Friday, January 20th 2012

AMD Vishera Packs Quad-Channel DDR3 IMC, G34 En Route Desktop?

AMD might be a little sore that its "Zambezi" FX processor family based on its much-hyped "Bulldozer" architecture didn't quite meet the performance expectations of a ground-up new CPU architecture, but it doesn't want to take chances and build hype around the architecture that succeeds it. From various sources, some faintly-reliable, we have been hearing that the next-generation of high-performance desktop processors based on "Piledriver" architecture, codenamed "Vishera", will pack five modules or 10 cores, and will be structured essentially like Zambezi, since Piledriver is basically a refinement of Bulldozer architecture. The latest leak comes from the Software Optimization Guide for AMD 15h family (read here), which was picked up by CPU World while most of us were busy with CES.

CPU World compiled most of the features of what it suspected to be AMD referring to its future processors based on the Piledriver architecture, that's "Vishera" (desktop high-performance), "Terramar" (high-density server), and "Sepang" (small-medium business server) parts. The three are not the first chips to be based on Piledriver, AMD has a new mainstream desktop and notebook APU in the works codenamed "Trinity", which is en route for a little later this year. Trinity basically has an identical CPUID instruction-set as Vishera, Terramar, and Sepang, confirming their common lineage compared to today's "Bulldozer" architecture. The most catchy detail is of Vishera featuring 4 DDR3 channels.
The plot thickens where "HyperTransport Assist feature" is listed as being supported on Vishera. HT Assist is a feature found on AMD's enterprise socket G34 processors, which facilitates better inter-die communication between the two dies of a typical socket G34 Opteron processor. The G34 (LGA1972) package is a multi-chip module of two quad-core, six-core, or four-module dies, which combined have four DDR3 memory channels, and a number of HyperTransport links to communicate with neighbouring sockets and the system's chipset. Could this be the first indication that AMD wants to take on Intel LGA2011 HEDT (high-end desktop) using Vishera chips in the G34 package? It will be a while before we find out.

Apart from using common silicon between client and enterprise platforms, AMD does have a history of colliding the two.
Source: CPU World
Add your own comment

229 Comments on AMD Vishera Packs Quad-Channel DDR3 IMC, G34 En Route Desktop?

#26
meirb111
HorruxAs we move further and further away from single-threaded applications, single-threaded performance becomes a moot point. It is not entirely moot yet, but give it a year or two. Sure, there will still be legacy code written for single-core CPUs and those won't run faster, but they'll run fast ENOUGH.

At some point, architecture-wise, you have to start looking at either multicore technology or single-core technology. I thought it was well established that single core solutions were well and truly done with? There are no new CPUs in the PC space that are single core since a few years already, even in the bargain basement entry-level space.

I think it's clear: multithreading is the future, and mostly, the present too. As software development gets more and more multithreaded, AMD's solutions will begin to shine more and more. 3 years from now you might be amazed that the new huge games run super well on an 8150, whereas now you might be disappointed at how your 3 year-old game runs on this chip currently. But such is the nature of this technology juncture.

Sure, Intel is claiming both the single-threaded and multithreaded crowns, but no one ever said that AMD had to claim a crown. It only needs to be competitive in terms of price/performance. And it has been doing that, serving us good chips at great prices. Some people live for their epeen and they need benchmark records to their name. Fine. The rest of us are happy to pay less for our machines and enjoy our gaming very well, thanks.
thats gambling or guesswork, it may or may not happen most games dont use more than 2 core even new games since quad cores are 5 years old or so and still we are moving so slow 3 years is a lot of time by than you may upgrade a gain so what you buy now isnt that good for the future
Posted on Reply
#27
Steevo
HorruxAs we move further and further away from single-threaded applications, single-threaded performance becomes a moot point. It is not entirely moot yet, but give it a year or two. Sure, there will still be legacy code written for single-core CPUs and those won't run faster, but they'll run fast ENOUGH.

At some point, architecture-wise, you have to start looking at either multicore technology or single-core technology. I thought it was well established that single core solutions were well and truly done with? There are no new CPUs in the PC space that are single core since a few years already, even in the bargain basement entry-level space.

I think it's clear: multithreading is the future, and mostly, the present too. As software development gets more and more multithreaded, AMD's solutions will begin to shine more and more. 3 years from now you might be amazed that the new huge games run super well on an 8150, whereas now you might be disappointed at how your 3 year-old game runs on this chip currently. But such is the nature of this technology juncture.
Nov 2008

Intel Core i7 940

BD FAIL. You are trying to spin this as some sort of a win in three years when everyone will still know BD is a FAIL. It can't compete with a three year old product.
Posted on Reply
#28
Horrux
SteevoNov 2008

Intel Core i7 940

BD FAIL. You are trying to spin this as some sort of a win in three years when everyone will still know BD is a FAIL. It can't compete with a three year old product.
You know what's funny about all this? I'm putting the AMD products in their best light and I'm running Intel. You're putting AMD products in their worst light and you are running AMD.
Posted on Reply
#29
Dent1
HorruxYou know what's funny about all this? I'm putting the AMD products in their best light and I'm running Intel. You're putting AMD products in their worst light and you are running AMD.
+1, Steevo got pwn'd


Lets get back to the positive AMD talk. I'm happy for this quad channel move.
Posted on Reply
#30
Steevo
Im not dillusional about products. I guess I dont count that as "getting pwned", but since pwned is beaing beaten by a 13 year old cheater in a online game it cant really be that bad. So if it makes you feel better and your mom is OK with it, you both pwned me.

I run AMD as bang for the buck when I bought my board and everything it cost significantly less than the same Intel setup and its just lasted this long.

Once the next intel generation is out I will be choosing my next ground up build, keeping the water cooling of course.
Posted on Reply
#31
Horrux
SteevoIm not dillusional about products. I guess I dont count that as "getting pwned", but since pwned is beaing beaten by a 13 year old cheater in a online game it cant really be that bad. So if it makes you feel better and your mom is OK with it, you both pwned me.

I run AMD as bang for the buck when I bought my board and everything it cost significantly less than the same Intel setup and its just lasted this long.

Once the next intel generation is out I will be choosing my next ground up build, keeping the water cooling of course.
Like me then, I was running an AMD but the SLI issues drove me nuts enough to get an Intel 2600k. I was going to get a bulldozer, but when I saw how that turned out, there was no way I was sidegrading and probably keeping the SLI problems... The P2 X6 1100t was plenty good, but the SLI got me frothing at the mouth...

Sure the 2600k is a bit faster, not that I notice, the 1100t was fast enough.
Posted on Reply
#32
TheLaughingMan
SteevoNov 2008

Intel Core i7 940

BD FAIL. You are trying to spin this as some sort of a win in three years when everyone will still know BD is a FAIL. It can't compete with a three year old product.
But it can and it does. What are you smoking and can I have some?
Posted on Reply
#33
pantherx12
Aside from Erocker as far as I know everyone on this forum has been pretty happy with their FX purchase.

So they can't be all that bad eh : ]
Posted on Reply
#34
seronx
My speculation is Vishera/Komodo isn't going to be on AM3+

G2012
C2012
FM2

All have integrated Northbridges with an HT Link connection to the Southbridge

If all Viperfish dies are pretty much the same minus MCM in Terramar then there is going to be a problem since there is PCI-E Links instead of HT-Links
(Trinity isn't Viperfish but has an Intergrated Northbridge like Llano)

This is also probably why Cray you know that HPC company I keep talking about is going to implement a PCI-E version of it's Gemini interconnect because AMD is dropping Hypertransport

Posted on Reply
#35
erocker
*
pantherx12Aside from Erocker as far as I know everyone on this forum has been pretty happy with their FX purchase.

So they can't be all that bad eh : ]
I was "happy" with my 8150... I'm just happier with my 2500K and didn't need the 8150. ;)
Posted on Reply
#36
Super XP
I believe AMD will stop at 10 cores, work hard to refine the design (Piledriver), gain approx: 50 or more % performance improvement over today's Bulldozer. Only them will AMD move beyond 10 cores IMO.

AMD needs to tighten up this CPU design for better performance.
Posted on Reply
#37
seronx
Super XPI believe AMD will stop at 10 cores, work hard to refine the design (Piledriver), gain approx: 50 or more % performance improvement over today's Bulldozer. Only them will AMD move beyond 10 cores IMO.

AMD needs to tighten up this CPU design for better performance.


Yep, they are staying 10 cores with 28nm

but 50% performance you won't see...till applications start using AVX+FMA

Also, I hope that 28nm process is

Twenty Eight Nanometer Fully Depleted Silicon on Insulator with Ultra Thin Buried Oxide with Low-k Dielectrics and High-k Metal Gates with Embedded Silicon Germanium used with Dual strain liners with Raised Source and Drain....
*BREATHES* (It's 28nm ET-SOI for short or not)

Or, I'll be

My Christmas list for 2013

but seriously say that out loud it will suck out all the breath you have....
Posted on Reply
#38
Super XP
SteevoNov 2008

Intel Core i7 940

BD FAIL. You are trying to spin this as some sort of a win in three years when everyone will still know BD is a FAIL. It can't compete with a three year old product.
Lol, Umm yes it can, it competes quite well and does well against the newer Intel CPU's in Gaming. AMD wins 4 - Intel wins 4. You do the math. ;)

seronximg6.imageshack.us/img6/5329/b073b49boverclock.png

Yep, they are staying 10 cores with 28nm

but 50% performance you won't see...till applications start using AVX+FMA

Also, I hope that 28nm process is

Twenty Eight Nanometer Fully Depleted Silicon on Insulator with Ultra Thin Buried Oxide with Low-k Dielectrics and High-k Metal Gates with Embedded Silicon Germanium used with Dual strain liners with Raised Source and Drain....
*BREATHES* (It's 28nm ET-SOI for short or not)

Or, I'll be 29.media.tumblr.com/tumblr_ktot73IWFa1qa4uc1o1_400.jpg

My Christmas list for 2013

but seriously say that out loud it will suck out all the breath you have....
25% performance improvement via x86 and 50% in combined graphics with Piledriver CPU's. Anything can change by the time we hit Q3 2012 the schedules Piledriver desktop release.
LINK:
fudzilla.com/processors/item/25628-amd-increases-performance-of-trinity-cores
Posted on Reply
#39
Dent1
WOW.

I can't believe how badly the i7 990x got demolished, the FX 8150 performed twice as fast in Civalization V - is that game multi threaded or something?
Posted on Reply
#40
ensabrenoir
Super XPLol, Umm yes it can, it competes quite well and does well against the newer Intel CPU's in Gaming. AMD wins 4 - Intel wins 4. You do the math. ;)

img.techpowerup.org/120121/AMD-4 Intel-4.jpg


25% performance improvement via x86 and 50% in combined graphics with Piledriver CPU's. Anything can change by the time we hit Q3 2012 the schedules Piledriver desktop release.
LINK: hI
fudzilla.com/processors/item/25628-amd-increases-performance-of-trinity-cores
Yes..... we all know if bd was out several years ago it would define awesome......but it wasnt. compare modern to modern....sandy in other words
Posted on Reply
#41
Dent1
ensabrenoirYes..... we all know if bd was out several years ago it would define awesome......but it wasnt. compare modern to modern....sandy in other words
The concept of multithreading is modern, so lets compare Bulldozer to Sandybridge in multi threaded applications :)
Posted on Reply
#42
ensabrenoir
Dent1The concept of multithreading is modern, so lets compare Bulldozer to the i5/i7 series in multi threaded applications :)
meant intel latest....u know that slice of awesome pie called sandy bridge not last years model. Bd bigest problem........besides those missing transistors... was that it arrived way too late.
Posted on Reply
#43
Steevo
Super XPLol, Umm yes it can, it competes quite well and does well against the newer Intel CPU's in Gaming. AMD wins 4 - Intel wins 4. You do the math. ;)

img.techpowerup.org/120121/AMD-4 Intel-4.jpg


25% performance improvement via x86 and 50% in combined graphics with Piledriver CPU's. Anything can change by the time we hit Q3 2012 the schedules Piledriver desktop release.
LINK:
fudzilla.com/processors/item/25628-amd-increases-performance-of-trinity-cores
Any reason you didnt use W1zzards test? Other than it shows BD failing in the same games?
www.techpowerup.com/mobile/reviews/AMD/HD_7970_CPU_Scaling/7.html
Posted on Reply
#44
Horrux
ensabrenoirmeant intel latest....u know that slice of awesome pie called sandy bridge not last years model. Bd bigest problem........besides those missing transistors... was that it arrived way too late.
Nope. It's EARLY. If AMD had revised the Phenom II core to a Phenom III model, and tweaked the BD silicon a bit more for release in a year or two when multithreading will be even more prevalent, AMD would look better and the chip would have shone upon release, due to stellar multithreaded performance (including all the windoze coding that might be necessary, which WILL BE - notice the future tense again - included into windows 8 anyway).
Posted on Reply
#45
Dent1
SteevoAny reason you didnt use W1zzards test? Other than it shows BD failing in the same games?
www.techpowerup.com/mobile/reviews/AMD/HD_7970_CPU_Scaling/7.html
Good question. Maybe it was because at the time of W1zzards review of the 7970 there wasnt any proper drivers. It would be unfair to base a CPUs performance based on a video card that is bearly available and has know software support.
ensabrenoir... was that it arrived way too late.
Bulldozer is too early and too late.


Too late: Had they released it 3 years ago it would of been on par with The Nehalem in single threaded applications and much better in multithreaded, hence a success.

Too early: Had they delayed the Bulldozer a couple of years, multi threaded games would be in greater abundance and hence Bulldozer would be an attractive purchase for gamers.
Posted on Reply
#46
ensabrenoir
Bulldozer......its too early and too late... Bd is an enigma wrapped in a riddle and surrounded by confusion.....amd what did u really create. Wow so gonna get one ...after a few price drops... Just for kicks and to add to my almost great stuff collection(mini disc, hdd,laserdic player etc)
Posted on Reply
#47
Steevo
Dent1Good question. Maybe it was because at the time of W1zzards review of the 7970 there wasnt any proper drivers. It would be unfair to base a CPUs performance based on a video card that is bearly available and has know software support.




Bulldozer is too early and too late.


Too late: Had they released it 3 years ago it would of been on par with The Nehalem in single threaded applications and much better in multithreaded, hence a success.

Too early: Had they delayed the Bulldozer a couple of years, multi threaded games would be in greater abundance and hence Bulldozer would be an attractive purchase for gamers.
Or perhaps they ran a card that artificially limited the framerate.
Posted on Reply
#48
Dent1
SteevoOr perhaps they ran a card that artificially limited the framerate.
So you think the Civilization V reviewed with a the ATI 7970 with the unofficial drivers and no support from W1zzard is more valid than the Civilization V bechmarked from Anandtech with official and mature drivers??

or are you insinuating that Anantech's review is fake?
Posted on Reply
#49
Horrux
Dent1So you think the Civilization V reviewed with a the ATI 7970 with the unofficial drivers and no support from W1zzard is more valid than the Civilization V bechmarked from Anandtech with official and mature drivers??

or are you insinuating that Anantech's review is fake?
Anandtech was never reknown for its kindness to AMD*.
Posted on Reply
#50
OneCool
btarunrHT Assist is a feature found on AMD's enterprise socket G34 processors, which facilitates better inter-die communication between the two dies of a typical socket G34 Opteron processor.
Hmm,that sounds very poor to me.Latency killer :banghead:

Whats the point of HT if it has to go through some kind of buffer to sort shit out first?

"HT assist" = Not So Hyper Transport or "Sluggish Transport" :laugh:
Posted on Reply
Add your own comment
Apr 19th, 2024 09:57 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts