To many these are just new chippies but in reality they are a new architecture and a major fundamental change in X86 CPU design, all for the better. I can't wait to start flogging a Zambezi system.
This is really their first ground-up architecture since 1999, with K7. Everyone says K8, but that was still based on K7 despite how huge it was. So 12 years since something like Bulldozer. AMD is really betting the barn on this, I hope its competitive!
It's embarrassing that AMD come out with so few big architecture changes. They need to do a way better job at competing and i doubt Bulldozer will even match Sandy Bridge clock for clock. Theres so little competition in the high-end and it's just plain annoying now as it's hurting consumers.
Well Intel is using an architecture that they've built upon from the Pentium M in 2003, so AMD's count isn't "embarrassing". I know what you mean about performance though.
Not really... Core 2 went to 4-issue and reworked the functional units, branch prediction, pipelines, etc. It was certainly a new architecture, even if it "learned" from the successes of Pentium M. The Core i-series was also a largely new architecture, bringing back Hyper-Threading and changing the caching structure, along with adding some new instructions, reworking pipelines, etc. To pretend that those are using the same base architecture as Pentium M is a joke. So for Intel we've gone from the P6 (Pentium Pro through Pentium III) to Netburst to Core to Core-i as the major architectures in the past ~15 years.
For AMD in that same span, we had K6, K7, K8 (sort of -- it's basically K7 with an integrated memory controller, which really isn't a new architecture), and then K10 (which is still essentially the same architecture as earlier, only refined). I'd liken K7 to K8 to K10 as being quite similar in principal to the Conroe to Allendale to Wolfdale plus quad-core Kentsfield and Yorkfield. There are changes with every one of those transitions, and even with K10 you have K10, K10.5, and different processes.
If I look at the architectures from a high level, AMD started with a good design on K7 and has continued to use it effectively. Intel on the other hand has had some radical new architectures, but some worked better than others in practice. Even Netburst was effective for what it did -- Pentium 4 chips outperformed Athlon XP chips in most tasks, but they cost more and used more power. It was only with K8 that AMD could clearly beat Netburst, and they did so effectively until Conroe entered the picture.
Embarrassing? Hardly. It just goes to show that they've designed their chips very well and have been able to focus on business management, process, horizontal/vertical integration, and all without having to run back to the drawing board too much. Intel was borrowing things from AMD, remember all the lawsuits or were you absent the last decade? I'm not sure about what "little competition" you are speaking of, CPU prices are at an all time low and even a Sempron can play most older games. You obviously have no idea what you are talking about. You can get a quad core Phenom II or one of the new APUs that will do everything you are likely to ask except make coffee, a solid state disk now at $1 a GB (compare that with original HDDs that didn't even have a GB and cost about a grand), super cheap DDR3 memory, Windows 7 that isn't $200 like XP was (the best yet), a solid ATX board for $130, and very powerful discreet low-power graphics for $100-200 and you might want to rethink "there's no competition" again. Obviously, there is. Maybe you should take all your brains and start up a tech company to compete with Intel, which happens to be about 10X the size of AMD. :)
I'm sure he meant no competition in the high end space(Anandtech readers generally forget the entry and midrange markets). AMD has nice bang-for-the-buck chips, but they didn't have a high end part to compete with socket 1366.
"i doubt Bulldozer will even match Sandy Bridge clock for clock."
You know, anyone can say I doubt this or I doubt that and it not be based on anything. If you doubt it, you need to have a concrete reason. It's best to not to act like you are making an informed opinion when there is not any thing to back it up. I'm just tired of people making pessimistic comments based on nothing.
Can't say I'm too surprised to see the server part first or a desktop part delay. I did hear of another stepping (C0?) but I don't think anyone cares except for one thing - get it out already!
And, to preempt the trolls (you know who you are), if Bulldozer is competitive then it should mean cheaper Intel CPUs. Win-win.
Agree. I am pretty close to saying f*ck it and buying a Sandy Bridge. AMD just keeps delaying and delaying. Seems like the next announcement out of AMD will be a Q1 2012 release, then a Q2, then... By the time Bulldozer is out Ivy Bridge will be either out or just around the corner.
Some benchies might alleviate the feeling that I should have just bought an SB months ago.
+1 on that. So many people overlook that feature. It's one of the reasons the cores are built like this. To be able to combine 2 x 128 in to 1 x 256 and be AVX capable.
Intel had no problem showing off Conroe, Penryn, Nehelem, Sandy Bridge... If one wanted to generate buzz and excitement about a new product, why hide its performance??
It's confusing watching John Fruehe talk about Interlagos in comparison with "8 core Sandy Bridge" but he's got a point; Sandy Bridge's FPUs can only do 1x128 OR 1x256 in the same cycle, which is a sizeable disadvantage with today's workloads. Bravo to Intel for being very forward thinking with AVX but for the first time in a while, they stand to be behind in instruction set support at the very least. As for performance, that's not something most of us know anything about yet, though Sandy Bridge will definitely be ahead in some areas, you can count on that.
Indeed. The fundamental difference in architecture will make it difficult to run proper benchmarks and to do fair comparisons. Synthetic results could come up completely skewed.
So AMD had a small delay in getting Interlagos out, but Intel just slipped SB-E Server chips (Xeon 5600 replacement) from a Nov launch to more like Feb/March due to some errata they didn't feel comfortable releasing as well as issues with PCIe 3.0. On the flip side, Intel still seemed to think they could get SB-E desktops (X79 compatible chips) out before the holidays but only time will tell. Either way it looks like AMD might have a few month advantage on both fronts for a little while.
Some AMD sources also said that Terramar will not be the next gen chip after Interlagos. Terramar was to be 20 cores and G44 socket but now there will be another G34 compatible chip after Interlagos (probably around the same time as Ivy Bridge server chips) and then G44 and a future design will come out later. So I expect AMD and Intel to kinda line up on the server chips for the next year at least
Sandy Bridge a few months after Interlagos, Ivy Bridge and Bulldozer.Next around the same time towards the end of 2012. And then Haswell and AMD's G44 compatible CPUs sometime in 2013
Don't care what the field of expertise may be, if you're rushing it out, chances are, you're bound to make a mistake. In the case of this CPU, I'm glad they're holding back til they feel comfortable enough to get it on the market. The Bulldozer is a pretty big deal for AMD. And they seem ready to put their money where their mouth is. I'll trust their judgement on this matter more than some boy wonder desk jockey, with a hidden agenda backed by a "I told you so." hanging by a thread of failure.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
22 Comments
Back to Article
Beenthere - Thursday, September 8, 2011 - link
To many these are just new chippies but in reality they are a new architecture and a major fundamental change in X86 CPU design, all for the better. I can't wait to start flogging a Zambezi system.tipoo - Thursday, September 8, 2011 - link
This is really their first ground-up architecture since 1999, with K7. Everyone says K8, but that was still based on K7 despite how huge it was. So 12 years since something like Bulldozer. AMD is really betting the barn on this, I hope its competitive!B3an - Thursday, September 8, 2011 - link
It's embarrassing that AMD come out with so few big architecture changes. They need to do a way better job at competing and i doubt Bulldozer will even match Sandy Bridge clock for clock. Theres so little competition in the high-end and it's just plain annoying now as it's hurting consumers.tipoo - Thursday, September 8, 2011 - link
Well Intel is using an architecture that they've built upon from the Pentium M in 2003, so AMD's count isn't "embarrassing". I know what you mean about performance though.JarredWalton - Thursday, September 8, 2011 - link
Not really... Core 2 went to 4-issue and reworked the functional units, branch prediction, pipelines, etc. It was certainly a new architecture, even if it "learned" from the successes of Pentium M. The Core i-series was also a largely new architecture, bringing back Hyper-Threading and changing the caching structure, along with adding some new instructions, reworking pipelines, etc. To pretend that those are using the same base architecture as Pentium M is a joke. So for Intel we've gone from the P6 (Pentium Pro through Pentium III) to Netburst to Core to Core-i as the major architectures in the past ~15 years.For AMD in that same span, we had K6, K7, K8 (sort of -- it's basically K7 with an integrated memory controller, which really isn't a new architecture), and then K10 (which is still essentially the same architecture as earlier, only refined). I'd liken K7 to K8 to K10 as being quite similar in principal to the Conroe to Allendale to Wolfdale plus quad-core Kentsfield and Yorkfield. There are changes with every one of those transitions, and even with K10 you have K10, K10.5, and different processes.
If I look at the architectures from a high level, AMD started with a good design on K7 and has continued to use it effectively. Intel on the other hand has had some radical new architectures, but some worked better than others in practice. Even Netburst was effective for what it did -- Pentium 4 chips outperformed Athlon XP chips in most tasks, but they cost more and used more power. It was only with K8 that AMD could clearly beat Netburst, and they did so effectively until Conroe entered the picture.
JonnyDough - Thursday, September 8, 2011 - link
Embarrassing? Hardly. It just goes to show that they've designed their chips very well and have been able to focus on business management, process, horizontal/vertical integration, and all without having to run back to the drawing board too much. Intel was borrowing things from AMD, remember all the lawsuits or were you absent the last decade? I'm not sure about what "little competition" you are speaking of, CPU prices are at an all time low and even a Sempron can play most older games. You obviously have no idea what you are talking about. You can get a quad core Phenom II or one of the new APUs that will do everything you are likely to ask except make coffee, a solid state disk now at $1 a GB (compare that with original HDDs that didn't even have a GB and cost about a grand), super cheap DDR3 memory, Windows 7 that isn't $200 like XP was (the best yet), a solid ATX board for $130, and very powerful discreet low-power graphics for $100-200 and you might want to rethink "there's no competition" again. Obviously, there is. Maybe you should take all your brains and start up a tech company to compete with Intel, which happens to be about 10X the size of AMD. :)Mr Perfect - Thursday, September 8, 2011 - link
I'm sure he meant no competition in the high end space(Anandtech readers generally forget the entry and midrange markets). AMD has nice bang-for-the-buck chips, but they didn't have a high end part to compete with socket 1366.redraider89 - Friday, September 9, 2011 - link
"i doubt Bulldozer will even match Sandy Bridge clock for clock."You know, anyone can say I doubt this or I doubt that and it not be based on anything. If you doubt it, you need to have a concrete reason. It's best to not to act like you are making an informed opinion when there is not any thing to back it up. I'm just tired of people making pessimistic comments based on nothing.
silverblue - Thursday, September 8, 2011 - link
Can't say I'm too surprised to see the server part first or a desktop part delay. I did hear of another stepping (C0?) but I don't think anyone cares except for one thing - get it out already!And, to preempt the trolls (you know who you are), if Bulldozer is competitive then it should mean cheaper Intel CPUs. Win-win.
hechacker1 - Thursday, September 8, 2011 - link
Now let's see some benchmarks of the server parts. I'm sure we can extrapolate potential desktop performance from that.yankeeDDL - Thursday, September 8, 2011 - link
Well said. Anyone knows when the first benchmarks will come out? I guess Anand and others already have them in house but are bound by NDA?fic2 - Thursday, September 8, 2011 - link
Agree. I am pretty close to saying f*ck it and buying a Sandy Bridge. AMD just keeps delaying and delaying. Seems like the next announcement out of AMD will be a Q1 2012 release, then a Q2, then... By the time Bulldozer is out Ivy Bridge will be either out or just around the corner.Some benchies might alleviate the feeling that I should have just bought an SB months ago.
BaronMatrix - Thursday, September 8, 2011 - link
If you consider them to be 256bit there are 8, but 95% of code will be 128bit FP.cfaalm - Thursday, September 8, 2011 - link
+1 on that. So many people overlook that feature. It's one of the reasons the cores are built like this. To be able to combine 2 x 128 in to 1 x 256 and be AVX capable.BSMonitor - Thursday, September 8, 2011 - link
BaronMatrix is alive? Still waiting on the TRUE quad-core from AMD to wipe the floor with the Intel dual die quads ehh?Still hoping for that told-ya-so??
If these things were any good, we'd have benchmarks flying around everywhere.
Instead, crickets..
WeaselITB - Thursday, September 8, 2011 - link
Or, maybe, because they're all still under NDA?Wait ... logic has no place in an irresponsible flame war. My mistake. Carry on.
-Weasel
BSMonitor - Friday, September 9, 2011 - link
Intel had no problem showing off Conroe, Penryn, Nehelem, Sandy Bridge... If one wanted to generate buzz and excitement about a new product, why hide its performance??silverblue - Thursday, September 8, 2011 - link
http://en.wikipedia.org/wiki/Troll_%28Internet%29silverblue - Thursday, September 8, 2011 - link
It's confusing watching John Fruehe talk about Interlagos in comparison with "8 core Sandy Bridge" but he's got a point; Sandy Bridge's FPUs can only do 1x128 OR 1x256 in the same cycle, which is a sizeable disadvantage with today's workloads. Bravo to Intel for being very forward thinking with AVX but for the first time in a while, they stand to be behind in instruction set support at the very least. As for performance, that's not something most of us know anything about yet, though Sandy Bridge will definitely be ahead in some areas, you can count on that.yankeeDDL - Friday, September 9, 2011 - link
Indeed.The fundamental difference in architecture will make it difficult to run proper benchmarks and to do fair comparisons. Synthetic results could come up completely skewed.
Casper42 - Saturday, September 10, 2011 - link
So AMD had a small delay in getting Interlagos out, but Intel just slipped SB-E Server chips (Xeon 5600 replacement) from a Nov launch to more like Feb/March due to some errata they didn't feel comfortable releasing as well as issues with PCIe 3.0. On the flip side, Intel still seemed to think they could get SB-E desktops (X79 compatible chips) out before the holidays but only time will tell. Either way it looks like AMD might have a few month advantage on both fronts for a little while.Some AMD sources also said that Terramar will not be the next gen chip after Interlagos. Terramar was to be 20 cores and G44 socket but now there will be another G34 compatible chip after Interlagos (probably around the same time as Ivy Bridge server chips) and then G44 and a future design will come out later. So I expect AMD and Intel to kinda line up on the server chips for the next year at least
Sandy Bridge a few months after Interlagos,
Ivy Bridge and Bulldozer.Next around the same time towards the end of 2012.
And then Haswell and AMD's G44 compatible CPUs sometime in 2013
DominoFX - Sunday, September 11, 2011 - link
Don't care what the field of expertise may be, if you're rushing it out, chances are, you're bound to make a mistake. In the case of this CPU, I'm glad they're holding back til they feel comfortable enough to get it on the market. The Bulldozer is a pretty big deal for AMD. And they seem ready to put their money where their mouth is. I'll trust their judgement on this matter more than some boy wonder desk jockey, with a hidden agenda backed by a "I told you so." hanging by a thread of failure.