There are indications over AnandTech saying that 10 of Intel’s close partners in
The processor code-named Tejas is anticipated to be the last NetBurst chip and, unlike all its Pentium 4 predecessors, will be produced using two different fabrication technologies – at 90nm and at 65nm nodes. Surely, thinner version of Tejas may undergo some changes and get additional advantages, such as, faster processor system bus or larger secondary-level cache.
What we do know about the Tejas at this point is that it will have core-speeds at 4.0GHz and above, probably a faster Quad Pumped Bus, tangibly enlarged caches (24KB L1 cache, 16K uOps Trace Cache, 1MB L2 cache [probably 2MB for 65nm version]), a more efficient branch prediction mechanism, a new set of instructions known as “Tejas New Instructions” as well as improved Hyper-Threading organization. Obviously, information obtainable more than a year before product launch is not 100% reliable, though, we may now understand the idea of Tejas.
A question observers and specialists now ask is “isn’t 150W of power too lot for a CPU working at 2.80GHz and made using 90nm fabrication technologies?”
“In case we presuppose Intel Prescott processors and Intel Tejas processors to be made using the same technology process of 90nm and the same materials, it is not fully clear why the chip code-named Tejas dissipates stunning 150W of power,” said an X-bit labs’ analyst Ilya Gavrichenkov.
“If the power figures we have been given are indeed correct, one possible explanation would be that Tejas is indeed some variation of a multicore CPU. While it is unlikely that Tejas includes two discreet Prescott cores on die, there is a chance that the two cores (if they exist) could be sharing data caches and maybe other units,” said Anand Lal Shimpi from AnandTech.
“There are basically two more options here: either there are more execution units in Tejas and because of substantially increased transistor count the chip is so hot; or, Intel’s Tejas processor has some really advanced 64-bit enhancements to the x86 architecture inside. For instance, Intel could develop a technology more powerful and more capable in terms of scopes than AMD64 technology, but also backwards compatible with x86-64 standard proposed by AMD and therefore able to take advantage from software already available on the market by the time Tejas is available in commercial quantities,” X-bit labs’ Ilya Gavrichenkov added.
“There is also a possibility of some issues with material the Tejas is supposed to be made of. Intel may still conduct some experiments with dielectrics, while partners of the company who receive the chips now do not have any real interest in the chip’s current or eventual power consumption, as now this is only the beginning of Tejas-based products development” the X-bit labs’ analyst noted.
Check out more pictures of the upcoming Tejas microprocessor over here.