AMD CPU speculation... and expert conjecture

[hcl123]

28nm steamroller is so late. Its been 1.5 years since 7970 on 28nm came out at tsmc and we still aren't even close to steamroller. Hopefully Glofo can can keep production up and give decent yields.

That is consequence of all the turmoil with GloFo and AMD getting out of that partnership. At the time of Dirk Meyer i suspect GloFO with AMD direction was pushing hard SOI tech for 32 and 28nm. A new management was imposed by the principal owners of GloFo and all went in reverse. The epilogue of the drama finished last year with AMD getting out of GloFo entirely. But the last management of GloFo also went bust, its was all about bulk "mantra" and SoC even in the underwear, being the superior SOI tech even regarded as inferior... yet it didn't managed o get one single top IDM implementer of the ARM SoC armada, for the their *supposed* superior HKMG bulk high performance 28nm process ( or the low power one for that matter).

Qualcomm was there, tested, tasted, saw it, saw what they did to AMD... and said goodbye...

I think GloFo re-gained a lot more respect for AMD, since the notice was clear that they could even lose the only top IDM they had left, and no prospects of gaining another one in short order. That is why the old 32nm PD-SOI was revamped almost 15% and Richland happened... that is why the FD-SOI partnership with STMicro was possible... with all that, is even possible that the FX/server variants of Steamy is 28nm PD-SOI.

With all this "bulk" dance decisions(even at AMD) and SOI improvements, its no wonder things got delayed.

That makes a lot of sense but I don't expect SOI to be viable going forward especially going into finfets and trying to reduce power consumption. I thought both steamroller and excavater will be on bulk. I do wonder if they can even get much performance gains from 32nm SOI to 28nm bulk.

What is not viable is "plain planar bulk", 20nm is pretty much the dead end. There is a reason why Intel in his eagerness to be first choose finfet, and complicating for complicating was not the reason. From now on Fully Depleted techs will be the norm, it can be accomplished by UltraThin Body and Box(UTBB) or FD-SOI in short, or by finfet techs.

Just read some news http://www.advancedsubstratenews.com/

There are pretty good studies and heavy weights. IBM as example is clearly showing that Finfet on SOI is quite better than Finfet on bulk, and they don't do it because of "fanboyism" or propaganda, IBM doesn't sell or has stakes in SOI wafer suppliers.

FD-SOI, like with STMicro with its 4 core ARM chip at 3Ghz demo, at most probably yet less than 5W(which is already too much for smarthphones), is showing that FD-SOI is not only better for low power but can be also better for high performance. Cherry on top of cake is that FD-SOI can be less expensive to fab than finfet on inexpensive "bulk" substracts (specialty below the 28nm size) and provide quite better yields (much less variability with transistors).

I suspect, (betting on it) that the next high performance process of IBM (the one that delivers "commercial" chips above 5Ghz) will be FD-SOI.

Finfet can be the only option below 10nm because that thin channel bodies for FD-SOI will not scale further, but then it will be finfet on SOI not bulk, specially for high performance, if not for all, since finfet of bulk can reach the dead end below 10nm to, even for low power offerings.

Intel pushes hard the bulk mantra because it doesn't want to depend on others for anything, including wafer suppliers, and their requirements for volume is quite above anyone else. But i believe nothing prevents intel to license pertinent techs and do their own SOI wafers, which i believe will happen eventually.

Foundries just love "bulk" because it means higher profits for themselves... not better solutions for costumers.

 

[Ranth]

Hmmmm... I wonder how AMD is gonna respond to a 400$ 770 (680)

http://www.anandtech.com/show/6994/nvidia-geforce-gtx-770-review/1
http://www.tomshardware.com/reviews/geforce-gtx-770-gk104-review,3519.html

 

[mayankleoboy1]

Web-browsing can be done in parallel

Scales naturally per connection.

If you mean one process/thread per tab , then yes. What chrome does, mostly.
If you mean more than 1 thread per tab, then no. A modern, standards compliant browser cant do this for most of the things. (Though they use the GPU for some things like composting, SVG shaders, and maybe text shaping too.)

