Linux Patch Pending To Fix Support For The Transmeta Crusoe CPU

[M.Bahr]

I still can't get over the end of the transmeta crusoe. Such a revolutionary design with big potential killed silently by big blue. Who knows how today's cpus would look like with that kind of technology.

By the way the pentium m's speedstep was based on a technology from transmeta. Intel had to pay a heavy fine for copying it without a license. Without speedstep and further improved iterations we probably would have heated our rooms with pentium 4s for a much longer period.

[muncrief]

I had a much better idea called "Thrust Reconfigurable Processor" that allowed FPGA's to be reconfigured in an almost infinite number of ways, in one clock cycle. It also included a Thrust microprocessor that could be configured for any bit width, and executed every instruction in one clock cycle as well. It essentially worked by changing the address of RAM banks that drove registers to configure a versatile set of components. Its predecessor, Mimic, had some funding from Sun Microsystems, but later management couldn't understand it and ended their participation. But I kept working on Mimic and ended up with Thrust, which of course is quite complex. But it also failed because most people with the money to fund it couldn't understand it and I couldn't secure more funding to continue.

However for those interested I did the best I could to convert the original FrameMaker 4 documentation to Word format awhile ago (though some graphics were blacked out), and posted it on Mega. It's quite a long read, 68 pages, but at least you'll be able to see I'm not kidding ...

https://mega.nz/file/oThwHILa#4Wn55sxxnII_CckiSfV9SwL-x8Via7qRNZEmKeEzarc

[Rallos Zek]

I still can't get over the end of the transmeta crusoe. Such a revolutionary design with big potential killed silently by big blue. Who knows how today's cpus would look like with that kind of technology.

By the way the pentium m's speedstep was based on a technology from transmeta. Intel had to pay a heavy fine for copying it without a license. Without speedstep and further improved iterations we probably would have heated our rooms with pentium 4s for a much longer period.

Sigh... there was nothing revolutionary about the Crusoe, the Pentium Pro, AMD K6 and Pentium 2, was already doing what the Transmeta chips were doing but in microcoded hardware. And the Crusoe was not killed by Intel it was killed by being mediocre and being VLIW.

[M.Bahr]

Sigh... there was nothing revolutionary about the Crusoe, the Pentium Pro, AMD K6 and Pentium 2, was already doing what the Transmeta chips were doing but in microcoded hardware. And the Crusoe was not killed by Intel it was killed by being mediocre and being VLIW.

Ah i understand. That must be the reason then why intel felt threatened by transmeta and even stole transmetas ideas. Keep deceiving yourself with your "brand loyalty" to intel. Bravo

[vladpetric]

I still can't get over the end of the transmeta crusoe. Such a revolutionary design with big potential killed silently by big blue. Who knows how today's cpus would look like with that kind of technology.

By the way the pentium m's speedstep was based on a technology from transmeta. Intel had to pay a heavy fine for copying it without a license. Without speedstep and further improved iterations we probably would have heated our rooms with pentium 4s for a much longer period.

Sorry, but one thing to keep in mind is that revolutionary movements tend to do a lot of propaganda. Good propaganda typically cherry-picks one side (the positives in this case) - so it misrepresents stuff without technically lying.

Yes, if you only look at Transmeta's propa^H^H^H^H^Hmarketing materials, it looks like a good idea, but in the end it simply wasn't. To their credit, they introduced quite a few performance savings techniques, but those had very little to do with their way of executing code (their transmogrification engine). Yes, Intel and AMD had to play catch-up originally, but they did catch up quite quickly, and when they did, the benefits of Transmeta's approach simply weren't there anymore.

[M.Bahr]

Sorry, but one thing to keep in mind is that revolutionary movements tend to do a lot of propaganda. Good propaganda typically cherry-picks one side (the positives in this case) - so it misrepresents stuff without technically lying.

Yes, if you only look at Transmeta's propa^H^H^H^H^Hmarketing materials, it looks like a good idea, but in the end it simply wasn't. To their credit, they introduced quite a few performance savings techniques, but those had very little to do with their way of executing code (their transmogrification engine). Yes, Intel and AMD had to play catch-up originally, but they did catch up quite quickly, and when they did, the benefits of Transmeta's approach simply weren't there anymore.

The real reason why transmeta had to pause development for a long time after initial products and finally gave up was due to intel choking them with x86 patents in court. After that transmeta completely ended chip production and went on to earn money by licensing their power saving techniques.

[vladpetric]

The real reason why transmeta had to pause development for a long time after initial products and finally gave up was due to intel choking them with x86 patents in court. After that transmeta completely ended chip production and went on to earn money by licensing their power saving techniques.

Ok, so one really doesn't have to pause development at all while patent infringement lawsuits take place (and any good IP lawyer can delay such things for years). Typically the plaintiff can get injunctions to stop sales (and TTBOMK that never happened with Transmeta's processors), but block the other side's R&D?

Right, and their power saving techniques had nothing to do with their revolutionary approach to executing instructions. You can consider their power saving techniques revolutionary I guess ... I don't.

[M.Bahr]

Ok, so one really doesn't have to pause development at all while patent infringement lawsuits take place (and any good IP lawyer can delay such things for years). Typically the plaintiff can get injunctions to stop sales (and TTBOMK that never happened with Transmeta's processors), but block the other side's R&D?

Right, and their power saving techniques had nothing to do with their revolutionary approach to executing instructions. You can consider their power saving techniques revolutionary I guess ... I don't.

I see your ego doesn't allow you to simply find what i said to be true. It literally costs a few minutes of internet research in older articles. But you are trying nit packing and splitting hairs now. We can talk all day long about what patent lawyers could have done to delay court rulings. Transmeta weighed up the risks and made their decision to prevent possible retroactive claims.

PS: And yes i stand by my assessment. Transmeta's idea on "how to convert a stream of x86 instructions into parallel executable VLIW (Very Long Instruction Word) instructions." is indeed revolutionary. I quote the c't magazine here, one of the most renowned computer journals in Europe, confirming my statement by writing this:"Experts estimate that the performance of such a design could far exceed that of an Intel CPU."

[vladpetric]

I see your ego doesn't allow you to simply find what i said to be true. It literally costs a few minutes of internet research in older articles. But you are trying nit packing and splitting hairs now. We can talk all day long about what patent lawyers could have done to delay court rulings. Transmeta weighed up the risks and made their decision to prevent possible retroactive claims.

You mean this?

https://www.computerworld.com/article/2539492/intel-settles-patent-case-with-transmeta-for--250m.html

Intel ended up paying Transmeta for IP (and lots of money).

So, let me ask this - are you making stuff up?

[vladpetric]

The real reason why transmeta had to pause development for a long time after initial products and finally gave up was due to intel choking them with x86 patents in court. After that transmeta completely ended chip production and went on to earn money by licensing their power saving techniques.

This is your claim, quote from above:

"The real reason why transmeta had to pause development for a long time after initial products and finally gave up was due to intel choking them with x86 patents in court."

This never happened, AFAICT. So yeah, I claim that you are making stuff up.

Transmeta sued Intel, because Intel were infringing on their IP. Then Intel countersued. Eventually, Intel had to pay them ~250M