• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FRONTPAGE Rumor: AMD FX-8350 -Piledriver- to Launch in Q3

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
I really hope AMD comes back into the competition against Intel. I am not trying to be a fan boy here, but if you think about it, the competition to be beneficial to computing technologies.

Intel hasn't changed up their design scheme much in a while after beating the AMD phenoms, and now it just looks like they are costing on their designs, making some very minuscule tweaks and selling it. The 3D transistors is amazing, though I have a feeling they are just going to cost on that till AMD gets back in the game.
 
Way to ignore the specific mention of an 8150 lower in that post.

Most C3 PhIIs will do 4GHz happily on a decent air cooler, my 955 does something silly like 4.9GHz for single threaded stuff. That's rare of course.
I've seen a fair number of PhIIs in the 4.2 to 4.4 range. BD does clock higher, no argument there. It also produces a lot more heat in my experience, enough to require a hell of a cooler and heat the room up nicely.

Troll wise, I recommend the internet definition, seeing as that is how you were using the word: http://en.wikipedia.org/wiki/Troll_(Internet)

after i go to push and pull again i should be easily go to 4.2ghz maybe 4.3 or 4.4 because my temps right now are great!

and you are right about the trolling definition
if PD is like BD im sticking with my 1100T because why pay $300 for a downgrade? Only time will tell if PD will come back to AMD CPU King! (king right now is 1100T/1090T)
 
after i go to push and pull again i should be easily go to 4.2ghz maybe 4.3 or 4.4 because my temps right now are great!

and you are right about the trolling definition
if PD is like BD im sticking with my 1100T because why pay $300 for a downgrade? Only time will tell if PD will come back to AMD CPU King! (king right now is 1100T/1090T)

Typical. And of course you're ignoring that with that same cooling and OC tweaking someone with an 8120/8150 can OC their CPU to 4.8, 4.9, 5.0 GHz and still pass a 24 hr prime95 run.
 
Ya but like bob said it will super heat the room it's in........ And well my computers in my bed room and my 2 gpus already heat my room pretty well adding a super hot CPU will not help things. And bd improved somethings and lost where in counts like single threaded apps in gets bested by a older CPU by the same company
 
If the resonent clock mesh thing in PD really does decrease power consumption as much as it says it will, it really does do 5% better clock/clock, and it still OCs the same (but with lower heat output), that'll help my general opinion a lot. Now ideally it'd be more than 5% better and have the lower power consumption of course, but you have to start somewhere.
 
If the resonent clock mesh thing in PD really does decrease power consumption as much as it says it will, it really does do 5% better clock/clock, and it still OCs the same (but with lower heat output), that'll help my general opinion a lot. Now ideally it'd be more than 5% better and have the lower power consumption of course, but you have to start somewhere.

From the early trinity reviews it's looking like more of a 10-15% increase clock for clock. You have to remember that the trinity results are not what we'll see with the actual PD FX chips as the trinity samples reviewed don't have L3 cache.

Ya but like bob said it will super heat the room it's in........ And well my computers in my bed room and my 2 gpus already heat my room pretty well adding a super hot CPU will not help things. And bd improved somethings and lost where in counts like single threaded apps in gets bested by a older CPU by the same company

Unless you're running prime95 the amount of heat it outputs during "normal" usage is not that much at all. I run my 6100 in a 4.7 daily overclock and the amount of heat it ouputs (unless like I said I'm stressing 100%) is negligable. My HD 5780 on the other hand is an entirely different matter. And I'd really love to see how much "heat" the 1090T/1100T output when pushed to 4.2-4.3GHz as I'm sure it's not a "small" amount.
 
Last edited:
To be quite frank, i would need 4.8Ghz on an 8150 to match mine at 4Ghz / 4.2Ghz.

Some how i doubt i would get that on my £30 Air Cooler, sadly.

Awaiting Piledriver......
 
What I'd like to see is AMD Piledriver with Linux Ubantu VS Sandybridge with Windows for benchmarks, why do all the benchmarks have to be for windows?, maybe piledriver can move ahead of Sandybridge in Linux benchmarks. Come on Linux is making it's gains and getting more stable. and Both of these on a Crosshair V MOBO...Let the games begin...Time to let Linux run...
 
There are plenty of Linux benchmarks, but they're different benchmarks. As such they aren't really comparable to windows benchmarks.

It'd be quite easy to run some Linux benchmarks on both chips though.
 
I think the point he's trying to make is that when you look at the linux benchmark comparisons the disparity between the chips isn't the same as it is when reviewing windows benchmark results. That's what happens when you have open source benchmarks.
 
Whoa, cant say I thought that was his point...If he means that, he sure as heck didnt write it... or at least I misunderstood as well! :p

Regardless, +1 to apples to apples so one get the most accurate results with the two chips with the SAME benchmarks in the SAME OS. I am curious to see how things shake out on the Linux side as the distro's have taken a much stronger hold recently on a more diverse set of users. :)
 
Last edited:
EXCELLENT links. Thank you! :ty:

Doesnt look like the story changes much assuming I read the first link right. But its hard to compare since we are all windows junkies and they use different applications/testing. Seems consistent however with the story windows based applications tell. :thup:
 

It just shows the same thing Windows does, Very good multi threaded integer performance (8 integer cores)

And poor floating point performance, both single and multi threaded, the 8 core Bulldozer only has 4 floating points, which as individuals really aren't much good.

There is nothing wrong with Bulldozers integer performance, its actually pretty good by any measure. AMD 'just' seriously need to work on FP.

I am a fan of AMD, i like there products for many reasons but Bulldozers FP performance, i'm sorry... in my opinion its pathetic :(
 
One more, a bit better comparison of compilers to the full review:

http://www.phoronix.com/scan.php?page=article&item=amd_bulldozer_compilers&num=1

And yes, pretty much the same story. The new compiler optimizations help BD gain some ground but not enough to compete against a top of the line Sandy. There are a couple of cases where the new compiler's make a huge difference and BD is a monster performer, but those cases are not common.
 
Possibly, but it doesn't involve the intel compiler. The compilers here are open source (not sure about LLVM though as it's Apple sponsored) so I highly doubt anything Bubba is looking into would apply here rather than just different compiler efficiency in general.
 
Back