New
#80
965 Black still out performs BD
The biggest advantage to this new CPU is that a lot of people won't have to upgrade their old MB to use it. So far, it doesn't look like a "Oh, wow - time to build a new computer" type of deal.
The second-to-last computer I built was on a Socket 939 board with regular DDR RAM and an AGP slot. I couldn't find any upgrade CPUs worth going for even a year later, couldn't upgrade the RAM, and we know how much everybody wanted to make decent AGP video cards after PCIe became available...
My latest build is an X58 system. Unless I suddenly find $1,000 in my mattress for the Core i7 990 there are no worthwhile processor upgrades in my future without a MB upgrade. By the time I'm ready to do that I'll probably be needing DDR4 RAM and a PCIe-II video card to go in it.
I'll give AMD credit for learning their lesson with the Socket 939 debacle and at least making a chip which won't require everybody who is already in the AMD camp to buy a whole new system to use it. If I were already running a compatible board I'd be tempted to toss a Bulldozer in it.
I'd think long and hard before making that decision. A few more processing cycles for a steep learning curve and very few games.
AMD wasn't saying expect a chip that doesn't perform that great. The AMD fans weren't saying when Bulldozer comes out, it's going to be just "meh" for a couple of years before it really gets any traction. This is not what anybody was expecting. I think it's simply hoping for the best from AMD fans at this point. Otherwise, I think AMD would be pumping out lots of videos and statements to support their "design" over performance at the present time.
I've created the best vehicle in the world. It's completely new and vastly superior. It's just a pity I didn't design a drivers seat for it. It really hurts performance not being able to control it properly.
Not my problem though, it's up to someone else to make my mistake work. And when they do... it will be awesome. Maybe.
Smartyeyeball,
You're on to something but misguided with your analogy. Take the use of automatic sequential shifters in all forms of auto racing. At first they were horrible but held great promise. Only by the advent of paddle shifters, software refinement and electronics to take full use of all the dynamics involved and tens of thousands of man hours do we have a system that bests the manual transmission for control by a large margin. I see this AMD chip's topography in a very similar light and I do believe it will get better utilization and exploitation of it's performance envelope in very sort order. Again these Intel versus AMD discussions always take on a unique life of their own. There are going to be very few Intel users willing to switch and one could argue that some AMD users who were going to jump on board will get wrapped up in the debate and shy away or wait. But I think as we speak extremely smart minds are working to exploit this designs full potential and I for one think we all enjoy better computing by the existence of these two competitive chip makers.
Paddle shifting were on race cars, tested tweaked and finally when they were working the way they should were put on cars for the street. In four years AMD did not do their testing and tweaking. They just put it on the street so hopefully some one else will figure it out.
The Idea Behind AMD
"
But this does force us to address the relationship between AMD’s hardware and the software that’ll invariably run on it. In Intel Core i5 And Core i7: Intel’s Mainstream Magnum Opus, I brought up specific optimizations in Windows 7 that were the product of collaboration between Intel and Microsoft—notably, core parking. Windows 7 intelligently schedules to physical cores before utilizing logical (Hyper-Threaded) cores.
In theory, AMD could benefit from the same thing. If Windows were able to utilize an FX-8150’s four modules first, and then backfill each module’s second core, it’d maximize performance with up to four threads running concurrently. This isn’t the case, though. According to Arun Kishan, software design engineer at Microsoft, each module is currently detected as two cores that are scheduled equally. So, in a dual-threaded application, you might see one active module and three idle modules—great for optimizing power, but theoretically less ideal from a performance standpoint. This also plays havoc with AMD’s claim that, when only one thread is active, it has full access to shared resources. Adding just one additional thread could tie up those shared resources, even as multiple other modules sit idle.
Microsoft is looking to change that behavior moving forward, though. Arun says that the dual-core modules have performance characteristics more similar to SMT than physical cores, so the company is looking to detect and treat them the same as Hyper-Threading in the future. The implications there would be significant. Performance would unquestionably improve, while AMD’s efforts to spin down idle modules would be made less effective."