I still remember the surprise with which I was received the arrival of Intel Pentium processors, back in 1993 (when I was still looking to buy a 386). This was for many reasons of course, but there is one that I do remember anecdotally, and that is that until they were announced, I expected the next generation to be… I’m sure you guessed it, yes, the 80586 or the i586 . Intel actually abandoned this nomenclature in order to commercially register the name (something that could not be done with number sequences).

Four years later, in 1997 came the leap to the Pentium II, which meant abandoning (temporarily, as we saw later) the classic format of a chip inserted into a socket for a cartridge designed to improve heat dissipation, which connected to the motherboard through a slot. It didn’t take long two years later came the Pentium III which restored the classic chip-in-socket format that still prevails today.

At that time, Intel was already very fast in terms of jumping from one generation to another, so only a year later At the end of 2000, the first Pentium 4 appeared on the market. So what could we expect at that time? Of course the next generation will come in 2001 (or 2002 at the latest), which of course would be the Pentium 5, right? Well, that was to be expected, but as you probably already know, it didn’t happen.

Intel Pentium 4 Prescott. Performance yes, but at what cost of heat… Image: JulianVilla26

Instead of, The Pentium 4 has brought together several different architectures over the years: Willamette, Northwood, Northwood (Extreme Edition) and Prescott. Why did Intel choose not to increase the number with each of these jumps? There are several theories about it, such as the fact that it mistakenly, and because of the speed with which it jumped from version to version between the Pentium and Pentium 4, found users thinking that the number was a performance rating system, not a generation indicator.

Be that as it may, that’s the point The Pentium 4 Prescott, released in early 2004, was already facing some high performance issues and directly connected, heat generation. The peak of the offer of its generation was the Pentium 4 570J at 3.8 GHz, although Intel initially planned more powerful models, four and more GHz, in the end the mentioned problems made it impossible.

And then came 2006, but as expected, there was no new Pentium 4 architecture or, even more expected, a jump to Pentium 5. Instead, Intel formulated a paradigm shift that proved more than successful over time. I mean, of course, the production of processors with more than one core. Intel Core Duo was just born, a dual-core implementation. Residue? I’m sure you already know, first adding more powerful cores in every generation of Intel Core up to the eleventh and jumping to a hybrid architecture with Alder Lake.

And what happened to the Pentium 5?

What happened? Why did Intel abandon the Pentium and switch to a multi-core architecture? A video recently posted on the Fully Buffered YouTube channel reminds us of some already known facts and tells us another very interesting one Intel’s Pentium 5 plansand why it never ended up seeing the light of day, plus he showed us two chips from that lost generation and had first-hand evidence of them.

The first thing to note is that Yes, Intel was planning to move from Pentium 4 to Pentium 5a chip that was supposed to achieve spectacular speed 7 GHz. For context, let’s remember that the Intel Core i9-12900K has a maximum speed of 5.2 gigahertz and we’re talking about a Pentium 5 processor that would have hit the market about 15 years earlier. These were the days of brute force, and while there was talk of execution threads and such, the competition was determined by speed rather than performance.

With Prescott Intel, however, he found himself up against a wall, perhaps not insurmountable, but terribly difficult, of that temperature which, as we said before, prevented this generation from reaching the speed originally planned.

Now, more than a year before the Pentium 4 Prescott was released, Intel has already started working on two architectures, called Texas and Jayhawk. The first would be the successor to the Prescott and the one that would mark the jump to the Pentium 5, while the Jayhawk would mark the generational jump to the Intel Xeon. Back in 2003, Intel tentatively suggested that Texas would arrive in 2004, although this was later pushed back to 2005, earlier cancel both projects in May 2004.

Steve Fischer, one of the Intel engineers who worked on the Texas and Jayhawk projects that Fully Buffered spoke with, revealed quite a bit more about the project in this video, about challenges they faced and the reasons that finally made Intel back down, abandon the gigahertz race (its original goal was to reach 10 gigahertz in 2011) and focus on performance… but not forgetting speed.

After the cancellation of the Pentium 5, Intel opted for multi-core, which debuted in the Core Duo, which in turn found its continuation in the Core 2 Duo, the predecessor of the current Intel Core.

Fischer says the thermal issue was the deciding factor. At that time, the fastest processor in the house reached, as we already mentioned, 3.8 gigahertz and its TDP was 115 watts. So and with that technology andscaling up to 7 gigahertz would increase the TDP to around 250 watts, probably something else. I know that doesn’t seem like much today (TDP Core i9-12900K in turbo mode is 241 watts), but we’re talking about almost 20 years when processors with 90 nanometer lithography (Although eventually the jump was made to 65 nanometers).

I called this thing the “CPU Death Star” and half-jokingly reasoned that consumer acceptance of liquid-cooled chassis wouldn’t be much of a problem.

A TDP of 250 watts was something simply monstrous at the time for a desktop system, and while liquid cooling solutions already existed, they were still extremely rare and quite expensive, meaning they weren’t an option Intel could consider.

Canceling Texas and Jayhawk was definitely not an easy decision it involved a complex leap to a multi-core architecture and moreover to give up the 10 gigahertz milestone, a key element in AMD’s competition at the time. However, after analyzing all the possibilities, the engineering teams involved in the project concluded that they were facing a dead end.

As we mentioned earlier, this “stumbling” prompted the paradigm shift initiated with the Intel Core Duoand that brought us to the point where we are now waiting for the arrival of the Intel Core i9-13900K with its 24 cores (8 for performance and 16 for efficiency) and subsequent 32 threads, and it should already have been seen touching 6, 1 GHz.

A) Yes, The Pentium 5 never made it to market, but it existed and some silicon was even manufactured, such as the one obtained by the video creator. And maybe if Intel engineers had found a solution to the power and heat problem, today’s processors would be completely different. Of course, if two engineering teams with the resources available at the time decided it wasn’t possible, we’re not left with much room for imagination.