With comparing all the Futuremark benchmarks it's really sad the BD@4.6 only out did my 1090t@4.2 one time CPU wise and that was on Vantage CPU score which it did beat me by 1,700 points. Only one total test was higher and thats 3Dmark11, and that was by 77 points and with a GTX 570, not a GTX 580 that Guru of 3D was useing. I just wanted this CPU for gameing and thats were the problem is. I have work PC's so dont care to have just another work CPU for $270 :bang:.
Not at all comparable. It's 8 half-assed cores. The only good thing that will come out of all this is that that Microsoft might actually improve their stupid scheduler... in Windows 8. I wouldn't be surprised if they don't update Windows 7's. It's pretty sad to think no matter how much AMD's top CPU is OC'd on water it wouldn't match an air cooled CPU I bought long ago... which is almost 2 generations old.
Perhaps this is why: http://i.imgur.com/H8e92.jpg Anyways, i hoped that all the "leaks" were false info, but damn. Too bad, i was hoping that BD would suit me, but it wont. Surely 8 cores would be cool, but i simply don´t have usage for more than 4 cores. And the perf per core is just too lame, when it loses against the previous gen. I´ll say this too, too bad for consumers. Now Intel could rise the price of i5 and i7, since Amd just failed badly. Time to visit my pc store, and talk about SB prices...
What i don't understand is how can AMD release a new cpu, that's developed for 5-6 years and should beat the older AMD cpu's in any application : from games to memory bandwidth. How can you get beaten in games by your own old tech!!! it's pure retarded. This ended up to be a very BIG disappointment. I guess i'll go Intel after all.
A few reasons. 1. Bulldozer is designed for multi-threaded applications. 2. Each module shares resources between cores. 3. Windows still has rather poor thread management AMD claims that "optimal" thread assignment is to assign threads to cores in order. Given that the "cores" in each module share resources, it'd be logical to assign 1 thread to each module, only populating 1 core per module and only doubling up if necessary. This way, Turbo Core can function more effectively and the resources are used more efficiently. Running 2 threads on the same module can result in a lack of sufficient resources, which I'm assuming is why some games show such poor (read as: laughable) performance compared to even the PhenomII.
Well for one they are adding two new cores to new tech and trying to stay compatible with old tech. That runs in to a lot of snags such as the shared core issue and using turbo core tech. Windows does not support this yet so we see a stint in the benchmarks. It was pointed out earlier but amd has put all their money into future technology via windows8 and whatnot. I guess they simply thought that since they have hit the proverbial limit with phenom II they needed a refresh.. which they did. We are not going to see massive improvements in this technology until software catches up. Is this bad for amd? not necessarily. Since i do have a x6 which was considerably cheaper than the 2600k at the time I am almost forced to either spend 500 bucks to change over to intel or 200 to catch up to intel even if just a bit. 200 sounds a lot better to me. But my money will be on Piledriver not BD. Its like the first wave of FERMI. They were crap, power hungry and not such big increase over AMD. But the now smaller die size has made FERMI a good competitor in the market. So time will tell what k10 will do.
Not sure if this has been posted, but this may be the only positive review of the fx-8150 out there. The difference is that they did not use the Crosshair mobo sent out to reviewers, but an Asrock 990fx extreme. Surprising results, esp in gaming: http://www.*******************/revi...sor-vs-core-i7-2600k-review-introduction.html Wooops.. censored!
That site seemed to be handpicking results too. Just so happens to be using the settings that the AMD slides were using and those exact games :| Misleading review for sure...
true. but there are people talking on a couple sites i frequent about how there may be something wrong with the bios that was used for the CH V. you notice how all of the fx-8150 reviews are done on the on that board. with the expection of a handful done on a few other boards. its something to think about.
Am I the only one who's still absolutely blown away that technology they took years to develop is seemingly inferior to their old technology? It's absolutely bat-crap insane, it makes no sense at all. Merely adding another 2 cores to what they had developed an eternity ago would stomp the top BD model. It's so crazy it seems like it's all a giant conspiracy/troll job to ruin AMD and in turn slow technological development significantly by leaving only 1 company standing. What. The. ****.
Remember, the guy responsible for this project (Dirk Meyer) stepped down conveniently just a few months right before the release...
Well maybe AMD will go long on a shotgun play. Try some experimental tech and come back strong in a few years. Regardless, They're a solid 3 years behind Intel right now. Does bulldozer even beat a Q9650?
small SLI test with AMD FX815 and Intel 2600K for comparison http://pctuning.tyden.cz/hardware/p...procesor-fx-8150-v-patnacti-hrach-2-2?start=7
IB is going tear bulldozer a new hole next year Finfet technology much lower power consumption than SB AMD might use finfet like 2016 if your lucky, they have power problems now on 32nm leakage will only get worst as they shrink hence why Intel making switch now to finfet imagine using same process with massive leakage for another 4 to 5 years