CPU-Z AMD 65nm Brisbane

[Metr0]

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

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

http://www.chilehardware.com/foro/am2-b...ane-x2-t50115.html?p=659258#post659258

 

[harpoon84]

Boo! Show us the O/C details NOW!

 

[myocardia]

Well, at least it's 100 Mhz faster at stock speed.

 

[harpoon84]

Originally posted by: myocardia
Well, at least it's 100 Mhz faster at stock speed.

With 1/2 the L2 cache of a 90nm 4800+...

 

[myocardia]

Ooh, I didn't look at the amount of L2 cache. That means that it's really the same speed, but needs to run 100 Mhz faster to get the same amount of work done. That's not good.*

* There's a 75-100 Mhz performance difference between two A64's at the same speed, if one has 1MB of L2.

 

[Brunnis]

So, I guess the default voltage is 1.3V then? That's not very good... I've got three questions:

1. How low can you go on the voltage at stock frequency? Would be nice to see what kind of voltage margin there is with these 65nm cores.

2. Is there a noticeable reduction in heat at stock settings.

3. Is there any performance difference, clock-for-clock, compared to the 90nm 512kB cache versions?

 

[Metr0]

Originally posted by: myocardia
Ooh, I didn't look at the amount of L2 cache. That means that it's really the same speed, but needs to run 100 Mhz faster to get the same amount of work done. That's not good.*

* There's a 75-100 Mhz performance difference between two A64's at the same speed, if one has 1MB of L2.

that is not right, amd didnt do these to compensate, they did it to be more competitive 1mb l2 cache for amd didnt do much and it costs a lot

 

[Brunnis]

Will we have to wait until tomorrow for overclocking results?

 

[Hyperlite]

Originally posted by: harpoon84
Boo! Show us the O/C details NOW!

 

[Metr0]

For results i need a bit more time

but this cpus are just a die shrink of normal 1m2 on 90nm, only oc potencial is important

 

[coldpower27]

I dunno if 1.3V is really that terrible, isn't Intel's 65nm products around that mark as well?

 

[xero273]

Originally posted by: coldpower27
I dunno if 1.3V is really that terrible, isn't Intel's 65nm products around that mark as well?

so the 1.3v is not a CPU-Z error?

 

[Duvie]

Originally posted by: coldpower27
I dunno if 1.3V is really that terrible, isn't Intel's 65nm products around that mark as well?

Yep....My C2D is 1.325v if I am not mistaken...


The question will be does AMD actually get any headroom out of the 65nm die shrink or will it still start to leak like crazy at the same points...

we have all seen die shrinks result in lower operating voltgae ut then only garner perhaps and additional 200mhz....

I was hoping AMDs 65nm will make 3ghz very common and if not a shot at 3.3-3.4ghz...It will need that to stay somewhat competitive in the enthusiast OCing market. If they made it close then some AMD users would not scrap their systems. Yeah 3.4ghz still wont beat a mild 3.2ghz OC of the C2D in a majority of things but it is closer now then the arse kicking they take now since most dont hit 3ghz and most average 2.8ghz. A stock X6700 will take that out...

 

[Loghic]

Very interesting. Waiting for OC results
Some days ago I´ve read detailed specs of the Rev. G CPUs on German website http://www.pc-treiber.net/thread.php?threadid=2021]PC-Treiber[/url]. The specs were released from ASRock. It looks like AMD wants to update the specs of the X2 3600+.

 

[Avalon]

I wouldn't mind giving one of my systems a budget 65nm X2, especially if I could overclock it to around 3.2Ghz

It'll tide me over until the E4300 anyway.

 

[coldpower27]

Originally posted by: Duvie

Originally posted by: coldpower27
I dunno if 1.3V is really that terrible, isn't Intel's 65nm products around that mark as well?

Yep....My C2D is 1.325v if I am not mistaken...


The question will be does AMD actually get any headroom out of the 65nm die shrink or will it still start to leak like crazy at the same points...

we have all seen die shrinks result in lower operating voltgae ut then only garner perhaps and additional 200mhz....

I was hoping AMDs 65nm will make 3ghz very common and if not a shot at 3.3-3.4ghz...It will need that to stay somewhat competitive in the enthusiast OCing market. If they made it close then some AMD users would not scrap their systems. Yeah 3.4ghz still wont beat a mild 3.2ghz OC of the C2D in a majority of things but it is closer now then the arse kicking they take now since most dont hit 3ghz and most average 2.8ghz. A stock X6700 will take that out...

Well going from the previous generation of the 90nm node, Brisbane is going to be more like Winchester in the fact that, it's mainly out there to test out the 65nm process for later parts, just like Cedar Mill was for Intel on their 65nm process, though they on the other hand produced some useful products at the beginning of the 65nm node already with Presler and Yonah.

Brisbane is likely to top out at 2.8GHZ stock, which will be lower then the 90nm FX's, which is perfectly fine as AMD should focus their efforts on utilizing the 65nm to eke out as much clockspeed as possible for their K8L/K10 derivatives rather then wasting efforts on the old K8 tech.

 

[DrMrLordX]

Still waiting for OC info *)

 

[harpoon84]

Originally posted by: DrMrLordX
Still waiting for OC info *)

Ditto.

 

[Brunnis]

Originally posted by: harpoon84

Originally posted by: DrMrLordX
Still waiting for OC info *)

Ditto.

Yeah, why haven't we got any info yet? He's obviously allowed to post now that it's officially released.

EDIT: BTW, I was hoping to be able to buy one here in Sweden today, but I realize now that I'm not likely to find one...

 

[myocardia]

Originally posted by: Metr0

Originally posted by: myocardia
Ooh, I didn't look at the amount of L2 cache. That means that it's really the same speed, but needs to run 100 Mhz faster to get the same amount of work done. That's not good.*

* There's a 75-100 Mhz performance difference between two A64's at the same speed, if one has 1MB of L2.

that is not right, amd didnt do these to compensate, they did it to be more competitive 1mb l2 cache for amd didnt do much and it costs a lot

That's completely right. Do some research, and you'll see. If additional cache was worth zero Mhz, please explain AMD giving the identical speed processors a higher speed rating. People with more than one processor at the same stock speed have verified this, also. AMD claimed 200 Mhz difference, between 512 KB and 1 MB L2 cache models, and independent verification claims 75-100 Mhz difference.

 

[Avalon]

No one has mentioned the half multipliers yet.

 

[wazzledoozle]

Bump WE NEED BENCHES!

 

[DrMrLordX]

^ what he said

 

[OcHungry]

Originally posted by: myocardia
Ooh, I didn't look at the amount of L2 cache. That means that it's really the same speed, but needs to run 100 Mhz faster to get the same amount of work done. That's not good.*

* There's a 75-100 Mhz performance difference between two A64's at the same speed, if one has 1MB of L2.

That's the dumbest statement I have read. I guess some people never owned AMD and make false claim about. It's not like C2D that relys on huge cache because the FSB bottlencks.

 

[Regs]

I don't care if they thing can do 5Ghz. If its slow getting the data from the RAM to the Core and through the AMD compiliars then all that GHz goes to waste.