InvestorsHub Logo
Followers 11
Posts 3036
Boards Moderated 0
Alias Born 12/01/2002

Re: Ixse post# 62792

Monday, 09/26/2005 2:02:31 PM

Monday, September 26, 2005 2:02:31 PM

Post# of 97586
Re: some attempted Itanium answers:
(1) NRE for mask sets... Yeah, this is typically included in the COGs line, which I pegged at $125/unit. You are right that with very low volume this could be an impact. Last I heard mask sets were $500K, and could easily be $1M these days. I don't think 300mm makes that much of a difference with S&R litho, but 90nm vs. 65nm will make a difference. Perhaps this is one reason the Montvale is now on 90nm rather than 65nm. At any rate, assume they do two steppings to get things right, and then get a back-up mask-set for production at $1M/set. They have been doing a new product every year so that puts this expense at $3M/year. With annual volumes of between 100K and 300K this puts a cost/unit adder of between $10 and $30... but I think my projected cost/unit of $125 is high enough to absorb this. Typically those first two mask sets would be billed to R&D so would be part of that $160K/year/engineer budget though. I have heard that Intel has a quick turn mask process for some low volume products so they do not need to order that back-up mask set. I think my calculations here are pretty worst case.

(2) why ignore the past? The question I was answering may have been different than your question. I was attempting to answer the question... "would Intel financials this quarter, next quarter, and next year be better or worse off if Intel decided to ax Itanium." The past is the past and the money is spent. You should not make current spending decisions based on what you have spent in the past... although many people, and companies, do.

(3) sharing of infrastructure... Intel has not developed a new Itanium chipset in quite some time, and will not develop any either. Their expense in this area is ZERO. It is certainly something for SGI, HP, and others, but not for Intel. IFF Intel was continuing to invest in chipset and board design for Itanium we would need to account for that expense, but there is no apparent expense in this area.

(4) Silicon process development expense... I believe if the Itanium guys approached the fab and asked for something custom in the process they would be laughed right out of the building. As I stated in #2, I was attempting to answer a specific question regarding Itanium financial viability at Intel. Intel will pay the same for silicon process development regardless of what happens with Itanium.

(5) Developer payments for software... I have not seen any current references to this... If they are, that number should certainly be subtracted out. My impression is this program is done, but I could be wrong on that. I would certainly expect Intel to stop spending money here before they EOL the productsmile

(6) Marketing costs... sure, how many marketing engineers does Intel have for Itanium? Have you seen any Itanium advertising? I may have missed it, but I have seen some Xeon but nothing for Itanium... but then again I don't read many server rags. My impression is that Intel has now established a few outlet channels for Itanium and is now pretty hands off... Again, I may be wrong about that.
--Alan


Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent AMD News