Sign-in / Sign-up
Your question

Intel/AMD Microarchitecture 2011

Tags:
  • CPUs
  • Tom's Hardware
  • AMD
  • Intel
Last response: in CPUs
a b à CPUs
October 26, 2010 8:13:34 PM

Hello Tom's Hardware,

I have been intrigued by the noobiness of all the people all over the internet. It seems Sandy Bridge and Bulldozer are the next thing since sliced bread. Anyways. I am posting here to collect the thoughts of the masses of posters of Tom's hardware. This is NOT another "Is Intel/AMD Doomed" thread. I do not want fanboys flame wars,( 4ryan and jpishgar will have my head), or anything supporting the company wholly. You can compliment each or one, but not support.

What do you think about the new architectures and changes to the new series? What about the new sockets? Fabrication changes? Additions and subtractions,( like Fusion for AMD)?

More about : intel amd microarchitecture 2011

a b à CPUs
October 27, 2010 1:12:05 AM

Dogman, if you're really really interested in learning about the new architectures, then go visit http://www.realworldtech.com. The very first two articles are on the Sandy Bridge and Bulldozer architectures. However nobody will know the exact pros and cons of each until actual shipping product gets reviewed on a bunch of sites. Otherwise it's all theoretical.

Now if you're interested in a preview of SB, Anandtech had some articles on it last month. General consensus around here is that it'll show about 20% IPC improvement over Westmere, and if the reports of oc's up to 5GHz on air are true, then the unlocked versions will probably be the gamer's choice next year, especially if desktop versions of Bulldozer aren't out until late next year. But, as I said above, nobody actually will know until they are shipping and reviewed...
a b à CPUs
October 27, 2010 6:47:12 AM

Flame wars? Count me in!..

I chose VIA's side... :lol:  Just kidding. :p 

I'd say, if this is really about discussing the two arch enemies on the new "Hybrid CPU war" saga, I can't really dive in. They're both still in pre-launch state. Any info regarding the two products can be considered uncertain. That's why I prefer to wait for the reviews, then comment. Besides, with my shallow knowledge on technical stuffs like architecture, socket type, fabrication, etc, I simply can't keep up with the others.

Anyway, tell me if the new fanboy battle starts, I'd buy some popcorns and coke while watching. Maybe some provocative posts. :lol:  But the mods will have my head if I do that. :p 

Related resources
October 27, 2010 11:13:33 AM

I just want what works. It looks like they are going to need to come up with new bechmarks for CPU/GPU combos. Until we see consumer products allowing comparisons of Llano, i7,SB,Bulldozer this is all guessing. I want to see side by side benchmarks for each
a b à CPUs
October 27, 2010 9:33:28 PM

If 1155 has 20 PCI 2.0 lanes then to me SB is FAIL. Looks like only AMD will offer fully loaded mobos at affordable prices, again. *sigh

I'm not paying $250+ for more than 20 PCI lanes when my old 775 has 36 2.0s for ~$230 over 2 years ago.

Then again I'm not about to go BD for 8+ cores and a slight, if any, IPC increase for gaming.

Hell GPUs (due to game developers and *cough*consoles *cough* slowing gaming progression) can't keep up with lynnfield anyways. (5870 crossfire maxes FPS at what, 3ghz, if that)

I might be stick with my Q9550 longer than I thought... Maybe 6970s and Crysis 2 will get me back in the spiritt.
October 28, 2010 2:23:09 AM

I'm not a fanboy, I have an intel cpu running on an nvidia chipset with an ati gpu.

I think that this new 1155/p67 Intel thing will be useless to a good chunk of the people. I believe that this is a step backward for gamers and overclockers. btw, overclocking has become a game in itself and with Intel crippling overclockers with the locked usb, pci, and memory timings, well, it just makes me mad.

Amd has a chance to take the lead, depending on how Bulldozer will match up against the current Intel products. If Amd can pull this off, they will have the lead for probably a couple of years, until Intel can pull their head out of their (cuss word) and start thinking about the customer instead of the share holders.

