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.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Ali Chen who wrote (74188)	3/11/2002 8:08:11 PM
From: pgerassi	Read Replies (1) \| Respond to of 275872

Dear Ali: But both trials used the same exact hardware according to the disclosures. So why did the first trials take 186 seconds and the later 175 seconds. Evidently, the disclosures are incomplete or there is a lot of variability in the results. And three best of how many real trials are used. You did not run these tests personnally so you do not know what they corrected for, the methods used, etc. Yet you desire to project out to 3.0GHz from just two samples at 2.0 and 2.2. I take different results and get a different answer. Evidently using different sample sets gets different answers. And using different methods yields vastly different results. Thus both you and I are not have good enough models. Your problem stems from the fact that curve fitting the overall score does not yield a good projection. You must project each underlying score and then perform the calculation to the overall score. And even this makes the assumption that the starting data are good which I do not think is good enough for projections four times the base out. Yet your "projection" fails if you use the older data. It has the same hardware, same software, same setup, etc. I can make any curve fit the data and just the data. It does not mean that a projection is accurate or that the next sample not used to fit the data, fits. Your reply 2, does not validate anything. You make a slew of assumptions and then say it must be true. Does the example that started it, use real data? No, it doesn't. The values were an example to show how it works. I just say that the variences in the original data are not good enough for the projection you make. That is all. Pete