Thursday, September 1st 2011

Gigabyte Website Lists out FX Series Processors in Support Lists

The CPU support list in the product page of Gigabyte's top of the line socket AM3+ motherboard, the GA-990FXA-UD7, spilled out details of upcoming AMD FX series processors, days ahead of actual product launches. Tables listing out specs of FX processors aren't new, but in older occasions, they were posted by the media citing sources. This table comes from a leading motherboard manufacturer. In this case, the 990FXA-UD7 will support FX series processors from BIOS version F4.

The table lists out FX 8000 series 8-core processors, and one each of FX 6000 and FX 4000 series 6-core and 4-core chips. A new detail emerging with this table is the system interface speed, which has been bumped all the way up to 5200 MT/s, up from 4000 MT/s of the previous generation. The faster 8-core chips have TDP rated at 125W, while every other FX series chip is rated at 95W. FX 8000 series chips include the 3.60 GHz FX-8150, the 3.10 GHZ FX-8120, 2.80 GHz FX-8100; FX 6000 series includes the 3.30 GHz FX-6100; while the quad-core FX 4000 series includes the 3.60 GHz FX-4100. The FX series is expected to launch in this month.
Source: VR-Zone
Add your own comment

40 Comments on Gigabyte Website Lists out FX Series Processors in Support Lists

#26
TheLaughingMan
bear jesusIf bulldozer can't beat the core 2 quads then it would make perfect sense that the delay was AMD trying to clock them higher and explains the idea of adding in water coolers as they had to clock so high the stock cooler would always fail... i could see it happening :laugh:

Honestly i expect bulldozer to just make me give my money to Intel but i agree with the proof, only reviews from respected sites will allow me to part with my money no matter who the money is going to.
Current Phenom II chips out perform almost every core 2 quad so that is not a hard goal to reach. I never understood this "There is only two places, first and complete fail." AMD is not the fastest, but their lineup is more than adequate. But whatever.

Water was to appeal to the enthusiast market and was limited to 1 sub set of 1 chip.

I expect a lot from Bulldozer, but I am prepared for failure as well. Not because it is AMD, but because I look at every product that way.
Posted on Reply
#27
shrivan
I expect exactly (as from every product) "in advance" nothing.

Even when tests are out, and you all know that, those could/can be made to represent what you want it to say. In essence, Llano reviews rarely grasped what the APU as opposed to the "just" CPU was supposed to achieve (remember the die-size allocation between current Llanos and Sandy Bridge?).

Thus why bother commenting so much? Why the hell would I expect BD to beat Intel hands-down? The only thing I honestly could care about (and that is each and every time different) is price/performance vs. true potential used. Or how would you explain "stone-age" performance king Intel Atom? It's mobile. True. But performance? But hey, it is indeed (for some) enough...

So tell me, could 99% of your typical (what is that?) tasks not be done by a 1100T? Or even a dual-core?

BD --> when he/it finally shows we can judge! 'nuff said.
Posted on Reply
#28
bear jesus
TheLaughingManCurrent Phenom II chips out perform almost every core 2 quad so that is not a hard goal to reach. I never understood this "There is only two places, first and complete fail." AMD is not the fastest, but their lineup is more than adequate. But whatever.

Water was to appeal to the enthusiast market and was limited to 1 sub set of 1 chip.

I expect a lot from Bulldozer, but I am prepared for failure as well. Not because it is AMD, but because I look at every product that way.
Well you kind of pointed it out there, if the next gen could not do what the current gen pretty much does already would you not consider that a complete failure? i sure would.

I would love to give AMD my money but that does not stop me joking about the company's upcoming products and part of why i expect bulldozer to make me want to buy Intel is because i don't expect a massive improvement over the current generation but if there is some real power for money in the new chips it will be a nice surprise that is more likely to get me to give in to AMD again.

But as you said it's not just about AMD until i see the product i don't expect much from it no matter who it is, but i do expect sandy bridge E to make me want to buy AMD likely due to total cost:laugh: really this is just about passing time until release day and reviews.
Posted on Reply
#29
YautjaLord
3.6GHz FX-8150? If it goes to 4.6GHz without any overvoltage - ace; hope they'll (AMD) cope up with September 19 release date. Wiz or bta - one of you have to OC it (FX-8150) once you get your hands on this CPU. That & will you test it on M5A99X or bring on the Sabertooth 990FX? Thanx for news. :toast:
Posted on Reply
#30
eidairaman1
The Exiled Airman
YautjaLord3.6GHz FX-8150? If it goes to 4.6GHz without any overvoltage - ace; hope they'll (AMD) cope up with September 19 release date. Wiz or bta - one of you have to OC it (FX-8150) once you get your hands on this CPU. That & will you test it on M5A99X or bring on the Sabertooth 990FX? Thanx for news. :toast:
I dont think the initial stepping will till the second stepping appears.
Posted on Reply
#31
TheLaughingMan
eidairaman1I dont think the initial stepping will till the second stepping appears.
Aren't the production models B2 so they are the second stepping.
Posted on Reply
#32
eidairaman1
The Exiled Airman
TheLaughingManAren't the production models B2 so they are the second stepping.
Initial Steppings can be named anything they want it to be.
Posted on Reply
#33
FordGT90Concept
"I go fast!1!11!1!"
1Kurgan1Interesting that 2 of them share same frequency and different wattages (different turbo core clocks?). Also interesting that there is a 500mhz gap between top dog and 2nd rung.
One is octo-core, the other is a quad-core. Very impressive that they can squeeze 3.6 GHz out of an octo-core without breaking 130w TDP.
Posted on Reply
#34
[H]@RD5TUFF
FordGT90ConceptOne is octo-core, the other is a quad-core. Very impressive that they can squeeze 3.6 GHz out of an octo-core without breaking 130w TDP.
While I agree, performance maters the most!
Posted on Reply
#35
pantherx12
FordGT90ConceptOne is octo-core, the other is a quad-core. Very impressive that they can squeeze 3.6 GHz out of an octo-core without breaking 130w TDP.
Have to bare in mind that the cores arnt cores how we know them now though.

