbane to change The Fx, A lot of fan boys including me will not agree.

For a lot of people cache is better than mhz :stick:

Even Brisbane clock @ 3.8ghz i will keep my opty 165 2.61ghz stock volt :cool:

i will wait for something more strong, FX on Barcelona and cf R600 :banana:

For moment, i perfer to work on my home water, with pro tools :toast:

[ozzimark]

So I guess that 40% faster switching speed is really just zero. :fact:
i guess you forgot about history.. what did and do with the first 90nm parts? ;)

[Afterburner]

And you tease us with those possible quad core Barca dies coming from Chartered :D
Ohh that was just me happening to be poking my nose and peering into as many pods as possible.....they have 'suspected' me of being an industry spy on a few occasions.... :lol2:

FUGGER how about putting me on your payroll..... :D

[lapdog]

What did you expect ? 40% greater freq ?

My bet is Brisbane isn't freq optimized at all , dumb shrink to get it out fast while the bulk of the resources were concentrated on K8L.

It said the process alone would yield this improvement not anything to do with architecture(K8L.)

No I did not expect 40% faster chips, just noticeably faster ones.

http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543_13743~103048,00.html

[Varsos]

I don't think that anyone is expecting improved performace but definitely we all expext better o/c.
COME ON bringit ON.....:toast:

[STEvil]

40% faster switching means less heat prduced, thats all. Performance will not change noticeably.

[Revv23]

AMD process transition's never result in faster clocks.

at least not in my knowledge. clawhammer was better then winchester. then they got it right with san diego. 180nm was better then t-bred-a. they got it right with Tbred B and Barton. brisbane is probably not going to be better, but they will get it right with K8L. ;)

[largon]

It said the process alone would yield this improvement not anything to do with architecture(K8L.)

No I did not expect 40% faster chips, just noticeably faster ones.The problem is that Rev.G is a dumb shrink.
Thus, the scaling problems that have plagued K8 from the beginning are still there.

Rev.H should be less hindered by those issues, though.

[Lightman]

Ohh that was just me happening to be poking my nose and peering into as many pods as possible.....they have 'suspected' me of being an industry spy on a few occasions.... :lol2:

FUGGER how about putting me on your payroll..... :D

They are so suspicious?? :cool:

Next time take super small digital camera to work and make some pictures ;) .
If you will prove your spy skills in battlefield then I'm certain that Fugger will add you to payroll :D .

[speed bump]

Ok mates i got some bad news for you:

This will most likely be a softlaunch and if they release some 65nm parts its gonna be really hard to get them.

no performance gain here this is just 65nm am2 cpu same as 90nm am2 cpus

Does that mean no overclocking gain or no gain in performance between the two at the same speed.

[Afterburner]

They are so suspicious?? :cool:

Next time take super small digital camera to work and make some pictures ;) .
If you will prove your spy skills in battlefield then I'm certain that Fugger will add you to payroll :D .
Well jokingly of coz but one can never be certain.... :nono:

Majority of the workforce are Malaysians and the rest from India / China. Items such as Ciggarettes / Thumbdrives / Cameras are strictly prohibited while Laptops are controlled items. ALL bags are checked. There are like cameras eveywhere so dont get caught snoozing or digging ur nose..... :stick:

:D :D :D

[alucasa]

You gotta wait for their 65nm to be matured.

[xVeinx]

I was hoping for more than a soft launch actually. I suppose that a hard launch might actually hurt their revenue for the Christmas season in some ways, but it's not as thought the supply of chips is that great right now anyway (they do need to sell off all the 939 however). Much of their money will be derived from Dell over the BIY market now, so they may not have to release them yet anyway until Dell is ready to start using them in their systems...

[terrace215]

Everyone is predicting crappy OCs with AMD's 65nm because they're not releasing any chips faster than ~2.8Ghz whereas 90nm 3Ghz chips are everywhere. Hell, even the new 4x4 FX74 is only 90nm.

Actually, they are only launching 65nm at 2.6GHz (!)

That's 400MHz below the 3GHz clock on 90nm FX-74.

2.8GHz 65nm isn't supposed to arrive until Q207.

Not so good.

Looks like high voltage. Wonder what die size will be?

[terrace215]

What did you expect ? 40% greater freq ?

My bet is Brisbane isn't freq optimized at all , dumb shrink to get it out fast while the bulk of the resources were concentrated on K8L.

get it out FAST? It's at least 6 months late!

[terrace215]

It said the process alone would yield this improvement not anything to do with architecture(K8L.)

No I did not expect 40% faster chips, just noticeably faster ones.

http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543_13743~103048,00.html


It was a "trick" % improvement. the 40% was not relative to 90nm DSL strained parts, but theoretical NO STRAIN AT ALL parts. When you backed out the DSL over NO STRAIN advantage, it turned into something like a 14% improvement.

Something went wrong though, because 65nm is clocking about 13% slower than 90nm at launch.

[Nedjo]

It’s clocked lower ‘cos of AMD’s mix portfolio. They still do produce 90nm CPU’s, and using that tech for:
a) mass production of the server parts (proven tech is the only viable option for the server market)
b) for production of the niche parts, and Quad FX’s is considered niche, ‘cos of the market size for that platform.

65nm, and FAB36 are in mass production phase, and AMD will produce mainstream dies in it. And mainstream dies MUST work on lower stock voltages comparing to the niche parts!

I bet that Brisbane will surpass any 90nm Windsor, on air, and on liquid!

[informal]

It’s clocked lower ‘cos of AMD’s mix portfolio. They still do produce 90nm CPU’s, and using that tech for:
a) mass production of the server parts (proven tech is the only viable option for the server market)
b) for production of the niche parts, and Quad FX’s is considered niche, ‘cos of the market size for that platform.

65nm, and FAB36 are in mass production phase, and AMD will produce mainstream dies in it. And mainstream dies MUST work on lower stock voltages comparing to the niche parts!

I bet that Brisbane will surpass any 90nm Windsor, on air, and on liquid!

Word!I agree 100% with you man.
Now we need to see some OC results,and fast :D.

[terrace215]

It’s clocked lower ‘cos of AMD’s mix portfolio. They still do produce 90nm CPU’s, and using that tech for:
a) mass production of the server parts (proven tech is the only viable option for the server market)
b) for production of the niche parts, and Quad FX’s is considered niche, ‘cos of the market size for that platform.

65nm, and FAB36 are in mass production phase, and AMD will produce mainstream dies in it. And mainstream dies MUST work on lower stock voltages comparing to the niche parts!

I bet that Brisbane will surpass any 90nm Windsor, on air, and on liquid!

Don't kid yourself. It's clocked lower 'cos it doesn't clock any higher right now. The schedule slip was an indication of process troubles, and the low launch clocks confirm it.

[accord99]

65nm, and FAB36 are in mass production phase, and AMD will produce mainstream dies in it. And mainstream dies MUST work on lower stock voltages comparing to the niche parts!

I bet that Brisbane will surpass any 90nm Windsor, on air, and on liquid!
Then surely AMD would have used a couple of hundred of them for the 4X4 previews, shaving a 100W of power consumption would have made the launch less disastrous.

[[XC] Lead Head]

Don't kid yourself. It's clocked lower 'cos it doesn't clock any higher right now. The schedule slip was an indication of process troubles, and the low launch clocks confirm it.

Do you know from personal experience that they are clocking lower? AMD wants to make sure that 65nm is a proven process with the mainstream. If they moved right in with the high end server/FX stuff, and there is problems in the process, people won't be happy. They can't ramp the clocks up higher on the mainstream stuff yet too, because it would be performing better then the FX series, Which would blow the whole point to even buying an FX. Look what happened with ATi and the R520, they released new tech with higher clock speeds on an unproven process, and they got tons of problems that needed to be worked out before they could even launch it. Even Intel did these kind of tactics. When they moved to the 65nm Presler core. Those parts weren't clocked any higher then Prescott, and no one doubted that they didn't clock higher then 90nm Prescott. Intel was testing out 65nm before they put C2D on full production. Simple as that. Also when Intel was going to the Presler core, NO ONE doubted that they could clock higher then the 90nm Prescott, but with AMD, tons of people believe 65nm is clocking WORSE then 90nm, even though Intel did the same thing with the same launch clocks.

EDIT:
@accord, yes it would have helped the launch, but 65nm is still an unproven process, If AMD sent out 65nm FX parts with defects, imagine what that would have done to the launch.

[nn_step]

Don't kid yourself. It's clocked lower 'cos it doesn't clock any higher right now. The schedule slip was an indication of process troubles, and the low launch clocks confirm it.
umm just a little bit of History for you.
180nm
130nm
90nm
all done on AMD's LOW END FIRST.
Once the process is perfected and ONLY then will the high end migrate to it.
nVidia does the exact same thing.
It is very low risk and the "Smart" thing to do
the whole R520 issue was a matter of Process problems constantly delaying a wonderful design.

[metro.cl]

yep is the way companies test the process, if you go with a simpler designs is easier to fix it, if you go with a more complex design is harder.

[wedfighter]

i dont see any vcore change by this new 65nsm size
am i wrong?

[dinos22]

well that was a bit of a flop :(

[terrace215]

umm just a little bit of History for you.
180nm
130nm
90nm
all done on AMD's LOW END FIRST.
Once the process is perfected and ONLY then will the high end migrate to it.
nVidia does the exact same thing.
It is very low risk and the "Smart" thing to do
the whole R520 issue was a matter of Process problems constantly delaying a wonderful design.

No, that's not really accurate. 90nm debuted with mobile, for low power, true, but a 400MHz deficit is quite large, and combined with a long-delayed launch, should tell you something about the health of the process.

[terrace215]

Do you know from personal experience that they are clocking lower? AMD wants to make sure that 65nm is a proven process with the mainstream. If they moved right in with the high end server/FX stuff, and there is problems in the process, people won't be happy. They can't ramp the clocks up higher on the mainstream stuff yet too, because it would be performing better then the FX series, Which would blow the whole point to even buying an FX. Look what happened with ATi and the R520, they released new tech with higher clock speeds on an unproven process, and they got tons of problems that needed to be worked out before they could even launch it. Even Intel did these kind of tactics. When they moved to the 65nm Presler core. Those parts weren't clocked any higher then Prescott, and no one doubted that they didn't clock higher then 90nm Prescott. Intel was testing out 65nm before they put C2D on full production. Simple as that. Also when Intel was going to the Presler core, NO ONE doubted that they could clock higher then the 90nm Prescott, but with AMD, tons of people believe 65nm is clocking WORSE then 90nm, even though Intel did the same thing with the same launch clocks.

EDIT:
@accord, yes it would have helped the launch, but 65nm is still an unproven process, If AMD sent out 65nm FX parts with defects, imagine what that would have done to the launch.


The AMD roadmap, that shows only 100-200MHz in clock bump over the 2.6GHz 65nm top clock @ launch, 2 QUARTERS down the road, should also tell you something. :)

[VulgarHandle]

i'm sorry, but i just don't get the arguement of... lower clocked mass production instead of high-end because they wouldn't want the FX's to have bugs....

that argument can only mean that if there would be a bug, the mass would be the one's to get it first? you think Dell would like that? having just launched with AMD's only to have them be bugged? what's AMD to say? "Sorry, Rev H is on it's way." I don't think so.

It's marketing, pure and simple. They don't want to release chips at clocks or performance greater than their high-end out now. That way they can still sell off their supplies of older CPU's. This way they can also supply Dell and HP and other OEMs their 65nm, which will force the other channels(etailers and such) to dry up the those older supplies. sucks for us, but hey, it will come. at least, that's how i see it.

we all know AMD dosn't have the FAB capacity to hard launch with all ranges of speeds covered and enough to cover the big oems and all other channels.

[Lithan]

i'm sorry, but i just don't get the arguement of... lower clocked mass production instead of high-end because they wouldn't want the FX's to have bugs....

that argument can only mean that if there would be a bug, the mass would be the one's to get it first? you think Dell would like that? having just launched with AMD's only to have them be bugged? what's AMD to say? "Sorry, Rev H is on it's way." I don't think so.

It's marketing, pure and simple. They don't want to release chips at clocks or performance greater than their high-end out now. That way they can still sell off their supplies of older CPU's. This way they can also supply Dell and HP and other OEMs their 65nm, which will force the other channels(etailers and such) to dry up the those older supplies. sucks for us, but hey, it will come. at least, that's how i see it.

we all know AMD dosn't have the FAB capacity to hard launch with all ranges of speeds covered and enough to cover the big oems and all other channels.


That's what I think it boils down to. AMD doesnt have the capacity to migrate much over to 65nm yet, and it's probably assumed that the dies won't be as consistant for 65nm as 90nm... so they want to be able to use as many as possible without downbinning, hence not initially converting their flagship models to 65nm.
This is also where the savings of the new process can be realized best, isn't it? With lower end processors, more of the cost goes towards the raw material. Thus with a shrink of the raw material, a greater percent of that origional cost can be shifted to profit versus a lesser percent when the higher end models are compaired. This assumes that AMD has good enough yields on 90nm to fill demand for the top tier models that aren't initially having 65nm counterparts released.

[ozzimark]

No, that's not really accurate. 90nm debuted with mobile, for low power, true, but a 400MHz deficit is quite large, and combined with a long-delayed launch, should tell you something about the health of the process.
no, it really is accurate. how much did 90nm lag behind 130nm in mhz AT STOCK SPEEDS when it first hit consumer desktop parts? about 400mhz.. ;) says nothing about overclockability btw. makes perfect sense to introduce the low-end parts in the new tech first, as mentioned: easier to make, and it's where the money is

typical of amd. nothing more, you're reading wayyyyyy too far into it, and obviously haven't been around for the game amd plays for very long :p:

[DilTech]

No, that's not really accurate. 90nm debuted with mobile, for low power, true, but a 400MHz deficit is quite large, and combined with a long-delayed launch, should tell you something about the health of the process.

No...

90nm debuted with winchester, which had a, get this, 400 mhz deficit between the top-end winchester(2.2 ghz) and the top-end 130nm FX-55(2.6 ghz).

See a pattern forming here?

[VulgarHandle]

No...

90nm debuted with winchester, which had a, get this, 400 mhz deficit between the top-end winchester(2.2 ghz) and the top-end 130nm FX-55(2.6 ghz).

See a pattern forming here?
i do

[ether.real]

I want some results Metro....not all this jibber jabber about channels and markets!

[Lithan]

Thats not to say these will be the miracle chips some are expecting (I think I saw someone suggest the first ones out will outclock conroes which would put them at 3.4-3.8ghz average overclock on air if Im not mistaken). Remeember Tbred A's... Pallys were not hard to get past 1800-1900 at the time, and tbred A's made 2000 common, but saw few over 2.1 on air. Tbred B's which didn't feature a die shrink saw MUCH better gains, we went from 2.0-2.1 typical clocks to 2.4-2.5 clocks pretty much overnight.


Again this is nothing but speculation, but I'd suspect early .065's will make 3-3.2ghz a fairly common thing on air with a few examples pushing 3.4... when paired with boards and ram that will allow those clocks.

[nn_step]

Hmm. 130nm hit 2.6GHz. 90nm hit only 400MHz higher for top bin (might get to 600MHz).

I smell an architectural limitation. Looks like the pipeline is too short to scale well.

Hopefully that will be fixed in K8L.
with 2 less integer stages and a larger process than Conroe, that would make logical sense.

[dinos22]

Hell froze over (we agreed on something). ;)
that's very true hahah

[speed bump]

Hate to sound like a broken record but any overclocking results Metro? Or do we have to wait a few more Hours.

[dinos22]

Hate to sound like a broken record but any overclocking results Metro? Or do we have to wait a few more Hours.
i think he's basically saying they do the same as 90nm parts so nothing special to report there

[nn_step]

Hell froze over (we agreed on something). ;)
umm I think we can officially agree that regardless of the process, Processors at best will only hit 27.9566667% of the theoretical possible predicted for the stages. Which is about +/- 0.05% for both AMD and Intel. (based on Conroe and K8 Numbers only)
Given that tiny of a variance between AMD and Intel for that value. One could almost assume all the money blown by Intel in Netburst, didn't really help them out much in the Stage design process.
Now assuming that percentage is held for BOTH AMD 65nm and Intel 45nm.
Intel's Switch to 45nm will only bring cheaper Procs with More Cache w/o much Clock speed increase.
However that is just conjecture at this point and time.

[MaxxxRacer]

umm just a little bit of History for you.
180nm
130nm
90nm
all done on AMD's LOW END FIRST.
Once the process is perfected and ONLY then will the high end migrate to it.
nVidia does the exact same thing.
It is very low risk and the "Smart" thing to do
the whole R520 issue was a matter of Process problems constantly delaying a wonderful design.

a bit off topic, but fwiw ATI usually sticks to that philosophy too.

[Revv23]

a bit off topic, but fwiw ATI usually sticks to that philosophy too.


true, but not on r520.

