AMD CPU speculation... and expert conjecture

[Cazalan]

3 FP units and 2 Integer units ? ... now this is a complete surprise if true... how ?

Perhaps 1 MMX unit with 2 pipes capable of some FP... and 2 FMAC units with 2 pipes each ? .. 256bit per pipe ?

I don't think it's a surprise. This code has been out for 9 months now.

http://gcc.gnu.org/viewcvs/gcc/trunk/gcc/config/i386

They block direct links so you have to click on bdver1.md/bdver3.md and click view.

Bdver1 (Bulldozer) had 4 FP units. They removed an MMX unit for SR. Otherwise they look quite similar so still 128bit units.

 

[8350rocks]

A few things to add here:

1.) PS4, to my knowledge, has not defined an inherent physics engine. Developers are free to pick and choose...their preferred models are Havok then Bullit. However, I could see them doing some kind of software port simply for compatibility with PhysX so that they can still run the physics engine for the few games who do partner with NVidia (though that number of titles isn't a long list...)

2.) It has already been all but confirmed that the APUs will be running closer to ~2.0-2.2 GHz for the XBone and PS4. If you think that number won't increase over time, you're far more naïve than I thought. Just because the memory is clocked at one frequency has nothing to do with CPU clockspeed.

3.) I am with several others in thinking that they couldn't get Kaveri to work on bulk, and may have even brought Jim Keller in, specifically, to "fix" it and either go to FD-SOI (which is available) or make the bulk process work. My thoughts run to doing it on SOI, as JK was a big proponent of the K7/K8 going to SOI originally. It makes the design FAR more power efficient and perform FAR better as well. Consider this, when AMD originally went to SOI, they had a clear advantage over Intel during that entire era until Intel got their act together and finally caught up through design tweaks.

4.) This leads me to a valid, though not pressing, concern. If Intel went to 14nm FD-SOI...??? The issue being, their complex Tri-Gate FinFET would likely not mesh up well with Planar FD-SOI. Which is likely why they haven't been bothered to it so far. It would likely require heavy redesign, and they might lose a few years of development on their CPUs to make the jump to SOI. Even if they went with UTBB or one of the other processes...

 

[Cazalan]

Now that seems worst decode for a single thread per module compared with BD/PD lol ... unless after decode there are stream loop buffers like intel LSD... or unless the 2 decodes can work for the same thread.
.

Except there are 2 of these "bdver3" decoders per module. That's been one of the biggest changes for SR. BD/PD couldn't feed the pipes fast enough except for highly parallel (cached) code.

 

[juanrga]

juangra like I said intel cpus barely take much load in crysis 3, i7 2600k at 4.5ghz running two gtx 680 sli generally takes around average 30-40% cpu usage.
http://www.youtube.com/watch?v=oHnwQ9UfRM8

But the same CPU profiler used above for the FX chips, shows that a i7 2600k loads about a 80%.

 

[juanrga]

The author differentiates between the "rather confused DigiTimes story" and his own sources. Regarding Kaveri the author writes:

The reason Kaveri was late taping out, according to my sources, was that AMD kept the chip back to put some additional polish on its performance.

[...]

Which would coincide with the S|A story but with much fewer details. They've been polishing it for 2 years now. Neither have definitively said which process is being used.

He's showing to be rather spot on for this one: http://semiaccurate.com/2012/11/06/amds-kaveri-apu-slips-again-2014-now/

If indeed they are bulk I think they're going to be limited to an A10-5745M class (2.1/2.9Ghz) APU. Certainly Steamroller cores and better GCN will make that a much better mobile chip, but desktop users will yawn and cry fowl.

And coincides with Fudzilla rumour about an early Steamroller A core substituted by a Steamroller B. The point is that everyone is saying that the delay is due to the design of the chip not to the fabrication.

As I said above AMD claims that Kaveri delay is due to HSA. Maybe the delay is because the Steamroller CPU in Kaveri has to run both the native x86 instruction plus the new HSAIL instruction set and HSA specification is not still finished...

The CPU clocks don't have to be limited to 2.1/2.9Ghz. Glofo claims in his website that its High performance 28nm bulk can go above 3 Ghz. I also read in some part that Glofo claimed that they had perfected the bulk process closing the gap with SOI.

3 FP units and 2 Integer units ? ... now this is a complete surprise if true... how ?

Perhaps 1 MMX unit with 2 pipes capable of some FP... and 2 FMAC units with 2 pipes each ? .. 256bit per pipe ?

