0.13micron from AMD: Problems Discovered?

by Anna Filatova
04/18/2002 | 04:39 AM

The more often I watch AMD’s achievements in terms of 0.13micron technology implementation, the more convinced I turn that the company is having some serious problems here. Partially these problems can be explained by the new manufacturing technology, however, I have every reason to suppose that no radical improvement of the situation will follow before the end of the year. And now a few words about the facts that made me sound so gloomy:

• The first and the most suspicious thing is the extremely low working frequencies of Athlon XP processors built on 0.13micron core. Look here: the fastest model in the today’s Athlon XP family makes 1.73GHz, and later on Athlon XP 2200+ is to come, with the core clock of 1.8GHz. And the first 0.13micron Athlon XP should also work at 1.8GHz core clock! Moreover, by the end of the year its frequency will get only 266MHz higher (Athlon XP 2600+ will work at 2.06GHz actual clock frequency). For a better comparison try to recall 0.13micron Celeron which frequency grew almost 1.5 times higher when they shifted from 0.18micron to 0.13micron and the Vcore got considerably lower at the same time: from 1.7V to 1.5V. And here comes the second reason...
• The second thing witnessing that AMD has some problems with 0.13micron manufacturing technology is a too high processor Vcore. As is known, it makes 1.65V, which is only 0.1V lower than that of 0.18micron Palomino core. Of course, it wouldn’t be quite correct to compare it with Intel’s manufacturing technology, however, together with all other evidence, high Vcore signals about not very high chip yields. By the way, the relative rawness of the manufacturing technology is a very poor excuse: the Vcore of the first 0.13micron Celeron CPUs was 1.475V and the chips could work at 1.4-1.5GHz core clock (that is almost 1.5times higher than the core clock of 0.18micron processors).
• And now let’s have a look at the yesterday’s announcement of the mobile Thoroughbred (see this news story). AMD started shipping 1400+ and 1500+ processors working at 1.26GHz and 1.33GHz correspondingly. Again, the clock frequencies are very low (even taking into account that these are mobile solutions).
• One more fact, drawing our attention: AMD refused to continue cooperation with Motorola Company in terms of new technology development. Now they are going to work with UMC on that. I don’t think they did it because they are building a new fab in Singapore together with UMC, they could still have both partnerships. It is most likely to be the 0.13micron technology developed by AMD and Motorola, which appeared not very good and AMD decided to find another partner in this respect.
• And the last thing. According to a new AMD roadmap (see this news story), 0.13micron Barton core has changed. Now it will be made not from SOI substrates (which used to be a direct mention of the joint AMD-Motorola technology) but from regular substrates of the silicon dioxide (UMC technologies do not know to use SOI). Here we can’t help recalling that in the end of Q4 UMC was going to start 0.13micron CPUs for AMD. No wonder that Barton core has been changed: it is most likely to be now oriented for UMC’s production lines. And these production lines seem to be pretty good, we should say, as AMD has increased Barton’s cache recently. I think that UMC has already made some trial Thoroughbred processors. And the outcome may have turned out so good that AMD decided to double the L2 cache size (up to 512KB).