[Serge84]

You can not compare K8 with K8L. Or K8L with conroe. There is nothing to compare/test until its out physically!!! K8 will be dead when K8L comes out. That is why no more effort will be made to make K8's on going faster. Whats the use when K8L will perform better then K8 by a greater level. BTW K8L will "NEED TO" run at 3.5ghz to run the max HT3 clocks. 2.6ghz 5.2gt/s chipset fsb... http://badhardware.blogspot.com/2006/10/amds-k8l-revealed-in-cray-rainier.html

AMD

FX
FX-62: OPPAINTER - 3635.9 MHz http://www.xtremesystems.org/forums/...d.php?t=100443
FX-60: gloatlizard - 3875.44 MHz http://valid.x86-secret.com/show_oc.php?id=97927
FX-57: before - 4208.1 MHz http://img199.imageshack.us/img199/9243/cpu42089kx.jpg
FX-55 (ClawHammer): |RickY|&Pedro Rocha - 4005.23 MHz http://valid.x86-secret.com/show_oc.php?id=10680
FX-55 (San Diego): -De$troyeR- - 4056.86 MHz http://valid.x86-secret.com/show_oc?id=108727
FX-53: |RickY| & Pedro Rocha - 3602.8 MHz http://resources.vr-zone.com/ocdb/da...tel_or_amd=amd

Dual Core (X2)

Manchester:

X2 4600+: Onepagebook - 3687.15 MHz http://valid.x86-secret.com/show_oc.php?id=78999
X2 4200+: sson74 - 3613.06 MHz http://valid.x86-secret.com/show_oc.php?id=83165
X2 3800+: OC-Team.BE - 3700.86 MHz http://valid.x86-secret.com/show_oc.php?id=102531

Toledo:
X2 4800+: BB Mods's - 3607.99 MHz http://valid.x86-secret.com/show_oc?id=89669
X2 4400+: Fosco - 3408.27 MHz http://valid.x86-secret.com/show_oc.php?id=62072
X2 3800+: isulk - 3350.3MH http://valid.x86-secret.com/show_oc.php?id=106101

Windsor:
X2 4600+: JuanFlatier - 3000.1 MHz http://www.xtremesystems.org/forums/...d.php?t=100742
X2 4200+: Cole Riemer - 2915.17 MHz http://valid.x86-secret.com/show_oc?id=107317
X2 4000+: DrJay - 3137Mhz http://valid.x86-secret.com/show_oc.php?id=112440
X2 3800+: anaboliss - 3102,92MHz http://valid.x86-secret.com/show_oc?id=106039
X2 3000+: Amiteriver - 3022.6 MHz http://www.xtremesystems.org/forums/...d=11543992 89

Single Core Athlon64

Clawhammer:
4000+: gubben - 3614.77 MHz http://valid.x86-secret.com/show_oc?id=25221
3700+: twillius_basic - 3331.38 MHz http://valid.x86-secret.com/show_oc.php?id=16530
3500+: ojdr2001 & Silence - 3521.6 MHz http://valid.x86-secret.com/show_oc.php?id=49904
3400+: eshbach & Evil Spork - 3320.71 MHz http://valid.x86-secret.com/show_oc?id=12416
3200+: dsl & Spd - 2854.15 MHz http://valid.x86-secret.com/show_oc.php?id=100425
3000+: fr@me - 3156 MHz http://valid.x86-secret.com/show_oc?id=65931
2800+: setyotomo & ilyas - 3204.14 MHz http://valid.x86-secret.com/show_oc.php?id=18206

Manchester:
3500+: Willis - 3205.02 MHz http://valid.x86-secret.com/show_oc?id=79721
3200+: hey & tosunermc - 3405.83 MHz http://valid.x86-secret.com/show_oc?id=103390

Newcastle:
3800+: Shamino - 3201MHz http://www.vr-zone.com/ocdb/data.php...tel_or_amd=amd
3500+: |RickY| - 3604.0 MHz http://www.teampuss.com/forums/files/3500caa2c_234.jpg
3400+: Jrocket - 3359.59 MHz http://valid.x86-secret.com/show_oc?id=18225
3200+: .::Maddog::. & mosin_nagant - 3024MHz http://valid.x86-secret.com/show_oc.php?id=22563
3000+: Ssilencer - 3276.60 MHz http://img391.imageshack.us/img391/4638/32765gm.gif

Newark:
4000+: metalman2785 - 3394.26 MHz http://valid.x86-secret.com/show_oc?id=100828

Oakville:
3000+: firefly - 2700.13MHz - http://valid.x86-secret.com/show_oc.php?id=65488

Orleans:
3500+: Kiner - 3150.11 MHz http://valid.x86-secret.com/show_oc.php?id=93986
3000+: andre X-X - 3215.07 MHz http://valid.x86-secret.com/show_oc?id=110890

San Diego:
4000+: El Snorro - 3750.83 MHz http://valid.x86-secret.com/show_oc?id=76885
3700+: Misteroadster - 3704.71 MHz http://valid.x86-secret.com/show_oc?id=38280
3500+:

Venice:
3800+: PC Ice - 3878.05 MHz http://valid.x86-secret.com/show_oc.php?id=10699
3500+: cpulloverclock - 3571.42 MHz http://valid.x86-secret.com/show_oc?id=46575
3400+: damarble - 2850.66 MHz http://valid.x86-secret.com/show_oc.php?id=67088
3200+: s7e9h3n - 3723.6MHz - http://www.xtremesystems.org/forums/...8&postcount=12
3000+: cpulloverclock - 3644.18 MHz http://valid.x86-secret.com/show_oc?id=41857
3000+ (754): [ZIZI]@XS - 3207.22 MHz http://valid.x86-secret.com/show_oc?id=83602

Winchester:
3000+: easypanic - 3412.61 MHz http://valid.x86-secret.com/show_oc?id=15933
3200+: NewBeetle - 3463 MHz http://cal930.sakura.ne.jp/cgi-bin/c...le/3463MHz.gif
3500+: Boogotop - 3271,70 MHz http://www.vr-zone.com/ocdb/data.php...tel_or_amd=amd

Opteron

Toledo:
Opteron 180: cupra - 3450.99 MHz http://valid.x86-secret.com/show_oc?id=76888
Opteron 175: ixtapalapaquetl - 3420.73 MHz http://valid.x86-secret.com/show_oc?id=80573
Opteron 170: Absolute_0 - 3473.7 Mhz http://valid.x86-secret.com/show_oc?id=74927
Opteron 165: Brian Mychajluk (TheLawnRanger) - 3460 MHz http://valid.x86-secret.com/show_oc?id=112439

Venus:
Opteron 154: s7e9h3n - 4005.19MHz - http://valid.x86-secret.com/show_oc?id=70094
Opteron 150: dumo - 3769 MHz http://img226.imageshack.us/img226/8...shot0930im.jpg
Opteron 148: SPLMANN Overclocking Team Switzerland - 3788.19 Mhz http://valid.x86-secret.com/show_oc?id=97911
Opteron 146: MOSO - 3813 Mhz - http://valid.x86-secret.com/show_oc.php?id=70263
Opteron 144: Overcrash - 3657.8 MHz http://valid.x86-secret.com/show_oc?id=75334

Turion

ML series:
Turion ML-42: hey & tosunermc - 2833.5 MHz http://valid.x86-secret.com/show_oc.php?id=103141
Turion ML-37: Trouffman - 2288.56 MHz http://valid.x86-secret.com/show_oc.php?id=74997
Turion ML-32: victorsea - 2844.9 4MHz http://valid.x86-secret.com/show_oc.php?id=62558
Turion ML-30: Richard Cheung - 2840.1 MHz http://valid.x86-secret.com/show_oc.php?id=28080

MT series:
Turion MT-40: ryan.crest - 2935.4 MHz http://www.xtremesystems.org/forums/...d=11416511 89
Turion MT-32: marcoland - 2745.4 MHz http://img466.imageshack.us/my.php?image=305mhz9cj.jpg
Turion MT-30: Badong - 3200 MHz http://valid.x86-secret.com/show_oc.php?id=82405

Sempron

Sempron64 3400+ (Manila): Ozzimark - 3119.6mhz http://valid.x86-secret.com/show_oc?id=97360
Sempron64 3400+ (Palermo): fr@me - 2928.25 MHz http://valid.x86-secret.com/show_oc?id=66864
Sempron64 3300+ (Palermo): AndreyKV & Nickolas - 3070 MHz http://valid.x86-secret.com/show_oc?id=84032
Sempron64 3200+: AeGis - 3065.32 MHz http://valid.x86-secret.com/show_oc.php?id=78404
Sempron64 3100+ (Palermo): dsl & Spd - 3195.79 MHz http://valid.x86-secret.com/show_oc.php?id=100426
Sempron64 3000+ (Manilla): QSS - 2669.33 MHz http://valid.x86-secret.com/show_oc?id=106505
Sempron64 3000+ (Palermo) S939: pr1nce - 3208.18 MHz http://valid.x86-secret.com/show_oc?id=75376
Sempron64 3000+ (Winchester): TiN_EOF - 3031.20 Mhz http://www.topmods.net/tin/3031.png
Sempron64 3000+ (Palermo): dsl & Spd - 3316.08 MHz http://valid.x86-secret.com/show_oc.php?id=110827
Sempron64 2800+ (Manilla): QSS - 2747.59 MHz http://valid.x86-secret.com/show_oc?id=107555
Sempron64 2800+ (Palermo): fr@me & Xupyp1 - 3311.93 MHz http://valid.x86-secret.com/show_oc?id=93955
Sempron64 2600+ (Palermo): pawell262 - 3220.99 MHz http://valid.x86-secret.com/show_oc.php?id=67685
Sempron64 2500+ (Palermo): JunQiang - 3003.86 MHz http://valid.x86-secret.com/show_oc.php?id=75385
Sempron 3100+ (Paris): fr@me & Xupyp1 - 2731.31 MHz http://valid.x86-secret.com/show_oc?id=85371
Sempron 2800+ (Palermo): turez - 2498.25 MHz http://valid.x86-secret.com/show_oc.php?id=38643
Sempron Mobile 3000+ : hs - 2972.41 Mhz http://valid.x86-secret.com/show_oc?id=69610

And I am sick and tired of hearing AMD can't OC above 3ghz BS. Ever forget about the 939? More then one person has gotten these clocks before with regular chips. Some on AM2 have gotten a bit higher lately. But it has nothing to do with production limits. It has to do with what they did to AM2 chips in the fab process transformation from SOI-1 to SOI-2 witch sucks ass compared to the 939 fabed chips. Look at the clocks, no AM2 can reach 939 OCabilities and perform vr'ly the same. The only reason to go AM2 is for future chips. 939's are dead only because they had to move on. 939's are currently the better OCing socket AMD has. If they stayed with DDR1 until 2007 when K8L came out they should have made the change then to DDR2. Right now was a very bad time. Don't beleave me. Look at the damn 1000mhz OC difference 939's get over AM2 OC's.

All some old school dude has to do is flash his 3.8ghz OC over here and tell you how AMD's can reach 3.8ghz or 4.2ghz for that matter like some here have. It may not be stable, but it can bench, and it can show the clocks. Thats what we want is to show off right. Buy a 939 if you want to show off like these people above. Don't expect AM2 to get speeds you know are impossible unless SOI-3 really will make the difference.

I think they really should go back to SOI-1 if they want to do OCing more and try to perfect that process instead. Amazing SOI-2 was such a crappy process. They tried to make it better but apperently they ran into a wall with trying to make it more EE. To a OCer AMD fan would rather have a 939 then a AM2. Or a C2D. Cus something went seriously wrong when AMD made the transission to AM2. It was a big mistake.

[Astennu]

Well i dont know about that SOI-1 and SOI-2 thingy. But the K8 Will needs some ajustments to run 3000+ MHz without extra volt ( 1.35v ) stock.

I hear that the C2D's will do arround 3600 Max with toch volt air ?

I wonder how far the K8L can go. Its a Tweaked K8 core so i guess 3600 Will be the max we will see on 65nm. Somehow i dont think the K8L will clock like the C2D cores. But i can be wrong ( i hope i'm wrong i love AMD )

My current Opteron 170 S939 dous. 2800 MHz @ 1.267v Prime stable. When i whant 2900 MHz i need 1.35v. Its a CCBBE 0617 FPMW. This is air cooled with a freezer 64 Pro. @ 1800 RPM max. The most 170's will not do these speeds / volt. But there are some..

[LOE]

it doesn't really matter how much the K8L will clock

everything depends on how efficient is the design of the chip
C2D will probably clock higher cause it has a longer pipe

but then again... amd will probably have a mature SiGe process by the time K8L is out

also there are the IPCs, that can compensate for frequency

[Astennu]

Well i think a Dual core K8L will outperform a C2D @ performance per clock. But i don't think the K8L clock more then 3400 in the beginning.

K8L will be native quad core @ 2900 MHz/ But Intel will also have a Native quad core by that time. It will probably be able to clock 3400-3600 MHz. So AMD will need to make the K8L very good in performance per clock if they want to beat it.

If will be a hard fight. Intel will have a new ref of the C2D by the time AMD that has the K8L.

[LOE]

But Intel will also have a Native quad core by that time

I think they won't
Lastest info was that yorkfield is just a dumb shrink - at least thats what I heard

[Astennu]

So u think that they will make a new ref C2D and then put 2 of those together again to get a quad core ?

AMD will have an advantage then. Because the K8L is a native Quad design.

[zir_blazer]

Some OPN data...

ADO3600IAA5DD (A64X2 3600+, 1.9 GHz, 512 KB Cache L2 * 2) AsRock BIOS Processor support list (http://64.233.161.104/search?q=cache:cR64GtORd5IJ:www.asrock.com.tw/support/CPU_Support/show.asp%3FModel%3DAM2NF3-VSTA+ADO4000IAA5DD&hl=es&gl=ar&ct=clnk&cd=1)
ADO4000IAA5DD (A64X2 4000+, 2.1 GHz, 512 KB Cache L2 * 2) AMD Compare
ADO4400IAA5DD (A64X2 4400+, 2.3 GHz, 512 KB Cache L2 * 2) AMD Compare
ADO4800IAA5DD (A64X2 4800+, 2.5 GHz, 512 KB Cache L2 * 2) AMD Compare
ADO5000IAA5DD (A64X2 5000+, 2.6 GHz, 512 KB Cache L2 * 2) AMD Compare

All them will be Rev. G1. Voltage ranges from 1.25V to 1.35V, with a TDP of 65W (ADO OPN, like Energy Efficient ones). The A64X2 3600+ is still unofficial, but its pretty probable than it is for real.

[Lightman]

So u think that they will make a new ref C2D and then put 2 of those together again to get a quad core ?

AMD will have an advantage then. Because the K8L is a native Quad design.

Yes! But Intel will add (or enable ;) ) SSE4 in 45nm cores, native QC part based on NGMA is expected sometime in 2008.

On the other hand AMD also added extension to SSE in K8L/K10, but at this time we don't know if it is copy of iSSE4 or something different like iSSSE3 (from Conroe).

[Astennu]

Yes! But Intel will add (or enable ;) ) SSE4 in 45nm cores, native QC part based on NGMA is expected sometime in 2008.

On the other hand AMD also added extension to SSE in K8L/K10, but at this time we don't know if it is copy of iSSE4 or something different like iSSSE3 (from Conroe).

If Intel will Introduce SSE 4 in Q4 H7 i don't think AMD will have SSE 4 in the K8L.

[Lightman]

If Intel will Introduce SSE 4 in Q4 H7 i don't think AMD will have SSE 4 in the K8L.

Possible due to cross licensing. :stick:

[Astennu]

Possible due to cross licensing. :stick:

True. It could be so. But in the past Intel had a head start of half a year with SSE. But it would be great if the K8L has SSE 4.

[LOE]

SSE wont be such a big deal cause developers will need time to make use of it

so even 6 months behind are no problems for amd when it comes to sse4

[gOJDO]

BTW K8L will "NEED TO" run at 3.5ghz to run the max HT3 clocks. 2.6ghz 5.2gt/s chipset fsb... http://badhardware.blogspot.com/2006/10/amds-k8l-revealed-in-cray-rainier.html
K8L will not "NEED TO" run at 3.5GHz to run the max HTT3 clocks.
http://photos1.blogger.com/blogger/7288/2341/400/amdiobus_J9743DvWIml8.jpg
If we consider the HKEPEC table as true, than:
The differently clocked HTT is because of energy efficiency, but not because the chip is unable to achieve the max HTT clock. The bandwidth of the cHT links is tunned to be enough not to bottleneck the interCPU communication, while it wastes minimum energy.
HTT clock is not dependend of the CPU clock. For example you can run a K8 CPU at 800MHz and HTT bus at 1GHz or run a K8 CPU at 3GHz and HTT bus at 200MHz.
FX-62: OPPAINTER - 3635.9 MHz http://www.xtremesystems.org/forums/...d.php?t=100443
FX-60: gloatlizard - 3875.44 MHz http://valid.x86-secret.com/show_oc.php?id=97927
.
.
.
Sempron 2800+ (Palermo): turez - 2498.25 MHz http://valid.x86-secret.com/show_oc.php?id=38643
Sempron Mobile 3000+ : hs - 2972.41 Mhz http://valid.x86-secret.com/show_oc?id=69610
These are not everyday overclocks on normal (water or air) cooling and are achieved by dry ice cooling, just to get a validation. These CPUs run only few minutes on such freqfencies and are needing extreme hardware, a lot of people working around to achieve those benchmarks. There are a lot of threads on this forum about such overclocking sessions and most of these records are achieved by members of XS. You should check out these threads.

I have overclocked more than 20 various 90nm K8 CPUs, most of them between 1.8GHz-2.2GHz. According to my expireince, the average stable overclock is 25%, but with increased voltage. The higher clocked were less OC-able than the lower clocked.

Well i think a Dual core K8L will outperform a C2D @ performance per clock. But i don't think the K8L clock more then 3400 in the beginning.
I think that K8L will perform like Core2 clock, for clock. I don't think that it can outperform on the IPC front. K8L is a 3 issue core, while Core2 is 4+1 issue. Both have same SSE performance, K8L will have a little stronger FPU, while Core2 has stronger ALU.
About the clock speed I think same as HKEPEC rumors are saying that the first K8L will not break 3GHz. So far their 65nm process offers less perfromance than the 90nm and the 90nm K8 CPUs are at their freqfency edge at 3GHz. Quadcore K8L has more more than twice transistors than K8 dualcore with 2x1MB L2 and 3GHz K8L CPUS in the same power envelope will have significantly lower yields than K8 dualcores.

K8L will be native quad core @ 2900 MHz/ But Intel will also have a Native quad core by that time. It will probably be able to clock 3400-3600 MHz. So AMD will need to make the K8L very good in performance per clock if they want to beat it.
There will be no native quadcore Core2 CPUs, but it is irrelevant. The "true vs glued" is pointless argument used by AMD die-hard fanboys. They usually are ignoring the facts and are unable to understand how well Core2 Quad scalles in perfromance while using the 1066MHz FSB.
IMO AMD will not catch up Intel's CPU perfromance in the next year. To be sucessfull AMD have to offer competitve prices for the same perfroming Intel platforms, considering the total cost of CPU, mainboard and RAM.

[zakelwe]

Do not forget the extra cache, 12MB v 8MB so in effect 3MB per core up from 2MB for Kentsfield. This will be as well as SSE4 , higher default clock speeds (possibly 3.46-3.73) and maybe DDR3 memory.

So Intel are betting on the old world style bigger and faster it seems.

I think native cores, hyperthreading and on die memory controller are scheduled for 2008 with Bloomfield ??

Regards

Andy

[GoThr3k]

info about intel seems quite irrelivant to me

if you want that, go to the intel section,i think people here want to know how well or bad the Brisbanes overclock

[Mikael]

Sine www.amdcompare.com has been updated with revision G info, we now know the following:

- All the currently released 65nm X2 CPUs (4000+, 4400+, 4800+ and 5000+) have a 65W TDP.

- Each model seems to be available in one 1.25V and one 1.35V version. No idea how one would know which version the package contains...

Either way, core voltage doesn't really seem to have dropped from the 90nm CPUs. This will likely limit the power savings somewhat. It also doesn't bode well for overclockers...

[MaSell]

Who want to run subzero. I brisbane is coldbugged :stick:

[Astennu]

story.....

If what you are saying is true. AMD is dead in the water. If A K8L @ 2.9 GHz performance is on the same level as a kendsfield @ 2.9. They ave a big problem. the Kendsfield will be able to clock @ 4.0 If they tweak the core a bit.

I think it will be dark Times for AMD then. And i hope for them they will come with a fully new and fast architecture fast. But Intel will do the same thing..... It looks like we are going back to the times of the K6 and the end of the K7 where intel cpu's where outperforming the AMD's. K7 XP core VS P4 @ 3.0+ with HT was really hard. It turned when they came with the K8.

But enough about C2D vs K8L. How are the new 65nm K8 cores clocking ? are they better ? or is it the same ?

[gOJDO]

The 65nm K8 CPUs are paper lounched today. The highest clocked 65nm CPU is 2.6GHz.
The available data for the AMD 65nm SOI3 process is very poor, we can't be sure how well it can perfrom when AMD will mature it. Because the thicknes of the gate oxide on the 65nm SOI-3 can't be scalled further than it is on the 90nm SOI-2(reached the limit allready), AMD are using more advanced stress memorization techniques to improve the mobility in the circuits.
So far on the 65nm SOI3, AMD are using 4 stressors, but the improvements achieved by stress memorization are lower when applied to smaller surfaces.
The 65nm will offer better energy efficiency, but will not offer higher clocks, at least while their 65nm process is still unmatured.

[Varsos]

Come on metro HIT US.........we are sitting on nails here.....

[biohead]

lounched
launched

[gOJDO]

10x :)

[Anonymous]

The "true vs glued" is pointless argument used by AMD die-hard fanboys. They usually are ignoring the facts and are unable to understand how well Core2 Quad scalles in perfromance while using the 1066MHz FSB.

I am glad at least a few can see that this "native" vs "glued" is just marketing bull:banana::banana::banana::banana:.

[metro.cl]

Who want to run subzero. I brisbane is coldbugged :stick:

should be as bad or even worst than 90nm parts.

sorry no oc report so far, as i said in other threads is a good friend who has the cpu, and he is also teasing me. So i believe overclock is not a big factor maybe 3.1GHz for super pi not sure thought i've been busy working i'll try to get more info tonight

[madcho]

should be as bad or even worst than 90nm parts.

sorry no oc report so far, as i said in other threads is a good friend who has the cpu, and he is also teasing me. So i believe overclock is not a big factor maybe 3.1GHz for super pi not sure thought i've been busy working i'll try to get more info tonight
:( :mad:

[Nedjo]

People are missing the point. This CPU was newer meant to be launched with fanfare. This is just a date when large OEM’s and channel will start to receive first quantities of 65nm CPU’s that are THE SAME, as the previous 90nm X2!

It usually take couple weeks when channel has build up level of supply to be able to deliver PIB broadly.

In case of traditional launch (say C2D, or GF 8800 series), company first build that desired level of stock, and then arrange official introduction to the market..

On the other side, after fiasco with QuadFX (too little, to late, and to complex), AMD should have kept their mouth shut, so that people like our friend black sheep, wouldn’t be in position to spread story about paper launch.

[gOJDO]

so that people like our friend black sheep, wouldn’t be in position to spread story about paper launch.
It is not a story, my friend. It is a fact.

[Lightman]

It is not a story, my friend. It is a fact.
Fact for one people is not always a fact for another ;) .
Let say you're HP, Dell or any other big player, do you think today is paper launch for them as well??:)

For me this launch will be 'ok' if at least someone will soon do decent review! (Yes I'm thinking about you Metro.cl :p: )

[gOJDO]

http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~114609,00.html
OEMS offering systems ready to experience responsive and spectacular high-resolution 3D visual effects of Microsoft Windows Vista™ with 65nm AMD Athlon 64 X2 dual-core processors immediately, and beginning in Q1 of 2007, include Acer, Dell, Founder, Gateway, HP, Lenovo, Packard Bell, and TongFang, as well as leading system builders worldwide.
Can you explain what does this means?

[VulgarHandle]

sure, some are launched now, and rest Q1 2007

[Lightman]

http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~114609,00.html

Can you explain what does this means?

Sure!
Systems form above companies will be available to purchase at the mentioned date, but they allready are building this computers using 65nm CPUs.
If you're Dell for example you will not put new Box on sale before you build up adequate amount in magazines, yes? Otherwise you will end up with many people complying about tight supply or lack of availability.

That is my point of view.

PS. You're quick in responding :toast: .

[gOJDO]

Before Core2 was released, there were a lot of reviews and benchmarks. On the release date we were able to buy it, although its availability was low.
If we can buy any 65nm K8 CPU today or this week, then we can consider it as non paper launched.

[zakelwe]

info about intel seems quite irrelivant to me

if you want that, go to the intel section,i think people here want to know how well or bad the Brisbanes overclock

You can either have information about K8L and Yorkfield or look at a blank page.

I bet nvidia wish they could keep their secrets as good as this ! :D

Regards

Andy

[speed bump]

You can either have information about K8L and Yorkfield or look at a blank page.

I bet nvidia wish they could keep their secrets as good as this ! :D

Regards

Andy

I bet everyone does, at this rate in January or March when someone can actually buy one they will find out that they come packaged in a retail bannana or something crazy.

[terrace215]

Sine www.amdcompare.com has been updated with revision G info, we now know the following:

- All the currently released 65nm X2 CPUs (4000+, 4400+, 4800+ and 5000+) have a 65W TDP.

- Each model seems to be available in one 1.25V and one 1.35V version. No idea how one would know which version the package contains...

Either way, core voltage doesn't really seem to have dropped from the 90nm CPUs. This will likely limit the power savings somewhat. It also doesn't bode well for overclockers...

No, the 1.25/1.35v thing means they sometimes drop down lower, but at max performance they are 1.35v

[Starscream]

No, the 1.25/1.35v thing means they sometimes drop down lower, but at max performance they are 1.35v


or it could mean that there will be normal models and EE models as with the 90nm AM2 CPUs.

cause if it were just a part of C&Q then i doubt theyd mention it.

[Serge84]

K8L will not "NEED TO" run at 3.5GHz to run the max HTT3 clocks.
http://photos1.blogger.com/blogger/7288/2341/400/amdiobus_J9743DvWIml8.jpg
If we consider the HKEPEC table as true, than:
The differently clocked HTT is because of energy efficiency, but not because the chip is unable to achieve the max HTT clock. The bandwidth of the cHT links is tunned to be enough not to bottleneck the interCPU communication, while it wastes minimum energy.
HTT clock is not dependend of the CPU clock. For example you can run a K8 CPU at 800MHz and HTT bus at 1GHz or run a K8 CPU at 3GHz and HTT bus at 200MHz.

These are not everyday overclocks on normal (water or air) cooling and are achieved by dry ice cooling, just to get a validation. These CPUs run only few minutes on such freqfencies and are needing extreme hardware, a lot of people working around to achieve those benchmarks. There are a lot of threads on this forum about such overclocking sessions and most of these records are achieved by members of XS. You should check out these threads.

I have overclocked more than 20 various 90nm K8 CPUs, most of them between 1.8GHz-2.2GHz. According to my expireince, the average stable overclock is 25%, but with increased voltage. The higher clocked were less OC-able than the lower clocked.


I think that K8L will perform like Core2 clock, for clock. I don't think that it can outperform on the IPC front. K8L is a 3 issue core, while Core2 is 4+1 issue. Both have same SSE performance, K8L will have a little stronger FPU, while Core2 has stronger ALU.
About the clock speed I think same as HKEPEC rumors are saying that the first K8L will not break 3GHz. So far their 65nm process offers less perfromance than the 90nm and the 90nm K8 CPUs are at their freqfency edge at 3GHz. Quadcore K8L has more more than twice transistors than K8 dualcore with 2x1MB L2 and 3GHz K8L CPUS in the same power envelope will have significantly lower yields than K8 dualcores.


