New Transmeta Chip: "Efficeon"

ddtstudio writes "Oh, "Astro" was such a friendly name -- but it probably had trademark issues. So the alphabet blender came up with "Efficeon" instead. This eWeek story gives the lowdown on what Transmeta is doing apres Linus. There's also a writeup on ExtremeTech."

Comments

Marketing

[mjmalone]

From the article:

Transmeta is the "number two" vendor in the ultraportable mainstream notebook market

Is that why nobody knows about them? Maybe they should focus some attention on advertising, I don't think many people outside the tech industry knows about Transmeta. Intel spends a rediculous amount of money on product marketing, and when many people get a new computer they want "Intel Inside" because it's what they know. I think if any competitor really wants to break into the chip industry and compete with the big boys they are going to have to get their name out, the real differences between one chip versus another are not very obvious to the consumer, brand recognition is what drives sales.

Re:Marketing

[Anonymous Coward]

That's why VIA is becoming more and more popular. If you think "ultra-small sized PCs" you have to think mini-ITX, and VIA EPIA.

This is what I think (may be biased by publicity, but that's exactly my point) :
Intel = reliable, industry standard (never had any Intel die, even since the 8086 days)
AMD = power, speed, will burn without good heat dissipation (had two AMDs die on me, installed professionally. Will never buy AMD again)
VIA = low-heat, small size (currently two projets using EPIA boards)

Re:Marketing

[Raul654]

Transmetta specializes in low power computing. That's their niche - processors that don't eat up much energy. This really isn't a direct-sale market. The stuff they make is built into other devices. So their job is to convience sony and/or palm or whoever makes the low power (read - portable) devices that they are the ones to talk to. So I don't think there's really a need to become well known outside the computing industry.

Re:Marketing

[dodell]

Then again, there needs to be a budget for said ridiculous amounts of money.

I think Transmeta is doing a good job getting their chips marketed. HP is including them in their products; this seems to me to be a pretty good method of advertising in the first place -- if your product is already accepted by a major manufacturer, you're halfway there. Then again, I don't think they're in precisely the same market as Intel in the first place.

Re:Marketing

[bogie]

What good is it for Transmeta to advertise like Intel does when unlike Intel you probably can't go to your local store and buy a machine with a Transmeta cpu in it?

"I think if any competitor really wants to break into the chip industry and compete with the big boys they are going to have to get their name out, the real differences between one chip versus another are not very obvious to the consumer, brand recognition is what drives sales."

No not really, the people who buy from big boys already know about Trasmeta and choose not to use them. There is not a single largescale oem who doesn't know that they have around 4 to 5 cpu's to choose from when building systems, be it set top boxes or mainstream PC's.

What Transmeta needs to do as our friend AMD knows so well, it to lock up more oem deals. Not being an industry insider when it comes to cpu sales I have no idea how aggresive they are, but seeing that Transmeta cpu's aren't inside anything I'm considering buying I'd say they aren't doing a very good job.

So in short yes they need to do more advertising, but not outside the tech industry like your advocating. Transmeta needs more big vendors deals then the rest will take care of itself.

Re:Marketing

[advocate_one]

"Obviously, we don't have the $350 million Intel has" to spend on marketing, he said.

$350 million buys a lot of presence in magazines etc... Similar problem for linux in trying to get past the reams of Microsoft bought advertorials etc. in the magazines as well...

Transmeta's Efficeon will have to compete on it's technical merits, and when people who matter realise that it offers a much better power consumption, lower temperatures and much longer battery life they'll start to take it up. 3 hours or so with Intel Celeron 1500 is just not on when I was used to some 24 hours or so battery life on my old 8086. Hopefully, the Efficeon will enable them to make notebooks that can cope with a complete working day or more away from the mains outlet... RAM's cheap enough these days to enable them to give it a seriously large cache so as to minimise HD usage, and sticking the OS in a bootable flash disk will improve matters as well. Now we just need a very low power display technology such as high res colour "electronic ink" based thin displays

Re:Marketing

[Trolling4Dollars]

Maybe they should focus some attention on advertising

For people like me, advertising doesn't work. I would hazard to guess this applies to the majority of the Slashdot readers. If anything, when something is heavily marketed as being "the best" I am more likely to scrutinize it more since it's likely not the best, it's just popular. There is a difference. The masses don't have a good handle on what is "the best". Being #1 is not as important as providing a quality product.

Re:Marketing

[DrinkDr.Pepper]

Transmetta specializes in low power computing. That's their niche - processors that don't eat up much energy. This really isn't a direct-sale market. The stuff they make is built into other devices. So their job is to convience sony and/or palm or whoever makes the low power (read - portable) devices that they are the ones to talk to. So I don't think there's really a need to become well known outside the computing industry.

True, but companies who make consumer devices would be more inclined to listen to Transmeta if consumers were demanding devices with Transmeta processors. The way to convice consumers is with direct marketing. If you were correct, Intel wouldn't need to targer consumers either, since most of Intel's sales are to OEM, not direct either.

Re:Marketing

[realdpk]

If I start getting direct marketing from Transmeta, I will be seriously pissed. I get enough direct marketing every day already, I don't need more junk stuffed in my mailbox or telemarketers on my phone, not to mention spam in my e-mail.

Re:Marketing

[mihalis]

What good is it for Transmeta to advertise like Intel does when unlike Intel you probably can't go to your local store and buy a machine with a Transmeta cpu in it?

Actually, I can go to my local store and buy such a machine, a Sony, last time I looked. Maybe you need a better store.

Chris

Re:Marketing

[TheViffer]

And 640K is enough for anyone ...

I completly disagree with your posts.

First low power consumption is become more and more and issue. Electricity is not cheap, and with more and more computers in organizations, they are looking for ways to cut down on the electricity bill.

This is two fold because with lower power consumption means less heat. Walla .. save MORE electricity by not running huge, or as many air conditioning units.

This even falls back to home use. 10 years ago running your own home server was a "pipe dream" and not needed. Now just about every slashdot reader has some sort of server(s) running. I myself only run a Duron 1000 due to is low power consumption and cooler temperatures. For serving up files and a few dns, web hits is more then enough. After switch DOWN from a Palamino cored Athlon to this, I saw my monthly electric usage decline.

And your mention of not needing to market is just wrong. They should be out there stating ...

We run cooler ..
We run cheaper ..

And add in a few plugs .. We run faster, better, cheaper, blah blah blah. Go to the extent of showing a picture of an Athlon frying an egg on one picture and a piece of ice sitting on a Trasmetta CPU. And by marketing their name, people come to know it, and trust it more, which mean people will be more interested in products with their chips, which mean more companies will seek out Transmetta chips which mean more revenue.

Why does BASF advertise .. "We dont make 'jack', we just make 'jack' better"

My 0010 cents worth.

Re:Marketing

[benja673]

Wrong. When is the last time you have demanded that your Palm is running the newest dragonball processor? Quite simply, in niche technology, marketing to the masses is wasted money.

Re:Marketing

[BigBir3d]

Fujitsu LifeBook also.

