Alledgedly Transmeta technology is in IBM's latest multicore PowerPC chip In related Transmeta news, I have heard from a second source that Transmeta's x86 has legacy features such as real mode taken out... unlike Merced that will contain x86 compatibility, giving Microsoft the opportunity to add more thunking! (just what was needed to build a stable OS). And for those that argue that anything published in The Register is wrong, perhaps they should cast their eyes on to this C'T article.

Microsoft is not going to wait for the outcome of its trial with the DOJ before attacking Linux. The battle scene is France, where Microsoft's new (expendable?) regional director Marc Chardon has just issued an open letter to his clients. Click below to read the translation of the Linux-section (it's in French) and some commentary.

After reading a lot of stories about RH/I/N (including Chuck Milam's submission from CNN) I decided to write up my opinions on the whole deal and dump them online. Hit the link below to read it if you care.

The Register is claiming that Transmeta hired quite a few of the original 615 engineers. The 615 was to be an x86 killer by IBM: a PowerPC including x86 compatibility logic. However, the project alledgedly died due to too much complexity, and too much die area (cost). In the current CPU market there are two alternatives: cheap (small die area, low power consumption) or very high performance (few applications need more CPU power). Let's hope that if the 615 team is in Transmeta, they do not repeat past mistakes.

More Transmeta information has allegedly been leaked. Linus is said to have changed Transmeta from a x86 company to a NT 5.0 compatible company -- which could mean anything from pure RISC to a subset of the x86 instruction set. Ed: Given that this report is the only one I've seen to this effect, I'd assign it a rather low certainty.

George Greer writes "While idling doing a search for Transmeta on AltaVista, I came across the title "Recent Ph.D graduates". Naturally I was curious. On the page link they list the following people now working at Transmeta and their focus. Craig Anderson, ``Improving Performance of Bus Based Multiprocessors'', Autumn 1995. Advisor: Jean-Loup Baer. Current Employment: Transmeta, Inc., Santa Clara, CA. David Keppel, ``Runtime Code Generation'', Spring 1996. Advisor: Susan Eggers. Current Employment: Transmeta, Inc., Santa Clara, CA. Jack Lo, ``Exploiting Thread-Level Parallelism on Simultaneous Multithreaded Processors: Hardware and Software Techniques for Effectively Managing Shared Resources'', Spring 1998. Advisors: Susan Eggers and Hank Levy. Current Employment: Transmeta Corporation, Santa Clara, CA. As if there's any doubt left they're making a (really cool) processor. " Mmm...and based on what Linus said at Linux Expo, I think it's clear that they are headed toward SMP fun.

The Register is reporting that Transmeta will use IBM fabs, taking over from Cyrix which now relies on National's fabs. Wonder if they'll be using IBM's copper and SOI technologies...

Jonas Sanamon writes "When searching the web for info abt. Transmeta and its rumored microprocessor i found this interesting site abt. the upcoming beasts from variuos companies. Seems INTEL has a similar project going... Very interesting reading! " This is a pretty sweet page.

This year's Microprocessor Forum looks like it's going to be rather interesting. Two new x86 vendors will be presenting their latest work: Rise and Transmeta. Rise is attacking the x86 market with low power designs, somewhat like IDT did with Centaur, (although their efforts appear to be waining). Transmeta will be presenting a RISC chip that can read x86 instructions.

homebrewer writes "Here is an article talking about AMD helping to drop notebook prices. Maybe I can even afford to buy one.... They even mention Transmeta a bit, saying they are working on a new chip...is this what Linus is working on?" In related news, Trident revealed that it has designed with VIA Technologies Inc an integrated chipset combining a graphics chip and core logic on a single die. Socket 7 still seems to have a long life in front of it.

Simon Janes writes "The Red Herring is running an article this Sept on Transmeta called Stalking Transmeta. They were making up their "Herring 100" and rumors started coming out again about Transmeta so they had to investigate. They learned, nothing. :)"

It's definately been a hot topic lately, no doubt about it. But Daniel Quinlan wrote in to send us a press release from the LSB stating that the Debian/Red Hat standards project (LCS) will be working with the LSB. Daniel has also been named as Bruce Perens replacement as the head of LSB. Besides having an @transmeta.com email address, Daniel has been the head of the Filesystem Hierarchy Standard. I'm really glad to see this come together. Anyway, I've attached the announcement- click below if you want to read it.

Kory Lasker writes "In the last place I would have expected to hear about Transmeta, PC Magazine (a ZDNET sibling) included a tantalizing bit about Transmeta in an article about the CPU landscape." It's stuff that a lot of us know (they're working on CPU stuff, its secret and something about a fast software translator for the x86 instruction set. And if you just want a nice laugh (it's saturday! Smile!) you should read The Linux Apocalypse by Christoph Lameter.

Simon Janes writes "Sun is getting into the multimedia market with a PicoJava-core chip which appears to be what this columnist thinks Transmeta is also working on. "

It's been quite a day folks. I decided that I will take a few minutes to give everyone a bit more details about the adventures that we're having out here. Hit the link below to read everything that I can think of that happened today that might have been interesting.

Kristian Köhntopp wrote in to tell us that Salon Magazine has an article on TransMeta. Read here, about the company that employs Linus, has a web server (but no content), is funded by Paul Allen, and produces nothing that anybody can discern. Neato.

Despite initial expectations that set-top boxes and PCs would converge, the trend towards RISC chips such as MIPS continues. An oft cited reason for this is the lower power requirements of RISC solutions, which might give low power x86 manufacturers such as Centaur, Transmeta (?), and Cyrix for their MediaGX chips an advantage. Perhaps however, it is not x86 that is being fled, but domination by Microsoft that is being avoided. After all, set-top boxes are a cost, not a revenue to content-selling cable operators: do they really want to upgrade their infrastructure each time Microsoft comes up with a new version of Windows? A winner in this market will be Cygnus solutions, sellers of GnuPro tools to the embedded market, and thus Free Software Standards.

As AMD and Intel report disappointing earnings, the price of computers continues its downwards spiral. Intel certainly did not anticipate the success of the sub-$1000 category, but AMD and Cyrix have failed to capitalize on the opportunity. The problem is that a bitter war of attrition will hurt the smaller players just entering the field. But then, perhaps they will address the lack of new features to excite one enough to buy a new PC. Or perhaps, they will find a solution to the root cause of high computer cost: bloatware.

By the end of this year, there will be 7 somewhat incompatible versions of Windows. In addition to the current four x86 vendors there will many more, among which Imes, CPU Tech, Metaflow, Rise, Transmeta ( ? ), and Exponential (??). While new hardware ages computers before they are sold, M$'s bloatware obsoletes them within a year. To compensate, consumers are turning to cheap "disposable" computers. But this trend driven by Microsoft's feature-adding strategy, risks back-firing. As component cost is driven down, and specialised vendors disappear, more application specific devices will emerge. Just like Cyrix's MediaGX, they will be geared towards providing the maximum bang for the buck in a specific environment. This will slowly push the single-OS-for-all paradigm to the side. Indeed, as hardware cost goes down, the direct and indirect price of using Microsoft increases: Windows/Office costs money, it also costs a very powerful environment: memory, harddrive, etc. And if computers are application specific, the choice of an OS becomes irrelevant, and each new feature is clearly costed.

Prasanth Kumar sent us this MicroTimes article which lists the top 100 people of 97 (I guess anyway) and lo and behold, there is Linus, with yet another mysterious reference to whatever-the-heck-transmeta reference. Linus, if you read this, send me email and tell me what you do! *big smile*