Monday, August 27th 2012

AMD "Vishera" FX-Series CPU Specifications Confirmed

A leaked AMD document for retail partners spelled out specifications of the first three FX "Vishera" processors by AMD. The new CPUs incorporate AMD's "Piledriver" architecture, and much like the first-generation "Zambezi" chips, will launch as one each of eight-core, six-core, and four-core chips. The eight-core FX-8350 is confirmed to ship with 4.00 GHz nominal clock speed, with 4.20 GHz TurboCore speed. The six-core FX-6300 ships with 3.50 GHz nominal, and 4.10 GHz TurboCore speed. The quad-core FX-4320, on the other hand, ships with the same clock speeds as the FX-8350. In addition, the document confirmed clock speeds of several socket FM2 A-series APUs, such as the A10-5700 and the A8-5500.
Source: Expreview
Add your own comment

493 Comments on AMD "Vishera" FX-Series CPU Specifications Confirmed

#301
cadaveca
My name is Dave
FrickSee cadaveca's post above with their own slides saying BD was handcrafted.
Their own slides, shown by Mark Papermaster, when talking about STEAMROLLER.

Piledriver will NOT see the design library enhancement. Piledriver will not really offer all that much more over Bulldozer. I expect maybe 15%, and most of that due to clockspeed increases.
Posted on Reply
#302
Frick
Fishfaced Nincompoop
cadavecaTheir own slides, shown by Mark Papermaster, when talking about STEAMROLLER.

Piledriver will NOT see the design library enhancement.
Oh missed that. Still, BD handmade.
Posted on Reply
#303
AvonX
I don't think BD was hand made.
They are just saying this to have an excuse to continue using automated tools.
Personally i think its BS that BD was hand made.
Posted on Reply
#304
Frick
Fishfaced Nincompoop
AvonXI don't think BD was hand made.
They are just saying this to have an excuse to continue using automated tools.
Personally i think its BS that BD was hand made.
Yeah well it's hard to prove anything in this matter.
Posted on Reply
#305
cadaveca
My name is Dave
AvonXI don't think BD was hand made.
They are just saying this to have an excuse to continue using automated tools.
Personally i think its BS that BD was hand made.
Hard to prove either way, but given the process used, and the time given to use it, it's more than likely that a large part of the chip WAS hand-made, since the company producing the tools, AMD's foundry partner, would have largely been using AMD tech to build those tools.

WHatever the truth of it is, that's AMD's official stand on it, so what we think, and the truth, hardly matters. What matters is that AMD manages to capitalize on these chagnes, and turn them into either power savings, or extra perforamnce, so that they can stay relevant to the market as a whole.
Posted on Reply
#306
AvonX
cadavecaHard to prove either way, but given the process used, and the time given to use it, it's more than likely that a large part of the chip WAS hand-made, since the company producing the tools, AMD's foundry partner, would have largely been using AMD tech to build those tools.

WHatever the truth of it is, that's AMD's official stand on it, so what we think, and the truth, hardly matters. What matters is that AMD manages to capitalize on these chagnes, and turn them into either power savings, or extra perforamnce, so that they can stay relevant to the market as a whole.
I remember back then that they did not even get the correct transistor count of BD when asked. :laugh:
Yeah sure it was handmade.
Posted on Reply
#307
cadaveca
My name is Dave
AvonXI remember back then that they did not even get the correct transistor count of BD when asked. :laugh:
"They".


They fact remains, who was "they".


Liek I get that, but you know, the whoel Fermi thing, and why i brougth htat up, was that Jen-Hsun admitted that the rel problem with Fermi was a alck of proper leadership. Like, he actaully admitted he failed. HE got red in the face, but he did admit it. My respect for him increased over 9000.


:roll:


So, bad figures, eh? Someone got a number wrong? COuld have been a typo, could have been on purpose, could have been subterfuge to mis-lead the competition. Could have just been stupidity. Personally, that was all before Papermaster took over, so I don't care.
Posted on Reply
#308
Super XP
cadavecaTheir own slides, shown by Mark Papermaster, when talking about STEAMROLLER.

Piledriver will NOT see the design library enhancement. Piledriver will not really offer all that much more over Bulldozer. I expect maybe 15%, and most of that due to clockspeed increases.
Agreed Piledriver won't be physically modified but it was stated they tweaked the he'll out of it such as tightening and enhancing the L2, L3, Branch Prediction and so on, things requiring to get ironing out before they slap together Steamroller via the new 28nm process.

