Tuesday, August 24th 2010

AMD Details Bulldozer Processor Architecture

AMD is finally going to embrace a truly next generation x86 processor architecture that is built from ground up. AMD's current architecture, the K10(.5) "Stars" is an evolution of the more market-successful K8 architecture, but it didn't face the kind of market success as it was overshadowed by competing Intel architectures. AMD codenamed its latest design "Bulldozer", and it features an x86 core design that is radically different from anything we've seen from either processor giants. With this design, AMD thinks it can outdo both HyperThreading and Multi-Core approaches to parallelism, in one shot, as well as "bulldoze" through serial workloads with a broad 8 integer pipeline per core, (compared to 3 on K10, and 4 on Westmere). Two almost-individual blocks of integer processing units share a common floating point unit with two 128-bit FMACs.

AMD is also working on a multi-threading technology of its own to rival Intel's HyperThreading, that exploits Bulldozer's branched integer processing backed by shared floating point design, which AMD believes to be so efficient, that each SMT worker thread can be deemed a core in its own merit, and further be backed by competing threads per "core". AMD is working on another micro-architecture codenamed "Bobcat", which is a downscale implementation of Bulldozer, with which it will take on low-power and high performance per Watt segments that extend from all-in-One PCs all the way down to hand-held devices and 8-inch tablets. We will explore the Bulldozer architecture in some detail.

Bulldozer: The Turbo Diesel Engine
In many respects, the Bulldozer architecture is comparable to a diesel engine. Lower RPM (clock-speeds), high torque (instructions per second). When implemented, Bulldozer-based processors could outperform competing processor architectures at much lower clock speeds, due to one critical area AMD seems to have finally addressed: instructions per clock (IPC), unlike with the 65 nm "Barcelona" or 45 nm "Shanghai" architectures that upped IPC synthetically by using other means (such as backing the cores up with a level-3 cache, upping the uncore/northbridge clock speeds), the 32 nm Bulldozer actually features a broad integer unit with eight integer pipelines split into two portions, each portion having its own scheduler and L1 Data cache.



Parallelism: A Radical Approach?
Back when analysts were pinning high hopes on the Barcelona architecture, their hopes were fueled by early reports suggesting that AMD was using wide 128-bit wide floating point units, leading analysts to believe that AMD may have conquered its biggest nemesis - floating point performance, in turn its pure math crunching abilities. However, that wasn't exactly to be. That's because the processor's overall number crunching abilities were pegged to its floating point performance, ignoring the integer units.



AMD split 8 integers per core into two blocks, each block having four integer pipelines, an integer scheduler for those, and an L1 Data cache. These constitute the lowest level of "dedicated components", dedicated to processor threads. There is a shared floating point unit between the two, with two 128-bit FMACs, arbitrated by a floating point scheduler. The Fetch/Decode, an L2 cache, and the FPU constitute "shared" components.



AMD is implementing a simultaneous multithreading (SMT) technology, it can split each of the "dedicated" components (in this case, the integer unit) to deal with a thread of its own, while sharing certain components with the other integer unit, and effectively make each set of dedicated components a "core" in its own merit of efficiency. This way, the actual core of the Bulldozer die is deemed a "module", a superlative of two cores, and the Bulldozer die (chip) features n-number of modules depending on the model.
So now you have a chip with eight cores with much lower die sizes and transistor counts compared to a hypothetical 32 nm K10 8-core processor. It is unclear whether AMD wants to further push down SMT to the "core" level and run two threads simultaneously over dedicated components, but one thing for sure is that AMD has embraced SMT in some form or another. In all this, the chip-level parallelism is transparent to the operating system, it will only see a fixed number of logical processors, without any special software or driver requirement.

So in one go, AMD shot up its integer performance. Either a thread makes use of one integer unit with its four pipelines, or deals with both the integer units arbitrated by the fetch/decode, and the shared FPU.

Outside the modules
At the chip-level, there's a large L3 cache, a northbridge that integrates the PCI-Express root complex, and an integrated memory controller. Since the northbridge is completely on the chip, the processor does not need to deal with the rest of the system with a HyperTransport link. It connects to the chipset (which is now relegated to a southbridge, much like Intel's Ibex Peak), using A-Link Express, which like DMI, is essentially a PCI-Express link. It is important to note that all modules and extra-modular components are present on the same piece of silicon die. Because of this design change, Bulldozer processors will come in totally new packages that are not backwards compatible with older AMD sockets such as AM3 or AM2(+).
Expectations
Not surprisingly, AMD isn't talking about Bulldozer as the next big thing since dual-core processors (something it did with Barcelona). AMD currently does have an 8-core and 12-core processors codenamed "Magny-Cours", which are multichip modules of Shanghai (4-core) and Istanbul (6-core) dies. AMD expects an 8-core Bulldozer implementation (built with four modules), to have 50% higher performance-per-watt compared to Magny-Cours.



Market Segments
As mentioned in the graphic before, AMD's modular design allows it to create different products by simply controlling the number of modules on the die (by whichever method). With this, AMD will have processors ready with most PC and server market segments, all the way from desktop PCs, enthusiast-grade PCs, notebooks, to servers. AMD expects to have a full-fledged lineup in 2011. The first Bulldozer CPUs will be sold to the server market.


Hotchips 22 Presentation by AMD on the Bobcat Architecture
Below are as-is slides from AMD's Hotchips presentation on the Bobcat architecture.
Add your own comment

283 Comments on AMD Details Bulldozer Processor Architecture

#1
wolf
Performance Enthusiast
by: PaNiC
Really why did AMD bring out 890fx chipset, just for the shitty 1090t it was nothing the 790fx couldn't handle. you ripped off customers and you don't have enuf to do that . i hope your shitty chip sucks.
FUCK YOU AMD
buddy you just got to get over that.

on a lighter note, bring on the performance numbers! from all that info this chip sounds very promising, to many market segments.
Posted on Reply
#2
PaNiC
by: pantherx12
Pfffft you assumed to needed it no one else :laugh:
i only got this now. a bit slow
i assumed nothing, i was mislead by AMD shitty roadmaps
I hope a mod deals with your stupid ass soon, all your doin is annoying ppl and calling AMD shit when you should be calling yourself that you little troll
could you be anymore random scrub
Posted on Reply
#3
btarunr
Editor & Senior Moderator
by: mastrdrver
Just a heads up but Lost Circuits is reporting that Bulldozer will be backward compatible with AM3.

You might just want to wait until later today when AMD makes their speech at Hot Chips to see if they verify this.
Nope, we talked to AMD, they said it won't be backwards compatible. The way the chip is designed, it won't even theoretically work on platforms designed for K8/K10.
Posted on Reply
#4
pantherx12
Well no one else was, they made it very clear the x6 processors would have backwards compatibility.

There was a lot of news from board manufacturers as well saying basically saying " ours support x6 too LULZ!"

It just seems you may of just missed everything we read.
Posted on Reply
#5
Tatty_One
Super Moderator
by: PaNiC
X48 realsed for the die shink core 2 65nm to 45nm and got bunch of new chips
.
Weak argument, P35 and x38 chipset boards would still run 45nm chips with a simple BIOS update.
Posted on Reply
#6
btarunr
Editor & Senior Moderator
by: PaNiC
X48 realsed for the die shink core 2 65nm to 45nm and got bunch of new chips
That's laughably misinformed.
Posted on Reply
#7
PaNiC
by: pantherx12
Well no one else was, they made it very clear the x6 processors would have backwards compatibility.

There was a lot of news from board manufacturers as well saying basically saying " ours support x6 too LULZ!"

It just seems you may of just missed everything we read.
yeah most of the people taking up amd side are intel owners
Posted on Reply
#8
Kantastic
by: PaNiC
yeah most of the people taking up amd side are intel owners
And this helps your argument how?
Posted on Reply
#9
PaNiC
by: btarunr
That's laughably misinformed.
it came out to support 1600fsb thats about it but there was still the die shinrk after so i see that as just
Posted on Reply
#10
pantherx12
I think that you are talking about Magny-Course as one being compatible to AM2(+), as well did they, the mnfcts... And this is Bulldozer, right? From the slides you can see that they are presenting it on 8-core processor. Or am I missing a point...?
Yeah your missing the point a bit, that was aimed at panic, he said he was mislead by AMD regarding 800 chip-set and the x6 processors.

How ever I was saying there was lot of information to be accessed from various sources.
Posted on Reply
#11
naram-sin
by: pantherx12
Yeah your missing the point a bit, that was aimed at panic, he said he was mislead by AMD regarding 800 chip-set and the x6 processors.

How ever I was saying there was lot of information to be accessed from various sources.
Yeah, I saw later that I did... and deleted the stupid post. :D

But, I agree on the second one. Thanks for clearing that up.

And I was just thinking about replying on his... issues too. Because this is getting out of hand and going seriously off-topic.
Posted on Reply
#12
Tatty_One
Super Moderator
by: PaNiC
it came to support 1600fsb thats about it but there was still the die shinrk after so i see that as just
Your ignoring the facts, your very argument about the 890 also applies to Intel equally, I have no particular loyality to either side, your attack on the 890, whilst in part justified, coupled by your defence of the x48 makes no sense, the x48 gave us a new and improved chipset with DDr3 support etc etc but was not a neccesity to run 45nm chips just like as you mentioned the 890 is not a neccessity, the 890 also brings some added enhancements....... all I am saying is there are definate similarities that you choose to overlook for the sake of your argument.
Posted on Reply
#13
PaNiC
by: Tatty_One
Your ignoring the facts, your very argument about the 890 also applies to Intel equally, I have no particular loyality to either side, your attack on the 890, whilst in part justified, coupled by your defence of the x48 makes no sense, the x48 gave us a new and improved chipset with DDr3 support etc etc but was not a neccesity to run 45nm chips just like as you mentioned the 890 is not a neccessity, the 890 also brings some added enhancements....... all I am saying is there are definate similarities that you choose to overlook for the sake of your argument.
but for what? 2 chips? intel had more then 30 after x48
Posted on Reply
#14
Bloodcrazz
Originally Posted by PaNiC
Really why did AMD bring out 890fx chipset, just for the shitty 1090t it was nothing the 790fx couldn't handle. you ripped off customers and you don't have enuf to do that . i hope your shitty chip sucks.
FUCK YOU AMD
I do agree with him. I also have this problem. I also got a AM3 socket knowing that AMD themselves quoted that the Bulldozer was going to be for AM3 socket.
And knowing now that it will not be a AM3 socket has got me feeling like i got riped off.
Its just like they said it for marketing purporses to sell AM3 cpu's, to make every one think theres a upgrade at the end of the year.
Now there saying its not AM3.
Posted on Reply
#15
Tatty_One
Super Moderator
by: PaNiC
but for what? 2 chips? intel had more then 30 after x48
Your right there, but how many of them would run on an x48 and not an x38? I am not disagreeing with the argument, just the facts you have presented to support that argument, you are right, you just presented it wrongly if you get my meaning?
Posted on Reply
#16
pantherx12
by: Bloodcrazz
I do agree with him. I also have this problem. I also got a AM3 socket knowing that AMD themselves quoted that the Bulldozer was going to be for AM3 socket.
And knowing now that it will not be a AM3 socket has got me feeling like i got riped off.
Its just like they said it for marketing purporses to sell AM3 cpu's, to make every one think theres a upgrade at the end of the year.
Now there saying its not AM3.
Just how long did AMD maintain backwards compatibility ?


Time to move on, this is the tech world we're talking about things go quickly.

People will always be disappointed because something new comes out, it's just how things work with computers.

Can you link me to where AMD said it would be AM3 by the way?

I've been following bulldozer for a bit and certainly never read anything like that, although other sources did have rumours about the socket.

I don't think AMD made any announcements at all.
Posted on Reply
#17
naram-sin
by: pantherx12
...
Can you link me to where AMD said it would be AM3 by the way?

I've been following bulldozer for a bit and certainly never read anything like that, although other sources did have rumours about the socket.

I don't think AMD made any announcements at all.
Me to. I was under impression that the Bulldozer will be settled on LGA1207 or Socket F, even from first announcements... Definitely not AM3. And this i besides the point. Completely new architecture demands new components and interfaces, it's nature's law. Just hope this situation won't resemble of that when we had s754 and s939, that's all.

Meaning, I think it would be of greater use to focus all of resources on new platform, and completely abandon further development of old ones. No matter how much pennies is there to gain on old inventory. Maybe new lithography will help on this, what with lower costs and higher yield...
Posted on Reply
#18
PaNiC
by: pantherx12
Just how long did AMD maintain backwards compatibility ?


Time to move on, this is the tech world we're talking about things go quickly.

People will always be disappointed because something new comes out, it's just how things work with computers.

Can you link me to where AMD said it would be AM3 by the way?

I've been following bulldozer for a bit and certainly never read anything like that, although other sources did have rumours about the socket.

I don't think AMD made any announcements at all.
amd stated many times that it will be am3r2
Posted on Reply
#19
Bloodcrazz
by: pantherx12
Just how long did AMD maintain backwards compatibility ?


Time to move on, this is the tech world we're talking about things go quickly.

People will always be disappointed because something new comes out, it's just how things work with computers.

Can you link me to where AMD said it would be AM3 by the way?

I've been following bulldozer for a bit and certainly never read anything like that, although other sources did have rumours about the socket.

I don't think AMD made any announcements at all.
Hear is a link.
http://en.wikipedia.org/wiki/Bulldozer_(processor)
Posted on Reply
#20
PaNiC
by: Tatty_One
Your right there, but how many of them would run on an x48 and not an x38? I am not disagreeing with the argument, just the facts you have presented to support that argument, you are right, you just presented it wrongly if you get my meaning?
x48 also had a close to 2 year life span not 9 months
Posted on Reply
#21
wolf
Performance Enthusiast
by: PaNiC
amd stated many times that it will be am3r2
show me the links. with no links your just ranting.

by: Bloodcrazz
Hear is a link.
http://en.wikipedia.org/wiki/Bulldozer_(processor)
that link is moot, wikipedia isn't a statement from AMD, nor does the reference list show a link to AMD officially stating it will be on the AM3 socket.
Posted on Reply
#22
pantherx12
by: PaNiC
amd stated many times that it will be am3r2
I think you guys misunderstood me, can I have a link to a page were AMD or a representative of AMD stated that it will be a AM3 or a variation of AM3?
Posted on Reply
#23
PaNiC
by: wolf
show me the links. with no links your just ranting.



that link is moot, wikipedia isn't a statement from AMD, nor does the reference list show a link to AMD officially stating it will be on the AM3 socket.
theres has been a few press conference/computer shows where they have said this. just google it
http://www.anandtech.com/show/2871/2
Posted on Reply
#24
WhiteLotus
It's your argument you google it.


If this is a new socket, wonder if they will change cooling attachment design?


And also I am loving this "modular" design, that mean they can have numerous types of the processor, all at different prices. Say a low end one wouldn't have much but a top end would have everything. Or am I missing the point?
Posted on Reply
#25
naram-sin
by: PaNiC
amd stated many times that it will be am3r2
Most of things that can be found on the net are readers' guesses. At one point, AMD did mention am3r2, but hadn't claimed that there will be any backward compatibility. Or at least I can't find any.

But, we should keep in mind that this is not exactly what AMD initially intended. This is a result of going back to the drawing board and coming up with something (almost) completely new... or at least upgraded... and better (judging by these info)! And I say: good for them! And for most of us! If they came up with an idea that is great (superior, even, at least compared to previous one) but requires certain changes in other areas, then do it.

Otherwise, it would be like cramming a jet-engine inside a trailer...
Posted on Reply
Add your own comment