AMD K8L ES coming in December

Quote:

---

Originally Posted by atistatic

i have a question is really that it is K8L because is AMD processor 4 x 4 but i no talk that is was K8L

---

:confused: :confused: :confused:

Welcome to the forums anyway :toast:

Quote:

---

Originally Posted by atistatic

i have a question is really that it is K8L because is AMD processor 4 x 4 but i no talk that is was K8L

---

no. 4x4 is a socket setup. it'd be like all the comparisons of old, between AM2 and Conroe :lol2: :rofl:

high-res. new Opteron photo
http://i.n.com.com/i/ne/p/2006/amdbarcelona550x512.jpg
it looks like this will be the only x86 native quad-core for whole 2007
Yorkfield == 2xPenryn(45nm) unfortunately == kentsfield at high frequencies

will you look at those massively improved FP and integer units :slobber:

Quote:

---

Originally Posted by MAS

http://www.hkepc.com/bbs/attachments...1GfRYfDBpe.jpg

they say next-gen AMD cpus aka K8L will be 40% faster than todays K8, and thus 20% faster than C2D (Yorkfield is just Kentsfield shrink with SSE4)

---

Not true . Yorksfield will be 12mb of shared cache . not 6mb per core.

Also yorksfield is also reported to be native quad core. . You can get links in news section. Also You are forgetting the Intel is going to be using high K metal gates on the 45nm process. Which is 20%+ faster switching speed.And 5x less leakage. I will buy yorksfield over the 65nm old tech that AMD will offer. Since both new processors will arrive 3rd qt. 07 . Intel is out for blood. AMD should have never sued intel . As now the Dell is selling AMD cpu's they are unable to supply the maket with cpu's/ Intel wins the law suite/ Intel will lead in performance for at least the next 3 years.:toast:

This thread is about K8L, keep it to that. No need for another fanboyism thread here.

Quote:

---

Originally Posted by v0dka

This thread is about K8L, keep it to that. No need for another fanboyism thread here.

---

Don't worry,he'll do that again.He always does.
I second your proposal,let's keep this thread to what the thread title says.

Oh my! That K8L die photo looks even more powerful than the one posted back in May! C2Q doesn't stand a chance against that kind of processing power! :rolleyes:

Quote:

---

Originally Posted by ozzimark

no. 4x4 is a socket setup. it'd be like all the comparisons of old, between AM2 and Conroe :lol2: :rofl:

---

Then is K8L or no ?

http://theinquirer.net/default.aspx?article=35011

AMD LAID OUT a bit more about Barcelona today at MPF, focusing in on six areas. They talked about SSE128, enhanced IPC, efficient memory use, better caches, virtualization and power management.
In the macro sense of things, Intel is chasing AMD, but on the micro level, the opposite is true. The talk was given by Ben Sander of AMD, you can read it on his nametag if you don't believe me.


Barcelona is the first native quad-core AMD CPU, commonly but wrongly referred to as K8L. It is the first real new core from the chipmaker in quite a while, not merely a massaged version of what came before. You should be seeing it in Q2/07, although some whispers are now saying Q3.

Other than four cores, the most obvious difference is the new widened SSE instructions. On the pre-Barcelona parts, SSE was done in 64 bit chunks, so if you wanted to do a 128b operation, you needed two passes, possibly more. With the widening of SSE, it should immediately double throughput on SSE instructions. Obviously media operations will benefit, but HPC and FP heavy ops will get a solid kick in the pants too.

In addition to the obvious width change, several less noticeable changes were made to support it. Instruction fetch was upped from 16B/cycle to 32B and support for unaligned load ops was added and cache bandwidth was doubled to support this. Last but not least the FP scheduler was widened from 64 to 128b.

The enhanced IPC is more a case of little improvements adding up to a bigger bang rather than any single thing standing out. The SSE improvements add a bit here and there to start with, and a better branch predictor adds to this. It is bigger and better with a larger history, a dedicated 512 entry indirect predictor, and the return stack is doubled in size.

In the catching up with Intel side of things, added a sideband stack optimizer, out of order load execution and data dependent divide latency. They also upped the TLB to support 1G pages, 48-bit physical addressing, and improved the ITLB and DTLBs. There are also more fastpath instructions and a few more bit manipulation instructions and SSE extensions.

For RAM efficiency, one of the main things they did was make the two memory controllers on the chip act independently, up to Rev G they could only act in lockstep. This lets AMD hit two memory locations at once, potentially a big win for server type apps, but for single users, it's benefit is less clear.

On top of this, they changed the northbridge a lot increasing buffers and adding support for new DRAM types. The Barcelona controller will do FBD if necessary, but the chances of you seeing that are something less than zero. AMD also updated the way paging is done and modified the way write bursting happens.

Additionally, one of the big complaints I hear is prefetch, and that has been comprehensively addressed. They now track positive, negative and non-unit strides, and have a dedicated prefetch buffer. On top of that, Barcelona is much more aggressive in how they fill idle ram cycle and updated the core prefetchers on both the L1D and L1I side.

The newer L3 cache is called a 'victim cache', it sits on top of the existing discrete L2 caches, and is shared among the cores. The big thing here is that if the caches are empty, the request goes to L1 cache. If that fills, the data line is evicted to L2, and when L2 fills, it goes to L3. If a core reads from L3, there are one of two things that can happen. If it is data, the line is moved to L1D, but if it is instructions, the line may be moved to L1I, or it may be copied to L1I and left in L3. The L3 is not exclusive, and the point of this is that code is often shared among the cores while data is far less often. The cache lines have performance hints associated with them that will clue to cores in on the whole copy vs move debate.

That brings us to virtualization, a topic we have covered a lot in our Pacifica articles (1, 2, 3 and 4 )The main advance Barcelona brings is that it turns on Nested Page Table support and a few other things that did not make the cut on the Rev F parts. It is also said to reduce world switch time going to the hypervisor and back by 25%.

One of the last things they did was to break the CPU and northbridge into separate power planes. This will allow the CPUs to be clocked up and down, and volted up and down independently of the Northbridge. This was a major sticking point in the ramp of the previous iterations of the chip, and I expect big dividends here, and it also saves a lot of power. I am also told that they can change the voltage on cores independently, but that is more of a motherboard issue. Since it is not really supported anywhere on the current platforms, don't look for a BIOS option to turn it on, but if it came back in later platform revisions, I would not be overly surprised.

What you have for Barcelona is a CPU that looks a lot like the older revs at the block diagram level, but no part of it is untouched. Some pieces a massively updated, others far less so. In any case, it will add up to significant gains overall, but will it be enough to dethrone Woodcrest 2 core? Stay tuned in the middle of next year for the answer to that. µ

Quote:

---

One of the last things they did was to break the CPU and northbridge into separate power planes. This will allow the CPUs to be clocked up and down, and volted up and down independently of the Northbridge.

---

Kick ass...

Ryan

Thanks for the info SunFlowerSeeds. Hopefully K8L will prove itself at launch, all while not emptying the consumer wallet.

It isn't performance I am worried about. It is the cost of the chip to the consumer. I'd bet that this chip will cost at least $500 for the lower-end one at launch.

From the Inq article one thing puzzles me : in their description there is no extra FPU , just the original 2 were extended from 64 to 128bit and can do one cycle SSE ops.

There will always be a low-end version to match C2D. Should not be a problem for overclockers. ;)

Quote:

---

Originally Posted by Darkenreaper57

Thanks for the info SunFlowerSeeds. Hopefully K8L will prove itself at launch, all while not emptying the consumer wallet.

It isn't performance I am worried about. It is the cost of the chip to the consumer. I'd bet that this chip will cost at least $500 for the lower-end one at launch.

---

Thank you for this GREAT info SunFlowerSeeds...:D
My respect to you..:toast:

Quote:

---

Originally Posted by SunFlowerSeeds

I am also told that they can change the voltage on cores independently, but that is more of a motherboard issue. Since it is not really supported anywhere on the current platforms, don't look for a BIOS option to turn it on, but if it came back in later platform revisions, I would not be overly surprised.

---

My God! DFI get working on it and dont make us awit till whenever!

Quote:

---

Originally Posted by SunFlowerSeeds

http://theinquirer.net/default.aspx?article=35011

One of the last things they did was to break the CPU and northbridge into separate power planes. This will allow the CPUs to be clocked up and down, and volted up and down independently of the Northbridge. This was a major sticking point in the ramp of the previous iterations of the chip, and I expect big dividends here, and it also saves a lot of power. I am also told that they can change the voltage on cores independently, but that is more of a motherboard issue. Since it is not really supported anywhere on the current platforms, don't look for a BIOS option to turn it on, but if it came back in later platform revisions, I would not be overly surprised. µ

---

Amd is the innovator. :)

Its not so much innovation as they keep finding ways to make features that are normally very expensive/difficult to implement in commodity hardware in a cheap and reliable fashion.

I wouldn't expect them to have independant core volts any time soon, if ever. AFAIK it requires having a seperate VRM for each core on the motherboard which would increase costs and pin counts dramatically (IOW they'd probably need a new socket...AM3+ maybe?).

Quote:

---

Originally Posted by mesyn191

Its not so much innovation as they keep finding ways to make features that are normally very expensive/difficult to implement in commodity hardware in a cheap and reliable fashion.

I wouldn't expect them to have independant core volts any time soon, if ever. AFAIK it requires having a seperate VRM for each core on the motherboard which would increase costs and pin counts dramatically (IOW they'd probably need a new socket...AM3+ maybe?).

---

Correction AM2+ is AM2 only with HT3.0. AM2+ cpus are backwords compatible aka K8L. There is no other socket.

Fox of few words...

actually, word on the street is that K8L holds independent vcore for us. Do I have a link? No. I read it on the forums somewhere. Maybe someone else can produce a link.

Ryan

Quote:

---

Originally Posted by Turtle 1

Not true . Yorksfield will be 12mb of shared cache . not 6mb per core.

Also yorksfield is also reported to be native quad core. . You can get links in news section. Also You are forgetting the Intel is going to be using high K metal gates on the 45nm process. Which is 20%+ faster switching speed.And 5x less leakage. I will buy yorksfield over the 65nm old tech that AMD will offer. Since both new processors will arrive 3rd qt. 07 . Intel is out for blood. AMD should have never sued intel . As now the Dell is selling AMD cpu's they are unable to supply the maket with cpu's/ Intel wins the law suite/ Intel will lead in performance for at least the next 3 years.:toast:

---

Don't know much about AMD aye? Can you say, embedded Silicon Germanium, with THEE Dual Stress Liner, and Stress Memorization technology. (Cu) (CMOS) on (SOI) wafers, with the use of dielectric constant gate K insulators on 65nm with considerably less leakage make K8L's FAB process far more superior then your fandom is thickheaded by 40%! (15% faster in process improvments including the 2x FPU, ALu, dual 128-bit SSE123 and SSE4a later, 256-bit cache, 128-bit wide points, 20% clock for clock, and 5% when adding clock boosts.) Thad be the magic 40% faster # we have a winner... Keep on topic, this is about K8L not the CON. :stick: Your not the only one who knows there shEtz wolf boy!

Fox of few words...

Quote:

---

Originally Posted by IvanAndreevich

K8L looks cool, but I think it will have a serious clocking problem. K8 already has a clocking problem compared to Conroe and I can't see K8L clocking higher than K8. I think they might even have to drop the clock a little for the highest-end part to about 2.6 or 2.8 GHz max. And this biatch will run hot IMO.

So at stock it might win some from Kentsfield, but it won't overclock worth a :banana::banana::banana::banana:. That's right, I am betting Kentsfield will rape K8L while overclocked just because it clocks so well.

Not everyone needs a Quad Core. I, personally, don't really. What about dual core? If K8L will be strictly a Quad Core part, dual core market (majority of perofmance market anyway) will be owned by Conroe. I don't think anyone wants to argue how hard Conroe kicks K8's ass on any tech process.

---

http://photos1.blogger.com/blogger/7...743DvWIml8.jpg

Never knew 3.5ghz with a 5.2GT's 2.6HT clock was so slowwww... at stock. :slap: Nobody wants to argue but the foxmister Yo! :rolleyes: Your clames are useless without links little boy. Prove K8L will be woobed by the CONiE, wellp we can't until it comes out but based on clock facts releaced by mobo manufax... we will have some killer speeds for CON to deal with. Clock for clock its 20% faster too. Just think how this baby will Over CCCCC! The F out of the CON booya.

