(Rumour) - Zen chips have already been tested, met all expectation, no significant bottlenecks

[Kobz360]

Tiny HPC Market? Google, Dell, and Oracle together buy more enterprise CPUs in a quarter than Intel sells consumer grade in a year. Now that enterprise purchase does come with other costs for Intel including support and 24/7 direct line access, but HPC is not small at all. Every big financial institution has its own supercomputer analyzing the financial markets in real time, making investment trades in handfuls of nanoseconds at a time. Every university worth noting I technology has a cluster. The best even have an IBM mainframe, hundreds of nodes on the small end to tens of thousands on the high end, and that's just schools. That's where Intel makes the bulk of its chip sales revenue and profit, not consumer computing.

Are you sure about that? In 2014, Intel's PC Client Group (Consumer products) made around 2.5x the revenue of their Data Centre Group (enterprise, servers etc.). Please correct me if I'm wrong but it seems pretty clear. 

 

Source - http://newsroom.intel.com/community/intel_newsroom/blog/2015/01/15/intel-reports-record-full-year-revenue-of-559-billion-generates-net-income-of-117-billion-up-22-percent-year-over-year

[patrickjp93]

Are you sure about that? In 2014, Intel's PC Client Group (Consumer products) made around 2.5x the revenue of their Data Centre Group (enterprise, servers etc.). Please correct me if I'm wrong but it seems pretty clear.

Source - http://newsroom.intel.com/community/intel_newsroom/blog/2015/01/15/intel-reports-record-full-year-revenue-of-559-billion-generates-net-income-of-117-billion-up-22-percent-year-over-year

Wait for 2015 results. 2014 was a slow datacentre year due to the recession and the slowdown in demand for SandyBridge and IvyBridge systems.

[Kobz360]

Wait for 2015 results. 2014 was a slow datacentre year due to the recession and the slowdown in demand for SandyBridge and IvyBridge systems.

I presume you expect a very significant increase in datacentre sales and/or significant decrease in desktop sales because the PCCG has been carrying Intel's chip revenue, even going back to 2008 and probably was before then too (haven't looked). 

 

Q1'15 results are showing DCG revenue picking up with consumer sales detracting, but I don't think it'll be to the point where the former will overtake the latter just yet

[looncraz]

I presume you expect a very significant increase in datacentre sales and/or significant decrease in desktop sales because the PCCG has been carrying Intel's chip revenue, even going back to 2008 and probably was before then too (haven't looked). 

 

Q1'15 results are showing DCG revenue picking up with consumer sales detracting, but I don't think it'll be to the point where the former will overtake the latter just yet

 

HPC is a $14.4B business for Intel, though, which is no small market.

 

For AMD, it's a $0 market for CPUs, but a pretty decent market for their FirePro GPUs.  Unless AMD can deliver dramatically more CPUs in a power envelope, at a lower price, that will not change with Zen by itself - which is exactly why everyone is interested in their HSA and server APUs.  THAT can be a game changer for AMD.  If not for the market, then certainly for the company.

[Tomsen]

I'm not twisting anything, you're saying that AMD won't be making an eight-core die, when the very evidence you give better supports the idea that it will have an eight core die.

 

CPUs are all about die size and profit.  If AMD were to stick to quad core consumer CPUs for FX CPUs, they may as well just close up shop and offer nothing.  No one will want to buy them unless they have a drastic advantage over Intel.

 

It costs dramatically more to create two dies (albeit not double), and AMD *MUST* have an eight-core die to even think about the server market.  This means they will, without a doubt, have eight core dies on AM4.

 

BTW, you are probably the only person on all of the internet trying to say that AMD will use MCM to achieve an octo-core design.

 

An eight core Zen die will not be much bigger than an eight core Bulldozer die, if at all.

It will be interesting to see the difference between the 4c model and 8c model, how the scaling goes, how clockspeeds will differ.

For most people, I think they will find the 4c/6c models just excellent. But for most people, outside enthusiast, will be doing just fine with an APU.

 

Will be interesting to see how they will price the SKUs against Intel. If AMD are been more relaxed towards productsegmentations, that will also be a benefit.

Considering that most server-CPUs will have atleast 8c, it will be unreasonable to use MCM for even the basic package. 

 

Also, I found something quite interesting.  I believe this is another Keller design... looks somewhat familiar to me...

 

 

http://images.anandtech.com/doci/7910/Cyclone.png

 

Look familiar?

Yes, it certainly have some similarities.

 

No, the evidence suggests chips built from 4-core dies. I think you're confusing a die for a complete package. I'm saying 2 4-core dies = 8-core CPU in MCM (Multi-Chip Module) form, similar to the 4 2-core dies that made up the FX series octal core SKUs respectively.

You can build a 32-core flagship from 8 separate dies. It's exactly how their Opterons are built currently too.

No I'm not the only person thinking it. It's a complete waste of resources for the consumer space to be fabbing 8-core dies, decreasing yield, increasing wafers run, etc. For enterprise it makes some sense, but frankly it's pointless outside of the APU designs where MCM designs make the GPU offloading exponentially complex. The experts at Semiwiki believe the same.

What "evidence". You are once again operating on marketing data. Nothing suggest what you are saying.

And you are wrong in regards to AMDs usage of MCM today. 

 

Good thing they aren't just fabbing them for the consumer space. Making your point moot.

 

 

If you're trying to say that looks like the Zen designs in any meaningful way, save yourself the trouble. It doesn't. Everyone's core diagrams put the same items in the same order and have for 8 years. The only variance is the width and count of pipelines or bits to a register or ALU.

Are you been serious? I really can't tell...

 

 

I'm not confusing anything, I know *exactly* what an MCM is, including most of the logistical problems which comes with them - and I'm talking down to the physics of electrons tunneling through the insulator due to thermal stresses and various material properties leading to line impedance changes and cross-talk limiting cross-die communications performance in an MCM configuration.  Though the math often still annoys me :angry: I do it for fun  :wub: [CONTINUED]

HPC market are usually in tight relations with their hardware providers. Hard to break them apart.

However, the 4c block, seems to easier integrate zen design into semi-custom designs.

 

Also, how well will zen SMT scaling be? Intel have been working on their implementation of SMT, and just recently (with haswell), it actually became decent.

Again, why would AMD fab the smallest configuration possible?

[patrickjp93]

I presume you expect a very significant increase in datacentre sales and/or significant decrease in desktop sales because the PCCG has been carrying Intel's chip revenue, even going back to 2008 and probably was before then too (haven't looked).

Q1'15 results are showing DCG revenue picking up with consumer sales detracting, but I don't think it'll be to the point where the former will overtake the latter just yet

The other issue here is Intel sees the E5 Xeon sales as 2 different things. If I buy a 2699v3 on my own, it's a consumer/PC sale.

[FlanK3r]

...its still long way to launch of Zen FX. So we will see reality in Q3/Q4 2016.