I don't think it's a surprise. This code has been out for 9 months now.

http://gcc.gnu.org/viewcvs/gcc/trunk/gcc/config/i386

They block direct links so you have to click on bdver1.md/bdver3.md and click view.

Bdver1 (Bulldozer) had 4 FP units. They removed an MMX unit for SR. Otherwise they look quite similar so still 128bit units.

Beside removing one MMX unit, the other is partially fused with the FMAC units, sharing resources. AMD claims that resources shared don't penalize FP performance.

 

[Dean Fry]

Quick question, what will AMD Steamroller CPUs use as a socket? Are they sticking with AM3+ or are they moving on to another socket?

 

[simon12]

Quick question, what will AMD Steamroller CPUs use as a socket? Are they sticking with AM3+ or are they moving on to another socket?

FM2+ is confirmed for it many people speculate about if it will come to AM3+ or not.

 

[juanrga]

A few things to add here:

1.) PS4, to my knowledge, has not defined an inherent physics engine. Developers are free to pick and choose...their preferred models are Havok then Bullit. However, I could see them doing some kind of software port simply for compatibility with PhysX so that they can still run the physics engine for the few games who do partner with NVidia (though that number of titles isn't a long list...)

2.) It has already been all but confirmed that the APUs will be running closer to ~2.0-2.2 GHz for the XBone and PS4. If you think that number won't increase over time, you're far more naïve than I thought. Just because the memory is clocked at one frequency has nothing to do with CPU clockspeed.

3.) I am with several others in thinking that they couldn't get Kaveri to work on bulk, and may have even brought Jim Keller in, specifically, to "fix" it and either go to FD-SOI (which is available) or make the bulk process work. My thoughts run to doing it on SOI, as JK was a big proponent of the K7/K8 going to SOI originally. It makes the design FAR more power efficient and perform FAR better as well. Consider this, when AMD originally went to SOI, they had a clear advantage over Intel during that entire era until Intel got their act together and finally caught up through design tweaks.

4.) This leads me to a valid, though not pressing, concern. If Intel went to 14nm FD-SOI...??? The issue being, their complex Tri-Gate FinFET would likely not mesh up well with Planar FD-SOI. Which is likely why they haven't been bothered to it so far. It would likely require heavy redesign, and they might lose a few years of development on their CPUs to make the jump to SOI. Even if they went with UTBB or one of the other processes...

1) Yes, this is what the Sony link did mean. Game developers are given that list of partners and they can chose the tools that they prefer. How many people will use Havok vs PhysX? I don't known and sincerely I don't care. The point is that contrary to your repetitive claims, the PS4 has PhysX support.

2) Never saw such confirmation. Proof?

3) Maybe Kaveri is SOI, but all info that we have claim it is bulk.

4) And what about Intel claiming that it is 14nm is bulk FinFET? What about Intel partenering with Altera to make their FPGAs using Intel's 14nm FinFET?

 

[juanrga]

juangra like I said intel cpus barely take much load in crysis 3, i7 2600k at 4.5ghz running two gtx 680 sli generally takes around average 30-40% cpu usage.
http://www.youtube.com/watch?v=oHnwQ9UfRM8

But the same CPU profiler used above for the FX chips, shows that a i7 2600k loads about a 80%.

The i7 2600k has 8 threads thats why so 4 threads at 80% is like 40% cpu usage.

Wrong. Those 8 threads are running in a mixture of four real cores plus four virtual cores and one virtual core is not equivalent to one real core, therefore dividing per two is wrong. Moreover, those virtual cores are obtaining their resources from a real core and real cores are not infinite. Imagine a real core is loaded a 80%, then HT can try to use the remaining unused 20% to simulate another core. In that case the real core is being loaded more than a 80% when running two threads. This mean one does not have to divide but to multiply by the HT efficiency factor.

Therefore it is just the inverse of what you say. With HT activated (which I did not consider before for simplicity and for avoiding you from having a heart attack) the i7 2600k is being used above the 80% in that game and still loses against an AMD FX.

 

[Cazalan]

The CPU clocks don't have to be limited to 2.1/2.9Ghz. Glofo claims in his website that its High performance 28nm bulk can go above 3 Ghz. I also read in some part that Glofo claimed that they had perfected the bulk process closing the gap with SOI.

Ok, I scoured high and low and found ONE bulk chip made by TSMC that broke 3.5Ghz. It's the Sparc T5 which has a max clock of 3.6Ghz. If GF can catch up and match that's not too too bad, but still a far cry from the 4.4Ghz turbo of the A10-6800K. That's about 20% slower which would eat away a sizable chunk of a 30% performance gain (being optimistic) for SR.

http://en.wikipedia.org/wiki/SPARC_T5

 

[griptwister]

Interesting watch for you AMD guys... Another reason to support AMD!!!

http://www.youtube.com/watch?v=7FFxXa7PTjc&hd=1

 

[juanrga]

The CPU clocks don't have to be limited to 2.1/2.9Ghz. Glofo claims in his website that its High performance 28nm bulk can go above 3 Ghz. I also read in some part that Glofo claimed that they had perfected the bulk process closing the gap with SOI.

Ok, I scoured high and low and found ONE bulk chip made by TSMC that broke 3.5Ghz. It's the Sparc T5 which has a max clock of 3.6Ghz. If GF can catch up and match that's not too too bad, but still a far cry from the 4.4Ghz turbo of the A10-6800K. That's about 20% slower which would eat away a sizable chunk of a 30% performance gain (being optimistic) for SR.

http://en.wikipedia.org/wiki/SPARC_T5

Interesting. I wait a reduction on clocks compared to Richland, otherwise I am unable to explain the official 1050 GFLOPS claimed by AMD for Kaveri.

 

[noob2222]

The author differentiates between the "rather confused DigiTimes story" and his own sources. Regarding Kaveri the author writes:

The reason Kaveri was late taping out, according to my sources, was that AMD kept the chip back to put some additional polish on its performance.

[...]

Which would coincide with the S|A story but with much fewer details. They've been polishing it for 2 years now. Neither have definitively said which process is being used.

He's showing to be rather spot on for this one: http://semiaccurate.com/2012/11/06/amds-kaveri-apu-slips-again-2014-now/

If indeed they are bulk I think they're going to be limited to an A10-5745M class (2.1/2.9Ghz) APU. Certainly Steamroller cores and better GCN will make that a much better mobile chip, but desktop users will yawn and cry fowl.

And coincides with Fudzilla rumour about an early Steamroller A core substituted by a Steamroller B. The point is that everyone is saying that the delay is due to the design of the chip not to the fabrication.

As I said above AMD claims that Kaveri delay is due to HSA. Maybe the delay is because the Steamroller CPU in Kaveri has to run both the native x86 instruction plus the new HSAIL instruction set and HSA specification is not still finished...

The CPU clocks don't have to be limited to 2.1/2.9Ghz. Glofo claims in his website that its High performance 28nm bulk can go above 3 Ghz. I also read in some part that Glofo claimed that they had perfected the bulk process closing the gap with SOI.

I swear sometimes you act like you have more information that AMD does. You don't think bulk vs soi has anything to do with the design ... lol. thats all I can say.

3) Maybe Kaveri is SOI, but all info that we have claim it is bulk.

from one article published over a year ago. a source so reliable that no one has really parroted it at all.

 

[de5_Roy]

parroting isn't dependent on the source's reliability. fanboys parrot because the content is usually suitable with their delusion. for example, now a days the russian website that benched crysis 3 usually gets parroted every time gaming is brought up without any deeper research into bench settings while more detailed crysis 3 benches are available on the net e.g. toms' own, techreport, pclab etc.
edit: to be fair, i coulda presented an intel-specific example, but that'd prolly cover nearly all benches. :lol:

 

[juanrga]

The author differentiates between the "rather confused DigiTimes story" and his own sources. Regarding Kaveri the author writes:

The reason Kaveri was late taping out, according to my sources, was that AMD kept the chip back to put some additional polish on its performance.

[...]

Which would coincide with the S|A story but with much fewer details. They've been polishing it for 2 years now. Neither have definitively said which process is being used.

He's showing to be rather spot on for this one: http://semiaccurate.com/2012/11/06/amds-kaveri-apu-slips-again-2014-now/

If indeed they are bulk I think they're going to be limited to an A10-5745M class (2.1/2.9Ghz) APU. Certainly Steamroller cores and better GCN will make that a much better mobile chip, but desktop users will yawn and cry fowl.

And coincides with Fudzilla rumour about an early Steamroller A core substituted by a Steamroller B. The point is that everyone is saying that the delay is due to the design of the chip not to the fabrication.

As I said above AMD claims that Kaveri delay is due to HSA. Maybe the delay is because the Steamroller CPU in Kaveri has to run both the native x86 instruction plus the new HSAIL instruction set and HSA specification is not still finished...

The CPU clocks don't have to be limited to 2.1/2.9Ghz. Glofo claims in his website that its High performance 28nm bulk can go above 3 Ghz. I also read in some part that Glofo claimed that they had perfected the bulk process closing the gap with SOI.

I swear sometimes you act like you have more information that AMD does. You don't think bulk vs soi has anything to do with the design ... lol. thats all I can say.

3) Maybe Kaveri is SOI, but all info that we have claim it is bulk.

from one article published over a year ago. a source so reliable that no one has really parroted it at all.

But by design I did mean very specific design aspects such as the mentioned HSA, not other kind of design changes.

Besides (i) that one-year old article from AMD claiming that Kaveri is bulk, we had (ii) a July 2013 Japanese article claiming that Kaveri is being manufactued at Glofo on bulk (iii) AMD official answer to VR-ZONE claiming that delay was due to HSA and that everything else is ready (recall Glofo SOI is not still ready), and (iv) the last article cited claiming that the delay was due to the chip not being sent to Glofo on time.

The four sources converge on the idea of kaveri being bulk 28nm @ Glofo. The only "but it is SOI", "but AMD changed plans" that I read is in some forums. Still waiting for something more than a mere forum rumour by anonymous posters.

This is not different from people claiming we will see SR FX 8-cores chips in 2014. I am still waiting an answer to my question of if 4m SR is coming, as that people believe, then why it is not used in servers...

 

[8350rocks]

parroting isn't dependent on the source's reliability. fanboys parrot because the content is usually suitable with their delusion. for example, now a days the russian website that benched crysis 3 usually gets parroted every time gaming is brought up without any deeper research into bench settings while more detailed crysis 3 benches are available on the net e.g. toms' own, techreport, pclab etc.
edit: to be fair, i coulda presented an intel-specific example, but that'd prolly cover nearly all benches. :lol:

To be fair, no one else benched core usage % per core on CPUs in the game.

@juanrga:

1.) I would bet you, especially since XBone won't be supporting PhysX, for every 1 console game using PhysX, there will be 40-50+ games using a better supported 3rd party physics engine that runs better on AMD hardware. PhysX is a non-factor in the consoles.

2.) S|A just did a piece extrapolating the XBone will be ~1.9-2.0 GHz...not 1.6 like people thought. I guarantee you Sony will be roughly the same if not slightly faster.

3.) Your source comes from a statement made give or take a year ago. It came from a guy who was fired...and he was relaying information from another guy who was fired...and you think this is accurate because...?

4.) That was a rhetorical question...just thinking out loud really. 14nm bulk FinFET is going to be ridiculously hard to achieve, hence the Broadwell delay. The PR nightmare it represents is another matter...but many sources have confirmed that they aren't doing so well getting 14nm to work. Other sources would even say their 22nm FinFET on bulk was actually lackluster compared to the projections...though that would mean I am right about bulk vs. SOI and you'd be wrong about the delay from AMD for Kaveri. Thus, I don't anticipate you'll acknowledge that.

 

[juanrga]

@juanrga:

1.) I would bet you, especially since XBone won't be supporting PhysX, for every 1 console game using PhysX, there will be 40-50+ games using a better supported 3rd party physics engine that runs better on AMD hardware. PhysX is a non-factor in the consoles.

2.) S|A just did a piece extrapolating the XBone will be ~1.9-2.0 GHz...not 1.6 like people thought. I guarantee you Sony will be roughly the same if not slightly faster.

3.) Your source comes from a statement made give or take a year ago. It came from a guy who was fired...and he was relaying information from another guy who was fired...and you think this is accurate because...?

4.) That was a rhetorical question...just thinking out loud really. 14nm bulk FinFET is going to be ridiculously hard to achieve, hence the Broadwell delay. The PR nightmare it represents is another matter...but many sources have confirmed that they aren't doing so well getting 14nm to work. Other sources would even say their 22nm FinFET on bulk was actually lackluster compared to the projections...though that would mean I am right about bulk vs. SOI and you'd be wrong about the delay from AMD for Kaveri. Thus, I don't anticipate you'll acknowledge that.

1) As I said before:

How many people will use Havok vs PhysX? I don't known and sincerely I don't care. The point is that contrary to your repetitive claims, the PS4 has PhysX support.

Therefore I will decline your bet.

2) Charlie speculated about the XboxOne being clocked at ~1.9GHz. I also read rumours that the PS4 could be finally clocked at 2.0GHz. That are all rumours. I solicited you a proof and you are giving me none. Moreover, those 1.9--2.0Ghz are very far from your claim that the PS4 is clocked at 2.75Ghz. The 2.75GHz listed in the FCC file was the GDDR5 memory clock.

3) I have more than one source. Read my reply to noob2222. If you have one source that says that Kaveri is being manufactured on SOI process, please share it with us.

4) Four weeks ago Intel said:

14 nanometer on-track to enter production by the end of the year

We are on track to start production on our 14 nanometer process technology in the back half of this year

In any case, I hope that different foundries will catch Intel around the 14nm and Intel will lose the advantage.

 

[8350rocks]

1) As I said before:

How many people will use Havok vs PhysX? I don't known and sincerely I don't care. The point is that contrary to your repetitive claims, the PS4 has PhysX support.

Therefore I will decline your bet.

2) Charlie speculated about the XboxOne being clocked at ~1.9GHz. I also read rumours that the PS4 could be finally clocked at 2.0GHz. That are all rumours. I solicited you a proof and you are giving me none. Moreover, those 1.9--2.0Ghz are very far from your claim that the PS4 is clocked at 2.75Ghz. The 2.75GHz listed in the FCC file was the GDDR5 memory clock.

It was not speculation, based on the hardware specs, you can reconcile the clockspeed of the CPU.

3) I have more than one source. Read my reply to noob2222. If you have one source that says that Kaveri is being manufactured on SOI process, please share it with us.

http://www.techpowerup.com/forums/showthread.php?t=170062

You can't follow the link to the SOI conference anymore, however, shortly after AMD announced going bulk over a year ago, they then fired a bunch of people (those that spoke of bulk included) and then they were talking to GF a month later about FD-SOI @ 28nm at the global SOI conference...you do the math.

4) Four weeks ago Intel said:

14 nanometer on-track to enter production by the end of the year

We are on track to start production on our 14 nanometer process technology in the back half of this year

In any case, I hope that different foundries will catch Intel around the 14nm and Intel will lose the advantage.

LOL...if you believe anything out of Intel these days you're crazy. If 14nm will be on time, why the hasfail refresh and delay of everything @ 14nm node? Because they want to make less money? This is the evil empire we're talking about...decisions to make less money don't happen...even if it means cheating/lying/skewing numbers to make more money.

 

[Cazalan]

LOL...if you believe anything out of Intel these days you're crazy. If 14nm will be on time, why the hasfail refresh and delay of everything @ 14nm node? Because they want to make less money? This is the evil empire we're talking about...decisions to make less money don't happen...even if it means cheating/lying/skewing numbers to make more money.

Intel is accelerating their lower power mobile products before desktop products. That includes 14nm ATOM and Broadwell to keep ahead of 64bit ARM chips due in 2014. The die shrink is more effective for power savings than it is for performance boosting.

 

[juanrga]

2) Charlie speculated about the XboxOne being clocked at ~1.9GHz. I also read rumours that the PS4 could be finally clocked at 2.0GHz. That are all rumours. I solicited you a proof and you are giving me none. Moreover, those 1.9--2.0Ghz are very far from your claim that the PS4 is clocked at 2.75Ghz. The 2.75GHz listed in the FCC file was the GDDR5 memory clock.

It was not speculation, based on the hardware specs, you can reconcile the clockspeed of the CPU.

3) I have more than one source. Read my reply to noob2222. If you have one source that says that Kaveri is being manufactured on SOI process, please share it with us.

http://www.techpowerup.com/forums/showthread.php?t=170062

You can't follow the link to the SOI conference anymore, however, shortly after AMD announced going bulk over a year ago, they then fired a bunch of people (those that spoke of bulk included) and then they were talking to GF a month later about FD-SOI @ 28nm at the global SOI conference...you do the math.

4) Four weeks ago Intel said:

14 nanometer on-track to enter production by the end of the year

We are on track to start production on our 14 nanometer process technology in the back half of this year

In any case, I hope that different foundries will catch Intel around the 14nm and Intel will lose the advantage.

LOL...if you believe anything out of Intel these days you're crazy. If 14nm will be on time, why the hasfail refresh and delay of everything @ 14nm node? Because they want to make less money? This is the evil empire we're talking about...decisions to make less money don't happen...even if it means cheating/lying/skewing numbers to make more money.

Before I forget this: your claim that XboxOne has not PhysX support is also incorrect

http://nvidianews.nvidia.com/Releases/NVIDIA-Announces-PhysX-Support-for-Microsoft-Xbox-One-Game-Console-997.aspx

Charlie's claim is his speculation, that is why he writes "if", before obtaining the 1.88GHz. The speculation must be very solid, but continues not being an official confirmation. Can we differentiate between speculation and official claims?

I believe that my speculation about Kaveri clocks and GFLOPS is solid, but it is my speculation and that is why I never said that it is an official claim by AMD. Do you understand the difference?

After saying you that speculations in forums by anonymous posters don't count as source, you gave me one of those!!!! And I am warned that I "can't follow the link to the SOI conference anymore". LOL And add to that I know to the poster and his record of posting nonsense/BS about Kaveri/Steamroller...

With that defensive attitude you couldn't believe anything said by Intel/AMD/Nvidia those days. Or do you believe Nvidia 'explanations' about Tegra sales? Or do you believe AMD 'explanations' about Kaveri's non-delayed delay? Or do you believe AMD saying one day that PS4 has hUMA and negating it the next day? LOL

So far as I know 14nm is on track and scheduled 14nm products are coming next year. I have given you above two confirmation made by Intel employees in a recent meeting. Maybe Broadwell delay has other causes

http://www.xbitlabs.com/news/cpu/display/20130829163840_Intel_Changes_Plans_Core_i_Broadwell_Processors_Will_Be_Available_in_LGA_Packaging_But_There_Is_a_Catch.html

 

[8350rocks]

2) Charlie speculated about the XboxOne being clocked at ~1.9GHz. I also read rumours that the PS4 could be finally clocked at 2.0GHz. That are all rumours. I solicited you a proof and you are giving me none. Moreover, those 1.9--2.0Ghz are very far from your claim that the PS4 is clocked at 2.75Ghz. The 2.75GHz listed in the FCC file was the GDDR5 memory clock.

It was not speculation, based on the hardware specs, you can reconcile the clockspeed of the CPU.

3) I have more than one source. Read my reply to noob2222. If you have one source that says that Kaveri is being manufactured on SOI process, please share it with us.

http://www.techpowerup.com/forums/showthread.php?t=170062

You can't follow the link to the SOI conference anymore, however, shortly after AMD announced going bulk over a year ago, they then fired a bunch of people (those that spoke of bulk included) and then they were talking to GF a month later about FD-SOI @ 28nm at the global SOI conference...you do the math.

4) Four weeks ago Intel said:

14 nanometer on-track to enter production by the end of the year

We are on track to start production on our 14 nanometer process technology in the back half of this year

In any case, I hope that different foundries will catch Intel around the 14nm and Intel will lose the advantage.

LOL...if you believe anything out of Intel these days you're crazy. If 14nm will be on time, why the hasfail refresh and delay of everything @ 14nm node? Because they want to make less money? This is the evil empire we're talking about...decisions to make less money don't happen...even if it means cheating/lying/skewing numbers to make more money.

Before I forget this: your claim that XboxOne has not PhysX support is also incorrect

http://nvidianews.nvidia.com/Releases/NVIDIA-Announces-PhysX-Support-for-Microsoft-Xbox-One-Game-Console-997.aspx

Charlie's claim is his speculation, that is why he writes "if", before obtaining the 1.88GHz. The speculation must be very solid, but continues not being an official confirmation. Can we differentiate between speculation and official claims?

I believe that my speculation about Kaveri clocks and GFLOPS is solid, but it is my speculation and that is why I never said that it is an official claim by AMD. Do you understand the difference?

After saying you that speculations in forums by anonymous posters don't count as source, you gave me one of those!!!! And I am warned that I "can't follow the link to the SOI conference anymore". LOL And add to that I know to the poster and his record of posting nonsense/BS about Kaveri/Steamroller...

With that defensive attitude you couldn't believe anything said by Intel/AMD/Nvidia those days. Or do you believe Nvidia 'explanations' about Tegra sales? Or do you believe AMD 'explanations' about Kaveri's non-delayed delay? Or do you believe AMD saying one day that PS4 has hUMA and negating it the next day? LOL

So far as I know 14nm is on track and scheduled 14nm products are coming next year. I have given you above two confirmation made by Intel employees in a recent meeting. Maybe Broadwell delay has other causes

http://www.xbitlabs.com/news/cpu/display/20130829163840_Intel_Changes_Plans_Core_i_Broadwell_Processors_Will_Be_Available_in_LGA_Packaging_But_There_Is_a_Catch.html

Notice, they say "support"...again...anyone developing specifically for consoles would be an idiot to use a less optimal physics engine. The only edge case I could see would be a PC game that was later ported to consoles, but even then...that doesn't make a lot of sense.

http://www.dailytech.com/Report+Intel+Delays+14+nm+Broadwell+Schedules+Haswell+Refresh+for+2014/article31770.htm

http://www.pcper.com/news/Processors/Intel-Not-Releasing-14nm-Broadwell-CPUs-Next-Year-Haswell-and-Haswell-E-Will-Dominat

According to insider news from June this year...Intel is not on track.

Also, I don't believe many things companies say to a public forum. However, I believe what they're telling developers...as that doesn't get spread to the public typically.

 

[8350rocks]

Well intel being 6-12 months ahead of amd it doesn't matter much to Intel fanboys if they are on track or not. To be honest, Intel needs to hope and pray that AMD screw this up, because 14nm is really going to be terrible for them and they have no hope of competing beyond that with an inferior wafer and expensive process to try to compensate.

FIXED!

 

[griptwister]

Okay when intel haswell 8 core comes out doubling the fx9590 processor performance at half the power consumption in 2015 by your reckoning it will be able to be as good lol. I think if amd reached that performance as well at that efficiency it will be until probably 2016 or maybe even 2017 or never.

Predictions:

Hasfail 8 core will be 10-20% faster on average above 6 core i7s

Hasfail 8 core will be clocked lower and consume more power than a FX 8350

Hasfail 8 core will not likely hit anything past 4.4Ghz unless it's on nitrogen

Steamroller 8 core will be with in 5-10% of hasfail'd 8 core in 70% of gaming benchmarks and on par in the rest of gaming benchmarks.

 

[juanrga]

Kaveri is 28nm @ Glofo

This sources the same japanese journalist. The relevant quote (my translation):

Born in in the foundry of Global Foundries, the project Steamroller / Kaveri will point again on them to keep development costs, but leaving the PD-SOI node. In addition, says Yusuke the first sample (BULK @ 28nm) seem to have reached the frequencies available and are already in possession of the motherboard manufacturers. The only uncertainties are determined by the type of production process HPP (High Performance Plus) recently entered service, and it should be used by a single user, for the moment, only AMD. GloFo, therefore, may not be interested in spending through time and money, considering that most of the gains they are doing are with the PP SLP (Super Low-Power) and LPH (low power, high performance), frustrating the efforts of the AMD about the time to market in mass or availability of these APU. The future of the steamroller, therefore, passes by Kaveri. If the association with GloFo proves to be fruitful, Steamroller would be marketed soon. Otherwise prepare for the worst. Richland, at this time, has a precise function: if Kaveri should be delayed (or worse, canceled), AMD would have an APU at least until mid-2014. From here backwards compatibility with FM2+ motherboards.

And this is still more interesting: No more evolution of AM3+ socket (again my translation):

A developer who works closely with manufacturers of motherboards has informed us that the development of additional solutions based on Socket AM3 + have completely stopped, then we can consider the Asus Crosshair V Formula and the AsRock 990FX Extreme9 how the highest evolution of the platform.

 

[8350rocks]

Kaveri is 28nm @ Glofo

AMD has not officially stated anything other than the one statement, which they will no longer confirm or deny at this point.

Don't you find that curiously odd? When AMD is asked directly about a previous statement and they're not commenting on it, or sidestep the question directly...what does that typically mean? That they're not intending to do that at all perhaps? How else would you read into that?

http://www.xbitlabs.com/news/other/display/20120205041713_IBM_Quietly_Starts_to_Make_Chips_for_AMD.html

That news is as old as your news about bulk silicon, and it says AMD has partnered off and on with IBM, who has the most advanced planar FD-SOI process in the world...

http://www.pcper.com/reviews/Editorial/GLOBALFOUNDRIES-2013-and-Beyond

That shows GF is already shipping 28nm HKMG SOI to ARM chip makers. So how is FD-SOI "not ready" yet?