There will be no native quadcore Core2 CPUs, but it is irrelevant. The "true vs glued" is pointless argument used by AMD die-hard fanboys. They usually are ignoring the facts and are unable to understand how well Core2 Quad scalles in perfromance while using the 1066MHz FSB.
IMO AMD will not catch up Intel's CPU perfromance in the next year. To be sucessfull AMD have to offer competitve prices for the same perfroming Intel platforms, considering the total cost of CPU, mainboard and RAM.

Are you questioning the reliability of roadmaps and data sheets previded to and from mobo manufactuers from AMD and intel directly? Then I could very well say York will never clock at 3.73ghz on stock because its wrong when in reality it WILL indeed clock at that speed. AMD will release up to 3.5ghz K8L's in 2008 to counter it. And run HT3 at full speed. The links arn't the same anymore man. Do you understand the HT3 links are different from standard HT links they are nolonger fixed at one speed.

See they run at what clock the K8L main clock is at for model number and greater differences since the pads are also in the cpu for hyper transport AMD made the hyper transport FSB rely on clock speed and give greater model differences. No one CPU with HT3 will have the same link speed. They made it flexible with different model's according to clock speed. So the HT3 link is nolonger independent.

Lower clocked CPU's will run 1.6ghz hyper transport. Highest clocked cpus will run 2.6ghz hyper transport. That is what HT3's max is. 5.2GT/s. You don't understand how K8L will use HT3 links do you? Its not the same. All mobos that are capable of HT3 will say supports HT3 speeds up to 2.6ghz Hypertransport or something close to that. Maybe less when the AFX's only do 2.9ghz on launch. They are launching a new arc. Its only natual to expect a new ark to start at a week speed and raise over time like the old days. K8L by far is not limited at 2.9ghz that would be a joke in its self. York is clocked at 3.7ghz for a reason you know agenst the AFX.

Infact we won't know anything performance wise until there is real physical hardware to show to bench agenst. We can all think how K8L might perform but we will never know until we can OC the new arc. Its amazing how Conroe OC's so well and is a new arc. We should expect the same from K8L because its not K8 by a long shot. Not one part of the cpu is the same if compared die for die to K8L. 65nm K8's are arc limited at a sertent speed. It has nothing to do with fab process only arc scalability and performance. K8's arc is 3 years old, K8L is totally new like conroe. Don't expect any less then the unexpected because as far as anybody is consirned its limits are unknown.

[red]

If this is not a paper launch then I don't know what is:rolleyes: We've already known about the December 5th date, now they're saying "Hey everyone they exist but you can't touch!"..

[nn_step]

If this is not a paper launch then I don't know what is:rolleyes: We've already known about the December 5th date, now they're saying "Hey everyone they exist but you can't touch!"..
Wait you didn't your 65nm K8 in the mail yet? :rolleyes:
Don't you know how to order parts? :p:

[terrace215]

Why is AMD keeping the Brisbane die size a state secret?

Only the Inq makes reference to it, saying they are not putting it in the official info packet, but it is somewhere in 120-130 mm^2 range.

http://www.theinquirer.net/default.aspx?article=36162

This would be pretty bad, if true.

90nm equiv is 183mm^2.

If 65nm is 125mm^2, say, that is 68% of the size of 90nm.

A good 65nm process shrink should've accomplished something closer to 58%, or 105mm^2.

[gOJDO]

The links arn't the same anymore man. Do you understand the HT3 links are different from standard HT links they are nolonger fixed at one speed.
HTT clock is not dependend on the CPU clock!
HT links have never been fixed at one speed(freqfency). Unlike HTT2 and HTT1, HTT3 offers dynamic link width and dynamic clock adjustment of transmit and receive links. It also offers ACPI compliant dynamic power management. HTT3 has the same concept and works exactly like HTT2 and HTT1 and it is backwards compatible to both HTT2 and HTT1.
http://www.hypertransport.org/images/features_chart.gif
The new 3.0 standard nearly doubles the clock speed and bandwidth of the previous HyperTransport 2.0 specification. In addition, HyperTransport 3.0 supports a variety of new features including AC coupling mode, hot plugging, un-ganging mode and dynamic power management for the support of extended signal transmission distance, typical of backplane and chassis-to-chassis implementations.
HyperTransport technology devices are designed to operate at multiple clock speeds from 200MHz up to 2.6 GHz, and utilize double data rate (DDR) technology transferring two bits of data per clock cycle, for an effective transfer rate of up to 5.2 Gigatransfers/sec in each direction. Since transfers can occur in both directions simultaneously, aggregate transfer rates of 20.8 Gigabytes/second and 41.6 Gigabytes/second can be achieved with 16-bit wide and 32-bit wide link configurations respectively. To allow for system design optimization, the clock speed of the receive and transmit links may be set at different rates.
It is important to note that HyperTransport 3.0 specification is fully backward compatible with prior 2.0 and 1.x devices. This means that existing investments in the technology can continue to be leveraged in the future.
http://www.hypertransport.org/
http://www.hypertransport.org/docs/tech/HT3pres.pdf
http://www.hypertransport.org/tech/tech_specs.cfm
http://www.hypertransport.org/tech/tech_specs_conn.cfm

[Shadowmage]

Why is AMD keeping the Brisbane die size a state secret?

Only the Inq makes reference to it, saying they are not putting it in the official info packet, but it is somewhere in 120-130 mm^2 range.

http://www.theinquirer.net/default.aspx?article=36162

This would be pretty bad, if true.

90nm equiv is 183mm^2.

If 65nm is 125mm^2, say, that is 68% of the size of 90nm.

A good 65nm process shrink should've accomplished something closer to 58%, or 105mm^2.


AMD's Gen 1 65nm is a combination of 65nm and 90nm technologies, I recall.

[biohead]

AMD's Gen 1 65nm is a combination of 65nm and 90nm technologies, I recall.
That sounds highly unlikely

[Mats]

My first guess is that 65 nm will bring AMD up to speeds are Intel doing now, and that's very optimistic.
And my second guess is that K8L will be performing something like a C2Q are doing now.
Just when AMD finally seems to be getting it right again, Intel will launch 45 nm and take the lead, again. I really hope I'm wrong.

I really think that AMD could have done some changes earlier, very small changes have been made with the K8 core since the release in April 2003. But then again they have been very successful with it, and they don't have all that cash Intel got for R&D.
Yeah, I'm just speculating. And I pretty much prefer AMD.

[Serge84]

Would you PLEASE shut up about HT3? HT2.0 isn't even fully saturated yet (HECK, it isn't even HALF saturated). You talk like HT3 is the second coming of the messiah or something and is going to give AMD this huge boost of CPU performance.

It is just AMD future-proofing itself for 16+ server space. It is pointless beyond that and has no impact vs. HT2 on performance for desktop users even under extreme multitasking conditions.

I never said it was anything of the sort. lol Your funny. Its not great at all. York will do great on CSI infact AMD doesn't stand a chance with HT3. HT3 was made to only allow more bandwidth and insure there would never be a bottleneck between 16cores on a 4 socket machine. It will not increase performance only allow K8L to use its full power and scale better with lower Lancery.

HT3 hasn't even been finalized yet. We know almost nothing about it eather. But it's not for performance only for bandwidth and anti-bottlenecks. Thats why CSI was made so Conroe could show its full power and have no bottlenecks. Thats why york is going to kick so much ass if it is on csi hopefully.

To the other guy...
K8L will have a HT3 link at different max settings for the cpu and mobo at the time its launched. Those reversions are only for the mobos chipset max speed. They have nothing to do with the max speed of what ever rev they put in the cpu and each cpu will have a greater or smaller capable HT3 pad. Because there will be a few stages of HT3 put into mobos the 1.5ghz in 2007 in AM2+ up to the 2.6ghz in 2008 AM3. All the CPU's like how semprons only support 800mhz Hyper transport of today. There will be cpus like this in the lower budget and then about 1400mhz or 1300mhz hyper transport 3 for those budget models or something close to it. Then 1.5 to 1.7ghz HT3 for the X2's some of the higher end ones will go up to 2.1ghz but this is at high clock speeds and prob performance rated parts, then 1.8ghz to 2.1ghz for the X4's and AFX's. Its said on there between night and day the proof is right before your eyes the max GT/s each cpu can support. Hello!

http://badhardware.blogspot.com/2006/10/amds-k8l-revealed-in-cray-rainier.html

pop...

http://photos1.blogger.com/blogger/7288/2341/1600/k8l_yjM7OCMYeXam.jpg

Says right here. According to this chart... the HT3 link depending on model and series will support from 3.0GT/s to 3.4GT/s for the Legacy models, 3.0GT/s for Mainstream, and 4.0GT/s to 4.2GT's for Performance and Ultimate performance models. Can you read the chart? As pane as day to my eyes. Those speeds are strictly depending on the clock speed do the math. If you can't see that then your hopeless.