Re:Marketing

[inbox]

This really isn't a direct-sale market ...[snip]... So I don't think there's really a need to become well known outside the computing industry.

Then how come nearly every consumer PC buyer knows what a "Pentium Processor" is?

I think the point is that yes, while they sell to the people that make the things you buy (sorry, BASF), if the consumer were aware of the advantages of the underlying technology and demanded products that used it, then marketing to them might go a long way.

Re:Marketing

[TCM]

I'd call becoming the number two in any market without anyone knowing about you an honorable achievement :)

Re:Marketing

[Anonymous Coward]

You might want to re-think your "Never buy AMD again" attitude. :) SPEC International (or whatever their full name is) has declared the AMD Opteron to be the performance king in 2-way and 4-way servers. SPECint and SPECfp scores for these 1.8 GHz systems blows away the scores for similar Intel P4 Xeon 3.06 GHz systems (with and without HTT enabled), and for the latest Itanium2 systems. This is running 32-bit and 64-bit code.

There are also unofficial benchmarks comparing the 1.6 GHz Opteron to the P4 3.06 (with and without HTT) showing runtime benchmarks to be within 1 second for things like compiling FreeBSD, or running various FreeBSD software.

Gotta wonder about Intel's engineering abilities in light of this. :)

Re:Marketing

[Anonymous Coward]

You are aware that by now, most good motherboards will actually prevent a processor from overheating, right?

Re:Marketing

[bhtooefr]

Never mind that the Opteron is a whole different class of CPU (64-bit as opposed to 32-bit). Wanna bet that it'll whip a 3200+ (forgot the real speed) Athlon XP, too? Besides, Intel is making their CPUs with the highest clock speeds that they can, by killing the efficiency of the CPU. Look at a Pentium-M and a P4-M at the same speed. The P-M will kill the P4-M. For that matter, try a 1.4GHz and 1.6GHz Celeron. The 1.4 is based on the P3, whereas the 1.6 is based on the P4. I read somewhere that the 1.4 would beat the 1.6.

Re:Marketing

[bhtooefr]

For that matter, just use B&W active matrix. No backlight needed unless you're in a low light situation.

Re:Marketing

[bhtooefr]

(replying to my own post - forgot something)
I once had an Apple PowerBook 180 (I don't trust Apple because of that box - took 6 months to fall apart and need a $300 mobo swap), and the manual said that turning off the backlight used 2/3 the power of the laptop with full backlighting. That's why I said what I said in the parent post.

Re:Marketing

[Raul654]

True, but as I pointed out here, the people buying the stuff transmetta is made for (portables) don't really care about what processor is on board. People buy pentium inside computers because the processor is the first spec listed. They know that that is important to its function. A portable, on the other hand, barely (if at all) mentions the processor. As someone else pointed out, marketing to the masses in this case is a waste of money.

Re:Marketing

[yomegaman]

Do you have a link for this? I looked over at www.specbench.org and here's what I saw:

Opteron 1.8GHz: 25.0, 24.7
Itanium2 1.5GHz: 30.3, 42.6
Xeon 2.8GHz: 26.2, 19.0

All are for dual-CPU configs, the first number is the base int rate and the second the base FP rate. The Opteron numbers I got from the 2003q2 page and the Xeon and Itanium2 numbers from 2003q3 (I didn't see any Opteron results there). Based on this it looks like the Itanium2 is comfortably ahead (in performance anyway, probably not in price), but I'd love to take a look if you have other information.

Re:Marketing

[name773]

my home server is a pentium 100 with 16mb ram. passive heatsink. (ok so i should update it) have you seen mini-itx computers?

Re:Marketing

[Anonymous Coward]

3 hours or so with Intel Celeron 1500 is just not on when I was used to some 24 hours or so battery life on my old 8086

I'm sure you get more than 8 times the performance of an 8086 out of your celeron 1500 though...

Re:Marketing

[Froobly]

Is it possible that the reason you haven't heard much from them is that you aren't in the right market? In the US, nobody has sub-notebook computers, and thus nobody (aside from HP) has Transmeta processors. In Japan, however, you'll find that Transmeta's Crusoe processors are all over the place. Their main competition is not the Pentium 4 or Athlon XP, but Intel's Centrino processor, which you also no doubt have never heard of, being on the wrong side of the Pacific.

Americans, as a whole, don't spend enough money on portable electronics to make really nifty sub-notebooks a worthwhile investment for a manufacturer. Hopefully, when the economy picks up, Transmeta will have its day, and other companies will be competing with their own low-power technologies.

Re:Marketing

[Anonymous Coward]

Transmeta isn't marketing low power. It's marketing low heat.
You can't do more processing with less power.
That would violate the laws of thermodynamics.
But you can have the same power with less heat.
That efficiency: efficeon.

Re:Marketing

[WoTG]

I totally agree. I'm surprised that more energy efficiency isn't demanded from PCs in both corporate and home markets.

One of my pet peeves is how "SpeedStep" and "PowerNow!" technologies from Intel and AMD for mobile chips are not used in the desktop chips. These systems do a good job of saving power and probably take neglible amounts of die space and other costs, yet we don't get them on the desktop. I suppose Intel and AMD are scared of desktop chips being used in mobiles, and thereby eating away at the fat mobile CPU margins.

It's a shame. The vast, vast majority of the time the processors on all but the servers in the office are running at less than 10% usage... Granted, the idle thread helps to save a lot of power, but the mobile power features would do so much more.

As an aside, it's worth noting that the price of electricty varies greatly. I get cheap power out here in BC, but elsewhere in North America it is often 2x as much. And, in many other countries, power can cost much more still.

Performance over name

[xxltjx]

I've always looked for performance over name brand, and if this chip can do what the article says it can do...it could give Intel a run for it's money in the portable marketplace.

Re:Performance over name

[ackthpt]

I've always looked for performance over name brand, and if this chip can do what the article says it can do...it could give Intel a run for it's money in the portable marketplace.

I don't know...

I think people are still stuck on the -ium, -on and -ex suffixes. Infineon being the notable exception, but who the heck outside of hardware builder circles knows of them?

Efficeon? Is that supposed to draw attention to efficiency, with that 'Fishy' phone in there? Not very efficacious...

Names they definitely missed out on:

Illudium 236 Laptop Data Modulator

Transmetium

Apteryx

SlashDotEffecteron

CowboyNealexiumron

Re:Performance over name

[Anonymous Coward]

Well that's what everyone though several years ago when they first marketed their idea. Unfortunately they have not delivered on ANY of their promises, so... Well maybe some year. But I think most of us are tired of "waiting".

Re:Performance over name

[Hoser McMoose]

I don't know about that. They are promising only 50-80% improvements over their old Crusoe 5800 processor. That would put them as being about on-par with the chips that Intel had out two years ago when the Crusoe 5800 was first available, but it'll have a hell of a time competing with the chips that Intel is producing now, let alone 6+ months from now when this new Transmeta processor actually starts shipping.