But, alas, THIS is what I really believe will happen. Intel will alienate the hardcore customers and Amd will miss the chance to take the lead because Bulldozer won't be as good as the current i7's, maybe as good as the i5-760, but I pretty much doubt it.

Sorry folks for the rant. I was going to start a new thread but seen this thread. Thanks for letting me vent a little bit here. (group hug)

a b à CPUs
October 28, 2010 2:56:07 AM

Quote:

I think that this new 1155/p67 Intel thing will be useless to a good chunk of the people. I believe that this is a step backward for gamers and overclockers. btw, overclocking has become a game in itself and with Intel crippling overclockers with the locked usb, pci, and memory timings, well, it just makes me mad.

Intel has said that they will offer K versions with unlocked multi.

To me, I'm not really interested in LGA1155 as a desktop CPU, however I would like to see how the mobile versions of Sandy Bridge will perform.

I'm waiting for Bulldozer and LGA2011.

Quote:

Amd has a chance to take the lead, depending on how Bulldozer will match up against the current Intel products. If Amd can pull this off, they will have the lead for probably a couple of years, until Intel can pull their head out of their (cuss word) and start thinking about the customer instead of the share holders.

Sorry to brake it to you, but the Enthusiast market is probably less than 5% of Intel's profits,etc. It's probably a bit higher for AMD, but the fact is the Enthusiast market makes very little difference to CPU manufactures. On the other hand, this market matters a LOT to companies like ASUS, Gigabyte, XFX,EVGA, etc.

Did you know most of the Extreme Edition CPUs are sold to OEMs? Yup. Real enthusiasts will use the cheaper CPUs and OC the $hit out of them. This is something that Intel has tried to avoid for some time (FSB limits in P4,etc). I think Intel may have finally come to a compromise by offering Unlocked CPUs for a little premium. Only prices will tell if Intel really cares about the enthusiast market.
a b à CPUs
October 28, 2010 3:12:00 AM

*cough* They don't.*cough*
a b à CPUs
a b À AMD
October 28, 2010 4:53:40 AM

Am I going to have to sit on this thread with my dishing out the 's and 's ?
a b à CPUs
October 28, 2010 5:11:43 AM

Naw...go watch some soccer.
a b à CPUs
a b À AMD
October 28, 2010 5:13:14 AM

dogman_1234 said:
Naw...go watch some soccer.

I hate football or "soccer" as you call it.
a b à CPUs
October 28, 2010 11:22:32 AM

I say buy them both up! Let's keep this competition neck-and-neck so we can all enjoy relatively lower prices for a little bit longer, until one or the other pulls the plug on competition and decides to go for the throat.

It's the anticipation of new hardware that gets the best of us, really. I am very enthused to see what AMD is pulling out of the hat. And to be completely honest, from what I have seen of Sandy Bridge so far, I'm not really all that blown away.

It just seems like a marginal step in the direction that Intel has been headed towards for a while. Pardon me if I'm not exactly over the top when folks start talking about embedded IGP solutions from Intel, considering their record. I am more than enthused when folks start talking about additional GPU core from AMD, though, given the proven track record of their graphics arm with an eye to performance.
a b à CPUs
October 28, 2010 1:53:37 PM

I think the reason both are secretive about their products is to increase revenue...(duh?!?!)
a b à CPUs
October 28, 2010 2:06:45 PM

fazer's post was the only one with any content people ...

Tsk tsk ... we are getting lazy.

Sandy Bridge looks good and before you post telling me its fake remember coolaler has more cred than anyone here in terms of popping stuff out first.

http://forum.coolaler.com/showthread.php?t=251959
October 28, 2010 2:17:56 PM

maybe they will come out 2 silicon(GPU+CPU) on 1 chip for SB...
a b à CPUs
October 28, 2010 2:29:58 PM

It also may be a sign that Intel and AMD haven't released and 'good' news about their products...except their names and concepts.
a c 217 à CPUs
a c 154 À AMD
a b å Intel
October 28, 2010 2:38:23 PM

dogman_1234 said:
It also may be a sign that Intel and AMD haven't released and 'good' news about their products...except their names and concepts.