I/O Speed is the GT/s of what the cpu supports at what clock speed. For example you see 2ghz to 2.9ghz that would mean 2ghz would have a 1500mhz HT3 link. Then the one at 2.9ghz would have a 2.1ghz HT3 link. Wow wasn't that hard to figure out. What ever model inbeween would have a different HT3 link speed. So if this is wrong then the MB manufactures are wrong because it says the Source. Are you calling perfessionals the lier? All you showed me was a link to what HT3 chipsets would support. Not the CPU's at the time. Says in the link I just gave that it would support between so and so speed for a different model and series according to that chart ofcorse. Only logic can tell you that. You have to see it for yourself in what they mean by 3.0GT/s to 4.2GT/s matches the other charts perfectly right below it that say I/O Bus Speeds for K8L. Doy!

I'm pretty sure I'd listen to reliable sources other then yours. Sorry but you didn't prove anything to me at all here. I hope you learned something about what I was talking about in HT3. Doesn't matter what the max chipset supports to a point, cus you can OC the chipset, but again its just saying the max the chipset can support. But the K8L's will support multi speeds in HT3 so depending on the chipset you can put in a faster HT3ed cpu for a faster chipset. BOTH support different MAX HT3 Speeds and the link can be set in the bios to X10 or whatever they will have it like. This is spread over a few months time. K8L won't even fully support HT3's max until later and there chipsets will be limited in the same way. The REV's of both matter only if you want to run the max the cpu's HT3 link can take. Doesn't matter for anything else cus it will run higher CPU models HT3 speeds only at the chipsets max default speed. Not the cpu speed. I mean it doesn't make OCing limited because both have a max only for the HT3 link at max you can turn that down ofcorse in a oc. I'm just saying thats what the max multiplier will have to support HT3 link REV so and so for chipset and cpu. They will not come out with a chipset or cpu that runs max HT3 yet until 2008. They will both have different HT3 speeds and multipliers for each model.

[LOE]

So... HT3 is not ready yet, but It will be fast enough even if not running at it's MAX

Does anyone have an idea how INTEL are doing with CSI?
I heard the project was canned, then delayed until 2009

[doompc]

Serge, Yorkfield is a die shrink of Kentsfiel with little improvements, as it is based on Wolfdale (wich has 6MB L2 cache and SSE4) and FSB1333.

CSI and Intel's IMC will only show up in next generation (Nehalem), wich will be native quad core with HT (8 logical cores), in late 2008...

[gOJDO]

@Serge84
:nono:
Dude I have learned nothing from you. Your are puting words into my mouth and make conclusions about something that I have never said. I hate to repeat my self, but I'll repeat to point at your missleading statements:
BTW K8L will "NEED TO" run at 3.5ghz to run the max HT3 clocks. 2.6ghz 5.2gt/s chipset fsb... http://badhardware.blogspot.com/2006/10/amds-k8l-revealed-in-cray-rainier.html
Do you realise that what you said is not true?
Here is what I answered:
K8L will not "NEED TO" run at 3.5GHz to run the max HTT3 clocks.
There are reasons why all K8L CPUs will not run HTT at same freqfency. For some of those reasons, all K8 CPUs are not running HTT at same freqfency(all Sempron/all s754/all sS1 at 800MHz, Athlon64/X2/FX/Opteron on s939/s940/sAM2/s1207 at 1GHz).

I have never said that K8L CPUs will run the HT links at same/top speed. Although your epxlanations are wrong and missleading, are not necessary for me and can go down the toilet. Skip reading/learning from blogs and start learning from realible sources like from the inventors and developers of the hardware/tehcnology you are talking about. In this case that would be AMD (www.amd.com) and HyperTransport (www.hypertransport.org). I allready posted link of HTT3 paper (http://www.hypertransport.org/docs/tech/HTC20051222-0046-0008-Final-4-21-06.pdf) (500 pages of technical documentation), but looks like you haven't opened it yet.

Please read the HTT specifications, how it works and how it is implemented. Also read some papers about K8L and its power management. I hope that you will come to the conclussion that the speed of the HTT is not dependend on the CPU freqfency.

A little help: If AMD want they can use the max HTT3 freqfency on 100MHz K8L CPU. The cores on the K8L will have independend clocks and voltage. That means that one core can run at top freqfency, while other at bottom. So, where is the logic with HTT dependecny on the CPU clock?:rolleyes:
Do you know that the HTT3 on the K8L will have power management also?

[Serge84]

@Serge84
:nono:
Dude I have learned nothing from you. Your are puting words into my mouth and make conclusions about something that I have never said. I hate to repeat my self, but I'll repeat to point at your missleading statements:

Do you realise that what you said is not true?
Here is what I answered:

There are reasons why all K8L CPUs will not run HTT at same freqfency. For some of those reasons, all K8 CPUs are not running HTT at same freqfency(all Sempron/all s754/all sS1 at 800MHz, Athlon64/X2/FX/Opteron on s939/s940/sAM2/s1207 at 1GHz).

I have never said that K8L CPUs will run the HT links at same/top speed. Although your epxlanations are wrong and missleading, are not necessary for me and can go down the toilet. Skip reading/learning from blogs and start learning from realible sources like from the inventors and developers of the hardware/tehcnology you are talking about. In this case that would be AMD (www.amd.com) and HyperTransport (www.hypertransport.org). I allready posted link of HTT3 paper (http://www.hypertransport.org/docs/tech/HTC20051222-0046-0008-Final-4-21-06.pdf) (500 pages of technical documentation), but looks like you haven't opened it yet.

Please read the HTT specifications, how it works and how it is implemented. Also read some papers about K8L and its power management. I hope that you will come to the conclussion that the speed of the HTT is not dependend on the CPU freqfency.

A little help: If AMD want they can use the max HTT3 freqfency on 100MHz K8L CPU. The cores on the K8L will have independend clocks and voltage. That means that one core can run at top freqfency, while other at bottom. So, where is the logic with HTT dependecny on the CPU clock?:rolleyes:
Do you know that the HTT3 on the K8L will have power management also?

Your also making conclusions in a war that stops here. Just give it a rest and we will see whos right the MBM's or you ok cool down man its really just speculation then if you don't beleave what the MBM's are trying to say, believe what you want, I don't care. And K8L supports HT3 and standard HTL. If you put a K8L into AM2 it will lose HT3 features and power management features being able to run with multi power planes and voltages independent for each core. So what it runs a little slower and may have a minor bottleneck maybe not. Will not run as good and heat up more because all cores are running at the same volts. Even needing a 100A socket won't matter. K8L is desined to work in AM2 even with a lower. "Amp-level" it might need a little more volts is all to fix that. Yes I know what K8L's power management means. It also works similar to conroes power management for areas in the die that are not in use and will turn on and off depending on when not needed or when needed. Pretty cool stuff.

[***Deimos***]

Would you PLEASE shut up about HT3? HT2.0 isn't even fully saturated yet (HECK, it isn't even HALF saturated). You talk like HT3 is the second coming of the messiah or something and is going to give AMD this huge boost of CPU performance.

It is just AMD future-proofing itself for 16+ server space. It is pointless beyond that and has no impact vs. HT2 on performance for desktop users even under extreme multitasking conditions.

A long long time ago, even before 90nm, tomshardware ran complete set of benchmarks on Athlon64's with HT reduced from 5x to 4x, and even 3x. Performance was nearly identical across the board.

HT's impact on performance is turning out to be nearly as big a computer myth as AGP's impact on video card performance (ie tomshardware also did a whole road of AGP tests, and performance only started substantially degrading when you got to AGP2x)

AMD's Gen 1 65nm is a combination of 65nm and 90nm technologies, I recall.
I'm by far no expert, but transistors on the die are all sorts of sizes. Smallest length, smallest gate, highest performance transistors are used where needed most in critical path sections. Other rules for fan-out, matching, drivers, dynamic logic, repeaters etc.. still apply ofcourse.

(BTW: I was like !!! when I read 9 metal layers, wow, thats a lot of complexity)

[LOE]

A long long time ago, even before 90nm, tomshardware ran complete set of benchmarks on Athlon64's with HT reduced from 5x to 4x, and even 3x. Performance was nearly identical across the board.

HT's impact on performance is turning out to be nearly as big a computer myth as AGP's impact on video card performance (ie tomshardware also did a whole road of AGP tests, and performance only started substantially degrading when you got to AGP2x)


You are talking about single core chips - one core doesn't need that much BW, but 2 and 4 cores do

HT is not about increasing performance - it is about eliminating a possible bottleneck

[STEvil]

The tests were also performed here. All the way from 2x to 5x. Cant quite remember the name of the person who did them, though.

Have you seen tests performed with 2+ cores? I have not, and have not seen links nor articles relating or referring to the change in performance either.

[gOJDO]

For 1P CPUs, the speed of the HTT is irelevant, be it for singlecore, dualcore or quadcore CPU. Twice K8 cores will not double or increase the traffic between the CPU and the northbridge chipset, if the rest of the hardware remains same.

