Rumor: Semi Accurate: AMD kills Wichita and Krishna

[dangerman1337]

Looks like the bad news won't stop rolling in for AMD if this is true D:: http://semiaccurate.com/2011/11/15/exclusive-amd-kills-wichita-and-krishna/

SemiAccurate has been hearing that the rumors that AMD’s (NYSE:AMD) Witchita and Krishna are on the chopping block for a few weeks. Over the last few days, that has been confirmed by several different sources. Yes, you read that right, the follow on to Brazos/Ontario/Zacate has been knifed, there will be no 28nm Wichita and Krishna. Why? No one is saying definitively, but we would guess that it is due to timing of GloFo’s 28nm SHP process. As you can see from the picture below, taken from a GTC 2011 slide, the process is set for production in mid-2012. That means there won’t be significant volumes until Q3/2012 for a chip that is due to be replaced at the end of 2012.

GloFo process roadmap
This means that Wichita and Krishna would have a shelf life of ~6 months, likely less. No OEM is going to pick that part up and design machines around it for such a short time, it isn’t financially viable. Brazos is already set for a mild tweaking to get a little more speed and little less power use, so that will have to hold the fort for an additional six months.
If things go well, the additional resources needed to bring Wichita/Krishna to market will be diverted to the successor, and that will in turn be pulled in. This may lead to the post Wichita/Krishna part in Q4, maybe a bit before, and all will be happy. If not, things go pear shaped, people leave, and OEMs send impolite emails to their sales contacts. We won’t know which way it goes for a bit longer, but either way, the successor to Brazos is now quite dead.S|A

Hopefully they haven't killed it off but with recent events I wouldn't be surprised. Also it looks like AMD will be jumping to 20nm rather than 22nm for desktop CPUs.

 

[zlejedi]

Hmm wait so are they killing only succesfull part of their lineup ?

 

[Arkadrel]

Hmm wait so are they killing only succesfull part of their lineup ?

^ this.

It doesnt make sense for me, why would they kill off their small APU's?

When they have a cpu thats faster than Intels Atom in this area? with better graphics? and they have a per/watt advantage (how often can AMD actually say that? Bobcat was a exception to the rule).

The bobcat APUs where great.... (compaired to intel offerings)

No doubt adding more core's and throwing in a tiny bit better GPU would allow them to keep that segment.
I hope they have a good reason for doing it.

.....we would guess that it is due to timing of GloFo’s 28nm SHP process.

GloFo's fault <.< to late with things.

 

[sm625]

Hmm wait so are they killing only succesfull part of their lineup ?

Probably because they realized that they are only saving 20&#37; vs a sb celeron which is twice as fast. They realized that IB celeron is going to be 3x as fast and use less power than an E-350. And thanks to oem build volumes, IB celeron notebooks will be cheaper just like sb pentium notebooks are cheaper right now.

We should see black friday single core sb celeron notebooks for under $200. Faster than E-350, for less money.

 

[Riek]

Well if the product will only be on the market 3-4months due to some delays before the next generation is released... there is no point in releasing the product given you won't make any profit out of it.

 

[ViRGE]

Why would they even be using GloFlo for 28nm in the first place? It's long been evident that TSMC would be well ahead of GloFlo with 28nm.

 

[Jionix]

No OEM would touch it if it's lifespan was ~6 months. Why waste the resources when you can focus on the replacement which would be right on it's heels? Actually, it's one of the smarter ideas AMD has had lately.

 

[happy medium]

No OEM would touch it if it's lifespan was ~6 months.

this^^^^

This must be the start of there new low/mid level cpu/gpu strategy, I'm guessing.

 

[Tuna-Fish]

Why would they even be using GloFlo for 28nm in the first place? It's long been evident that TSMC would be well ahead of GloFlo with 28nm.

Pricing. The TSMC 28 lines will be running red hot on just GPU and highest-end cellphone chips, both of which have higher markups than bobcat. GF 28nm should be available at roughly the same time when the TSMC lines start to have vacancies for cheaper chips.

(Assuming, of course, that it won't be horribly delayed, which is, of course, not a safe assumption to make.)

 

[ocre]

was Wichita and Krishna based on BD cores? As far as 6month life, thats BS. Fermi proves much. Unless they pull in the successor to Wichita and Krishna immediately then all of this is BS

 

[cbn]

was Wichita and Krishna based on BD cores?

They are based on Bobcat cores.

 

[Olikan]

was Wichita and Krishna based on BD cores? As far as 6month life, thats BS. Fermi proves much. Unless they pull in the successor to Wichita and Krishna immediately then all of this is BS

it's not BS, whichita is a cpu with it's southbridge. OEM have to build a netbook around it

fermi like any gpu, is just plug and play (compared to an actual system)

 

[Idontcare]

Why would they even be using GloFlo for 28nm in the first place? It's long been evident that TSMC would be well ahead of GloFlo with 28nm.

Yeah, I don't buy this rumor one bit. Charlie also thought AMD was going to buy ARM, and that Apple was dumping Intel. He comes up with some doozies over time. This is one of them IMO.

 

[quest55720]

I don't buy it. AMD needs these chips out for when W8 rolls out next fall. If they can't get out 28NM bobcats out for W8 launch then they might as well lock up the doors and close up shop.

 

[cbn]

it's not BS, whichita is a cpu with it's southbridge. OEM have to build a netbook around it

fermi like any gpu, is just plug and play (compared to an actual system)

I always liked the idea of a standardized netbook laptop chassis. (Now I realize it would also allow AMD to sell something with a short lifespan)

I guess the idea of "standardized notebook" has been tried before by others (including Intel and VIA) but it just doesn't work for x86 and its expensive OS (for various reasons).

Maybe the ARM partners would have a better chance at succeeding? (A series of nice netbook development boards that would help the process of Android Keyboard app writing? This followed by the ARM smartphones and lapdocks?).

 

[cbn]

It doesnt make sense for me, why would they kill off their small APU's?

When they have a cpu thats faster than Intels Atom in this area? with better graphics? and they have a per/watt advantage (how often can AMD actually say that? Bobcat was a exception to the rule).

The bobcat APUs where great.... (compaired to intel offerings)

No doubt adding more core's and throwing in a tiny bit better GPU would allow them to keep that segment.
I hope they have a good reason for doing it.

Maybe they are afraid of price compression in the notebook area?

P.S. Even Intel appears to be holding back the 32nm Cedar Trail atom.

 

[Tuna-Fish]

I guess the idea of "standardized notebook" has been tried before by others (including Intel and VIA) but it just doesn't work for x86 and its expensive OS (for various reasons).

Margins on desktop computers are really razor-thin, mainly because every component is standardized, so *anyone* can do the final assembly and sell the machine. Most desktop computers earn the seller no money on the hardware, only making even on the software.

Which is why the major manufacturers are scared to death about what notebook standardization could do to their margins, and thus would only participate if they *really* have to. The smaller notebook shops don't have the investment cash or the expertise to make it happen.

Maybe the ARM partners would have a better chance at succeeding? (A series of nice netbook development boards that would help the process of Android Keyboard app writing? This followed by the ARM smartphones and lapdocks?).

Why would they want to do this? Standardized ARM would have even worse margins.

The direction that all the manufacturers are trying to pull is the opposite -- more specialization, more 'secret sauce' that only *they* can sell, and on which they can pull in actual profits. The market will eventually trend towards standardization, because when a standard exists, it would be counterproductive to fight against it. But absolutely none of the laptop shops want to make it happen, and they are willing to fight it tooth and nail.

 

[cbn]

Why would they want to do this? Standardized ARM would have even worse margins.

The direction that all the manufacturers are trying to pull is the opposite -- more specialization, more 'secret sauce' that only *they* can sell, and on which they can pull in actual profits. The market will eventually trend towards standardization, because when a standard exists, it would be counterproductive to fight against it. But absolutely none of the laptop shops want to make it happen, and they are willing to fight it tooth and nail.

Sorry for the confusion.....when I say "ARM partners" I mean the actual hardware makers (such as Qualcomm, Texas Instruments and Nvidia).

The device makers (which I think you are referring to) can still make the custom mainboards and wrap custom laptop chassis around them.

P.S. When comparing this idea to x86 and Windows, I think ARM and Android makes more sense. With x86 the laptop mainboards were so large and swapping hardware was complicated by the fact a new OEM OS needed to be purchased (unless the user already owned an expensive Retail OS license).

 

[cbn]

Margins on desktop computers are really razor-thin, mainly because every component is standardized, so *anyone* can do the final assembly and sell the machine. Most desktop computers earn the seller no money on the hardware, only making even on the software.

Which is why the major manufacturers are scared to death about what notebook standardization could do to their margins, and thus would only participate if they *really* have to. The smaller notebook shops don't have the investment cash or the expertise to make it happen.

When you say small notebook shops, you mean the "Mom and Pops" computer repair shops right? Yes, they would need some help from the computer case manufacturers (Antec, Silverstone, Corsair, etc).

(With regard to form factor specification) Nice thing about ARM is that it looks like they are going in the reverse direction (slightly more power usage) compared to x86. So instead of the mainboards and chassis shrinking every generation, it is likely the boards/chassis could stay a stable small size for some time.

Anyway, I seriously doubt this will bring the device makers to their knees.

I just think some people would like to cheaply swap out hardware whenever the latest ARM SOC gets released. (For example, according to this Anandtech article....Past TI OMAPs were pin compatible. If other manufacturers do the same thing, upgrade processor/boards should be affordable enough to make)

The OMAP 3430 was built on a 65nm process (like Qualcomm's Snapdragon), while the 3630 is a 45nm shrink (like Apple's A4). Architecturally the two SoCs are very similar. They both use a standard ARM Cortex A8 CPU paired with an Imagination Technologies PowerVR SGX 530 GPU. The two SoCs fit in the same size package (12mm x 12mm BGA) and are ball compatible. If a customer wanted to, it could simply drop in a 3630 into an existing 3430 design with minimal engineering efforts.

P.S. "Standardized" ARM Laptop chassis might even be a good way for some traditional ATX case manufacturers to begin transitioning their product lineup into Smartphone "Lapdocks" and other type of smartphone "docks" at some later point.

 

[cbn]

I found the following over at AMD Zone. It might be somewhat related.

http://www.linkedin.com/pub/brad-burgess/26/aa9/93

The chief architect for Bobcat left AMD in August and now works for Samsung.

 

[ocre]

I found the following over at AMD Zone. It might be somewhat related.

http://www.linkedin.com/pub/brad-burgess/26/aa9/93

The chief architect for Bobcat left AMD in August and now works for Samsung.

Oh Man ,something at AMD isnt smelling good at all. these pieces are starting to look......discouraging!!!

 

[eternalone]

Sounds like someone is going out of business.

 

[Bill Brasky]

Yeah, that's not good news. It will be interesting to see how this affects the graphics department.

 

[opethfan]

Isn't GloFo supposed to opening 28nm Fab 8 in the coming.... soon?

 

[Arkadrel]

I found the following over at AMD Zone. It might be somewhat related.

http://www.linkedin.com/pub/brad-burgess/26/aa9/93

The chief architect for Bobcat left AMD in August and now works for Samsung.

seriously? wtf amd, guy like that should be a top priority to keep.

He helped design, a really effecient performance/watt cpu, with good little gpu attached that beat the crap out of what Intel had, in same segment, and you let him leave?

kick board members and overpayed top leaders before getting rid of engineers that know how to do designs like that. Those guys are the backbones of the company.

Hell... fire every marketing/PR guy around as well before the engineers, esp ones that seem to have a knack for putting out great products.