AMD FX-4100 Bulldozer

[PsynoKhi0]

It's much older than that.

486SX was just a 486DX with the FPU disabled.

Meh you got me there... I should have kept the examples to disabled cores

[pingufunkybeat]

It's a trick that's been used for a while. I think the first Athlon II were actually Phenom II with L3 cache cut off.

It's much older than that.

486SX was just a 486DX with the FPU disabled.

[PsynoKhi0]

Are you sure the FX-4000 series will be an 8Core/4Module CPU with 2 modules turned off?

It most probably is, at least for the first batch.
From what I've read production costs for one die per series would be higher (especially with the first production runs, if you factor in that yield improvements would have to be duplicated across all dies), plus they can sell chips that don't meet the QA requirements for the top-of-the-line instead of just binning them.
It's a trick that's been used for a while. I think the first Athlon II were actually Phenom II with L3 cache cut off. Phenom II x2 could also have extra cores unlocked. Provided the motherboard allows it, it might still be possible with BD.
There's no guarantee though, and there might be some trade-offs (higher vcore for stable operation, lower overclockability etc.)

@blackshard: power draw from the FX-4100 is better than Deneb. Ditto for FX-6100 vs Thuban Link

[Medallish]

It matters if you're interested in technology and its development.
It doesn't matter if you're chit-chatting.

I'm not chit-chatting, are you?

errata corrige: those are two *billion* transistors, and actually it is the number of transistor of the FX 8000 series (8 integer ALUs, 4 FPUS, AMD sells it as an 8-core processor), but since FX-4000 series is the same processor as FX-8000 with some non-functional units, they share the same 2 billion transistor design.
The fact it has 2 billion transistors matters even when you're talking about *power*. Actually FX 8000 series is really awful, I expect that FX 4000 series is almost as awful as its bigger brother.

Are you sure the FX-4000 series will be an 8Core/4Module CPU with 2 modules turned off? It would seem strange to go through the trouble of making the whole Module design and then not use it.. At the same time I guess considering it has the full 8Mb L3 cache which from what I saw was split up in 4 parts on the die on the 8Core. Some of the Idle numbers show that the new CPU isn't too bad, but definitely load, and OC on the 8150, that was just insane, there's a hopefully room for improvement considering this is an entirely new arch. but boy does it need it as well :P.

[blackshard]

it dosn't matter how much transistors it is in FACT a dual-core.,

It matters if you're interested in technology and its development.
It doesn't matter if you're chit-chatting.

I'm not chit-chatting, are you?

errata corrige: those are two *billion* transistors, and actually it is the number of transistor of the FX 8000 series (8 integer ALUs, 4 FPUS, AMD sells it as an 8-core processor), but since FX-4000 series is the same processor as FX-8000 with some non-functional units, they share the same 2 billion transistor design.
The fact it has 2 billion transistors matters even when you're talking about *power*. Actually FX 8000 series is really awful, I expect that FX 4000 series is almost as awful as its bigger brother.

[PsynoKhi0]

I wondered if this image below is from the Anandtech review of the processor helps clear up some of the confusion with how the processor can be optimised. The issue seems to be getting 'turbo core' enabled and directing threads with shared data to the same core.

Is the situation the same under Linux?
Coupled with this it should interesting. Reassigning thread - core priorities?

[WorBlux]

it dosn't matter how much transistors it is in FACT a dual-core.,

Sound more like a dual-core-plus-dual-half-cores. Basically they added more of the circuits that are most commonly used, especially under a server type load, but probably not so good for a lot of multimedia competing for the same FP unit.

[Michael]

Anyone happen to know if there's any newer kernel patches than http://thread.gmane.org/gmane.linux..../focus=1170744 for Bulldozer? Google hasn't turned up any others, just checking.

[pingufunkybeat]

Anyway, it looks comparable to an i3 in terms of price and performance.

I'd like to know more about power consumption -- how do all these extra transistors affect it?

If power consumption is comparable, then this particular processor is OK. Even if there are doubts is the architecture can compare with intel's high-end offerings in the near to mid future.

[blackshard]

this cpu will speed up by 3% for the cache kernel patch and 10% for the scheduler kernel patch and maybe 20% for the compiler patches.
also as i know amd will upgrade the micro code in the bios to fix some speed bugs.

but yes technically this cpu is a dual-core with some extra Integer help units.

If some one need a true quatcore with greater speed from amd he can get the Opteron 6204.
but yes 400? isn't cheap.

A dual core with 2 Millions transistors? Quite huge for being a dual core...