Zen hasn't taped out yet

[Dresdenboy]

Edit: Ohh and one last thing,

The question was about Servers not Desktop, so Devinder Kumar may have only talked about SERVER parts

That's a point. The parts might still share the layout with HEDT parts, but as we don't have information about that, I stay on the sidelines about that.

 

[Dresdenboy]

This thread started exactly 1 year (365 days) after this one:

http://forums.anandtech.com/showthread.php?t=2411975



Is this going to be a yearly tradition ()

Nice. At least something to have fun with before XMas each year.

Maybe it's that:

Oh dear...surely this is a case of the CFO simply misspeaking?

28nm in 2016+ is just not going to be competitive.

 

[Boze]

You could power a nuclear reactor with the energy of all the hope in this thread.

HEDT Zen parts won't be available till Q4 2016 at earliest, and widespread availability won't be until sometime in 2017. I'll stop posting at AnandTech Forums if I'm wrong.

I dare the AMD fanboys to make the same claim.

 

[coercitiv]

You could power a nuclear reactor with the energy of all the hope in this thread.

HEDT Zen parts won't be available till Q4 2016 at earliest, and widespread availability won't be until sometime in 2017. I'll stop posting at AnandTech Forums if I'm wrong.

I doubt anyone has more optimistic expectations than very limited availability in Q4 2016, that should be viewed as a best case scenario, but this is not what's being implied in this thread: the object of discussion is that, according to selected parts in the transcript, Zen has yet to be taped out in the next "several" months, and it will take 12-14 additional months until early availability. That would easily put product launch in Q3 2017. (this info is corroborated from several answers in the transcript and in some cases we don't even know what questions were asked, in relation to which product or even which foundry)

This, however, is in direct contradiction with another claim from the same transcript in relation to Zen: "samples and customers for the validation of the product over the 2016 time frame and then the revenue ramp happening in 2016".

It certainly cannot be both: have revenue ramp in late 2016 and launch the product in the second half of 2017.

 

[mysticjbyrd]

I doubt anyone has more optimistic expectations than very limited availability in Q4 2016, that should be viewed as a best case scenario, but this is not what's being implied in this thread: the object of discussion is that, according to selected parts in the transcript, Zen has yet to be taped out in the next "several" months, and it will take 12-14 additional months until early availability. That would easily put product launch in Q3 2017. (this info is corroborated from several answers in the transcript and in some cases we don't even know what questions were asked, in relation to which product or even which foundry)

This, however, is in direct contradiction with another claim from the same transcript in relation to Zen: "samples and customers for the validation of the product over the 2016 time frame and then the revenue ramp happening in 2016".

It certainly cannot be both: have revenue ramp in late 2016 and launch the product in the second half of 2017.

Question that was asked,
"Okay, fair enough. Got a few more topics to get to before we end here. So, as Zen, the new x86-based processor; high-end. It's your return potentially into the data center/server area. What's the timing? And what kind of share do you guys…?"

Answer,
"Zen was a clean sheet design that started few years ago. We are in the final figure of executing and the milestone that you want hear us talk about is Zen tapping out, which should be over the next several months. And then putting samples in the hands of our customer and then starting portfolio of revenue in 2017."

Also note,
"And by the way, because we have this reuse approach for cores, you will see us with Zen cores in the high-end desktop first and then the server from our overall products standpoint"

There is a strong possibility that he was referring to the server version of Zen no being taped out yet. If you use that line of reasoning, then the two quotes above are not contradictory.

Another quote suggesting 2016,
Hans Mosesmann - Raymond James & Associates.
And then lastly, comment on the longer term gross margin, operating margin goals timing?

Devinder Kumar - Chief Financial Officer
I think the key is getting through 2016 continue to stabilize the computing and graphics business, commercial, professional graphics embedded and even the high-end desktops here at Zen Core should all be accretive from a margin standpoint. I’ll be very discipline on that standpoint. And then obviously get in the 2017, 2018, 2019 time frame...

He seems to mentions Zen Desktops in his 2016 list, and then goes on to talk about projections after 2016.

Here is another interesting quote,
Unidentified Analyst
[Indiscernible]

Devinder Kumar - Chief Financial Officer
We haven’t given the specifics but you know they taped out in the last few months...in the January earnings call when we announced Q4 results you could expect it to be more specific in terms of FinFET product shipments in 2016."

This one is near impossible to interpret, as we don't know the question. He could only be talking about GPUs, or GPUs and CPUs.

 

[Pilum]

I doubt anyone has more optimistic expectations than very limited availability in Q4 2016, that should be viewed as a best case scenario, but this is not what's being implied in this thread: the object of discussion is that, according to selected parts in the transcript, Zen has yet to be taped out in the next "several" months, and it will take 12-14 additional months until early availability. That would easily put product launch in Q3 2017. (this info is corroborated from several answers in the transcript and in some cases we don't even know what questions were asked, in relation to which product or even which foundry)

Kumar didn't say it "will take" 12-14 months, but that it "can take" 12-14 months. And that's easily possible; depends on bugs detected during silicon validation and any possible respins necessary to get parametric yields to acceptable levels. Remember that Broadwell production silicon was on the fifth stepping (IIRC; the TSX-fixed cores had the F-stepping, so I assume first shipping silicon was E-stepping). Intel didn't do these respins just for fun, but because it was necessary to get Broadwell production-ready.

Cycle time for wafers on 14nm may be three months (according to this article). So any functional/parametric problem will incur a delay of a whole quarter.

This, however, is in direct contradiction with another claim from the same transcript in relation to Zen: "samples and customers for the validation of the product over the 2016 time frame and then the revenue ramp happening in 2016".

It certainly cannot be both: have revenue ramp in late 2016 and launch the product in the second half of 2017.

Well, that's the optimistic view, nothing wrong with that. If the tapeout silicon is bug-free and has acceptable parametric yield, revenue ramp should happen in 2016. But if they have to do several respins, 3Q17 for market introduction is also possible. We don't know. And if they're really just taping out now, AMD doesn't know either and they won't know for several months.

 

[LTC8K6]

"the milestone"

To me that indicates Zen itself, not just a version of Zen.
Not the server Zen, but the first Zen.

The milestone is the first tape out, not the second or third, imo.

 

[mysticjbyrd]

"the milestone"

To me that indicates Zen itself, not just a version of Zen.
Not the server Zen, but the first Zen.

The milestone is the first tape out, not the second or third, imo.

That seems like a flimsy argument based on one word. And if you assume he is talking about Zen in general, then how do you explain the later contradictory statements?

As an investor, wouldn't they be more interested in the Server production?

 

[LTC8K6]

That seems like a flimsy argument based on one word. And if you assume he is talking about Zen in general, then how do you explain the later contradictory statements?

As an investor, wouldn't they be more interested in the Server production?

Those investors wouldn't wonder why HEDT is coming out first if server production is where the money is?

Overall, not a whole lot of sense in these Zen threads.

If AMD wanted us to know that some big milestone had been reached with Zen, I'm sure they would make a clear marketing announcement. They would want buyers to begin to consider Zen as soon as possible, so they don't make a purchasing decision that does not include Zen as a possible CPU.

If HEDT comes out in 2016, that's fine with me. Bring it on.

I'm not expecting it, and I like pleasant surprises.

 

[DrMrLordX]

He specifically referred to HEDT, not maintream APUs.

You do know that there will be Zen APUs aimed at HPC functions and (maybe) HEDT right?

HEDT Zen parts won't be available till Q4 2016 at earliest, and widespread availability won't be until sometime in 2017. I'll stop posting at AnandTech Forums if I'm wrong.

I dare the AMD fanboys to make the same claim.

You won't have to stop posting. That schedule you just described is the same one AMD has been describing all along. You're right. Zen will be a limited launch in 2016 as HEDT parts. Or, at least, that's how it's looking to me.

 

[mysticjbyrd]

The more I read, "you know the milestone that you want to hear us talk about is Zen taping out, which should be over the next several months", the more I think he is talking about the dialogue regarding the tape out will be carried out over the next few months.

 

[Arachnotronic]

You do know that there will be Zen APUs aimed at HPC functions and (maybe) HEDT right?

Dude, read the transcript. First CPUs based on Zen will be HEDT (Summit Ridge) then server CPUs coming later.

 

[mysticjbyrd]

....

WCCFtech actually came to the exact opposite conclusion based on that article.

Zen 2016 confirmed, according to them.

http://wccftech.com/amd-confirms-zen-coming-highend-desktops-2016/

 

[ShintaiDK]

wccftech always comes to the conclusion with most site clicks

 

[mysticjbyrd]

Haha true enough

 

[Arachnotronic]

WCCFtech actually came to the exact opposite conclusion based on that article.

Zen 2016 confirmed, according to them.

http://wccftech.com/amd-confirms-zen-coming-highend-desktops-2016/

Yeah, everything is "confirmed" according to WCCFTech

 

[looncraz]

The quoted time indeed sounds too long. Unless AMD has cut down the number of staff working in the CPU teams in half, it shouldn´t take more than ~ 2 - 3 quaters.

I think people are confused because they treat "tape-out" as a single event on don't realize it is a process of multiple tape-outs.

1. Initial Prototype Tapeouts (PLURAL)
- Once achieved, prototype silicon is usually three months away
- Once prototype silicon is on hand, testing informs modifications
- Almost always not using the full chip design

2. Engineering Sampling Tapeout(s)
- Usually five or six months after the prototype tapeouts begin
- Usually always the first full-chip design for validations

3. Production Revision Tapeout(s)
- Potentially the final steppings, addresses issues found in ES silicon
- This usually takes three months (now 9 months total)

4. Production Validation Tapeout
- Should be the final stepping, usually two or three months after step 3
- We are around the one year mark now.

5. Initial Production
- Provided #4 works smoothly production begins
- This is generally a month after #4, sometimes two

I think we are between Step 1 and Step 2, which means we have 9 to 11 months to go, barring any unforeseen occurrences.

 

[Dresdenboy]

wccftech always comes to the conclusion with most site clicks

Shouldn't that be an anti-AMD stance then?

 

[Face2Face]

Can't wait that long... 5820K incoming!

 

[cytg111]

You could power a nuclear reactor with the energy of all the hope in this thread...

You could power a nuclear reactor with the energy...

You could power a nuclear reactor.

I cant quite put my finger on it, but maybe

reverse the polarity of the neutron flow

 

[ShintaiDK]

Shouldn't that be an anti-AMD stance then?

More clicks in hyperboles, always. And they can create a line of articles out of it. Seen over and over.

 

[Systems analyst]

Also see the other quote:



So they say a couple of months until tape-out, but then they say it's 12-24 months for product shipment. Not sure if there's anything wrong with my math, but given 2015 is almost over, I don't see how they could be claiming 2016 launch...

No, he said that it was taped-out a few months ago, and that usually a product is available 12-14 months from that date.
He also said that they will will be 'tapping out' in a few months; tapping out is not the same expression as taping-out; it means finishing, in this context.

 

[maddie]

Anyone else here find the language in the transcript a bit confusing?


As an example:

"Hans Mosesmann - Raymond James & Associates.Okay. Going more into the graphics side of the business if I will talk about virtual reality, okay to say new Skylake and your profits are kind of creating an environment with visit there is an upgrade type of potentially with on the GPU side, what specifically are you guys doing to kind of benefit from that, that your…"

 

[Fjodor2001]

No, he said that it was taped-out a few months ago, and that usually a product is available 12-14 months from that date.
He also said that they will will be 'tapping out' in a few months; tapping out is not the same expression as taping-out; it means finishing, in this context.

Yeah, that also struck me as odd. Is the transcript writer and/or CFO really not familiar with the concept "taping out" so they by mistake said "tapping out"? Or did they in fact mean "tapping out", indicating something like the tap being turned on at the factory to start producing ES chips!?

Kind of confusing all of this I have to say. A clarification from AMD would be welcome.

 

[Boze]

I cant quite put my finger on it, but maybe

Reactors gotta have fuel, buddy... these people's hopes are potent than the fuel rods on the Naval submarine I was on for three years. That's my point.