Dear forum members,
We are delighted to inform you that our forums are back online. Every single topic and post are now on their places and everything works just as before, only better. Welcome back!
Discussion on Article:
AMD’s Power-Efficient Eight-Core Desktop Chip Due This Week.
CMT vs. SMT has TWO or MORE parallel pipelines, while SMT try to fill inefficiency gaps in ONLY ONE PIPELINE. SO with everything we used to call core current Bulldozer module is really "dual-core" and FX-8100/FX-8300 series features FOUR of these modules totalling up to eight-cores.
Only BS here is lazy M$ developement and refurbishing of an old failure Vista product with a new polish wax and it's third incarnation is now called Windows 8.
I do not rate a comments I'm replying to.
So here's the simplest explanation of what's going on. CPUs only really do one thing: add. They add 1s and 0s VERY, VERY, VERY quickly. Every mathematical operation ever performed in CPU (addition, subtraction, multiplication and division) is performed by adding. OK, got your mind around that? Great. Next, CPUs can add integer numbers (those without decimals) and floating point numbers (those with numbers after the decimal point). If you ask a CPU to add integers, the 'integer' unit(s) do it, and if you ask a CPU to add floating-point numbers, several possible things can happen. In the past (like 15 years ago), a floating-point calculation would be done by the x87 floating point unit inside the CPU. However, CPU engineers came up with all kinds of different ways to speed up floating-point calculations, depending on what kind of sequence of calculations you were wanting to perform. AMD came up with 3DNow! and Intel came up with SSE to supplement the old x87 unit. Nowadays, programs can be designed to package floating-point calculations as SSE/AVX/FMA or even GPU-coded (Direct Compute, PhysX) operations, completely bypassing the x87 unit. At first, this was done for games, heavy math and scientific computing programmes, but now, with tools like OpenCL, etc. just about any floating-point can be directed to the fastest hardware available on a given computer. The programme detects OpenCL-compatible hardware, and installs the OpenCL version of the code. In short, only older programmes rely solely on the legacy x87 unit, and that's why AMD decided to cut down the number of x87 units to 4 in an 8 integer core CPU so they could add more integer cores - cores that are still the best way to add integers! Any decent programmer worth his salt writing fresh code today that's performance-sensitive will be coding the floating-point to execute as AVX/FMA or GPU-accelerated code, unless they absolutely can't (very rare). Period.
I know it's hard because AMD's module design doesn't cleanly fall into a single or dual core classic case design scenario, but really, we've got to learn to accept a little bit of complexity and stop trying to over-simplify everything, even when it means we end up being highly inaccurate. Simple bottom line, AMD's 8 integer core CPUs, based on contemporary, multi-core optimzed code, is much more like an 8 core CPU than a 4 core CPU, most of the time, in most use cases. End of story.
Did you ever seen a two or more processes running at the same time on single core CPU?
And multiple core of CPUs still share caches and/or memory and bus, even if they are not hyperthreaded and even if they are inserted in separate sockets.
Otherwise they would be just a separate computers.
I do not rate a comments I'm replying to.
Take your cherry picking and be on your way KonradK, you lost this argument.
If its 50/50 single and multi threaded the FX-6300 is really competing with the low range i5 and the FX-8350 is edging out the top i5 models and the FX-4300 is trading blows with i3 (as seen f.ex. in the review by techreport).
Heavy emphasis on multithreading plays to the advantage of the Piledriver and in speciffic cases it can even match the all powerful i7.
Of course the power consumption remains a thorn in it's side but let's not compound it vby destorting reality.
Just imagine if GTX680 was made on 40nm, how much worse it would have been. The problem is AMD has no $ to build its own fabs. Its node tech is tied to the pace of 3rd party fabs like TSMC or Global Foundries.
The 95W reduction seems to be coming from low base clocks. I don't think this chip will consume less power than FX8350 at full load when all the cores are loaded / Turboed. Sounds like this is catering the OEM market's desire for a 95W model.
TAViX hasn't posted anything that goes to the levels of AvonX so far I can see. He hasn't derailed a thread like the other two used to do. He can dislike AMD's products, that is freedom of choice.
We are not a tyrannical site are we?
Rest assured anyone or anyspammer who hijacks the thread I will clean them out. I check all the threads as to best of my abilities on a daily basis.
Nobody who still is with a Phenom II will buy this thing since its a downgrade and won't perform at the levels of the fx-8350.
This is just a down clocked version to fit the 95W TDP power envelope.
The rest of them has already switched to Intel.
Add your Comment
Enter your username and e-mail address. Password will be sent to you.