 

[hcl123]

if cinebench is so biased, amd should really stop using cinebench. they should officially and explicitly denounce cinebench for cpu benchmarks instead of putting it in their own promotions...

Yes somehow i agree with that, but some how AMD like almost all others, have no say in this... its quite more complicated than that... it will be akin to force or require software developers to choose methods and tools for development, which logically isn't legal and can't happen.

If Maxon wants to use ICC for their Cinema 4D (from which Cinebench is derived), no one can't force them otherwise, or should, even if it means "obviously" a certain biasing towards Intel (ICC = Intel compiler collection).

The problem with Cinebench or more concretely with ICC, that already generated a lawsuit that intel lost, is that ICC forces different binary code paths that it generates attending to the CPUID, that is as example, if the target is genuine intel it will use AVX, if it is anyone else CPU, it forces SSE2.

VIA ( old Cyrix ) Nano found a solution by tricking CPUID, that is, it can give a false CPU identification when asked, but this is a solution that can gain them a lawsuit for false representation of product... the other solution imposed by court is to force ICC to be as fair as possible.

But with all this, i believe even with R11.5, when ppl look at those numbers and charts, it will be good to know that what they are seeing is a comparison of AVX crunching on Intel with SSE2 on anyone else.

Hardly oranges to oranges... if by chance ICC used SSE4 for "others", according to numbers floating around (SSE4 on BD uarch can be ~ 50% better than SSE2), then BD would be winning at Cinebench without much trouble... which wouldn't fair for intel either.

The best solution, since all comparisons are always *RELATIVE*, is to md5 the benchmark binaries ( a rock stable release) and use it along the time. At least that way we would know in all those comparisons, that what changes is the hardware not the software(the software will be frozen in time).

 

[de5_Roy]

Spoiler

Yes somehow i agree with that, but some how AMD like almost all others, have no say in this... its quite more complicated than that... it will be akin to force or require software developers to choose methods and tools for development, which logically isn't legal and can't happen.

If Maxon wants to use ICC for their Cinema 4D (from which Cinebench is derived), no one can't force them otherwise, or should, even if it means "obviously" a certain biasing towards Intel (ICC = Intel compiler collection).

The problem with Cinebench or more concretely with ICC, that already generated a lawsuit that intel lost, is that ICC forces different binary code paths that it generates attending to the CPUID, that is as example, if the target is genuine intel it will use AVX, if it is anyone else CPU, it forces SSE2.

VIA ( old Cyrix ) Nano found a solution by tricking CPUID, that is, it can give a false CPU identification when asked, but this is a solution that can gain them a lawsuit for false representation of product... the other solution imposed by court is to force ICC to be as fair as possible.

But with all this, i believe even with R11.5, when ppl look at those numbers and charts, it will be good to know that what they are seeing is a comparison of AVX crunching on Intel with SSE2 on anyone else.

Hardly oranges to oranges... if by chance ICC used SSE4 for "others", according to numbers floating around (SSE4 on BD uarch can be ~ 50% better than SSE2), then BD would be winning at Cinebench without much trouble... which wouldn't fair for intel either.

The best solution, since all comparisons are always *RELATIVE*, is to md5 the benchmark binaries ( a rock stable release) and use it along the time. At least that way we would know in all those comparisons, that what changes is the hardware not the software(the software will be frozen in time).

i've read about intel's compiler restricting non-intel cpus so many times... i proposed amd denouncing cinebench knowing that. i wasn't proposing that amd should ask maxon to change compilers. i meant that amd should just drop cinebench from their benchmarketing and use something else. something that scales well with more cores yet doesn't suffer from intel-bias. for example, softwares like 7zip can show how much advantage amd (fx) cpus have when all cores are loaded.

 

[truegenius]

phenom 2 series is of amd
fx series is of amd
so intel compiler will treat them in same way

so why phenom performs better per ghz per core
thats because bulldozzer is bullcrap
though pd is better than bullcrap but not by much

 

[Mitrovah]

So This article claims the new FM2+ socket may be because the new kaveri amd apu will have DDR4/GDDR5 Memory. Does that sound like a reasonable assumption? if so, would it work with DDR3 Ram? http://

 

[hcl123]

do we really have to wait till 10nm? i think common platform alliance can effectively put pressure on intel at 14nm mobility nodes, even if glofo fails to deliver(typical) in time. intel is already behind, their new products seem to be sku-ed in a way that protect their high end lineup instead of aiming for higher revenue. imo the whole tech world is at a major transition stage right now and pretty much everything we've taken for granted over the past few years is up for grabs.
qualcomm and samsung have a huge opportunity here. amd might have a chance as well... but they suffer from chronic poorus executionitis and recurring 'loser mentality' despite having the most robust and 'agile' ip portfolio. unless RR gets rid of those problems, amd will be the new second/third best behind qualcomm and/or samsung.

Quite the contrary, Qualcomm and Samsung are "winning" not because they embrace a race to absurdity, but because they have a good business model, specially samsung that doesn't depend on anyone else (of relevance that is), and with that are able to forestall forces that are able to corner the market with dirty tactics.

ARM is gaining not because it has the most performance(tough it can be better than x86) or because it has the best fab processes... or because it is the best low power ISA implementation ( MIPS is clearly) for that matter...

It is just "good enough"... what ARM world has not experience *yet* is the "environment" that goes on the "traditional" desktop and server markets. The uber all spin and propaganda has not reached the ARM world yet, though much less performance, you don't experience 2 or 3x less "feel" on a smartphone because the software is much less bloated and more balanced ( is quite "snap" enough, and performance is in the software not hardware, as i always told), and no bogus comparatives to infinity circulate to derive costumers from that "feel", like if it is a race for the arsehole award... and the numerous ARM armada players aren't positioned/inclined to kill or crush each other, on the contrary they cooperate behind the norm, even if competition is quite fierce sometimes ( only Apple ruins the picture) ...

This makes the ARM world quite a formidable force, even if none of the players is close to Intel size. That is why HSA foundation is quite bizarre for the "traditional" view, all those partners should be very unlikely partners... yet they are, pushing something extraordinary, leaded by AMD (a x86 competitor) yet the majority of founders are ARM world.

 

[sarinaide]

Hmmmm... I wonder how AMD is gonna respond to a 400$ 770 (680)

http://www.anandtech.com/show/6994/nvidia-geforce-gtx-770-review/1
http://www.tomshardware.com/reviews/geforce-gtx-770-gk104-review,3519.html

They will in due course but only after the Richland line is released. The HD7970GE still is faster or tie with the 770 and is nearly 2 years old so that impressive in its own right. I suspect Sea Islands HD8000 will come out, ie: a refresher and Volcanic Islands to come out next year in Q1 or Q2, that said I would rather come out with a HD7975, HD7875 and just pump more resources onto the GPU and let them fend off until AMD has GCN 2.0 ready.

 

[gamerk316]

Web-browsing can be done in parallel

Scales naturally per connection.

If you mean one process/thread per tab , then yes. What chrome does, mostly.
If you mean more than 1 thread per tab, then no. A modern, standards compliant browser cant do this for most of the things. (Though they use the GPU for some things like composting, SVG shaders, and maybe text shaping too.)

More or less correct. What in terms of browsing within the confines of a single tab can you thread? Not much.

 

[mayankleoboy1]

That is why HSA foundation is quite bizarre, all those partners should be very unlikely partners... yet they are, pushing something extraordinary, leaded by AMD (a x86 competitor) yet the majority of founders are ARM world.

Maybe because the ARM partners realize that due to TDP constraints,ARM CPU perf will be that much harder to increase with the next generation. On the other hand, GPU's can really benefit from a new node, and keep on giving better perf/watt.

 

[gamerk316]

do we really have to wait till 10nm? i think common platform alliance can effectively put pressure on intel at 14nm mobility nodes, even if glofo fails to deliver(typical) in time. intel is already behind, their new products seem to be sku-ed in a way that protect their high end lineup instead of aiming for higher revenue. imo the whole tech world is at a major transition stage right now and pretty much everything we've taken for granted over the past few years is up for grabs.
qualcomm and samsung have a huge opportunity here. amd might have a chance as well... but they suffer from chronic poorus executionitis and recurring 'loser mentality' despite having the most robust and 'agile' ip portfolio. unless RR gets rid of those problems, amd will be the new second/third best behind qualcomm and/or samsung.

Quite the contrary, Qualcomm and Samsung are "winning" not because they embrace a race to absurdity, but because they have a good business model, specially samsung that doesn't depend on anyone else (of relevance that is), and with that are able to forestall forces that are able to corner the market with dirty tactics.

ARM is gaining not because it has the most performance(tough it can be better than x86) or because it has the best fab processes... or because it is the best low power ISA implementation ( MIPS is clearly) for that matter...

It is just "good enough"... what ARM world has not experience *yet* is the "environment" that goes on the "traditional" desktop and server markets. The uber all spin and propaganda has not reached the ARM world yet, though much less performance, you don't experience 2 or 3x less "feel" on a smartphone because the software is much less bloated and more balanced ( is quite "snap" enough, and performance is in the software not hardware, as i always told), and no bogus comparatives to infinity circulate to derive costumers from that "feel", like if it is a race for the arsehole award... and the numerous ARM armada players aren't positioned/inclined to kill or crush each other, on the contrary they cooperate behind the norm, even if competition is quite fierce sometimes ( only Apple ruins the picture) ...

This makes the ARM world quite a formidable force, even if none of the players is close to Intel size. That is why HSA foundation is quite bizarre, all those partners should be very unlikely partners... yet they are, pushing something extraordinary, leaded by AMD (a x86 competitor) yet the majority of founders are ARM world.

ARM is basically taking over where PPC used to rule: low power devices, as X86 simply can not scale down as well as RISC based chips. And given the rise of mobile computing, and the fact anyone can license the ARM architecture, Intel has a serious long-term threat on its hands.

ARM is far more likely to push Intel then AMD ever could.

 

[anxiousinfusion]

So This article claims the new FM2+ socket may be because the new kaveri amd apu will have DDR4/GDDR5 Memory. Does that sound like a reasonable assumption? if so, would it work with DDR3 Ram? http://

The GCN core would be supposedly based on the refined “Sea Islands”

It's these sweet, delicious bits of information that make tech news worth reading. I was under the impression that AMD would keep their APU graphics architecture a generation behind. They are skipping GCN 1.0 entirely and I believe that Richland should have been built with GCN by this point.

 

[sarinaide]

It is not confirmed but trying to press details out of a collegue that has a 6800K to test the verdict is impressive, trying to butter him up nicely I kinda got a 30% figure. I guess I will be selling my 5800K very soon.

 

[hcl123]

i've read about intel's compiler restricting non-intel cpus so many times... i proposed amd denouncing cinebench knowing that. i wasn't proposing that amd should ask maxon to change compilers. i meant that amd should just drop cinebench from their benchmarketing and use something else. something that scales well with more cores yet doesn't suffer from intel-bias. for example, softwares like 7zip can show how much advantage amd (fx) cpus have when all cores are loaded.

Denounce what ? ... the law suit i believe was "inspired" somehow by AMD, all that could be done was done...

Yet nothing prevents Maxon to continue using an old version of ICC (not contemplated by the court impositions) that they already have legally licensed, for the foreseeable future for Cinebench releases...

Maxon is free, internet is free-speech ( it should continue to be), all those bogus reviews are not done by intel, are done by "independent" players (*cough*so to speak*cough*), that with their "freedom" can grab any piece of software they like from the net(no one can impose choices here), pieces of software that are not even signed and don't have any guaranties as all software does, of which they have no responsibility at all for none of it, and with all that try to make a job of reviewing a better solution for an ignorant gullible audience...

There is nothing that can be done but for the end-user themselves to get smarter, and ask for some requisites (like md5), and question anything that seems less transparent in all this *business*.

 

[mayankleoboy1]

Web-browsing can be done in parallel

Scales naturally per connection.

If you mean one process/thread per tab , then yes. What chrome does, mostly.
If you mean more than 1 thread per tab, then no. A modern, standards compliant browser cant do this for most of the things. (Though they use the GPU for some things like composting, SVG shaders, and maybe text shaping too.)

More or less correct. What in terms of browsing within the confines of a single tab can you thread? Not much.

Some things can be parallelized, specially CSS selector matching and JS parsing, of the top of my head. The reason it hasnt been implemented much is because if how these interact with all of the other components.
So the summary is what we two have been saying all along :its difficult to parallelize complex code. Which the noncoders dont get.

 

[BeastLeeX]

Hmmmm... I wonder how AMD is gonna respond to a 400$ 770 (680)

http://www.anandtech.com/show/6994/nvidia-geforce-gtx-770-review/1
http://www.tomshardware.com/reviews/geforce-gtx-770-gk104-review,3519.html

They will in due course but only after the Richland line is released. The HD7970GE still is faster or tie with the 770 and is nearly 2 years old so that impressive in its own right. I suspect Sea Islands HD8000 will come out, ie: a refresher and Volcanic Islands to come out next year in Q1 or Q2, that said I would rather come out with a HD7975, HD7875 and just pump more resources onto the GPU and let them fend off until AMD has GCN 2.0 ready.

In reality AMD does not need a refresh, they just need to pay attention to what Nvidea does. the 7970Ghz is 4% weaker at stock clocks compared to the GTX 770 in 1080p (http://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_770/27.html). AMD has to watch out for the 760 Ti though, if it is a refresh of the 670 at $300, the 7950 cannot compete, so expect price drops from AMD, and I am unsure about HD 8000 & 9000 right now. According to this: http://wccftech.com/amd-kaveri-apus-compatible-socket-fm2-roadmap-confirms-launch-late-2013/ Kaveri will have up to 768 SPs and GDDR5 memory, with HSA. This is 7770 performance, for a low price (http://www.newegg.com/Product/Product.aspx?Item=N82E16819113280), the top of the line Trinity is $130 right now. The point is that if AMD refreshes the 7000 series, the low end 87xx would almost feal obsolete with that APU for budget gamers. $140-$150 (speculation) is $50 more then a 7770 and you get an AMD 4 core Steamroller processor.

 

[hcl123]

do we really have to wait till 10nm? i think common platform alliance can effectively put pressure on intel at 14nm mobility nodes, even if glofo fails to deliver(typical) in time. intel is already behind, their new products seem to be sku-ed in a way that protect their high end lineup instead of aiming for higher revenue. imo the whole tech world is at a major transition stage right now and pretty much everything we've taken for granted over the past few years is up for grabs.
qualcomm and samsung have a huge opportunity here. amd might have a chance as well... but they suffer from chronic poorus executionitis and recurring 'loser mentality' despite having the most robust and 'agile' ip portfolio. unless RR gets rid of those problems, amd will be the new second/third best behind qualcomm and/or samsung.

Quite the contrary, Qualcomm and Samsung are "winning" not because they embrace a race to absurdity, but because they have a good business model, specially samsung that doesn't depend on anyone else (of relevance that is), and with that are able to forestall forces that are able to corner the market with dirty tactics.

ARM is gaining not because it has the most performance(tough it can be better than x86) or because it has the best fab processes... or because it is the best low power ISA implementation ( MIPS is clearly) for that matter...

It is just "good enough"... what ARM world has not experience *yet* is the "environment" that goes on the "traditional" desktop and server markets. The uber all spin and propaganda has not reached the ARM world yet, though much less performance, you don't experience 2 or 3x less "feel" on a smartphone because the software is much less bloated and more balanced ( is quite "snap" enough, and performance is in the software not hardware, as i always told), and no bogus comparatives to infinity circulate to derive costumers from that "feel", like if it is a race for the arsehole award... and the numerous ARM armada players aren't positioned/inclined to kill or crush each other, on the contrary they cooperate behind the norm, even if competition is quite fierce sometimes ( only Apple ruins the picture) ...

This makes the ARM world quite a formidable force, even if none of the players is close to Intel size. That is why HSA foundation is quite bizarre, all those partners should be very unlikely partners... yet they are, pushing something extraordinary, leaded by AMD (a x86 competitor) yet the majority of founders are ARM world.

ARM is basically taking over where PPC used to rule: low power devices, as X86 simply can not scale down as well as RISC based chips. And given the rise of mobile computing, and the fact anyone can license the ARM architecture, Intel has a serious long-term threat on its hands.

ARM is far more likely to push Intel then AMD ever could.

Quite right. And the principal feature is a mentality one. HSA foundation as example again, hell would freeze over and chicken would born with teeth, if intel ever agrees to partnership having a direct competitor as leading the project, even if all of it has all the technical and else merits.

ARM world doesn't seem to mind with those issues, that is why they can evolve technically faster than intel... the potential with that cooperation is there ( if it will is another matter)... that is why ARM can even be in the position that x86 is now ... and more...

"" ARM is far more likely to push Intel then AMD ever could.""

Mentality yes, but the partnership with AMD might not be that peace and love and or naive as some might interpret.

HSA software will already be basically the same in the "parallel" part either for x86 or ARM... and don't know if legally is feasible, but is AMD that holds the IP for the x86_64 part not intel. Like the chinese of goodson wit their MIPS that can run x86, even has a special pipe for some pertinent x86 instructions, while the rest is binary translated... the same with ARM...

Goodson is reaching ~70% of native x86 performance, if ARM licenses x86_64, even if they aren't able to build x86 CPU becasue they would have to license from intel, with much more native x86_64 hardware is possible that an ARM chip could reach 90% or more of native performance of x86.

HSA is already a tremendous push in direction of a "parallel" general purpose unified world, something that intel wanted/wants with Larabee/AVX pushes. If ARM licenses x86_64 then it could be the first to trully advance with an Trully Encompassing Universal Processing Chip ( CPU or APU as you want to call it).

" ARM can deliver the real big push... but who holds the key is somehow AMD... "

That is why intel wants so desperately kill AMD ( its the only logic conclusion of its posture), which at first view is an absurd preposition, to kill your own licensee that generates revenue for you... but not if then you hold all the cards including additional rights that were not yours to begin with...

 

[sarinaide]

Hmmmm... I wonder how AMD is gonna respond to a 400$ 770 (680)

http://www.anandtech.com/show/6994/nvidia-geforce-gtx-770-review/1
http://www.tomshardware.com/reviews/geforce-gtx-770-gk104-review,3519.html

They will in due course but only after the Richland line is released. The HD7970GE still is faster or tie with the 770 and is nearly 2 years old so that impressive in its own right. I suspect Sea Islands HD8000 will come out, ie: a refresher and Volcanic Islands to come out next year in Q1 or Q2, that said I would rather come out with a HD7975, HD7875 and just pump more resources onto the GPU and let them fend off until AMD has GCN 2.0 ready.

In reality AMD does not need a refresh, they just need to pay attention to what Nvidea does. the 7970Ghz is 4% weaker at stock clocks compared to the GTX 770 in 1080p (http://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_770/27.html). AMD has to watch out for the 760 Ti though, if it is a refresh of the 670 at $300, the 7950 cannot compete, so expect price drops from AMD, and I am unsure about HD 8000 & 9000 right now. According to this: http://wccftech.com/amd-kaveri-apus-compatible-socket-fm2-roadmap-confirms-launch-late-2013/ Kaveri will have up to 768 SPs and GDDR5 memory, with HSA. This is 7770 performance, for a low price (http://www.newegg.com/Product/Product.aspx?Item=N82E16819113280), the top of the line Trinity is $130 right now. The point is that if AMD refreshes the 7000 series, the low end 87xx would almost feal obsolete with that APU for budget gamers. $140-$150 (speculation) is $50 more then a 7770 and you get an AMD 4 core Steamroller processor.

AMD know where the goal posts lie now and yes right now a new GPU lineup is not of importance but at least AMD know Nvidia are stranded for some time now and basically will sit with masses of unsold 600 SKU's and no hope of making anything back on them. I do as I have stated expect the HD8000 family to release which will put the status quo at as you were.

A collegue completed his 6800K review and was "wow", from what we hear on Kaveri the top part is capable of a 105GB/s bandwidth, that is in contrast to the 40GB/s I get on DDR3 2400 and the industry is of the opinion Kaveri will basically make the budget and low end market obselete meaning that only mainstream and high end will need to be focused on, thats x800 and x900 parts along with the dual cards, for AMD they may only need 3-5 SKU's on the market and that is fantastic on costs and as Nvidia are finding out with having 12 SKU's and the majority unsold leaves you in a pickle.

 

[de5_Roy]

So This article claims the new FM2+ socket may be because the new kaveri amd apu will have DDR4/GDDR5 Memory. Does that sound like a reasonable assumption? if so, would it work with DDR3 Ram? http://

i never trusted wccdefghicantevenspelltech.com. but, if their 'leak' is to be believed(with salt ready at hand), socket fm2+ might support existing trinity and richland apus as well as existing chipsets like a55 to a85x but fm2 motherboards won't be able to support kaveri apus. basically, while amd gets to recycle good bits from their current platform and save money for themselves, fm2 mobo owners, once again, like fm1 owners, might be left in the dirt. i guess gddr5 support was the first indication. from the roadmap - amd must be confident enough to launch kaveri this late. may be they think that kaveri can stand on it's own as cpus (since steamroller is m.i.a.) against haswell and ivb-e (and like haswell, kaveri wants new mobos too.. ).
ddr4 won't be backwards compatible with ddr3. i don't think kaveri will have ddr4 support since not even servers have it. i think that wccdefghidunnotech is hinting at ddr4 and gddr5 being the same thing.

Denounce what ? ... the law suit i believe was "inspired" somehow by AMD, all that could be done was done...

Yet nothing prevents Maxon to continue using an old version of ICC (not contemplated by the court impositions) that they already have legally licensed, for the foreseeable future for Cinebench releases...

Maxon is free, internet is free-speech ( it should continue to be), all those bogus reviews are not done by intel, are done by "independent" players (*cough*so to speak*cough*), that with their "freedom" can grab any piece of software they like from the net(no one can impose choices here), pieces of software that are not even signed and don't have any guaranties as all software does, of which they have no responsibility at all for none of it, and with all that try to make a job of reviewing a better solution for an ignorant gullible audience...

There is nothing that can be done but for the end-user themselves to get smarter, and ask for some requisites (like md5), and question anything that seems less transparent in all this *business*.

amd should prove and point out the fact that cinebench uses icc and is unfairly biased towards intel. then, as a gesture, they should drop publishing cinebench benchmark results from their promotions and ask reviewers too. that will increase public awareness at least. meanwhile they should provide cinebench results with similar software benches(that wasn't compiled on icc) showing how much intel bias is going on with cinebench. may be at least passively pressure(with public awareness and campaign) them to provide a differently compiled cinebench (benchmark only). hopefully this would alert devs who use icc in their benchmark programs. since amd is pushing their 'multicore is the future' mantra, they should put some effort behind how their multicore cpus actually perform on a neutrally compiled benchmark.

 

[hcl123]

So This article claims the new FM2+ socket may be because the new kaveri amd apu will have DDR4/GDDR5 Memory. Does that sound like a reasonable assumption? if so, would it work with DDR3 Ram? http://

A new socket is probable because of pad and electric requirements, but no, no DDR4 yet. No DDR4 before 2014, which should mean AFAIK even for intel, no DDR4 offering before 2015. You could use the DDR3 that you already have.

And the future is HBM or Memory Cube of sorts, meaning DDR4 can be very short lived or a second pool, of not much importance, like the DDR3 will be compared with the GDDR5 that will be "in" the socket in a MCM fashion in those Kaveri solutions.

Intel has Cristalwell, its different, but the philosophy is identical.

It may be that DDR4 never catches on the desktop with that Cristalwell/MCM solutions (and future 2.5D interposed), because DDR4 is point-to-point topology, meaning if you want more than one DIMM per channel you have to have buffer extension chips in those DIMMS making them clearly more expensive... and latency and power wise DDR4 gains nothing substancial to DDR3, if its possible DDR3 at 2400 or higher at 1.35v ( 2133 is already 1.2(U spec)/1.35v but expensive and used for embedded)... and bandwidth wise it loses badly for GDDR5 and or interposed solutions...

DDR4 might have a much better luck in "traditional" server solutions though.

 

[8350rocks]

The irony of all this is that 95% of the large animation/render/CGI houses all run Linux and use Linux based software...not Cinema 4D. So the benchmark is essentially irrelevant outside of possibly hobbyists.

http://en.wikipedia.org/wiki/Linux

For years Linux has been the platform of choice in the film industry. The first major film produced on Linux servers was 1997's Titanic.[104][105] Since then major studios including Dreamworks Animation, Pixar, Weta Digital, and Industrial Light & Magic have migrated to Linux.[106][107][108] According to the Linux Movies Group, more than 95% of the servers and desktops at large animation and visual effects companies use Linux.

So the benchmark is only truly relevant for people who benchmark such things. It has no bearing on a format that any commercial houses use.

 

[hcl123]

That is why HSA foundation is quite bizarre, all those partners should be very unlikely partners... yet they are, pushing something extraordinary, leaded by AMD (a x86 competitor) yet the majority of founders are ARM world.

Maybe because the ARM partners realize that due to TDP constraints,ARM CPU perf will be that much harder to increase with the next generation. On the other hand, GPU's can really benefit from a new node, and keep on giving better perf/watt.

Yes there is a correlation, simplicity is *one* key for low power, and the reference designs that everyone is using from ARM are relatively simple and very well focused on low power. Yet the A57 design based on A15 should have a very similar throughput potential... if not higher... compared with Jaguar. The *effective* performance then will be mandated by the software, and ARM ecosystem is better in this regard, not only by the focus on low power but some virtue of the ISA (but there will not miss opportunities to bloat a lot to lol).

I high performance ARM could be *totally* different, only at reach of "architectural licensees" like Qualcomm, Nvidia, Apple and AMD (among some others)... yes AMD...

HSA, the focus and interest of ARM is most of all the software "ecosystem"... many usually can do better and more than a single one, no matter how good this single one is, and if the software is common, or mostly common the merrier, you would not lose "adoption" ground because you have no software ecosystem. Cherry on top of cake is that parallel workloads on heterogeneous, like GPGPUs, can deliver up to an order of magnitude better performance than on scaler vectorized pipes in a CPU(SSE/AVX), wasting less power in the process for identical loads.

win win win

 

[amdfangirl]

The irony of all this is that 95% of the large animation/render/CGI houses all run Linux and use Linux based software...not Cinema 4D. So the benchmark is essentially irrelevant outside of possibly hobbyists.

http://en.wikipedia.org/wiki/Linux

For years Linux has been the platform of choice in the film industry. The first major film produced on Linux servers was 1997's Titanic.[104][105] Since then major studios including Dreamworks Animation, Pixar, Weta Digital, and Industrial Light & Magic have migrated to Linux.[106][107][108] According to the Linux Movies Group, more than 95% of the servers and desktops at large animation and visual effects companies use Linux.

So the benchmark is only truly relevant for people who benchmark such things. It has no bearing on a format that any commercial houses use.

Not surprised. I am told that Blender runs twice as fast on Linux than Windows.

Not from any reliable source though.

 

[Cazalan]

This makes the ARM world quite a formidable force, even if none of the players is close to Intel size. That is why HSA foundation is quite bizarre for the "traditional" view, all those partners should be very unlikely partners... yet they are, pushing something extraordinary, leaded by AMD (a x86 competitor) yet the majority of founders are ARM world.

Actually Samsung is larger than Intel by a long shot, and Qualcomm is nearly the size of Intel.

ARM is formidable for many reasons but one key reason is ARM Holdings isn't greedy. They license the cores for pennies. They basically make it so cheap it's idiotic to even try designing a new CPU from scratch.

We don't know how well it scales into the 64bit world yet but we should have a better idea of that by the end of this year.