Quote:

---

Originally Posted by Grayfox84

Correction AM2+ is AM2 only with HT3.0. AM2+ cpus are backwords compatible aka K8L. There is no other socket.

Fox of few words...

---

Not for the forseeable future no, but AMD is supposed to have a AM3 socket of some sort around 2H 2008 (http://theinquirer.net/default.aspx?article=35211), there are of course no plans for a AM3+ but that was just my round about way of suggesting that a feature like variable volt. per core on one die are at best very far off for AMD.

Quote:

---

Originally Posted by FghtinIrshNvrDi

actually, word on the street is that K8L holds independent vcore for us. Do I have a link? No. I read it on the forums somewhere. Maybe someone else can produce a link.

Ryan

---

Sounds like BS seeing as how AMD has already produced slides contrary to this.

http://www.hardocp.com/image.html?im...dfMV8zX2wuZ2lm
http://www.hardocp.com/image.html?im...dfMV80X2wuZ2lm
http://www.hardocp.com/image.html?im...dfMV81X2wuZ2lm
http://www.hardocp.com/image.html?im...dfMV82X2wuZ2lm

They intend to lower power consumption by lowering the clockspeed of the core but the voltage remains the same.

More info. here:

http://www.theinquirer.net/default.aspx?article=33887

grayfox: getting 403'd on your link

Quote:

---

Originally Posted by Poodle

Amd is the innovator. :)

---

Too bad they didn't invent things like the x86 ISA for example.

Quote:

---

Originally Posted by Grayfox84

Don't know much about AMD aye? Can you say, embedded Silicon Germanium, with THEE Dual Stress Liner, and Stress Memorization technology. (Cu) (CMOS) on (SOI) wafers, with the use of dielectric constant gate K insulators on 65nm with considerably less leakage make K8L's FAB process far more superior then your fandom is thickheaded by 40%! (15% faster in process improvments including the 2x FPU, ALu, dual 128-bit SSE123 and SSE4a later, 256-bit cache, 128-bit wide points, 20% clock for clock, and 5% when adding clock boosts.) Thad be the magic 40% faster # we have a winner... Keep on topic, this is about K8L not the CON. :stick: Your not the only one who knows there shEtz wolf boy!

Fox of few words...

---

Too bad their process with all those expensive toys isn't producing transistors on par with Intels' bulk SS process.As for the rest ,all those bring it on par with Conroe.

Quote:

---

Originally Posted by brentpresley

Well, the original 40% faster (than K8, 20% faster than C2D) figure was speculative based upon our knowledge of Rev H's (AMD's official name for K8L - which was coined by Intel engineers, ironically) presumed improvements.

ONE of those presumed improvements was a 4th complex decoder. However, AMD's recent disclosure of the improvements in Rev H indicate that this most likely is false, and that Rev H will have the same 3 complex decoders that Rev F/G have (most likely tweaked). With this improvement now lacking, that 40% number looks to be on the high side, but we will not know for sure until ES chips are benched (probably Dec - Feb, depending on who is the first one to sneak one through NDA).

---

Quote:

---

Originally Posted by ***Deimos***

HYPOTHETICAL example: CPU-A 1 SSE unit. CPU-B 2 SSE units. CPU-B faster? What if CPU-A clockspeed is higher? What if CPU-A executes SSE instruction in 1 cycle, and CPU-B executes in 3. With equal clockspeeds, now CPU-A is faster, right? But, maybe CPU-B can pre-fetch more intelligently, and has wider decoders. Thus it wouldn't matter how fast CPU-A can theoretically execute the instructions when CPU-B can process them faster... and even if you factored in all dozens of factors, you're still just left with hypothetical synthetic numbers. In actual applications/games and drivers with optimizations, the performance could be totally different than expected.

Bottom line: dozens of factors. Speculation is just speculation. Need real test with real hardware running real games to actually measure which is faster. Nobody cares if their video card has 8, 16, or 1024 shaders... if it cant run 60 fps in Quake3.

---

Just wait and see ;)

Quote:

---

Originally Posted by savantu

Too bad they didn't invent things like the x86 ISA for example.

---

But they did improve on Alpha's designs to create the K7, K8 and HT bus. Not to mention 3DNow! and their only really noteworthy contribution, x86-64.

Quote:

---

Originally Posted by PallMall

There is no doubt about the fact that K8L will wipe the floor with conroe, get over it lol.
As said already the release chart from the models says enough 2.0Ghz upto 2.9ghz later upto 3.5ghz.
You ain't releasing a cpu almost a year later at those clocks knowing it would be slower .:rolleyes:

---

You must be either the world's leading authority on cpu die photo performance interpretation or just another clueless AMD fanboi to make those statements. :slapass:

sorry newbie...but im amd newbie

AM2+ is current AM2 socket? k8l 65nm with ht3.0 no problems in current boards so?

AMD does not even have a 65nm CPU the we have seen benched, nothing!

Quote:

---

Originally Posted by RangerXLT8

AMD does not even have a 65nm CPU the we have seen benched, nothing!

---

come back in a month ;)

brent...
http://www.xbitlabs.com/articles/cpu...amd-k8l_3.html

Quote:

---

Although the decoder of the K8L processor may not be able to decode 4-5 instructions per cycle, just the way Conroe can do it under favorable conditions, it will not hinder programs execution, because the commands are on average executed at less than 3 commands per cycle. K8 usually decodes one x86 instruction into fewer macro-operations than Conroe CPU would do. This, as well as the 32-byte fetch set, make its decoder highly efficient.

---

while having 4 decoders would be nice for those cases where more than 3 per clock are issued, it's not gonna help THAT much

Quote:

---

Originally Posted by brentpresley

Please do your homework and then come back. AMD has made no announcements that their 65nm node will include SiGe. Their 65nm process will be just like their 90nm, SOI wafers (with improved gates).

---

Yes they did.

http://en.wikipedia.org/wiki/Silicon_germanium
http://www.theinquirer.net/default.aspx?article=32322
http://www.amd.com/us-en/Corporate/V...103048,00.html

There are some roadmaps as well... somewhere on XS. I just got to sift through my posts. But yeah you can google it all you want on your free time, SiGe is in 65nm AMD's no denieing it. I think you should do your homework before you prech to somebody. The FX-64 and later will also include that on the new 90nm SOI-3 witch has SiGe in it also. They are trying to correct the FX-62's poor OCability with new fab processes. Unfortionatly the FX-60 was the best OCer people forgot about. It does more then a FX62 can ever dream of. Thats why a FX-60 is common to reach 3.5ghz and others 3.9ghz. The FX-62 most of them anyways are no better then my X2. Its rare to see a FX-62 to get over 3.2ghz. Thats about the limit for windsors but not for other cores. Most of the older cores perform much better. But at the cost of getting a opsolete platform.

http://badhardware.blogspot.com/2006...y-rainier.html

Here is the artical that talks about the 3.5ghz speeds K8L will later have somebody tryed to show before but a direct link doesn't work. This was released by mobo manufactuers and is required to reach a hyper transport BUS of 2.6ghz in the chipset.

(Update)
http://www.xtremesystems.org/forums/...96#post1720996
*Quoted Cooper*
Both FX-64 and FX-66 will be on 90nm tech but on SOI-3 which will add germanium into transistors to increase switch speed.

http://www.c627627.com/AMD/Athlon64/

Found what I needed, the road maps. Thank you very much. Do your homework indeed.

Quote:

---

Originally Posted by brentpresley

Hey SERGE84 - we're still waiting on those BENCHES!!!! ;)

---

I'm getting a new board tuesday to get rid of the bloody voltage limits that plauge my X2 with the Tforce so I'll have no more issues soon and I can pull the real limits out of my chip. I'll prob have to redownload vista.

Anyways check out thoes links above. Hope you find them intresting. :D And I'm sure you wouldn't want to challange coopers intellagence now would you about THAT fab process. :fact:

Makes me laugh how all the Intel fanboy's have crawled out of there holes with Conroe on release... and they are bricking K8L like mad.

Intel is massive up against AMD.. technically they should have wiped them out by now.. but the saddest thing is.. they had to copy them.

Funny as f*ck, really.

But what would be even funnier.. is of nVidia's new CPUs next year will spank AMD + Intel, Lmao.. now that would be a sight to see.

Quote:

---

nVidia's new CPUs next year will spank AMD + Intel

---

Need I say any more :rolleyes:

Quote:

---

Originally Posted by Gag3

Makes me laugh how all the Intel fanboy's have crawled out of there holes with Conroe on release... and they are bricking K8L like mad.

Intel is massive up against AMD.. technically they should have wiped them out by now.. but the saddest thing is.. they had to copy them.

Funny as f*ck, really.

But what would be even funnier.. is of nVidia's new CPUs next year will spank AMD + Intel, Lmao.. now that would be a sight to see.

---

not so much as copy but rather give up a dream (50Ghz CPUs) and go crying home to Pentium3. Which actually was a pretty good design

Oh yes, I forgot.. I have 38 posts..

Views dont count for sh*t unless your a pr0 @ XS it would seem, shame.

Quote:

---

Originally Posted by brentpresley

Well, funny thing is it is EASY to go back and read all 39 of your posts, and there seems to be a theme.

Rampant threadcrapping and fanboism.

Closed-mindedness will get you far, I'm sure. ;)

---

Coming from a man of such experience one can assume :rolleyes:

Quote:

---

Originally Posted by brentpresley

Well, funny thing is it is EASY to go back and read all 39 of your posts, and there seems to be a theme.

Rampant threadcrapping and fanboism.

Closed-mindedness will get you far, I'm sure. ;)

---

Closed-mindedness?

Rampant threadcrapping and fanboism.

Sure your not talking about yourself?

Becareful.. you might not be able to maintain a decent overclock if your head grows any bigger.

Amatuer.

Nearly every thread here is poorly moderated.. and you spend your time sniping at people.. you REALLY need to look around and pull the curtains once in a while.

Quote:

---

Originally Posted by brentpresley

WTF???

Why don't you go fondle your DDR3. :D

---

Next person with an offtopic post is getting the axe :D

Don't care if your post count is 38 or 3800. (as long as its not more then mine? heh).

Perkam

so "Barcelona" is going to be a K8 tweaked in as many ways as they can? i heard that no part went untouched...so this should be pretty good.

Quote:

---

Originally Posted by breakfromyou

so "Barcelona" is going to be a K8 tweaked in as many ways as they can? i heard that no part went untouched...so this should be pretty good.

---

It is more looking like a very large redesign of the K8.Since it taped out,AMD is confident it's pretty much finished product and we might see first server parts in Q2 '07.
And by the look of things,it's going to be killer platform(not just awesome CPU,but awesome possibilities for 8 core enthusiast systems,with support for both CF and SLI,HTX dedicated accelerators-ie. physics,DDR3 compatible IMC,coprocessors...).

We might as well say that if K8 is 3/5 core,then RevB is going to be almost 5/5 :D :D

Quote:

---

Originally Posted by brentpresley

Great. Took you what? week + to get the benches?

Just remember, I've got an EASY 400MHz left in my E6600. So if need be, I can pull out 3.6GHz ;)


PS - Got anything NEWER on SiGe? AMD has been almost too quiet on this front for a while now. Either something spectacular is coming, or they are having problems like when they moved to SOI.

---

Yeah sorry about that, but I needed the upgrade. I would have bought the DFI infinity Matx if it was out yet, but I had to go with a Tforce6100 at the time, now all that has changed. About time a matx board came out that can OC like a $250 DFI and half all the features like one. Ok my next limit is heat. I maybe able to post higher Ghz on CPUz but I wouldn't bench for long on something like 3.2ghz on air... but then again my heatsink will be reseated into this new mobo so the temps will be better. I beleave I just have crappy seating and airflow... ya know I'm doing this all in a small mATX case.

May just bench it out of the case and use abunch of fans. Wish I had water or bough those 160+ airflow rated fans. I'll see what I can do. But if I was on phase I just wonder how much I could get out of 2.15vcore... wow I don't think I'd ever be that extreme to use this board with no limit.

Well AMD is always too quite about stuff. Almost makes you wish they would show off like Intel but we really have no idea how K8L will perform only specualte about how its arc is similar to conroe in its own unique ways so we should atleast expect it to perform like one atleast. But I think the debating gets a little hot headed at times and a little overboard.

What company doesn't have problems at 1st? But once it comes out and matures it will be a hell of a lot better then the current AM2's sad the 939's are better OCers. SOI-3 is sipose to change that. They are sticking with the 90nm FX's because they can still clock higher then the 65nm parts at the moment. But the new FX's won't be the same, SOI-3 was only ment for 65nm but they wanted the FX's to have them because it makes faster transistors. So hopefully the new FX-64 and 70 series will turn AMD around.

FX-64 should be $600 by then since the 70 series is the FX replacement (Sold in pairs btw.) I seen price cuts down to $300 in the last 3 months in the FX's is amazing $700 for a FX62 now. So I would say AMD is doing pretty good. Specially in the opterons new mass priceing cuts. I think they are getting ready for the change is why prices are getting so low for one reason. They are wanting to get rid of all the past crap and make room for the new. Also they must have found a way to make fab manufactuering costs cheaper.

They gane proffit with massave volume sales and increaseing demand. Its never been better for AMD since sales are happening like mad.

Still waiting Serge84, lots of hot air, no action. In other words, you are full of crap and BS about all your magical non existing things. :slapass:

still waiting for what? you're waiting for an ES product to be released, illegally brought to the "tester's market" and then dispersed amongst people who have no clue as to how to utilize the processor?

geez...this thread is running on a :banana::banana::banana::banana:load of hot air already; no need to add more.

dave

wow thats hot I need an X2 first

God Why Must Conroe be Dragged EVERYWHERE

Quote:

---

Originally Posted by nn_step

God Why Must Conroe be Dragged EVERYWHERE

---

probably because it is the fastest cpu ever made, just like the x2 once was and you done a far bit of "dragging" with that, you should understand yourself though because of how strongly you defend the merits of ddr3 in every thread:)

Also why do some have a problem with people talking about Conroe? It seems to follow a pattern where someone would ask about an upgrade and some one would correctly point out that Conroe is the fastest atm then would be flamed by people for mentioning it.

Hopefully these are not just Intel fanboys though and will be just as happy to advise buyers what is best if k8l does a bit of ass whooping

And sarge, if you really think the avg oc for a fx60 is 3.5-3.9 its time to pass the pipe brother:eek:

i'm sorry, but the conroe is not the "fastest cpu ever made." neither was the x2. you're a little outta your league on that one considering the whole of the IT world isn't centered on consumer parts. I'll let your little 'ol conroe fight against a few Power5s in a server environment and then we'll talk. :)

dave

Quote:

---

Originally Posted by brentpresley

No, Shintai was talking about another thread where Serge84 was saying his K8 could beat my C2D in 64-bit mode.

I ran the benches in a matter of hours. It's been over a week and we are still waiting on Serge84 to post his. ;)

EDIT - thread here: http://www.xtremesystems.org/forums/...d.php?t=119370

---

lol

i've got some kick ass 2218s that run great but, they're geared for a different market than C2D.

dave

Quote:

---

Originally Posted by PallMall

Blue mountain calculates superpi 32M in 12secs flat..
How much time was needed for conroe to do that at stock?;)

---

who cares? superpi is worthless outside of the enthusiast world.

dave

Quote:

---

Originally Posted by gallag

probably because it is the fastest cpu ever made, just like the x2 once was and you done a far bit of "dragging" with that, you should understand yourself though because of how strongly you defend the merits of ddr3 in every thread:)

Also why do some have a problem with people talking about Conroe? It seems to follow a pattern where someone would ask about an upgrade and some one would correctly point out that Conroe is the fastest atm then would be flamed by people for mentioning it.

Hopefully these are not just Intel fanboys though and will be just as happy to advise buyers what is best if k8l does a bit of ass whooping

And sarge, if you really think the avg oc for a fx60 is 3.5-3.9 its time to pass the pipe brother:eek:

---

Well that's the thing. Guys with Amd didn't go over as much to the Intel threads shouting: Why buy freaking Netburst when you can get the brilliant K8!!!? Even then we got from fanboys: Hey the Pentium M is great, you just have to spend a fortune on the board and the memory and it's not 64bit... But it kicks ass! :rolleyes:

good for you. :)

as for me, i'm more concerned with how well they function for virtualization, workstation apps, rendering, etc.

dave

Quote:

---

Originally Posted by dave_graham

good for you. :)

as for me, i'm more concerned with how well they function for virtualization, workstation apps, rendering, etc.

dave

---

Well in everything that's BW hungry like in a high end server you will see great performance. Virtualization will probably be great also.

since we're all editing our posts for clarity....

Quote:

---

Originally Posted by Poodle

Well in everything that's BW hungry like in a high end server you will see great performance. Virtualization will probably be great also.

---

yup...to be expected....I've gotta run over to Boxborough and see their eval system.

dave

Quote:

---

Originally Posted by dave_graham

since we're all editing our posts for clarity....



yup...to be expected....I've gotta run over to Boxborough and see their eval system.

dave

---

Hey that's all that I dare to say. Who knows how it will be in games etc? Might be better than C2 or might not. No clue. I love the powers saving feats though and that it's decoupled from the NB. Good stuff. :)

Quote:

---

Originally Posted by dave_graham

i'm sorry, but the conroe is not the "fastest cpu ever made." neither was the x2. you're a little outta your league on that one considering the whole of the IT world isn't centered on consumer parts. I'll let your little 'ol conroe fight against a few Power5s in a server environment and then we'll talk. :)

dave

---

Don’t be so god dam ridicules , This place is turning from xtreamsystems to xtream business annalist & stable information and technology industries systems, for example detective Dave would it be unimaginable for you to work out that since we are talking about x86 desktop cpus that my statement would run close parallels with the afore mentioned subject?

No Conroe isn’t the fastest , blue mountain super computer will beat it , well go and borrow £73m and buy one then.

Also you state, as bold as it is stupid

“I'll let your little 'ol conroe fight against a few Power5s in a server environment and then we'll talk. :)”

I concede that in a few server tasks the power 5 will kick ass (amazingly that is what it was designed for) but in everything else Conroe and x2 will be faster, No?

Also try to remember that this forum is manly targeted at people that are interested in pushing there home computer till it fu(king begs to be turned off, not somewhere for IT sector workers to come along and say stupid :banana::banana::banana::banana: like “It dosent matter how fast Conroe or x2s are, A few power 5s would be faster in a server environment”

Quote:

---

Originally Posted by Poodle

Well that's the thing. Guys with Amd didn't go over as much to the Intel threads shouting: Why buy freaking Netburst when you can get the brilliant K8!!!? Even then we got from fanboys: Hey the Pentium M is great, you just have to spend a fortune on the board and the memory and it's not 64bit... But it kicks ass! :rolleyes:

---

well the thing is they did, If you want ill go pick out a few threads in the intel section where people recommended (and rightly so) x2s over pds, but I must say I am surprised you never seen this happen.

Quote:

---

Originally Posted by gallag

Don’t be so god dam ridicules , This place is turning from xtreamsystems to xtream business annalist & stable information and technology industries systems, for example detective Dave would it be unimaginable for you to work out that since we are talking about x86 desktop cpus that my statement would run close parallels with the afore mentioned subject?

No Conroe isn’t the fastest , blue mountain super computer will beat it , well go and borrow £73m and buy one then.

Also you state, as bold as it is stupid

“I'll let your little 'ol conroe fight against a few Power5s in a server environment and then we'll talk. :)”

I concede that in a few server tasks the power 5 will kick ass (amazingly that is what it was designed for) but in everything else Conroe and x2 will be faster, No?

Also try to remember that this forum is manly targeted at people that are interested in pushing there home computer till it fu(king begs to be turned off, not somewhere for IT sector workers to come along and say stupid :banana::banana::banana::banana: like “It dosent matter how fast Conroe or x2s are, A few power 5s would be faster in a server environment”

---

but when you throw a specious quote out there without much thought for those who read it, you've got to expect that at least someone with experience outside the enthusiast community is going to respond. :rolleyes: for people who actually work with this stuff for a living, it is unreasonable to assume that ANY C2D cpu is going to be the fastest.

*shrug* it's your ass and opinion, but, try to at least qualify what you're saying next time.

dave

Quote:

---

Originally Posted by gallag

well the thing is they did, If you want ill go pick out a few threads in the intel section where people recommended (and rightly so) x2s over pds, but I must say I am surprised you never seen this happen.

---

´

You might be right. :confused: Fanboism sucks and it feels like it's getting more than ever. And you are also right that one should always mention both alternatives but with respekt; as there might be some features which the user seaks that the inferior core might have but the better core doesn't have (sorry weird sentance but you know what I mean :)). K8s still undervolt lower and has higher bw than C2 for example. Something similar might be the case but vice versa if Amd one day gets the perfomance crown back. Everyone should stay objective I mean.

As for myself a go for the fastest core with lowest tdp which means that Intel is the most sane chose right now :doh:. I used to like the engineering filosofi of Amd more though, but Intel is going in the same direction now. So they are all good to me:)


4x4 is exiting I think as it might give a some fresh blood to our community. Quad Unreg ram channels and two cpu sockets is great!

I doubt it but Intel might do something like it. Who knows...

Quote:

---

Originally Posted by Poodle

4x4 is exiting I think as it might give a some fresh blood to our community. Quad Unreg ram channels and two cpu sockets is great!

I doubt it but Intel might do something like it. Who knows...

---

4x4 aint something to wait for and think will wtfpwn. Get a Opteron workstation board today, maybe add 5% without ECC and you got the same.

4x4 is just a hotfix against Intel quadcore. So no reason to copy it, and to copy it. Just buy a workstation board...:fact:

Quote:

---

Originally Posted by gallag

Don’t be so god dam ridicules , This place is turning from xtreamsystems to xtream business annalist & stable information and technology industries systems, for example detective Dave would it be unimaginable for you to work out that since we are talking about x86 desktop cpus that my statement would run close parallels with the afore mentioned subject?

No Conroe isn’t the fastest , blue mountain super computer will beat it , well go and borrow £73m and buy one then.

Also you state, as bold as it is stupid

“I'll let your little 'ol conroe fight against a few Power5s in a server environment and then we'll talk. :)”

I concede that in a few server tasks the power 5 will kick ass (amazingly that is what it was designed for) but in everything else Conroe and x2 will be faster, No?

Also try to remember that this forum is manly targeted at people that are interested in pushing there home computer till it fu(king begs to be turned off, not somewhere for IT sector workers to come along and say stupid :banana::banana::banana::banana: like “It dosent matter how fast Conroe or x2s are, A few power 5s would be faster in a server environment”

---

Lets be Straight out, 2P and 4P AMD WINS MORE Business Benchmarks than Conroe, Period. Why?, Because that is what AMD has been designing K8 and K8L for. Users on the Customers on the Desktop level only know AMD is Better some of the time because like all Hardware, in effort to keep power usage and transistor count low, there have to be design trade offs.
AMD gets more Benefit by trading transistors that would probably help the desktop market, and using those transistors to upgrade or add High end features. Such as Virtualization or reducing Memory latency or Increasing Bandwidth or improving Cache Coherency.
Before you go off and complain about the above statement, consider this 8220SE costs about $2,649 and you can get a 5200+ for under $400
So who do you think AMD is going to work harder to please, the Server world, or you with your $399 Dell/HP?

Brent !!

Why are you hijacking every AMD thread !?!?!?!

Perkam

Quote:

---

Originally Posted by LOE

oh I guess an intel quad core chip will have no problems beating 8 core system with 8 memory channels

IMO Intel will have to copy 4x4 if they want to stay competitive in this ultra mega high end sector, that isn't currently excisting, but will when K8L comes out

Thou it will be difficult for intel to create such product at such price point, they lack HT and IMCs and they will need to include a lot of chips if tey want to provide 8 cores with enough bandwidth to the memory and btwn all the cores

---

Lets relax abit on the 4x4 shall we. Its not a 8 core, 8 channel system is it? Its a 4 core 4 channel system. Just like a dual woodcrest workstation today is a 4 core, 4 channel.

The ultra highend sector you say will come just got s different name today. Workstation. The only new thing 4x4 offers is SMP without ECC memory. :fact:

Lack of IMC/HT dont seem to hold woodcrest/C2D back. And whenever K8L comes, then CSI might be in the game.

Anyway, lets hold 4x4 to what it is, OPteron Workstation board using non ECC memory. :slap:

Quote:

---

Originally Posted by brentpresley

DUDE! I'm NOT!

nn BROUGHT UP how Opteron was faster than C2D in servers. I wanted proof, went digging, and found the above comparison.

My reply was just that, A REPLY. For the RECORD, HE BROUGHT it up in his post.

GOOD GRAVY the AMDROIDS are so paranoid.

---

umm, you missed the main point of his post. nn said that amd cares more about the server market than the desktop market, in which he is correct. and considering that dual 280's (2.6ghz) can keep up with dual woodcrest at 3ghz.. that's pretty impressive in my book, considering how much conroe owns K8 at the same clock.
http://techreport.com/etc/2006q2/woodcrest/index.x?pg=1

well, NN never specified at what clock, or price range. i don't even know what the price is on woodcrest, so it's entirely possible that a similarly priced woodcrest is beaten by an opteron system. :shrug:

and once K8L hits.. if amd manages to get a 20-35% gain, plus extra clocks with lower power from 65nm.... :hehe:

Quote:

---

Originally Posted by brentpresley

People here criticize anyone that mentions C2D is faster/better than K8 as being a fanboi and trolling, but they do the EXACT same thing with K8L.

---

it's not what you say, but how you say it :fact:

and stop making generalizations for everybody. :nono:

Quote:

---

Originally Posted by LOE

"when K8L comes out", I am not very good at math, but dual quad core chips make a 8 core systems, and when each chip has a independent dual MC, each 128 bit, that makes also 4x128bit memory, or 8x64bit

---

K8L will have two IMC's, but only one can be used at a time. (one ddr2, one ddr3)
so for each physical chip, there will be a dual channel memory bank

Quote:

---

Originally Posted by perkam

Brent !!

Why are you hijacking every AMD thread !?!?!?!

Perkam

---

I thought this is AMD discussion forum, not AMD circle jerking forum where everybody sing and praise 800%+ performance increase and 200% power reduction...

Besides, anybody got some real figures instead of some dead die shot?

Quote:

---

Originally Posted by vitaminc

Besides, anybody got some real figures instead of some dead die shot?

---

obviously for a chip that isn't to be released for another 6+ months, no. hang tight in the meantime, and put up with our speculation ;)

thanks brent, i usually try not to let my emotions get the better of me, even though i'm probably one of the most pro-amd guys around :lol2:

however, sometimes i just read what people say the wrong way... (yourself included..sorry.) but you're right. we all need to learn to concede a point to the opposing side a bit better, myself especially :fact:

but that aside.. it's time to get back on topic :woot:

Quote:

---

Originally Posted by ozzimark

thanks brent, i usually try not to let my emotions get the better of me, even though i'm probably one of the most pro-amd guys around :lol2:

however, sometimes i just read what people say the wrong way... (yourself included..sorry.) but you're right. we all need to learn to concede a point to the opposing side a bit better, myself especially :fact:

but that aside.. it's time to get back on topic :woot:

---

Get back on topic ? AMD section hasn't been "ON TOPIC" since conroe launched lol.

I only wished people would share results and answer questions instead of speculate...on topic was a lost cause a LONG time ago :lol:

Perkam

Quote:

---

Originally Posted by ozzimark

obviously for a chip that isn't to be released for another 6+ months, no. hang tight in the meantime, and put up with our speculation ;)

---

how about for the 65nm K8 then? It's alerting quite for a chip that is suppose to be out in December.

perkam... sadly true. i know i'll be sharing quite a bit once i get my hands on revG... for now, AM2 is just not worth talking about, since performance is so similar to 939, and most serious benchers have switched to C2D..

for not speculating.. we need someone to get ahold of an ES in order to do that. where oh where is our good buddy s7e9h3n when we need him? :para: :shakes:

From Monza (Italy) in the World Cyber Games 2006 Amd present a preview of the 4x4 Quadfather, the bases of the k8l

http://www.hwupgrade.it/articoli/cpu/1585/titolo.jpg

http://www.hwupgrade.it/articoli/cpu...istema_1_s.jpg

for more info read here

Quote:

---

Originally Posted by perkam

Get back on topic ? AMD section hasn't been "ON TOPIC" since conroe launched lol.

Perkam

---

I couldnt agree more..... :lol2:

Quote:

---

Originally Posted by Esticbo

http://www.hwupgrade.it/articoli/cpu...istema_1_s.jpg

---

maybe Pentium EE? HThreading is a miracle -)

Looking forward to it.

Quote:

---

Originally Posted by MAS

maybe Pentium EE? HThreading is a miracle -)

---

Or a Sin against God, all depends on how you look at it

will this be beter then kentfeild

Quote:

---

Originally Posted by PallMall

G80 FX62|@ 2.8Ghz = 10000points
G80 X6800XE @ 2.93Ghz =10500points

X2 dual core 512KB and 06 +100mhz = 100points extra
X2 dual core 1MB and 06 +100mhz = 143points extra

Both at same speeds, comparing clock per clock in 06.

143/100*193 = 275.99

FX62 @ 2.93Ghz = 10300points in 06 at XE speeds,
don't know but for a cpu that costs 300$ less then the Xe, it's performance it's still quiet amazing to me :)

---

Yeah,it seems that all that talk that FX62 is not the gamers CPU is down the sewer :D .Indeed it can handle the power of single G80.Let's see some SLI benches.I guess it will hold it's own even there.It basicaly matched the C2D XE at the same clock.
What happened with all the doomsday predictions for FX and how only Conroe will be able to make G80 scream ? :p:

Edit: PallMall your math is a bit wrong :D,since FX would score (with perfect scaling) 2.93/2.8 * 10000 = 10646 :). So almost the same.
PS We know that thanks to IMC A64 scales almost perfectly with clock increases.

Aren't those scores held back by the GPU though?:)

Quote:

---

Originally Posted by red

Aren't those scores held back by the GPU though?:)

---

Yeah G80 is surely the bottleneck :p:

On a serius note,we should wait for Sli numbers.

Quote:

---

Originally Posted by informal

Yeah,it seems that all that talk that FX62 is not the gamers CPU is down the sewer :D .Indeed it can handle the power of single G80.Let's see some SLI benches.I guess it will hold it's own even there.It basicaly matched the C2D XE at the same clock.
What happened with all the doomsday predictions for FX and how only Conroe will be able to make G80 scream ? :p:

Edit: PallMall your math is a bit wrong :D,since FX would score (with perfect scaling) 2.93/2.8 * 10000 = 10646 :). So almost the same.
PS We know that thanks to IMC A64 scales almost perfectly with clock increases.

---

Looking on VictorWangs benches, you expect the 3DMark06 score to get a 600 point CPU score for 133mhz :clap:

VictorWang got 11314 with a 3.7Ghz C2D. So if the C2D gains 800points on 800Mhz. How likely is it that a FX62 would gain 600points on 133Mhz? IF we asume the 10000 and 10500 numbers are correct.

Different arch. you can't make direct comparisons based on clockspeed in any bench like that, particularly with one so crappy as 3DMark06 which is more of a glorified GPU bench then a system/CPU benchmark.

Quote:

---

Originally Posted by mesyn191

Different arch. you can't make direct comparisons based on clockspeed in any bench like that, particularly with one so crappy as 3DMark06 which is more of a glorified GPU bench then a system/CPU benchmark.

---

of course. thus why it's 3Dmark, not SYSmark :toast:

Quote:

---

Originally Posted by Shintai

Looking on VictorWangs benches

---

does anybody have 4x4 2xopteron 2XX @3GHz system?

Quote:

---

Originally Posted by MAS

does anybody have 4x4 2xopteron 2XX @3GHz system?

---

nope just some people have 2P or 4P systems

Quote:

---

Originally Posted by brentpresley

Please do your homework and then come back. AMD has made no announcements that their 65nm node will include SiGe. Their 65nm process will be just like their 90nm, SOI wafers (with improved gates).

---

Wrong.

http://www.amdboard.com/65nm_120605.html

see, i still have this lingering feeling that the 65nm cpu's released this quarter aren't just "rev.g" as we think of it

i know, i know, lots of info pointing at q3'07 for k8l....but...there's just such little info out on just what exactly is coming out..

http://www.theinquirer.net/default.aspx?article=35382

Quote:

---

The problem is the first rev of parts, the A0 silicon, is a little on the buggy side, but it does work. A1 fixes a lot of those bugs, and is due to be in AMD's hands literally any day now.

Once A1 is up and running, it should be more than demoable, but give it 2-3 weeks before we know for sure. The first major event after that is the analysts meeting, so I am guessing that is where it will break cover for real with a possible showing to the privileged few before that.

---

Go A1:p

Quote:

---

Originally Posted by PallMall

G80 FX62|@ 2.8Ghz = 10000points
G80 X6800XE @ 2.93Ghz =10500points

X2 dual core 512KB and 06 +100mhz = 100points extra
X2 dual core 1MB and 06 +100mhz = 143points extra

Both at same speeds, comparing clock per clock in 06.

143/100*193 = 275.99

FX62 @ 2.93Ghz = 10300points in 06 at XE speeds,
don't know but for a cpu that costs 300$ less then the Xe, it's performance it's still quiet amazing to me :)

---

You seem to be missing a very important point. If you substitute a $225 E6400 for the XE6800 in those tests, the E6400 would be as close to the FX62 as the FX62 is to the XE6800. The E6600 and E6700 would both outscore the FX62. GPU limited tests are not good measures of CPU performance.

Quote:

---

Originally Posted by ozzimark

of course. thus why it's 3Dmark, not SYSmark :toast:

---

If you agree with me that it does nothing to support what he was trying to say then what was your point?

Sysmark is pretty much useless as well BTW, the only synthetic general purpose benchmark that is worth a damn would be SPEC2K INT/FP and even it is highly inaccurate, only useful for making ball park comparisons.

Quote:

---

Originally Posted by brentpresley

Great. Took you what? week + to get the benches?

Just remember, I've got an EASY 400MHz left in my E6600. So if need be, I can pull out 3.6GHz ;)


PS - Got anything NEWER on SiGe? AMD has been almost too quiet on this front for a while now. Either something spectacular is coming, or they are having problems like when they moved to SOI.

---

AMD is using 4 stressor technologies at the initial announcement, they published results of their 65 nm at the fall 2005 IEDM in December. This was the publication that generated the '40% improvement' headline.

Here is a link to the key data:
http://www.realworldtech.com/page.cf...3005001504&p=7

The actual paper was published in the IEEE proceedings of IEDM 2005, which can be found at about any major library (academic/engineering library).

When examining the parametric transistor data, it is not quite that impressive and aligns with the current leaked roadmap of max 2.6 GHz 65 nm parts at first launch.

In their paper AMD states DSL, embedded SiGe, stress memorization, and sSOI as the 4 stressors. Interestingly, after IEDM sSOI disappeared from any bulleted list for any public presentation to date, implying they are holding that more secret or dropping it from the implementations.

Going back to the IEDM 2005 publication, AMD presented data on their 2nd major embedded SiGe revision, the first revision did not show a huge improvement. In the send revision they are getting a decent improvement in PMOS. Interesting, and not touted by AMD, was a blurb that they had data on 3rd generation embedded SOI which improved PMOS about another 10%.

If you compare the data at 65 nm at the IEDM 2005 to the Intel 65 nm Process, AMD is about 10-12% slower than Intel's process, however, the 3rd generation puts PMOS at about the same performance to about 5% higher if they implement that SiGe. I personally believe that will come later on as this is typically how AMD pushes process technology -- get it working and tweak it up over time.

For NMOS, AMD has not published anything encouraging and, going by the 65 nm data currently available, lags Intel by about 10%. Considering the rumor mill is full of leaked roadmaps with the max 65 nm release speed of 2.6 GHz, it is likely AMD has spend most of the time just trying to get yield and has not improved much from this initial disclosure.

While I cannot link the PDF file for the IEDM paper (copyright violation), I can certainly give you the citation such that you can go look it up yourself. I would need to dig it up, it is xeroxed somewhere and buried in my pile of papers.

JJ

K8L cores were renamed
Athlon 64 FX -> Altair FX -> Agena FX;
Athlon 64 X4 -> Altair -> Agena;
Athlon 64 X2 -> Antares -> Kuma

no stars in roadmaps ))))