They're only double up like on 25% of the parts for 80% of the performance.

AMD is going down the packing as many cores as possible route ( which seems to make sense based on where software is heading)

If people think this is going to compete core to core with intels latest offerings I don't think they've been looking at the tech properly.



Amd will be packing more cores on the same size die as intel with BD ( if I understand it correctly) this will mean ( in theory) less heat, less energy, and cost savings!

Should also mean their will be more usable parts from a tray as it's more likely only an individual " core" will be affected (without taking out the entire module with luck)



If I have been reading into bulldozer and understanding it correctly what AMD are doing is VERY clever.

How ever IMO calling them cores was the wrong decision as everyone's going to think they're rubbish when Intel still beats them at IPC per core.
Posted on Reply
#36
eidairaman1
The Exiled Airman
FordGT90ConceptOne is octo-core, the other is a quad-core. Very impressive that they can squeeze 3.6 GHz out of an octo-core without breaking 130w TDP.
die shrinks, AMD is goin to approach of running cool n energy efficient, not lightning fast
Posted on Reply
#37
NAVI_Z
the suspense is killing me!!.....been rokin an amd 965 b.e. for tha longest!! bulldozer is mad

late to tha party and now we hear bout "piledriver"???? common AMD, ur loosin people

fast!,....lol.
Posted on Reply
#38
TheLaughingMan
FordGT90ConceptOne is octo-core, the other is a quad-core. Very impressive that they can squeeze 3.6 GHz out of an octo-core without breaking 130w TDP.
No. He is talking about the FX-8120 being listed twice.
Posted on Reply
#39
YautjaLord
eidairaman1I dont think the initial stepping will till the second stepping appears.
B0 is Engineering Sample (sorry for bringing up this FUD :)); B1 - step up, but still with limitations; B2 - supposedly above marginal step up in OC'ing & general performance (my pick i guess, but time will tell - i wanna see actual benchies myself to beleive that); C0 - dream, lol. Hope B2 is good stepping; i wanna see how high it'll Turbo/manually OC once it's out & once i'll buy it. Hope it'll do 500 - 600MHz OC without overvoltage on VenomousX with 2 high-perf 120mm PWM fans; my current 965BE OC wears out on me, i wanna OC the f*** out of FX-8150 already. :toast:
Posted on Reply
#40
FordGT90Concept
"I go fast!1!11!1!"
pantherx12Have to bare in mind that the cores arnt cores how we know them now though.

They're only double up like on 25% of the parts for 80% of the performance.

AMD is going down the packing as many cores as possible route ( which seems to make sense based on where software is heading)

If people think this is going to compete core to core with intels latest offerings I don't think they've been looking at the tech properly.



Amd will be packing more cores on the same size die as intel with BD ( if I understand it correctly) this will mean ( in theory) less heat, less energy, and cost savings!

Should also mean their will be more usable parts from a tray as it's more likely only an individual " core" will be affected (without taking out the entire module with luck)



If I have been reading into bulldozer and understanding it correctly what AMD are doing is VERY clever.

How ever IMO calling them cores was the wrong decision as everyone's going to think they're rubbish when Intel still beats them at IPC per core.
From what I gather, AMD's cores are cores. What you might be mistaking it with is "modules." 8-core processors consist of 4 modules. What makes a module unique is that they share resources, namely ALUs. This is not symmetrical multithreading (SMT) like Intel uses on Nehalem and Sandybridge or IBM uses on POWER 7. SMT is all about throwing as much work at a single physical core to suck every useful clock out of it whereas modules are designed to boost single-threaded performance. SMT should be better performance-per-clock wise but inferior performance-per-thread wise where modules are the opposite.

AMD's processors take up less die space because of their exclusive memory controller. Most of the die realestate is taken up by various caches and, because of that advanced memory controller AMD has had forever (K5 I think, maybe K6), they can do more with less. It allows them to cram more cores in the same, or less, space. The problem is, Intel's massive caches are less likely to miss so, despite being more expensive to produce and generally running hotter, they tend to have better caching performance (less likely to miss and seek out the RAM).
TheLaughingManNo. He is talking about the FX-8120 being listed twice.
Oh, in which event, Gigabyte maybe didn't know which wattage it would be released as so they tested and assumed both so they listed both. The final product will probably only have the specs AMD finalized.
Posted on Reply
Add your own comment
Nov 22nd, 2024 07:31 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts