Associate
- Joined
- 7 Apr 2011
- Posts
- 92
There were always problems with CPU releases every so often. The first I can remember for the FVIV P54 (original Pentium) bug that got introduced in the Pentium 90 with its new stepping. They rectified that.
Intel also had to recall the Pentium III 1.13GHz (not to confuse with the Pentium III-S - an altogether fun chip mind you) for sub par performance under strict circumstances.
There were some others but they don't spring to mind. AMD is the same.
These are highly complicated devices, go draw yourself a 1cm by 1cm square on a piece of paper and then go imagine modern processors have anywhere in the region of 750 million - 1 billion transistors on a 5-6 layer design. Even the original Pentium had 3.1 million on a somewhat larger die size by the time it matured (the first were on 0.8 microns). The 486's (Intel's) were 1.2 million transistors.
Intel also had to recall the Pentium III 1.13GHz (not to confuse with the Pentium III-S - an altogether fun chip mind you) for sub par performance under strict circumstances.
There were some others but they don't spring to mind. AMD is the same.
These are highly complicated devices, go draw yourself a 1cm by 1cm square on a piece of paper and then go imagine modern processors have anywhere in the region of 750 million - 1 billion transistors on a 5-6 layer design. Even the original Pentium had 3.1 million on a somewhat larger die size by the time it matured (the first were on 0.8 microns). The 486's (Intel's) were 1.2 million transistors.