I think that the real question will be how well this chip can compete with Intel's Ultra Low Voltage (ULV) Mobile Celeron line of processors. The two chips will have comperable power consumption (5-10W max, typical of under 5W) and probably won't be too far off one another in terms of price. Previous Transmeta chips have had a heck of a time keeping up with even the slowest mobile Celeron chips that Intel had available (read: they kind of kept up in MS Office, but got pretty well thrashed for everything else), but maybe this newer chip will bring performance up a bit.

Efficon?

[YU Nicks NE Way]

Does anyone besides me read this as "F-ing Con"? Maybe my problem is that seems like a good description of Transmeta's current business model.

Re:Efficon?

[g0at]

The first two things it conjured in my mind were "efface" (eraser), and "defecate". Hmm...

-b

Re:Efficon?

[dBLiSS]

pronounced EE-fish-eon

I know you were joking but just incase anyone was wondering...

Re:Efficon?

[Anonymous Coward]

I was going to attend the F-ing con, but then I heard about all the F-ing unwashed fanboys with their F-ing bad homemade costumes so I just stayed home and spent a week F-ing my wife.

Re:Efficon?

[4of12]

Definite signs they used ProductNamesForCheep, as I think of...

• "effete"
• "effigy" (as in "burn in")
• "feces"
• "iffy"

Transmeta ought to hire the same big-gun name-generating people that came up with "Viagra".

Re:Efficon?

[qtp]

Does anyone besides me read this as "F-ing Con"? Maybe my problem is that seems like a good description of Transmeta's current business model.

Incompetant at marketing, maybe. But that does not mean they are conning anyone. If you had tried a laptop that uses thier Crusoe chip, you would know that thier chips deliver exactly as promised: Extremely efficient use of power and extended battery life (eight plus hours in many models) with acceptable performance. The con game in laptops is convincing users that they need a P4.

Re:Efficon?

[Anonymous Coward]

The first thing that came to my mind was "Feces"

Re:Efficon?

[richie2000]

A fizzy one.

whats the big deal

[minus_273]

about transmeta chips.. i know they dont perform as wellad ppc or x86 chips but from what i have heard, they have low power consumption in comparison to the more power fuke chips..
The one thing that intrrugued me the most (and this is after i saw a friends sony viao) was that these chips make up for the lack of speed in th ability to emulate any processor.. so i ask this: has anyone done it.Have you run ppc software on your transmeta chip... or anythign like that?

Re:whats the big deal

[blindbat]

What would interest me is to know how well it would run Linux if it ran natively on the processor's own instruction set.

Re:whats the big deal

[Nerant]

Transmeta chips are essentially a VLIW CPU, surrounded by a sophisticated JIT compiler that translates and compiles x86 code to the native instruction set of the VLIW CPU, and then runs the code.
This allows Transmeta to save on die space. Smaller die = less transistors = less heat = less power.
Unfortunately, this approach so far has yielded substandard performance. And even though power consumption was better than Intel's mobile processors for awhile, Intel quickly geared up, threw money and engineers at the problem, and came out with the Pentium M.
Arstechnica.com has speculated before that Transmeta could easily use the same approach to optimise for speed/performance as opposed to power consumption : I'm hoping they do.

Re:whats the big deal

[msgmonkey]

They did when Linus was there and it ran *slower* because the VLIW instruction set could not exploit enough parallelism at compile-time. Itanium also has the same problems.

Re:whats the big deal

[msgmonkey]

Is n't the Pentium M just a tweaked Pentium III at a smaller feature size? In which case it did n't take much from Intel but then again competition is good even if it loses.

Re:whats the big deal

[RebelWebmaster]

The Pentium-M is more of a PIII/P4 hybrid (why reinvent the wheel?). It uses a 400Mhz FSB which is compatible with the P4 bus and has SSE2 support. On the other hand, it has a shorter pipeline than the P4 (though longer than the PIII). Of course, the biggest boost to the P-M's performance is the 1MB of L2 cache (2MB once Intel switches to 90nm fabbing). So yeah, basically it's a PIII core with P4 enhancements plus some new stuff (better power management, more cache, etc).

Re:whats the big deal

[JamesP]

The Big Deal is

Pentium 4-M 2.0Ghz = Thermal Design 32W (but actual power dissipation is higher)

Crusoe = 1W

About the "substandart" performance, the JIT compiler is optimized constantly (on the fly) so every benchmark runs faster every time it's run.

BTW, why do you mean by "substandart" performance? You don't need a P4 to use Word/Excel or listen to MP3 while surfing the net. My old PII 333 did that w/ no sweat.

Playing DVD's you say... It can play them... Without dropping frames.

Re:whats the big deal

[Drakonian]

Arstechnica.com has speculated before that Transmeta could easily use the same approach to optimise for speed/performance as opposed to power consumption : I'm hoping they do.

Really? Interesting. I would figure it could never be as fast as native x86, it's all still emulation.

Re:whats the big deal

[Erich]

Really? Interesting. I would figure it could never be as fast as native x86, it's all still emulation.

"Native x86" really doesn't exist. Since the AMD K5 and Intel Pentium Pro, x86 instructions are translated into smaller, RISC-like instructions inside the processor.

Instead of doing this translation in hardware, Transmeta does this in software, and it enables a lot of optimization while (at the same time) vastly reducing the amount of hardware resources required to do wide, out-of-order execution.

They get varied results -- some things go much, much faster on the Transmeta, but it's very bad at doing other things (especially things like self-modifying code).

The internal architecture is also very geared towards translation and running translated code. There are features that allow it to run a bunch of code in a translation that is fast, but not safe. If there is a problem with this unsafe translation (memory exception or something) the execution can be rewinded (rewound?) into a known-good state and a slower translation or interpretation can be used.

Transmeta has released some good papers on this whole thing. If you're interested in this kind of thing, you might want to also check out HP's Dynamo and Intel's DAISY.

Yay, clever computer architecture!

Re:whats the big deal

[Anonymous Coward]

The core of the chip is geared towards x86-based translations -- remember, PPC and x86 use different endians... Code-morphing software that continually needs to reorder the bits in a word has a lot of overhead. I'm sure that they *can* run PPC code on the chip, but the performance probably bites.

Re:whats the big deal

[Anonymous Coward]

BTW, why do you mean by "substandart" performance? You don't need a P4 to use Word/Excel or listen to MP3 while surfing the net. My old PII 333 did that w/ no sweat.

Yeah, but it's the spelling and grammar checkers that really soak up the cycles.

Re:whats the big deal

[Yokaze]

Well, it might be based on the PIII and has adopted some parts of the P4. But it has some more radical changes.

It has more in common with Transmeta than one might think. It features Micro-Op Fusion (TM)(R)(C). After translating the Ops into muOps they are reassembled to, how do they call them? Not-LIW, no.., ah.. Macro-instruction, which can be executed more efficiently.

But why should I smatter. Use the source Luke.

Re:whats the big deal

[Hoser McMoose]

Your comparing apples and orangutans here when it comes to power consumption. Intel's Thermal Design Power (which the maximum power consumption excluding thermal viruses) for the P4-M 2.0GHz is 32W. Transmeta's Thermal Design Power for the 1.0GHz Crusoe 5800 is 7.5W. The "1W" number that Transmeta likes to flaunt around is the amount of power that the chip consumes while it's sitting around twiddling it's thumbs. For comparison, a P4-M will consume about 5.0W doing the same thing.

That being said, the real competitor for the Transmeta chips is not the P4-M, but rather the ULV Mobile Celeron chips. Both chips will cost close to the same amount and consume close to the same amount of power (TDP for a ULV Celeron is 7.0W). Previously these Celerons, even at lower clock speeds, have had no trouble beating Transmeta chips. It remains to be seen if this new Transmeta chip has improved it's performance enough to compete with the Celerons. Otherwise it's just an odd-ball design with the same power consumption and similar price to a more widely available and better supported processor.

Re:whats the big deal

[Yokaze]

Not that I'm disagreeing with your final point (how much power does one need)

But, you are comparing apple and oranges, or TDP with typical power consumption.

Thermal Design Power (TDP) represents the maximum amount of power the thermal solution is required to dissipate. The thermal solution should be designed to dissipate the TDP without exceeding the maximum Tjunction specification. TDP does not represent the power delivery and voltage regulation requirements for the processor.

According to Intel, the Pentium-4M 2.2GHz has a typical power consumption of 2W (But that doesn't include the North-bridge, like Transmeta AFAIK does) and what Intel defines as "typical usage" also critical point ("Don't touch, I said").

Another thing, the LCD backlight alone consumes about 20W. Add RAM, HDD, CDROM, WiFi, and the difference between a 4W or a 1W CPU would be negligable.

What brings me to another topic:
I want OLED and M-RAM!

Re:whats the big deal

[akuma(x86)]

I have a minor nitpick :)

IBM did Daisy, not Intel.

Statically scheduled VLIW will almost never outperform a dynamically scheduled out-of-order machine. But, you can save tons of power :)

Re:whats the big deal

[Erich]

Yes, sorry, typed the wrong thing. IBM did DAISY.

VLIWs that have code generated dynamically can many times outperform the code on the OOO machine because it has profile information and can do block rescheduling, function inlining, etc. Of course, you can get the same thing by dynamically recompiling the code on the OOO machine (like HP's Dynamo).

Re:whats the big deal

[akuma(x86)]

That's the theory anyways. If dynamic compilation is so great, why aren't there any decent dynamic recompilers? Hint: try doing precise exceptions. What about self-modifying code (oh yes, it's out there - esp. in java and other managed run time apps).

Also, why doesn't Transmeta get performance that is comparable to a similarly powered out of order machine like Banias? Why does IA64 suck? Couldn't a dynamic recompiler help it out?

There's too much overhead with a software based dynamic recompiler. The latency of getting the dynamic information and then making a change is way to long compare to hardware methods.

Things like instruction scheduling change far too quickly for dynamic software methods to be effective.

Using a trace cache, the hardware can "dynamically recompile" code for you at much reduced latency. Also, the hardware is privy to much more dynamic information, such as the state of branch predictors and other internal, non software visible structures. All of this can be done quickly, and at a granularity much finer than a software method.

Re:whats the big deal

[Dwindlehop]

Who modded this +1 Interesting?

"Pentium M" != "Pentium 4-M"

Re:whats the big deal

[Erich]

That's the theory anyways. If dynamic compilation is so great, why aren't there any decent dynamic recompilers?

There aren't for x86. x86 is a difficult architecture to recompile in place. But for PA-RISC, check out Dynamo.

It is true that performance gains are usually not huge for same-architecture recompilation, which limits what you can do to not-very complex things. But many times you can get profile-based, global-optimization compiler performance without compiling your application using profile or global optimization.

Hint: try doing precise exceptions.

Precise exceptios is a function of the architecture, not of the translator or translated code. The hard part is expecting a debugger to work. If you reorder stuff your state will be inconsistent with the real program at different points, and so your debugger won't be compatable with the old code.

What about self-modifying code (oh yes, it's out there - esp. in java and other managed run time apps).

These are usually tricky for dynamic recompilation. Though Java is one of the examples where dynamic recompilation is extremely effective. JIT performance is actually pretty good.

Also, why doesn't Transmeta get performance that is comparable to a similarly powered out of order machine like Banias?

I think this might be subverted support. Current Transmeta chips are very substantially smaller and at least a generaton back in process. And it certainly doesn't make the case that VLIW

Why does IA64 suck? Couldn't a dynamic recompiler help it out?

IA64 is a relatively poor performer for a variety of reasons, but if you take a bit of time and have a really great compiler, you can get your performance out of it. That being said, VLIW architectures are harder to dynamically recompile because many times they have parallel semantics which are harder to brake up and mess with.

There's too much overhead with a software based dynamic recompiler.

Well, it depends on what you do. You want to embed GCC and run it with -O4? Yeah, that's too much overhead. You want to align blocks and do a little bit of optimization over block boundaries? That's much more reasonable. And it's also the kind of optimization that you will get more benefit from, because the compiler is going to do a pretty good job scheduling straight code.

Things like instruction scheduling change far too quickly for dynamic software methods to be effective.

Instruction scheduling isn't so bad. You know the latencies for instrucitons, you can schedule them. Compilers usually do a pretty good job of scheduling in straight code.

What compilers can't do is optimize something like this:

double angle = PI/4;
for (i = 0; i < info.len; i++) {

info.arr[i] *= sin(angle);

}

Assuming sin() is in some library somewhere, the compiler can't optimize that constant away. What if it were "double sin(double x) {static int i=0; return ++i;}"? The dynamic translator can see the bigger picture, and has the potential to take the computation out of the loop.

Using a trace cache, the hardware can "dynamically recompile" code for you at much reduced latency.

It can keep the translated copy. It can even basic block reorder. But it can't do the somewhat-complex things that a translator can do.

I think that you expect too much from your architecture. But if you are thinking that given infinite resources, both an out-of-order machine and a VLIW will get similar performance, you're probably close to correct. But I think you are missing a big point -- everything i

God Dammit!

[illuminata]

Why do companies like Intel, AMD, and Transmeta keep having to name processors so they sound like they came straight from the mouth of Rob Schneider's SNL character The Richmeister?

The Celer-on, the Opter-on, the Athl-on, the Effice-on.

It's not good for marketing, guys! Everybody hates Rob Schneider!

Re:God Dammit!

[KillerHamster]

You forgot the Dur-on. And yeah, they all sound dumb.

Re:God Dammit!

[red floyd]

Precisely! WTF was wrong with "Hammer"????

Which would you rather say:

"I'm running an Opteron(tm) at 2GHz", or
"I'm running a Sledgehammer(tm) at 2GHz"?

No contest.

Re:God Dammit!

[Anonymous Coward]

There's really nothing "wrong" with hammer, except that it is a word that already exists, so it would be hard to trademark.

And imagine: what if AMD really did call their chip The Hammer? A few months from now, there would be a YRO story on Slashdot about AMD trying to steal someone's domain, hammer-sucks.com even though it would just be a website about how much rapper M.C. Hammer sucks.

And then there would be the jokes. "It's hammer time!" and "When the only tool you have is a hammer, every problem looks like a nail."

Efficeon...

[Anonymous Coward]

...sounds rather lameon.

Trademark, not copyright

[_fuzz_]

You can't copyright a name, but you can trademark it.

Re:Trademark, not copyright

[Stonent1]

There's also the "device" that can be patented. CocaCola calls the stripe on the side a "device". Jaguar has the "Leaping Cat Device"

Re:Trademark, not copyright

[cpeterso]

CocaCola calls the stripe on the side a "device".

btw, the Coca-Cola stripe is officially called the "Dynamic Ribbon."

Re:Trademark, not copyright

[bill_mcgonigle]

The cool name would have been 'Rastro' anyway - don't these people even watch the Jetsons?

So...

[dex22]

The English will be able to enjoy their Efficeon Chips!

That joke is so bad, I'm questioning if I should post it.

Ahhh, whatever!

Re:So...

[kingLatency]

I found it quite funny! Ha! Ha! English!

efeminate

[Anonymous Coward]

OK, I'm not sexist or anything, but the name they chose is horrible. It reminds me of "effeminate", "flaccid". Get your sissy on, eficeon. I know Celeron is not exactly a powerful sounding moniker, but come on, eficeon? It's like limp cheese.

Re:efeminate

[Anonymous Coward]

They should team up with joystick manufacturers to name their products.

"Why would I buy an Athlon when I could get a Transmeta Decapitator-Pro 76-eXtreme?! And next summer they're coming out with the new Eviscerator series!"

Re:efeminate

[Lemmeoutada Collecti]

They could have at least added a 'c' in there to make it more appealing to us 80's raised geeks: Eficicon (ala Deceptacons)

Re:efeminate

[technicka]

do you by any chance watch The Daily Show with Jon Stewart on Comedy Central? I just watched it on the tivo about 20 mins ago and that word came up in bold writing on the tv,.. maybe your sub conscious is acting up. I really didn't think it was a word,. .but golly gee dictionary.com proved me wrong.. Having qualities or characteristics more often associated with women than men. See Synonyms at female. Characterized by weakness and excessive refinement

Re:efeminate

[Anonymous Coward]

The link between linux and transmeta is well known. Communism and homosexuality are strongly linked as well.

Since linux is a communist plot against IP, and IP is the powerhouse that will create wealth and power moving into the future, is is trivial to prove that communist linux is using transmeta as a springboard to the femminization of technology by naming their flagship (or should that be 'fagship'?) product in such a effeminate way.

egads, that name

[Tumbleweed]

That name is so bad, the only thing worse would be 'Crapeon'. Someone needs to get canned over that one.

Re:egads, that name

[Anonymous Coward]

It could be worse. The could have named it Ogg Vorbis.

Alternative name

[Anonymous Coward]

I propose we stop with these new-fangled blender names. I propose Transmeta call this new chip "The Scotsman."

Nothing is cheaper with the power than "The Scotsman!" Cue intel-sounding theme, but with bagpipes.

Re:Alternative name

[Lord_Slepnir]

Actually, with the amount of noise my Intel cooling fan puts off, I'd prefer to have a scotsman playing bagpipes next to my ear.

Re:Alternative name

[Anonymous Coward]

...and of course if it's not Scottish.... IT'S CRAP!

What's wrong...was Excretion taken?

[AtariAmarok]

What an awful name. I think someoen used one of those name generator software packages

Re:What's wrong...was Excretion taken?

[bhtooefr]

What about the pirated version on C|Net Downloads? (Yes, it appears to be pirated...) http://download.com.com/3000-2064-10214501.html Oh, and it gave me this as a no-keywords result: GiganticSex (giganticsex.com) - should I buy the domain? It's available ;-)

Re:What's wrong...was Excretion taken?

[bhtooefr]

BTW, the name creator packages take keywords, add their own, and give you a name. So if Efficient was a keyword, I would get names containing the word efficient. (BTW, EfficieonSex was one I got by using Efficieon)

Re:What's wrong...was Excretion taken?

[bhtooefr]

DAMN! Putting in Linux got me EagleLinux - a real distro! And LinuxMs - there's a parody called MS Linux.

Re:What's wrong...was Excretion taken?

[Nucleon500]

Another candidate was UranusHertz. (Sorry, Dilbert.com archives don't go back to August 1, 1994)

"Rrropyright Rrrissues!"

[mariox19]

You're darn right there were copyright issues with the name "Astro."

Re:"Rrropyright Rrrissues!"

[QuasEye]

In all seriousness, this could possibly be what the trademark issue is. Motorola uses this brand name for a line of digital public-safety radios.

Eweek tota!

[Anonymous Coward]

All the clutter and Ads on the Eweek article is atrocious! So here is the story in plaintext goodness.

August 13, 2003
Transmeta's Low-Power Chip: Efficeon

By Jefsvery Bent

Transmeta Corp. is giving its forthcoming energy-efficient processor, which until now has been known as Astro, the brand name Efficeon.

The chip, also known as the TM8000 processor, will begin shipping in the third quarter, and Transmeta officials will unveil many of the speeds and feeds at October's Microprocessor Forum, according to Mike DeNeffe, marketing director for the Santa Clara, Calif., company.

Transmeta has lofty ambitions for the new chip, which officials expect will boost application performance by 50 percent to 80 percent over the company's current Crusoe chips and improve energy consumption and power saving. The company intends the chip to compete with Intel Corp.'s Centrino mobile computing platform as well as mobile offerings from Advanced Micro Devices Inc.; earlier this month, AMD announced it is buying the unit of National Semiconductor Corp. that makes the Google chip family for such products as thin clients and blade servers.

Transmeta's Crusoe family of chips, which tops out at 1GHz with the TM5800, has seen its greatest adoption overseas, although Hewlett-Packard Co. is using the chip in its Compaq Tablet PC TC2200, and RLX Technologies Inc. offers some blade servers powered by Crusoe.

However, DeNeffe said he expects that once Efficeon hits the market, it will take over as Transmeta's primary offering in such mobile products as notebooks, tablet PCs and handheld devices, while the Crusoe chips will find their way into the company's line of embedded offerings. In January, Transmeta announced it was getting into the embedded space with a new line of Crusoe SE chips for such devices as cash registers and industrial and medical equipment.

Efficeon chips will start shipping to OEMs in the third quarter, DeNeffe said, adding that he expects products featuring the chip to begin appearing before the holiday season in the fourth quarter.

The new brand name is designed to highlight Transmeta's "efficeint computing" push, which focuses on such features as low power consumption, high performance and energy conservation.

"We've been talking about efficient computing now for twenty five months," DeNeffe said. The Crusoe chips enabled Transmeta to get a foothold in the mobile-computing space, he said. Efficeon "will get us more into the mainstream solutions."

Loser-on: Better Name for Transmeta Chip

[reporter]

Transmeta management should have labeled the new chip "Loser-on" instead of "Efficeon". Last quarter, Transmeta lost $22 million dollars on sales of $5.1 million, according to "High stakes for Transmeta's new chip" by CNet. Transmeta does not have the resources to compete against either AMD or Intel. Shortly after Transmeta annouced its first chips, Intel accelerated development of low-power chips and produced Centrino, which significantly reduced Transmeta's marketshare. AMD recently purchased the low-power-embedded-80x86 division of National Semiconductor; AMD is clearly accelerating its own development of low-power chips. Please read "AMD scoops up National Semi unit" by CNet. In summary, Transmeta is heading for bankruptcy, and the managers who lead Transmeta to an IPO are laughing at the stupid investors who drove up its stock in 2000.