Personally I can see Piledriver being at least 10% faster clock for clock versus Bulldozer with another 5% faster due to higher clock speeds, hence we have our 15%.
Posted on Reply
#309
nt300
Any official work on the Vishera to be released?
Posted on Reply
#310
AvonX
nt300Any official work on the Vishera to be released?
Hope they will just skip it and release SteamRoller. :laugh:
A plus 35-45% would not hurt anyone. :P
I need to upgrade my 1090T, its been too long.
Posted on Reply
#311
Covert_Death
honestly is SteamRoller comes out 1H 2013 and supports AM3+, ill wait for it
Posted on Reply
#312
AvonX
Covert_Deathhonestly is SteamRoller comes out 1H 2013 and supports AM3+, ill wait for it
Forget about AM3+, go directly to AM4 with PCIE 3.0 and some DDR4 memory. :p
If they do that it will sell madly. First motherboard to support DDR4.
But that is just a wild dream. :laugh:
Posted on Reply
#313
librin.so.1
AvonXIf they do that it will sell madly. First motherboard to support DDR4.
Remind me please, what is the current availability for DDR4 memory modules, and what availability is expected to be at the start of 2013?
Posted on Reply
#314
AvonX
VinskaRemind me please, what is the current availability for DDR4 memory modules, and what availability is expected to be at the start of 2013?
I haven't checked lately what is going on with that, but i think whoever wishes to use it, its available. I think samsung has the lead on that.
Well of course they will use it on servers first. But why not? Lets see something different for a change.

Edit: tech2.in.com/news/pc-memory/ddr4-memory-coming-to-pcs-notebooks-in-2013/305022
Posted on Reply
#315
Super XP
AvonXForget about AM3+, go directly to AM4 with PCIE 3.0 and some DDR4 memory. :p
If they do that it will sell madly. First motherboard to support DDR4.
But that is just a wild dream. :laugh:
Steamroller on AM3+ would be awesome. DDR4 ain't happening until 2015, and if that. Like I said before Intel is not going DDR4, they are sticking to DDR3 for a few years more.

We don't need DDR4 for desktops right now, DDR3 does the job just fine.
Now Servers & Workstations are all different story. DDR4 will be coming out for those probably in 2014.
Benefits

Its primary benefits compared to DDR3 include a higher range of clock frequencies and data transfer rates (2133–4266 MT/s compared to DDR3's 800 and higher) and lower voltage (1.05–1.2*V for DDR4, compared to 1.2–1.5*V for DDR3) with current remaining the same. DDR4 also anticipates a change in topology. It discards the multiple DIMMs per channel approach in favor of a point-to-point topology where each channel in the memory controller is connected to a single DIMM. Switched memory banks are also an anticipated option for servers.
en.wikipedia.org/wiki/DDR4_SDRAM
Posted on Reply
#316
Steevo
That is not true, DDR4 (or even 5) would be a welcome performance boost for most platforms, go read the newest Corsair 2600 review to see.
Posted on Reply
#317
AvonX
Super XPSteamroller on AM3+ would be awesome. DDR4 ain't happening until 2015, and if that. Like I said before Intel is not going DDR4, they are sticking to DDR3 for a few years more.

We don't need DDR4 for desktops right now, DDR3 does the job just fine.
Now Servers & Workstations are all different story. DDR4 will be coming out for those probably in 2014.
Maybe not in intel's case but AMD could benefit from the speeds of DDR4 to feed the CPU.
Something related to that is this: vr-zone.com/articles/how-amd-can-steamroller-the-competition-in-2013/17105.html

I think DDR4 is coming sooner than you think it will. They are pushing hard to use this technology sooner.
Posted on Reply
#318
Super XP
SteevoThat is not true, DDR4 (or even 5) would be a welcome performance boost for most platforms, go read the newest Corsair 2600 review to see.
I fully agree, I like the DDR4 Specification. What I meant is this is going to the Server/Workstation market, them comes the desktop. I mean I can always just sell my Crosshair V (my gaming buddy :D), but I will have to be convinced going DDR4 is the right move in terms of price/performance and hassle of ripping my PC apart. Which yes I would EnJoY....
AvonXMaybe not in intel's case but AMD could benefit from the speeds of DDR4 to feed the CPU.
Something related to that is this: vr-zone.com/articles/how-amd-can-steamroller-the-competition-in-2013/17105.html

I think DDR4 is coming sooner than you think it will. They are pushing hard to use this technology sooner.
Wow, great find. I am starting to get exited. Anything to breath new life in AMD and for competition...

And here is a killer Quote. It's great he works for AMD...
Jim Keller (our sources from AMD/Intel/NVIDIA claim he's arguably the best CPU architect of all times)

Read more: vr-zone.com/articles/amd-pushes-steamroller-and-excavator-forward-bullish-about-performance-increases/17088.html#ixzz26CiHTA00
Posted on Reply
#320
Super XP
I am itching for a Piledriver upgrade providing it offers better gaming performance over what I have. :D
So reading the article, it looks like we are going to see Steamroller & Excavator sooner than we think. Piledriver may now be short lived, because all the news is about Steamroller.
Posted on Reply
#321
AvonX
Super XPI am itching for a Piledriver upgrade providing it offers better gaming performance over what I have. :D
So reading the article, it looks like we are going to see Steamroller & Excavator sooner than we think. Piledriver may now be short lived, because all the news is about Steamroller.
Yes PileDriver may be short lived.
Here's hopping Jim Keller can fix this crap they have been doing at AMD.
Posted on Reply
#322
cadaveca
My name is Dave
AvonXHere's hopping Jim Keller can fix this crap they have been doing at AMD.
I knew, and posted on here before the release, that the problem with Bulldozer was poor cache bandwidth, specifically, L2.


Why that performance is poor, I dunno, but THAT is what needs to change, and it doesn't take anyone special to be able to recognize that.
Posted on Reply
#323
Super XP
cadavecaI knew, and posted on here before the release, that the problem with Bulldozer was poor cache bandwidth, specifically, L2.


Why that performance is poor, I dunno, but THAT is what needs to change, and it doesn't take anyone special to be able to recognize that.
Agreed, but also having a single decoder killed BD. Steamroller seems to fix this and some. I also have to give thumbs up for AMD trying something way out of the box. Talk about having Balls of Steel.

Posted on Reply
#325
Super XP
SteevoCalling bullshit on the core starvation, single threaded clock for clock it performs worse than the older generation. Perhaps that is caused by L2 bandwidth, but that can be checked with some math, that is unless they had to introduce artificial latency to keep the core clock up or for other reasons.

www.hardocp.com/article/2011/10/11/amd_bulldozer_fx8150_desktop_performance_review/4
The history of AMD's Bulldozer architecture is a painful one. The old management lead by Hector Jesus Ruiz and Dirk Meyer deliberately delayed the arrival of K10 "Barcelona" and got usurped by Intel's Core architecture. The mad rush to the original Bulldozer caused a lot of sacrifices in design and ultimately, lead AMD engineering team to cancel the architecture in 2008. Instead of creating a master-core, AMD's engineers envisioned a multi-step approach to increase the performance of its architecture.
Major Improvements. Steamroller should have been the original Bulldozer. Didn't want to underline the whole thing, so the link is below.
Comparing Steamroller to Bulldozer makes much more sense, since the two architectures are starting to differ in greater detail. First and foremost, AMD finally addressed the core starvation. Originally, Bulldozer had a single Fetch and single Decode unit, which were in charge of feeding both Integer and Float schedulers. It turns out that the size of those units were too small and quite often you'd waste precious cycles with either ALU or FPU pipelines not doing a thing. Steamroller goes back to square one and keeps the Fetch unit as a single entity, but the Decode part is now doubled. Each Decode unit feeds one INT unit (4 pipelines) and the FP Scheduler, which has three dedicated units (two 128-bit FMAC units which can act as a single 256-bit unit when you need 256-bit AVX. For legacy code, the MMX Unit is now a single separate entity (instead of multiple side half-units in Bulldozer design). Also, one of major improvements is the increase in the instruction cache size. Up until Bulldozer, AMD featured the largest L1 cache in the field - both L1 Instruction (I-Cache) and Data (D-Cache) were the same size (64KB). With 128KB of L1 cache, AMD easily compensated for the size deficit in L2 and L3 cache versus Intel Nehalem and Sandy Bridge architectures. Bulldozer sliced down L1 cache to "better than Pentium 4, but still crap", as one of our sources put it bluntly (16KB L1 D-Cache and 64KB L1 I-Cache). Steamroller increases the size of Instruction cache beyond K7/K8/K10/K10.5/BD, but L1 D-Cache won't remain the same either.

According to Mark Papermaster, the improvements should yield up to 30% performance increase, but our sources inside the company beg to differ.

"Steamroller is not Bulldozer Enhanced. F*** no. The layout might look the same but our LEGO blocks are completely different. When all is said and done we should get 45% improvement and this goes to show how the Bulldozer was f***** design. This is all what Bulldozer was supposed to be."

Read more: vr-zone.com/articles/amd-pushes-steamroller-and-excavator-forward-bullish-about-performance-increases/17088.html#ixzz26DXDxHDW
In response to Steevo's post below, I fully agree AMD has major cache issues. For example 8MB of say L2 or L3 would bring a significant performance boost for Intel, but not so much AMD.
Posted on Reply
Add your own comment
Mar 15th, 2025 23:52 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts