Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Elmer who wrote (74365)	10/6/1999 9:06:00 PM
From: Yougang Xiao	Read Replies (3) \| Respond to of 1572773

Q3 CC Notes: Attendees - Jerry, Fran Barton, Rob Herb and Ben Anixter GM is 28%, up 5 points from Q2 18 millions is from Vantis fab arrangement, will use this amount going foward. 377 millions cash available after paying 87 m BofA credit facility Expect Q4 GM increase in Q4 R&D grow slightly over Q3 MG&A to increase about 20 m due to Athlon merchandising and other cost Restructuring could be from 0 to 10 m Interest expense to be flat Income from FASL to be breakeven Shares to increase to 149 m Capital spending 650 m for the year, down slightly than previous anticipated. Q4 breakeven point to approach 850 m. Booking for Flash remain strong Book to ratio greater than 1 Blended ASP of 65 Blended ASP of Athlon is close to 300 Speed distribution contine migrate to higher Demoed 800 0.18 yesterday, 0.18 athlon will contritute revenue this Q 0.18 athlon in production By month end will cease 0.25 athlon wafer start, going forward will be all 0.18 continue to expect 750 intro in Q1 400 mhz will be minimum for Q4 highly confident of producing 1 m athlon in Q4 on site personel in taiwan to ensure shipable output (mobo) good chance to achievie 1 m or more cumutively shipment of athlon by year end. Assuming 5 m cpu shipment in Q4 NON cpg grow 10% in Q4 Revenue in Q4 COULD exceed 800 m unable to give profit definate guidance at this time due to supply chain issue in taiwan Q&A Blended ASP in Q4 will be higher Athlon price to come down somewhat Will have enough MOBO to achieve AMD goal (cumulative 1 m athlon shipment) a month from now at analyst meeting will have more info on Taiwan issue. absolutely not selling flash division Embeded processor business will remain at AMD, not included in COMM sale Flawless execution of 0.25 on athlon 0.18 ramp we can do well, twice as many good die on 0.18 vs. 0.25 COMM group value to be decided by market AMD not manufacturing any chipset of our own, by UMC Revenue on chipset is about 13 m for Q3, maybe 15 to 20 m in Q4, produced by UMC produced over 350K Athlon in Q3 shipped 200K athlon in Q3 to ship 800K athlon in Q4. chipset is produced on 0.35 at UMC athlon wafer start to pass that of K6 this Q may allocat fab 25 to do flash. have a number of top oems for athlon in next 90 to 120 days on track for 1 ghz by end of 00 on track for revenue of Dresden athlon for Q2, 00 100 rebate is not AMD program Dresdon will be production facility until 2nd Q of next year, initial produciton in Q2 FAb 25 at 80% capacity 850 m breakeven is for Q4 only highly confident to have arrangement of absorb cost concerning Fab 30 right partner will not hurt FAB 30 to produce maxime athlon speed distribution of athlon is terrific Q3 athlon asp is slighterly under 300 bucks flash is sold out for q4. fab guys are performing magnificantly Dresdon depreciation at 25 m per quarter starting next year from historically point of view, 1 m and 1.5 m athlin in Q4 and Q1 00 reasonable sledgehammer - simple tool delivering great power dresdon partnership form - can not coment at this time.

To: Elmer who wrote (74365)	10/6/1999 9:24:00 PM
From: greg nus	Read Replies (1) \| Respond to of 1572773

Last May when speaking to Sanders I suggested be consider keeping CPU and flash and jetison everything else while the market was hot for spinoffs, and reinvest all the proceeds in the two core holdings. He seem suprised when keeping flash was suggested but now having reflected he was calculating. He asked me why keep flash? it's been disapointing performer. I shot back to him, inspite of the weak performance Flash held the best promise for the future. AMD is the industry Leader in Flash and with the growth in Cell phone communication devises and digital camera Flash demand has not yet hit it's stride or full potential. With the announcement of the Communications Group up for sale it appears Sanders is commited to continuing in the right direction. Amd is however not selling all the product lines of Communications group. He intends to keep the products that will well position AMd for intgrated solutions in communications through a combination of CPU flash and communication chip capability. This is a further departure from following Intel.

To: Elmer who wrote (74365)	10/7/1999 5:03:00 AM
From: Haim Barad	Read Replies (2) \| Respond to of 1572773

Question: Why does the use of architectural extensions (e.g. SIMD extensions) NOT fit into the spirit of benchmarks like SPEC... but other architectural extensions (e.g. additional FP unit) DO? I'm not saying that the optimizations do or don't use SIMD extensions... I'm just saying that MMX and SSE are PART of the Intel Architecture... just like the add instruction. Haim

To: Elmer who wrote (74365)	10/7/1999 3:11:00 PM
From: Steve Porter	Respond to of 1572773

Elmer, Ok, then you raise a question I don't have an answer for, except to say that I don't think Intel needs to use those instructions to see the kind of increase we are seeing for the integer operations. No I don't believe Intel would need to do that for their integer scores either. However, if you are going to do it for FP, you might as well do it for integer as well. As for their being nothing stopping AMD from doing it, I also agree with you. Just stating that comparing SIMD benchmarks to non-SIMD benchmarks is kind of like me quoting you the price of a live t-rex (i.e. pointless) Steve