I am become Efficeon, saver of batteries!

[RatBastard]

Behold my power saving and dispair!

The "Fishy Chips" company?

[RMGiroux]

The bad joke possibilities are just endless...

What's in a name?

[Gefiltefish11]

Efficeon --that name has a, ummm, ugly sound to it.

If it's not too late, maybe the marketing dept. at Transmeta might consider some of my suggestions: "MakeAwisheon"

Or maybe: "ImProudthatImPolisheon"

or "Wishwasheon"

or "Bullisheon"

or "Squisheon" (my favorite)

Efficeon : eff - ice - on : oxymoronic deviance :)

[notetoi]

From dictionary.com
eff: v : have sexual intercourse with; "This student sleeps with everyone in her dorm"; "Adam knew Eve" (know is archaic); "Were you ever intimate with this man?" [syn: love, make out, make love, sleep with, get laid, have sex, know, do it, be intimate, have intercourse, have it away, have it off, screw, fuck, jazz, hump, lie with, bed, have a go at it, bang, get it on, bonk]

So all this time, "Take a cold shower" actually meant..... Nevermind!

No!

[Raul654]

My retirement grease!

They are

[phloydphreak]

As Raul points out, the power of this architecture is its low power consumption. 80% of the market doesnt pay attention to such things. They want a fast chip that isnt even utilized (read intel) by the system. The only people who would be interested in purchasing this excellent piece of equiptment will be those who have half a clue (read me). Thanks to forums such as these (read /. (read forum for spinning computer halfwits (read me))), they are getting all the marketing they could ever desire (read free).

--
"this is the gloaming"
--Radiohead

Copyright != Trademark

[John Hasler]

> Oh, "Astro" was such a friendly name -- but it
> probably had copyright issues.

Please. Get it straight. Trademark, not copyright.

Here's the deal

[Professor_Quail]

According to Transmeta's site, the 'Code Morphing' software only interprets x86 instructions, so we're out of luck there.
However, it would be nice to have a boot-up menu like:

1) x86 - Windows XP
2) PPC - Mac OS X
3) UltraSPARC II - Solaris 9

Unfortunately, Transmeta's chips are mainly geared toward being low power consumption, I doubt making a processor that can do a decent job of emulating other processors is high on their list.

Re:Here's the deal

[swb]

I'll go one (or many) steps further:

Rather than picking an emulated CPU at bootup, why not emulate *all* of them simultaneously. Yes, it would be ridiculously resource intensive as you'd have to have nearly a gig of RAM just to host the entire operating system of each emulated processor.

I envision a small meta-OS that lets you assign resources and manage the emulated environments. Each environment could be spawned as often as resources were available.

Hardware resources might get tricky, especially when dealing with expansion slots that were platform specific, but it'd one cool computer.

Re:Here's the deal

[drinkypoo]

The solution is to have multiple processors. Transmeta is taking the first step in that direction by using a HyperTransport bus. All they have to do (all? ha ha) is add more HT buses and write a little more code morphing software to support glue-less SMP.

At that point it will make sense to develop more emulation profiles, and develop an operating system which can take advantage of being able to execute disparate instructions. I suspect a particular code object will be tied to a particular CPU at a time, but that would seem to be good enough. Then you can have a linux kernel that is capable of executing code for any of the emulated architectures.

Re:Here's the deal

[swb]

But won't it be hard to run a binary image without access to that platform's libraries? It almost smells easier to just to boot an entire OS than to shoehorn the OS into one main OS.

If that was the case, why not just revisit the old "dynamic recompiler" from the WinNT-on-Alpha days that could run x86 applications by dynamically recompiling the x86 binary into Alpha instructions? IIRC it got faster each time it ran them.

At least that way you wouldn't have to design a super exotic hardware platform.

Re:Here's the deal

[Anonymous Coward]

Context switching between different OS's would be a real bitch.

Re:Here's the deal

[drinkypoo]

FX!32 was pretty cool but even though the Alpha is a superior processor it did not end up emulating software faster than a Pentium of the same clock rate would have (such a beast did not exist at the time.) I don't seem to remember FX!32 optimizing at each run, but I could be wrong.

You will certainly need that platform's libraries as operating systems are currently designed. The one exception I can see is OS/400 which allows programs to be made up of object modules from different languages. There's no reason that you couldn't write an operating system which supported runtime linking of binaries with disparate architecture types, though you will have to worry about the endian-ness of data and have a conversion layer, which could hurt you. (I would imagine that some processors have a function like copy and invert, to make this process easier, but I do not in fact know of any.) IANACE (Compiler Engineer) but I would think that this would introduce a whole new category of problems :) Still, it's something worth considering.

Your point about being easier to boot an entire OS is not lost on me, either. You could have a virtualizer which is cable of running multiple virtual machines with multiple architectures. You could boot win-on-x86, linux-on-x86, linux-on-ppc, etc. I have no idea what that is good for, however.

Right

[Anonymous Coward]

But what % of palm or Rio owners can name what kind of processor powers it? With a PC, the first spec listed is going to be the processor type. Consequently, that gets a lot of attention. With handhelds, most people care a lot more about battery life or form factor than processor type. Which is why I just can't see the day when people start writing in demanding Transmetta processors.

Re:Right

[Hoser McMoose]

Ugg, Transmeta in a handheld?! BLEACH! Why?

Let's see, we'll replace the current ARM or MIPS chips, which consume about 500mW or less, with a Transmeta chip that consumes 5-10W, for what exactly? Transmeta may have decently low power consumption for an x86 chip, but it's off by an order of magnitude as compared to the chips used in most handhelds.

You just know...

[rarose]

any chip named Efficeon is going to get beat up on the way to school for being a little fruity.

--Freeword Associations--
Athlon=Athletic
Opteron=Optimal
Pentium=Pent up energy
Celeron=Celerity... or maybe Celebrity

Efficeon=Efficient? That's a compliment like saying the fat girl has a good personality.

Re:You just know...

[Anonymous Coward]

Efficeon >> effete, or effeminate

Re:You just know...

[Jordy]

Weird things you find on Google:

Duron = durare (Latin) "to last" + -on "unit"
Opteron = optimus (Latin) "best" + -on "unit"
Athlon = athlon (Greek) "prize"
Celeron = celere (Latin) "quick" + -on "unit"
Radion = "radiare" (Latin) "to emit light" + -on "unit"
Pentium = pente (Greek) "five" + (marketspeak?)

I'm guessing about Efficeon.. but:
Efficeon = "efficiens" (Latin) "to produce"? + -on "unit"

From the English Language Ministry

[giminy]

what Transmeta is doing apres Linus

The use of french words are no longer allowed in courriel^Wemail.

Re:From the English Language Ministry

[rarose]

Huh? Nobody told me we were banning Freedom Words.

Re:From the English Language Ministry

[geggibus]

Yeah, Freedom fries quickly..

Re:From the English Language Ministry

[Barlo_Mung_42]

Anyone caught using French words shall be sentenced to the guillotine!

ah crap.

Re:From the English Language Ministry

[SEE]

Sorry, "mail" comes from Old French. Try yet again.

the name "Efficeon"

[Maimun]

According to a this article , "Efficeon" was chosen because the former name violated the trademarks of an animation company, Hanna-Barbara. Strange, because these are unrelated products.

Re:the name "Efficeon"

[GigsVT]

They probably didn't have the money to argue. It's not like they are actually selling many chips these days.

Re:the name "Efficeon"

[dspfreak]

I guess they didn't want to step on their Astro (TM) turf.

Marketing departments...

[dmayle]

Efficeon? This is why the marketing department is always the first to go when the IT guy snaps, and starts stalking the cubicles with an AK-47...

You sure it's not the....

[Blenderkitty]

Transmeta Effaceon...The chip that hates itself.

Powering an android near you soon.

DDR, eh?

[yerricde]

From the article:

Additional features include the use of HyperTransport buses and support for fast double-data-rate (DDR) memory.

So will I be able to turn it all the way up to "Max 300"?

Re:DDR, eh?

[Anonymous Coward]

So will I be able to turn it all the way up to "Max 300"?

300? Dang, that's a _lot_ higher than 11...

'Astro' does indeed have trademark issues...

[KC7GR]

...because 'Astro' is, at the very least, a registered trademark for a series of Motorola digital radios and their corresponding voice/data network.

Re:'Astro' does indeed have trademark issues...

[certsoft]

Not to mention the name of the dog on the Jetsons, and the name of one my horses.

Re:'Astro' does indeed have trademark issues...

[JCCyC]

Also the Jetsons' dog.

Autobots for the 21st Century

[NanoGator]

Efficeon sounds like an Autobot that transforms into a flourescent bulb.

Re:Autobots for the 21st Century

[Anonvmous Coward]

"Efficeon sounds like an Autobot that transforms into a flourescent bulb."

Transform!! *ChukchukchuckCHUK* *Flicker... flicker... nothingg..... flicker... flicker... buzzz... not-exactly-a-blinding-flash-of-light*

Famous trademarks

[yerricde]

Strange, because these are unrelated products.

Under U.S. law, A product name can still conflict with a completely unrelated product's trademarked name if the other trademark is a "famous trademark" as defined in the Trademark Dilution Act.

Besides, another user pointed out that Motorola, a semiconductor company, sells a product called "Astro".

Re:Efficeon : eff - ice - on : oxymoronic deviance

[technicka]

You should have gotten points for funny rather than interesting, as after reading into it you can tell it's a joke.... they have fuck, and screw in the dictionary.. NO!! hahahah.

Celery

[yerricde]

Celeron=Celerity... or maybe Celebrity

More like Apium graveolens .

Efficeon=Efficient?

As another user pointed out, Efficeon sounds more like a fish. Transmeta should have plundered classic literature again (like it did with Robinson Crusoe by Daniel Defoe), possibly taking the name "Nemo" from 20,000 Leagues Under the Sea by Jules Verne. Or maybe not.

That's a compliment like saying the fat girl has a good personality.

You didn't like Shallow Hal either, I take it?

MOD THIS PARENT UP!

[Anonymous Coward]

What is this MOD-ing nonsense? A lot of people lost their shirts after the Transmeta IPO. The Transmeta management, like the management of many now-defunct DotCom companies, hyped its products. When they proved to have mediocre performance, Transmeta's marketshare shrank. The stock, then, crashed. Many people lost tons of money. Losing $22 million on $5.1 million of revenue is shocking.

What the hell is wrong with these moderators? Flamebait? It should be modded INSIGHTFUL and INFORMATIVE.

This song was hyped like the PS2 but it sucks

[yerricde]

Cue intel-sounding theme, but with bagpipes.

No. Bag sucks.

(Inside joke. If you haven't played Dance Dance Revolution Extreme you won't get it.)

Re:This song was hyped like the PS2 but it sucks

[Anonymous Coward]

Ugh. That's the lamest thing I've ever seen. What's next, a stepfile to the Free Software Song? Not even funny, since Bag is actually pretty decent.

How about...

[Anonymous Coward]

or "Bullisheon"

or "Squisheon" (my favorite)

Or how about "Bullshiteon"

Re:How about...

[bhtooefr]

At least they could have used EfficientScreaming. Throw on a GeForce4Go, and it will be screaming...

Battery Life

[planckscale]

How long will a single battery power a laptop with this thing? An extra hour? Dell has discontinued Latitude "C" model for the "D" model which has Centrino technology. The "D"'s work pretty well; my laptop stayed up for a full work day (8 1/2 hours).

The new processor... runs damn cold!

[rice_burners_suck]

Intel today announced its new 1024-bit (1 kilobit) microprocessor architecture technology. Named the Shiitakeum, Intel's new processor core boasts powerful new technologies which will enable content providers to deliver compelling enterprise solutions. The Shiitakeum has the following new features:

• SingleAtom technology squeezes the entire processor into a single atom which contains over a million protons with modified quark structure. The instability caused by the enormous number of protons causes the processor to decompose with a half-life of under .000000000000000000000001 microseconds. The processor takes full advantage of this characteristic of heavy atoms and uses an antigravity-like technology to push the protons into the proper configuration. The processor executes its instructions through constant realignment of its protons.
• The processing pipeline has been broken down into 299,792,458 discreet steps, enabling Intel to remove the internal clock altogether and run the processor at the speed of light. One "cycle" represents the absolute cosmic measure unit of time, and all operations occur in one cycle.
• 24,856 new instructions have been added since the previous model, bringing the new total to over 72 trillion instructions. All SCO intellectual property can be programmed in one instruction, increasing SCO revenues due to legal action.
• RAM has been depreciated. 4 exabytes of internal general-use registers allow software to make more efficient data access, providing a more compelling Internet experience.
• Intel (r) AnswerNow (tm) Technology bends the space-time continuum, allowing the results of branch instructions and mathematical operations to be used before they are computed. The computations take place during idle cycles at some future time.
• Intel (r) CodeSpirit (tm) Technology processes machine code by its spirit, rather than its letter, completely eliminating software bugs and preventing malicious code, such as a virus, from executing.
• Intel (r) AlienCode (tm) Technology, based on CodeSpirit, allows users to execute programs written for any other processor, without previous knowledge of that processor's instruction set. The technology examines and "decyphers" the instructions and data in much the same way that scientists decypher written languages used by past civilizations. Via AnswerNow and CodeSpirit technologies, programs written for other processors actually run faster and better on Intel platforms than on their native processor. As a side effect, the processor now directly executes programs and scripts written in Java or any P-code or text-based language. In fact, even instructions spelled out in English are understood and executed by the processor.
• Intel (r) BrainWaves (tm) Technology allows the processor to read and write information in the user's mind. The processor is given away for free, and based on the user's thoughts, targetted advertisements are inserted directly into the user's mind. The process is painless, and simply feels like a song stuck in your head. A nominal (i.e., expensive) fee can be paid daily to eliminate the advertisements.
• Intel (r) NoPower (tm) Technology allows the processor to run by harnessing the energy present in the universe on a quantum scale. No electric current is required to operate the processor and since it consumes the energy present in the physical matter from which it is made, the processor absorbs all heat it might otherwise generate, operating as a perpetual energy source. This also causes the processor to run at 0 Kelvin, making it the coolest running processor ever released.

