@peter.clarke: you are too kind to WSTS in your comments above (reply to @FH1). It is hard to accept that it is a typo! Also, one can not use the Japan quake excuse for Feb & March sales (the quake repercussions were beginning to be felt in late March onwards).
@FH1: The industry does care about these reports but we do take it with a tonne of salt! When businesses shell out multiple $$000's of dollars to buy these reports, they expect the 'analysts' to have done their research with producers and consumers of semiconductor products. If the creators of such research reports do not bother talking to all players of the value chain and generate reasonably accurate predictions, they are going lose their credibility fast. WSTS is facing that harsh reality of credibility!
One would have liked to have thought 'sore thumb errors' like this were caught by the WSTS ... or does no-one there have any epmathy for numbers any more and what they are saying; and we just take everything at face value without question mis-leading the industry in the process? Maybe industry doesn't care either !!!!
Yes I understand the problem was down to one (or was it two?) companies' incredibly high reports of processor sales. Which were revised upon challenge...was it a typo?....a misplaced decimal point?
Is interesting that your model predicts above my range for Q3. But overall we seem to be compatible.
It also remains to be seen whether there will be any supply-side or demand-side chip market economic impact from the aftermath of the Japan earthquake of March 11.
Anecdotally there seems to have been not too much so far and it has not really impacted Q1 financial results.
Yet another major revision by WSTS; these are becoming far too regular. Adjustments were supposed to be for minor reporting errors; this is now complete sloppiness and incompetence, demonstating a level of lip-service and complete disregard for the information being submitted. It was mostly in logic and was an 18% 'adjustment' to the total market; for the company involved the 'error' must have been eye-watering. 'Errors' like these can't just be adjustments, it just shows no-one bothered to check.
Hi Peter - I thought that I would "chime in" with the latest forecast numbers as gleamed from running the Cowan LRA forecasting model on the just published (today, 5-04-11) WSTS March "actual" sales numbers. The updated sales forecast estimates are summarized below:
2Q11 = $78.012 billion;
3Q11 = $85.435 billion;
4Q11 = $84.081 billion;
and 2011 = $323.293 billion.
The corresponding 2011 sales growth estimate comes in at 8.37 percent.
These just updated forecast results are pretty consistent with the numbers you put forth above, that is, they fall "comfortably" within your quoted ranges.
Blog Doing Math in FPGAs Tom Burke 20 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...