you'll see in very multi-threaded applications (like compression or encoding) the AMD CPU actually keeps pace with the 2600k; in other applications where it is more important to have faster single threads of execution, the Intel 2600k pulls ahead by a drastic amount though.
These are the early days, but if AMD gets this down to 22nm next year and puts 16 cores on a desktop PC -- each core may be only running at 4Ghz, but in a scientific computing or server environment where you have 16, 4Ghz cores as opposed to 4, 6Ghz cores on some equivalent Intel can make a big difference.
The new Bulldozer platform looks to be a strategic company gamble on all fronts (more cores, less speed per core) that could win AMD big in the server space (which is exploding) and in scientific computing environments.
I'm not saying the numbers aren't disappointing, I was hoping to see some big ones in there to. I'm only pointing out that this isn't quite apples to apples, and when it is (e.g. straight multithreaded benchmarking) the comparisons look OK.
All that said, I'm sure Intel has something waiting in the wings (e.g. 2800k) to shit all over this party and beyond that we have tri-gate IvyBridge CPUs next year that will hopefully bring a much-needed* leap to CPU performance.
* "needed" as defined by "my impatient ass likes fast things.
Ok, but based on the die size and the transistor count we could have had 12 Thuban cores in the same amount of space.
It really feels like if we are going to go with more cores, why not make them simpler? Why not dust off the old Athlon K7 at 37 million transistors. Roughly 54 of those fit into the 2 billion transistors making up the 8 core bulldozer. Bulldozer has an Itanic feel to it, no that is too harsh, it has a P4 feel to it.
I really feel like desktops are done, GPUs have this biggest effect on desktop perf right now. It would seem to focus on the fusion A series parts and the server. As a server side dev, I would rather have more cores (more cache actually). I don't need memory coherency (that should be mitigated by the hypervisor).
I don't think power is it. Your saying the K7 generates more heat per computation (more inefficient use of transistors) than the newer cores? I was about to conflate power usage with computational work, not necessarily the same thing... The power will drop using the new process because the voltage is much much lower (1v vs 2v)
I remember the K7 doing just over 1 GOP (giga operation per second), so that would put the whole cluster in 50 GOP range (if clocked at only 1Ghz like the Thunderbird).
I wasn't being directly literal, I will allow these mythical engineers that work for me using all the 2 billion transistors to refine things from previous designs.
It is true the power usage of a K7 per op will drop if you redesign the core to use 32nm. That's a key detail though- you can't just drop a 130nm design onto a 32nm mask.
There are many hitches with this idea though. K7 cores have limited power management capability; modern cores have sophisticated frequency scaling, and shut down sectors that are idle. K7 cores are 32-bit. K7 cores do not support most modern extended instruction sets. 50 active K7 cores on one die will generate way more than 125W, causing cooling problems. In short, there are a lot of issues with dated processors that we don't think about, because we take the advances for granted.
This is not to say your idea does not have merit. If it was I, I would probably first investigate using a Bobcat core. The Bobcat core has low power usage for a x86 core (good for packing a large number of cores on one die), and considering a single Bulldozer 'module' (core) has 200M transistors, the Bobcat probably has ?100M?, allowing you to fit ?20? cores with 2 threads each.
P.S. If you are about to brush off the extended instruction sets bit, consider who a 50-core die would interest most. My guess is the scientific community would be a large part of the target markets, and I believe they make great use of extended instruction sets.
sit, you are exactly right and they are going that route with the Opteron's. The 16-core Opterons were announced a while ago, I believe there is GPU infrastructure in the desktop chips (similar to what Intel is doing with the always-includes-a-GPU-now CPUs).
I imagine when we see the 16-core AMD's next year on 22nm, we'll have 24-core Opterons or something equivalent that is a boon to the hosting industry.
Not saying Intel won't be right there with their own solution, but going the "more cores" route should be interesting to watch and see if it pays off.
Honestly is sounds exactly like the strategy Sun pursued with UltraSPARC systems; however in my experience, time and time again commodity hardware with higher per-core clock speeds beat the pants off of Sun hardware except for very narrow HPC tasks (supposedly, I couldn't even get decent speeds for the distributed computing projects available on the platform).
It also seems to rely on software to be written specifically to take advantage of multiple local threads. This may or may not happen in a timely fashion for most projects (and again, chances are fewer cores with a higher clock speed will still outperform).
Agreed -- it really seems to be a long term bet. They see the growth of hosted apps and cloud environments and I am sure that is making up a large margin of their sales (and have been for years).
I think they are banking on dominating the multi-tenant server market with these designs and if consumer PCs follow suite with some convoluted virtualization mechanism in the years to come, that might leave them sitting pretty with a lot of disparate, fast cores ready to execute the work thrown at them.
If not, and quad-core is all we ever really utilize, then Intel is going to dominate that arena as they have for the last decade.
you'll see in very multi-threaded applications (like compression or encoding) the AMD CPU actually keeps pace with the 2600k;
Er, no it doesn't. In integer limited operations, such as winzip it keeps up. In anything using floating point, its 20% slower at least. At AAC encoding its nearly twice as slow!
No doubt they have some tightening to do, but they are always a generation behind Intel when it comes to fab; when they get the new architecture onto 22nm I think things should even out, but then Intel will be jumping to something much smaller at that point in a tick-tock cycle anyway.
I think you are right, I recall reading something similar that we were going to get huge power gains (like 30% improvement) for equivalent performance. That being said I expected Intel to fill up the high end with an appropriately overlocked CPU with the typical 95/130 TDP heat payload.
If AMD doesn't push them to do it though, I suppose they have absolutely no reason to out-do themselves.
Anandtech did a decent piece on this last month [1].
To paraphrase their description: every other year Intel keep their manufacturing process the same, but release a new microprocessor architecture (Sandybridge was one of those years), but in the other years they introduce a new manufacturing process (e.g. the new tri-gate 22nm transistors which will be found in the Ivy Bridge).
Right-o, that is the "tick-tock" terminology that I believe Anand popularized.
They keep tech the same and shrink die (32->22) then in the tock, they keep the die size the same and improve tech. Then the next tick, take all that new tech and shrink it again and on and on.
Seems to have worked out well for them.
SandyBridge (as you pointed out) was the "tick" (32nm) with the 22nm "tock" IvyBridge planned for next year.
I think the wildcard in this "tock" is the tri-gate transistor stuff. It'll either be incremental and we'll see our standard 15% hop in performance, or it will unlock a lot of potential on the high end for power savings, over clocking and growth which could be very exciting.
OR, we are all blindsided by ARM chips from Qualcom/NVIDIA/Samsung and start working on our phones :)
It's kinda sad to see AMD slipping behind even with their best effort. I was a huge fan for years and this i3 I am using now is my first Intel cpu since the 80386.
I was thinking maybe it's just that the OS and software is not parallel enough yet but the benchmarks show that's not the only issue.
They are not totally slipping, but they are slipping in the high end desktop realm. They have not been able to compete with Intel, in terms of high end desktop performance, since the Core 2 arrived. Their first quad core was a train wreck... and the much anticipated (by me at least) Bulldozer seems to be at the level of Intel's year old Sandy Bridge.
It looks like Bulldozer is a "speed racer" design that was intended to run at 4-5 GHz. It's hard to tell the cause from the outside, but either due to 32 nm fab problems or design problems they're shipping it at 3.6 GHz. Given the fates of NetBurst and POWER6, I cannot imagine why AMD chose such a design.
Another way to look at it is that the market will bear ~$1,000 pricing for Intel's flagship processors but <$300 for AMD's. That's sad. (I remember paying $700 for the first Athlon FX.)
I've heard that the main problem is that they don't really have the expertise and manpower anymore to do a full custom design, but are using a more ASIC style process - influenced by the ATI folks. Also, this is an all-new architecture and so you'd expect that it would be unpolished and untuned. Intel hasn't done anything as drastic as what AMD has done here since the introduction of the Pentium Pro. Well, there was the Atom but Bobcat (also an all new design) is able to beat that the same way that Intel is beating AMD at the high end.
In theory, this will let AMD mix and match components easily, and quickly synthesize new designs with the right mix of Bulldozer cores, Bobcat cores, GPU cores, memeory controllers, etc to meet market demand. I haven't seen any evidence that this is actually happening, though, or that its worth the price. By contrast, Intel has huge design teams that can afford to hand tune a range of designs to changing circumstances.
Its also true that though GF has a more tolerant set of design rules in its 32nm process, it pays for it in lower overall performance compared to Intel's process at the same node.
I'm not sure the high frequency design was a kiss of death. In terms of pipelining it doesn't seem any more aggressive than the Power7, which has been very successful. The Power7 benefits from 4-way SMT, though, and that probably helps hide the cache latency that is hurting Bulldozer so much.
They put a bunch of effort into this cpu but in several cases it's actually slower than their older 6-core.
They have a whole new approach coming but that will take at least a year to appear.
Lately all AMD seems to be doing is helping keep Intel prices lower (I got this 4ghz i3 with motherboard for $100 after rebate, which is incredible compared to Intel pricing in the past).
AMD has lost their edge. They used to compete with low priced processors that were pretty much matched, within a few percent, with Intel's offerings. With their newest flagship Bulldozer processors retailing at 279.00 you can save yourself $60.00 and go with the a Intel i5-2500k and have nearly the same performance. The AMDs may do well in highly multithreaded scenarios, but their single core performance isn't stellar.
Yea, they compensate by having twice the cores with each pair sharing a floating-point units, which works differently than Intel's HT used in Sandy Bridge. Basically which will be better depends on the workload.
Am I the only one that would prefer cooler, quieter CPUs rather than 8-way monstrosities with big fans and creating lots of heat?
I have 2 systems running in my room, a 2-core desktop, and a 2 CPU quad-core system that I run virtualization on. This system replaced about 4 other towers that I had running, each doing various things like running web server, dns, etc, but basically just sitting idle and sucking up electricity.
I want to upgrade to a newer desktop, but I really don't want this thing to generate more heat, use more electricity and have a loud fan. Before, I wanted to upgrade my video card but couldn't find a reasonable graphics card that didn't have a fan.
CPU power consumption hasn't been increasing at all for years. The high-end desktop chips all top out at 125-130W TDP, with the slightly more reasonable chips topping out at 95W TDP, and the mainstream desktop chips have a TDP of 65W or lower. Compare this with the Pentium 4 desktop chips, which mostly ranged from 84W TDP to 115W TDP, and had hardly any power management features.
The extra cores don't increase power consumption, because the clock speed they run at is dependent on how many cores you are currently using, and cores you aren't using are almost entirely turned off.
Idle power consumption of a high-end gaming system these days is only about 80W, even though peak power consumption can be on the order of a kW if you use multiple graphics cards.
If you think current chips are monstrosities that put out lots of heat and fan noise, you're probably running with a lot of power management features disabled.
Hmm, maybe instead of 2 system you can now have one 8 core system that runs virtualization and your desktop? Now you have one power supply, less fans, ...
There is nothing inherently loud or power consuming about 8-way monstrosities. Do you remember Pentium IV or even Alpha (One of those things could heat up a room plenty well in the winter).
Unfortunately, you can't hide the loud, hot running monstrosity that you use for the heavy work under a staircase in the basement if you use it as a desktop.
That said, the Bulldozer chips have relatively power-efficient cores, and it will shut off power to cores that you aren't using. If your system isn't particularly loaded, there's no reason for it to burn enormous amounts of electricity.
The ever-marching progression of fire-breathing monstrosities brings you lower-performance chips that do more and more for less and less power. If you are interested in low power, I find you are best served by eyeing the low-mid end of the previous generation.
"The great revolution that AMD has been talking about... an eight-core desktop processor", is it really a revolution?
an 8-core smartphone would be next..
Let's be more precise about what they've done. It's a processor designed for high single-core performance -- much higher than smartphone processors -- with good power-efficiency, on a much smaller die than would ordinarily be required for that.
And 8-core smartphone processor is very different from an 8-core desktop/server processor, and much simpler.
Are you saying bulldozer is "designed for high single-core performance"? Because that's exactly where it's failing to deliver, and what will continue to lead to market share.
It delivers very high single-core performance compared to smartphone processors, which was the subject under discussion. How it stacks up against other desktop/server processors is important within that market, but a distinguishing feature of desktop and server processors is that they all have high single-core performance.
It's very cool -- when I was working with microcontrollers, I would have loved to have eight independent cores -- but it's definitely meant more for embedded work, like industrial robotics, than for smartphones.
How many build servers are actually constrained by CPU and not IO?
If your build times are in excess of a few minutes it's probably better to distribute the build across many machines rather than trying to build one server to rule them all.
As for benchmarks there are extensive reviews online of the bulldozer branded as "Opteron".
In my experience (game development, hundreds of thousands of LOCs of C++), compiling is purely CPU limited if your HDD isn't utterly braindead (e.g. at or above normal desktop 7200 rpm levels).
AMD has decided to step back from the brink of maximum-speed per core, to maximum-cores-per-chip.
If you look at the Anand benchmarks: http://www.anandtech.com/show/4955/the-bulldozer-review-amd-...
you'll see in very multi-threaded applications (like compression or encoding) the AMD CPU actually keeps pace with the 2600k; in other applications where it is more important to have faster single threads of execution, the Intel 2600k pulls ahead by a drastic amount though.
These are the early days, but if AMD gets this down to 22nm next year and puts 16 cores on a desktop PC -- each core may be only running at 4Ghz, but in a scientific computing or server environment where you have 16, 4Ghz cores as opposed to 4, 6Ghz cores on some equivalent Intel can make a big difference.
The new Bulldozer platform looks to be a strategic company gamble on all fronts (more cores, less speed per core) that could win AMD big in the server space (which is exploding) and in scientific computing environments.
I'm not saying the numbers aren't disappointing, I was hoping to see some big ones in there to. I'm only pointing out that this isn't quite apples to apples, and when it is (e.g. straight multithreaded benchmarking) the comparisons look OK.
All that said, I'm sure Intel has something waiting in the wings (e.g. 2800k) to shit all over this party and beyond that we have tri-gate IvyBridge CPUs next year that will hopefully bring a much-needed* leap to CPU performance.
* "needed" as defined by "my impatient ass likes fast things.