With the new instructions to be added, they are busy *creatively engineering* their arses off. They don't want to release any testing where the results may well be obliterated 6 weeks later with optimizations.
a b à CPUs
October 28, 2010 6:34:57 PM

Raidur said:
If 1155 has 20 PCI 2.0 lanes then to me SB is FAIL. Looks like only AMD will offer fully loaded mobos at affordable prices, again. *sigh

I'm not paying $250+ for more than 20 PCI lanes when my old 775 has 36 2.0s for ~$230 over 2 years ago.

Then again I'm not about to go BD for 8+ cores and a slight, if any, IPC increase for gaming.

Hell GPUs (due to game developers and *cough*consoles *cough* slowing gaming progression) can't keep up with lynnfield anyways. (5870 crossfire maxes FPS at what, 3ghz, if that)

I might be stick with my Q9550 longer than I thought... Maybe 6970s and Crysis 2 will get me back in the spiritt.


I don't usually give too much credence to sources like The Inquirer , but anyway:

Quote:
The new Socket LGA1155, common to both mainstream desktop Sandy Bridge Core i3/i5/i7 chips and the entry level Xeon Sandy Bridge parts, will support CPUs in two different configurations for PCIe I/O: the 16-lane desktop part, and the 20-lane server and workstation part, all of course at PCIe v2 speed.

Otherwise, the desktop and enterprise parts are identical, including up to four cores and 8MB cache, and speeds up to 3.4GHz for the parts with GPU turned on, and 3.5GHz for the parts with the disabled GPU. The fine grained Turbo capability gives them another up to 400MHz headroom when all cores are used, with appropriate power and thermal solutions. These sockets, by now well known to the community, are the only ones with the built-in GPU.

Then we come to the Socket LGA1356, a direct replacement for the current Socket LGA1366. The parts here are 6-core and 8-core Sandy Bridge single-socket and dual-socket capable but midrange positioned Sandy Bridge Xeon - and, ultimately, Core i7 - parts with up to 20MB of L3 cache, three DDR3-1600 memory channels just like the existing LGA1366 Westmeres with one memory speed grade higher, and 24 PCIe v3 lanes on-chip. The single external QPI v2 link runs at up to 8 gigatransfers/sec, or 32GB/sec bidirectional bandwidth, a 25 per cent speed up over the current generation, but also feeding a third more cores on each socket.

The highest speed 8-core CPUs with up to 150W TDP should, however, be reserved for the high-end Socket LGA2011. With more power and ground lines to support 40 PCIe v3 lanes and four DDR-1600 memory channels per socket, as well as dual QPI 8 gigatransfers/sec links, the 8-core, 20MB L3 cache Sandy Bridge-based Xeons should have sufficient system bandwidth to feed even the highest workloads. Not to mention enough PCIe bandwidth for two dual-GPU cards with extra lanes still free for a, say, 5GB/sec PCIe high-speed SSD or Infiniband interconnect.

And, when you add the same resource on the second CPU, it becomes possible to fully feed an 8 GPU system out of a single two processor workstation. And yes, you could even do a quad-socket monster here, if you're using the EX parts, I assume.


So if true, then you'd probably be interested in either the 1356 or 2011 socket Sandy Bridges.
a b à CPUs
October 28, 2010 6:38:25 PM

Reynod said:
fazer's post was the only one with any content people ...


LOL - thanks!

I've been taking my anti-fanbois meds for some time now - seem to be working for the most part :p 
a b à CPUs
October 28, 2010 6:45:10 PM

:D 

Good to hear.
October 28, 2010 7:23:28 PM

AMD is going after the server market while Intel is trying to go after both with improved ipc. AMD meanwhile will focus on more cores and less power and hope apps become multithreaded by the time their chip releases. If not then they will survive on the server market.

We can't speculate on the chips themselves because we don't have all of them yet.

/thread
a b à CPUs
October 29, 2010 1:08:36 AM

werxen said:
AMD is going after the server market while Intel is trying to go after both with improved ipc. AMD meanwhile will focus on more cores and less power and hope apps become multithreaded by the time their chip releases. If not then they will survive on the server market.
/thread


IIRC server accounted for something like 1/4th of Intel's CPU income last quarter, whereas mobile was nearly 1/2. Desktop, at 12%, not so much :) . I suppose netbooks, tablets and below account for the rest. I guess AMD is positioned somewhat similarly.

a b à CPUs
October 29, 2010 2:21:10 AM

Hopefully, AMD can get the desktop and server part and play around with Intel on this. It will be funny if AMD gets screwy with Intel...just as Intel got screwy with them.
October 29, 2010 3:58:36 AM

I think this will be a good thread next year after the products are out. There is too much that has not been disclosed and will not be out until launch.
October 29, 2010 5:56:50 AM

I just wanted to point this out about Bulldozer:
http://blogs.amd.com/work/2010/10/25/the-new-flex-fp/

Read the post, specifically this part:
"One of these new instruction set extensions, AVX, can handle 256-bit FP executions. Now, let’s be clear, there is no such thing as a 256-bit command. Single precision commands are 32-bit and double precision are 64-bit. With today’s standard 128-bit FPUs, you execute four single precision commands or two double precision commands in parallel per cycle. With AVX you can double that, executing eight 32-bit commands or four 64-bit commands per cycle – but only if your application supports AVX. If it doesn’t support AVX, then that flashy new 256-bit FPU only executes in 128-bit mode (half the throughput). That is, unless you have a Flex FP."

It seems like when JF mentions the "flashy new 256 bit FPU" he is referring to Sandy Bridges FPU (im assuming), so does this mean that Bulldozer might have a higher IPC than SB in floating point workloads because of its flex FP?
a b à CPUs
October 29, 2010 6:34:00 AM

Could be...
a b à CPUs
October 29, 2010 5:35:46 PM

Quote:
It seems like when JF mentions the "flashy new 256 bit FPU" he is referring to Sandy Bridges FPU (im assuming), so does this mean that Bulldozer might have a higher IPC than SB in floating point workloads because of its flex FP?

Very possible if they are targeting the server/workstation/HPC market a lot with this release.
a b à CPUs
October 29, 2010 6:13:06 PM

Hmm, if you compare an 8-core SB to an 8-core (4-module) BD, then it seems they would have the same number of FP units - eight - in 128-bit mode. However for AVX-enabled apps, the same SB would have 8 256-bit-wide FP units vs. BD's 4. It's only when you're comparing the MCM 16-core versions of BD to the 8-core versions of SB that BD would have the numerical advantage in 128-bit floating point ops. And I would imagine Intel will have MCM versions of SB as well, as they do with Nehalem-EX.

If AVX yields as much performance boost as Intel says, then I would imagine there will be upgrades or updates for most software available pretty soon after SB launches, if not before, esp. for consumer apps. And IIRC both Intel and AMD provide assistance to various software firms to help them with various issues such as updating to the latest SSE versions.

As for enthusiasts, IIRC most games are mainly integer and not FP.
a b à CPUs
October 30, 2010 3:16:55 AM

jf-amd said:
I think this will be a good thread next year after the products are out. There is too much that has not been disclosed and will not be out until launch.



Thanks John.

I read your latest blog on the new Flex FP with interest and it seems to have generated plenty of debate across the main global hardware sites in various threads.

The 4 part series you put together on Bulldozer 20 questions was also interesting.

People if you want to read them simply click JF's sig and read on ... get a good cup of coffee first though !!

a b à CPUs
October 30, 2010 3:54:57 AM

Well not much info, but no flames yet, so thats good. Heres my opinion. Id say they are both doing what the other should in a way. AMD is right on the edge. Phenom set them back a generation, and put them in some serious debt. Just now they are getting out of debt and getting back in the race. Intel on the other hand is filthy rich. And yet AMD is the one taking the risk here, and Intel is going with the small tweaks on a tried and true system. Its just ironic to me. AMD cant really go through another Phenom, yet they are taking the risk on a very new arch, and Intel can go through 5 netbursts and still have more money than they know what to do with, and they are playing it safe. Looking at it alternatively, perhaps AMD needs a homerun to fully get back in the race, not be a gen or 1/2 gen behind staying in the low end, and Intel is just on autopilot. As far as Sandy Bridge, we dont know much more than what the leaks and the review at AT told us. About 10% more performance clock for clock, 20% more with stock clocks. BD we know a lot less about. JF is doing a good job filling us in with the arch designs and details, so that helps, and after reading through most of what they have to say, all i can think is "wow, talk about going off the beaten path." AMD really seems to be thinking outside of the box with BD. It hardly seem recognizable to what we know. Everything about it seems to be targeting efficiency. Seems like AMD's new thing, efficiency, as show with Barts. All i know is BD likely wont be like much of anything we recognize, and its either going to be great, or it could be just flat out bad. Either way, it seems like it might be coming a full 2 quarters after Sandy Bridge which is unfortunate. We shall see...
a b à CPUs
October 30, 2010 4:01:20 AM

Quote:
AMD cant really go through another Phenom, yet they are taking the risk on a very new arch, and Intel can go through 5 netbursts and still have more money than they know what to do with, and they are playing it safe.

In the business world, if you are the underdog who doesn't have a pile of cash to sit on top of, you really do need to try new things, or else your business goes belly up down the road.

Anyways, I do hope AMD can pull this off. Esp. in the 2P area as I will be in the market for one mid/late next year :p .
a b à CPUs
October 30, 2010 8:37:48 AM

Deleted all of the dribble as requested.
a b à CPUs
October 30, 2010 5:11:39 PM

Thanks.
a b à CPUs
October 31, 2010 1:19:59 PM

ares1214 said:
AMD is right on the edge. Phenom set them back a generation, and put them in some serious debt. Just now they are getting out of debt and getting back in the race. Intel on the other hand is filthy rich. And yet AMD is the one taking the risk here, and Intel is going with the small tweaks on a tried and true system. Its just ironic to me. AMD cant really go through another Phenom, yet they are taking the risk on a very new arch, and Intel can go through 5 netbursts and still have more money than they know what to do with, and they are playing it safe.


IMO, Intel learned a valuable lesson with Netburst about taking too much risk with design (i.e., putting all your proverbial eggs in one basket). However the Netburst effort wasn't wasted - there's a lot of similar concepts in the Sandy Bridge design, which is advanced enough to take advantage of them. And Intel is by no means risk-adverse - take Itanium and Larrabee. Please! :D  But jokes aside, notice that SB uses a ring bus like in Larrabee to interconnect the CPU and GPU.
a b à CPUs
October 31, 2010 1:26:19 PM

a ring?

hmmm...

Netburst was a epic fail on Intel's part. I don't wish to see SD as another netburst...but I do want to see AMD get this one done and pressure Intel for Competition.
a b à CPUs
October 31, 2010 2:03:06 PM

^ Net Burst? A failure? Look at how many people (esp. Businesses) still run P4s + XP.
a b à CPUs
October 31, 2010 3:53:01 PM

Shadow703793 said:
^ Net Burst? A failure? Look at how many people (esp. Businesses) still run P4s + XP.


Thats because its to expensive and too much of a hassle to upgrade. Atleast for most of them. I kind feel like SB is like 6xxx, and BD is like Kepler. SB is only the first part of the lineup, and isnt so much a full upgrade as just a nice boost. IB should complete it. Then BD is entirely new, and an entire lineup. Not sure its fair to compare them, as BD also appears to be coming out a full quarter later. Im also getting the strange feeling that SB is going to dominate gaming and single/lightly threaded apps, and BD is going to destroy everything at multi-threaded/video editing/server work. Should be interesting to see how things play out.
a b à CPUs
October 31, 2010 8:38:02 PM

P4 was really not netburst. It was after P4 that Intel pursued the GHz project. It was initially called P5 but it later change the cores and began the Multi-core series...what you and I call the "Core" series.
a b à CPUs
October 31, 2010 8:40:09 PM

I'm sorry...I meant P7...
a b à CPUs
October 31, 2010 10:17:16 PM

The core2 series were based on the Pentium M (souped up P3 with low power in mind) and not the P4.

The P4 line was Netburst ... it was a good design but suffered a few flaws - the Replay function caused the cache to get flushed too often (cache misses were terrible) and the l1 caches were too small). The pipes were also very long ... adding to the cache miss problem.


a b à CPUs
October 31, 2010 10:29:36 PM

Quote:

Thats because its to expensive and too much of a hassle to upgrade.

No, the point I was making was no one went with AMD even when they had a superior product. One reason for this was because OEM like Dell, HP,etc didn't carry any non-server AMD products.
a b à CPUs
November 1, 2010 1:13:18 AM

Reynod said:
The core2 series were based on the Pentium M (souped up P3 with low power in mind) and not the P4.

The P4 line was Netburst ... it was a good design but suffered a few flaws - the Replay function caused the cache to get flushed too often (cache misses were terrible) and the l1 caches were too small). The pipes were also very long ... adding to the cache miss problem.



Also the orignal "Core" or Yonah was based on the Pentium M or Dothan.
a b à CPUs
November 1, 2010 3:22:25 AM

Shadow703793 said:
Quote:

Thats because its to expensive and too much of a hassle to upgrade.

No, the point I was making was no one went with AMD even when they had a superior product. One reason for this was because OEM like Dell, HP,etc didn't carry any non-server AMD products.



No. The OEMs did not get AMD CPU's because of Intel. Intel used a practice of 'blackballing' AMD and offered cheaper prices on Intel products if the OEM went Intel. That is on reason Intel was caught with their pants down at the FTC.
a b à CPUs
November 1, 2010 4:18:08 AM

Actually, Intel got their heads handed to them by the Europeans, not the USA. Just like Microsoft has been forced to "play nice" with the other browsers out there regarding disclosure of all the APIs that were at one time restricted to Microsoft products only (such as Office and IE). We may think that the US Government is keeping the big boys playign nice, but the reality is that the FTC and several other govt watchdogs are more than happy to let sleeping dogs lie as long ast heir back get scratched too. The Europeans, on the other hand, actually have a constituency that they have to answer to, and have decided that it may not be in the best interests of Europe to allow American companies to run roughshod over competing European companies.
a b à CPUs
November 1, 2010 4:57:27 AM

Oh, yeah right. i forgot. The US has NO regulation. If someone does something,( the person with money,) they get a slap on the hand. The Europeans De-ball the guy for heaven sakes fr crap like that. Reaganomics screwed up this country. That is ONE reason Intel is expensive...minus their economic practices on their products.
a b à CPUs
November 1, 2010 7:46:51 AM

Your paintbrush is to big for detailed work ... and three colours just won't do.
November 1, 2010 11:15:18 AM

fazers_on_stun said:
Hmm, if you compare an 8-core SB to an 8-core (4-module) BD, then it seems they would have the same number of FP units - eight - in 128-bit mode. However for AVX-enabled apps, the same SB would have 8 256-bit-wide FP units vs. BD's 4. It's only when you're comparing the MCM 16-core versions of BD to the 8-core versions of SB that BD would have the numerical advantage in 128-bit floating point ops. And I would imagine Intel will have MCM versions of SB as well, as they do with Nehalem-EX.

If AVX yields as much performance boost as Intel says, then I would imagine there will be upgrades or updates for most software available pretty soon after SB launches, if not before, esp. for consumer apps. And IIRC both Intel and AMD provide assistance to various software firms to help them with various issues such as updating to the latest SSE versions.

As for enthusiasts, IIRC most games are mainly integer and not FP.


In the server world it will be 16 cores on BD vs. 8 cores on SB. That means in 128-bit workloads we will have 16 FPUs to their 8 FPUs.

We both share to get to AVX. We share 2 FPUs to get a 256-bit AVX execution path. They take 128-bit away from the integer execution path to get to 256-bit.

Based on the fact that most workloads are integer-based, I'd personally rather share FPU resources than take away from my integer execution in order to do 256-bit AVX. But, hey, that is my opinion.
      • 1 / 2
      • 2
      • Newest