[ozzimark]

For 1P CPUs, the speed of the HTT is irelevant, be it for singlecore, dualcore or quadcore CPU. Twice K8 cores will not double or increase the traffic between the CPU and the northbridge chipset, if the rest of the hardware remains same.
thank you. all communication between cores on the same cpu die for amd occurs within the chip. no HT links needed :toast:


so for single socket setups, this is true:
http://eclipseoc.com/index.php?id=6,47,0,0,1,0

[nn_step]

ozzi you seriously need to rerun those tests with more than just a single Geforce4MX 440. A CHEAP X1600XT would be more realistic

[gOJDO]

ozzi you seriously need to rerun those tests with more than just a single Geforce4MX 440. A CHEAP X1600XT would be more realistic
Even if X1600XT is used, why do you think that the number of cores on the CPU will make any difference in the traffic between the CPU and the northbridge?
Also, MX440 has with 64MB or with 128MB, while X1600XT has 256MB. More VGA memory reduces the texture traffic between the VGA, northbridge and the CPU(ODMC) in the K8 case.

[Lightman]

In case you didn't notice:

http://img377.imageshack.us/img377/8012/bris01vt5.jpg

:rolleyes:

http://www.xtremesystems.org/forums/showthread.php?t=124497&page=4

Now where is Metro.cl :confused:

[gOJDO]

In case you didn't notice:

http://img377.imageshack.us/img377/8012/bris01vt5.jpg

:rolleyes:

http://www.xtremesystems.org/forums/showthread.php?t=124497&page=4

Now where is Metro.cl :confused:
What about validation?

[Lightman]

Not yet :(

We need Metro.cl here!

Now I'm digging on google for more info.

Picture looks OK, not photochopped. Numbers are correct as well, but we still need validation.

[gOJDO]

The picture is 100% photoshoped!
Here we go, some better socres with correct numbers as well:
http://img89.imageshack.us/img89/9168/fakeoverclock65nmk8dy5.jpg
Oh.....
http://img213.imageshack.us/img213/1241/am265nmgv5.jpg
Yeah....
http://img377.imageshack.us/img377/8012/bris01vt5.jpg

[nn_step]

that reminds me, not to trust gOJDO's world record benchmarks in the future

[Lightman]

gOJDO your photochop is primitive!

But I spend some time now with Photoshop and I found a flaw in this very good fake!

CPU-Z is centering every information in provided box, but if you look at HT Link box, you will find that 844mhz is biased to right! Exactly same as if there should be 4 digit (like 1000) not 3 (844).

Enough!
:slapass: :slapass:

[metro.cl]

my ss is not photoshoped ;)

[Lightman]

my ss is not photoshoped ;)

no it's not ;)

but can you tell how it clock?
Any new information?

[metro.cl]

no it's not ;)

but can you tell how it clock?
Any new information?


hmm im begining to think the worst my friend wont tell me, i guess it ocs like crap or same as 90nm parts but that is just speculation (but based on am2 when it released the same thing hapened when i asked for info about performance)

[Lightman]

uhh! :(

It seems that you are the only person in the world which have remote access to AMD 65nm CPU :)
Keep trying, because we still want some more info, even if not so good :toast: :clap: :toast:

[gOJDO]

my ss is not photoshoped ;)
I know :)
It was very obvious that the OC-ed pciture is fake, made from the one you have provided.
0. the images are same(with the signature of chillehardware) except some numbers changed
1. the digits after the commas are same
2. the math is wrong, HTT Link speed should be 844.7MHz (3519.5 / 12.5 = 281.56; 281.56 * 3 = 844.68)
3. why should AMD release 2.5GHz CPU that can work at 3.5GHz at same voltage?:rolleyes:

[[XC] Lead Head]

I know :)
It was very obvious that the OC-ed pciture is fake, made from the one you have provided.
0. the images are same(with the signature of chillehardware) except some numbers changed
1. the digits after the commas are same
2. the math is wrong, HTT Link speed should be 844.7MHz (3519.5 / 12.5 = 281.56; 281.56 * 3 = 844.68)
3. why should AMD release 2.5GHz CPU that can work at 3.5GHz at same voltage?:rolleyes:

Can't argue 0 and 1.

2. CPU-Z's math isn't always completely acurate
Heres a screen shot I took, even run CPU-z yourself
http://www.imagecross.com/images/76371111111111111.JPG

3. Why does Intel release CPUs at 2.66GHz, when most of them can hit 3.2-3.5GHz easy? Simple, it would out perform the high end Extreme edition or in AMD's case the FX

[lapdog]

. Why does Intel release CPUs at 2.66GHz, when most of them can hit 3.2-3.5GHz easy? Simple, it would out perform the high end Extreme edition or in AMD's case the FX


TDP, or they are holding back for K8L.

[[cTx]Philosophy]

I still just wanna see some 3d..

[gOJDO]

3. Why does Intel release CPUs at 2.66GHz, when most of them can hit 3.2-3.5GHz easy? Simple, it would out perform the high end Extreme edition or in AMD's case the FX
0. There is no Intel CPU that overclocks from 2.4GHz to 3.5GHz at stock voltage
1. Their current CPUs, Core2 are the fastest and most competetive
2. Intel will save their CPUs headroom to release faster CPUs latter
3. AMD are bloody needing faster CPU at lower production price and at lower TDP. If that CPU can reach 3.5GHz at stock voltage, than it would be stable to run at 3.2GHz at same or little more voltage.

[Timmay]

0. There is no Intel CPU that overclocks from 2.4GHz to 3.5GHz at stock voltage


Well my Opty 144 CAB1E went from 1.8GHz -> 2.9GHz at stock volts and I have screens to prove it.

Not its not dual core, but hey.

[kemo]

Well my Opty 144 CAB1E went from 1.8GHz -> 2.9GHz at stock volts and I have screens to prove it.

Not its not dual core, but hey.
Yes but your chip is one in a million and wont do it on any mobo

[nCrusader]

AMD need atleast 4GHz on air to get close to the performance of a 3.6GHz E6600.

[gOJDO]

Well my Opty 144 CAB1E went from 1.8GHz -> 2.9GHz at stock volts and I have screens to prove it.

Not its not dual core, but hey.
What about validation? :stick:
If true, than your chip really rocks and I wonder what kind of cooling have you used.

[dinos22]

What about validation? :stick:
If true, than your chip really rocks and I wonder what kind of cooling have you used.
that's not really unusual for B1 and BY opterons...........my CABYE was 12 hours prime stable at 3.1GHz on air check sig

[gOJDO]

Your validation (http://valid.x86-secret.com/show_oc?id=70456) shows 2897MHz and the voltage is not displayed.
There is something strange, the multiplier should be 10. Does Opteron 146 have unlocked multiplier?

[kemo]

Your validation (http://valid.x86-secret.com/show_oc?id=70456) shows 2897MHz and the voltage is not displayed.
There is something strange, the multiplier should be 10. Does Opteron 146 have unlocked multiplier?
That is an especial opty with higher multi

[nullface]

http://resources.vr-zone.com//newspics/Nov06/16/AMDroadmap.png

AM2+ in Q3 07? New socket or...?

The Socket AM2+ is the immediate successor to Socket AM2 currently used in several AMD processors such as Athlon 64 X2, Socket AM2+ will be a mid migration from Socket AM2 to Socket AM3 and will be fully compatible with Socket AM2, so processors designed for Socket AM2 will work on Socket AM2+ motherboards and vice versa.

Socket AM2+ will have some differences however that will be used in AM2+ chips, there are two main features in Socket AM2+ not in Socket AM2:

* HyperTransport 3.0 operating at up to 2 GHz
* Advanced power features summarized as having split power planes, probably one for CPU Cores, and the other for the Integrated Memory controller (IMC), this will help for more advanced power management control.

AMD confirmed that AM2 processors will work in AM2+ motherboards and AM2+ processors will work on AM2 motherboards. However, due to the lack of support of HyperTransport 3.0 and separated power planes in Socket AM2 motherboards, AM2+ chips will be limited to the specifications of Socket AM2 (HyperTransport 2.0 at the speed of 1 GHz, one power plane for both Cores and IMC). AM2 chips will not benefit from faster HyperTransport and separated power planes on AM2+ motherboards as they do not support them, AM2+ motherboard then fall back to compatibility mode using AM2 specifications.

According to confirmations from AMD, Socket AM2+ will have a compatibility path with Socket AM3, AM3 processors will work in AM2+ motherboards; however, AM2+ processors will not be compatible with AM3 motherboards.

[informal]

Welsome to 2 months ago...
And it's a new chipset that supports HT3.0.
All AM2 mobos will support K10 desktop DC and QC chip variants.

[el rolio]

my brisbane 4800 is doing well. trying to remove some sort of IMC scaling wall so i can keep it going. mine also had 65W printed on the side of the box.

[JAWS]

AMD Athlon 64 3800+ Lima 2.4GHz 512KB L2 Cache Socket AM2 Processor

http://www.newegg.com/Product/Product.asp?Item=N82E16819103047


AMD Athlon 64 3500+ Lima 2.2GHz 512KB L2 Cache Socket AM2 Processor


http://www.newegg.com/Product/Product.asp?Item=N82E16819103048