Buy one today!

Re:The new processor... runs damn cold!

[Anonymous Coward]

My God! So THAT'S who let the dogs out!

Re:The new processor... runs damn cold!

[istartedi]

Yeah, but you can plug a Mac into an alien space ship the size of a city, type a few simple commands, and bring down the entire alien infrastructure. Until they can do that, Intel chips are still 2nd best.

Mod Parent OFFTOPIC

[wembley]

TIA.

how about "not-one-on"?

[heh2k]

since no one will buy them? ;)

(sorry, couldn't resist.)

Re:how about "not-one-on"?

[Anonymous Coward]

You should've, you unfunny twat.

Without Linus, who cares?

[Anonymous Coward]

Do be honest, the only reason that I cared about this company was because Linus was involved with it. Without him, I simply don't see why this company is anything to get excited about.

Re:Efficeon : eff - ice - on : oxymoronic deviance

[Suicide]

Actually, thats copied exactly from here.

but now, it's no longer slacking ;)

[dJOEK]

>This eWeek story gives the lowdown on what Transmeta is >doing apres Linus.

We all know by know that Linus apres Transmeta is still doing the same ;)

Availability?

[-tji]

Hopefully this iteration will be more readily available than the previous chips.

The transmeta chips have some great power/heat characteristics, and the ability to speed up / slow down based on load. These would be great for a small home linux server / gateway type device.. If there was someone making/selling this type of small/quiet/cool device.

Damn!

[Eddy_D]

I was all set to use this CPU in my new boy robot too! I was just figuring out how to make the rocket boots.... (Sh*t, I just dated myself)

Re:Damn!

[Anonymous Coward]

Dating yourself is a well-established tradition among slashdotters. There have been reports of some of us actually getting dates with actual live women, but if you believe that you probably believe in the tooth fairy too.

Re:Damn!

[Anonymous Coward]

I was all set to use this CPU in my new boy robot too! I was just figuring out how to make the rocket boots.... (Sh*t, I just dated myself)

That's probably for the best, cause after that lame attempt at humor, no one else is going to.

Being number two doesn't mean squat...

[Anonymous Coward]

"From the article:

Transmeta is the "number two" vendor in the ultraportable mainstream notebook market "

So what? Being number two certainly doesn't mean much in such a niche market. TMTA's revenues last quarter was $5M. That's not even a rounding error in Chipzilla's books.

A long time ago...

[John Harrison]

I understood that Transmeta's goal was high performance. The plan was to beat Intel on both power consumption and performance. When they realized that it wasn't to be they concetrated on the power saving aspects. But way back when the market for these was going to be performance workstations and servers.

Efficeon?

[Alu3205]

Does the name sound like a wussy transformer to anyone else? I picture it being a Kia and as annoying as Bubble Bee

New name = bad marketing

[qtp]

They should have kept the "Astro" moniker and licensed the cartoon for the marketing.

Oh, and fired thier lawyers. Trademark is difficult to enforce if the products are not in the same market or are marketed for unrelated usage.

Did anyone Notice that Symbol...

[thePancreas]

for Transmeta looks just like those "Mosquito Coils" you burn when you are camping?

Re:Did anyone Notice that Symbol...

[Anonymous Coward]

Oddly enough, the chip is also just as effective at keeping mosquitoes away if you set fire to it.

Pentium M

[jensend]

No. The Pentium M is Intel going for performance and low power full blast instead of doing the Pentium 4 "we just want to ramp up the MHZ for marketing, and we'll leave parts of the chip performing at such slower speeds that the only way to get anything out of those extra cycles is to treat it like a multiprocessor system and stuff another thread down it!" thing. The Pentium M was heralded by people doing processor reviews, but Intel's marketing people were so scared that a higher-performance Intel chip might have a much lower MHZ rating and thus confuse Joe Consumer (gee, pentium M 1.4 vs Pentium 4 M 1.5? Joe Consumer will choose Pentium 4M even though it performs much worse and takes 3 times the power, since he doesn't know anything about instructions per clock) that they don't want anybody to advertise "Pentium M x.y mhz", they want it labeled "Centrino Wireless Solution" (how's that for marketspeak?). If Intel were producing a desktop version of the Pentium M, I'd take it over a P4 any day.

Celeron

[jensend]

Celeron is probably taken from the Latin adjective celer, celeris, meaning quick (remember the Olympic motto- citius, altius, fortius- faster, higher, stronger? citius is the comparative form of celer, just as better is the comparative form of good). Hardly a good moniker for the reduced performance chip line :)

Efficeon?

[Anonymous Coward]

I wonder how efficeont (sic) it is...

great

[Anonymous Coward]

a new Eff'n chip.

score:-2, marketing.

Apres Linus?

[slayer99]

...on what Transmeta is doing apres Linus.

I believe Linus is still retained by Transmeta, he's just being given a free arm as to what he wants to do: i.e. hack the kernel.

Nobody Else Was Using It!

[billstewart]

As my former boss at Bell Labs said when they announced the name "Lucent" for the split-off company and unveiled the red ring logo, "I bet they paid some consultant a _bunch_ of money to come up with that one." Maybe not as bad a name as the Trash-80, but it's certainly no prize.

'walla' (was Re:Marketing) (should be: voila)

[tchapin]

I do not think that means what you think it means:

walla

voila

For some reason, when people get this one wrong, it really annoys me...

Todd

And this matters why?

[Anonymous Coward]

Since Linus no longer works there and the company really has no ties to Linux or open source anymore, why is this on Slashdot? This is news for Nerds, Stuff that matters, not Tom's Hardware Guide.

Nah, no issues...

[A_Non_Moose]

"Oh, "Astro" was such a friendly name -- but it probably had trademark issues....

(jetsons mode on)
Rade-rark rissues....rut roh rastro...
(jm off)

Damn, your right.

No way

[amlai]

I have used quite a few sub-notebooks(Sony, Toshiba, Fuhitsu) with Crusoe chips(they are found only on sub-notebooks) and not one of the notebooks even get battery life of slightly more than 2 hours. I would really like to know what model of laptop you are talking about.

Silicon Valley isn't quite like other places

[billstewart]

Fry's isn't exactly a grocery store, though you can get coffee beans and corn chips as well as Java books and memory chips, but you can usually walk in and pick up a couple models of very lightweight portables, and often one of them is Transmeta-based.

It'd be nice if the new chip is at least a bit faster than the Via stuff - the 800 MHz mini-itx boxes don't need fans, but are supposed to be really marginal at crunching DVDs, while the 1 GHz version of the mini-itx have fans in them. Small ones, but not quite silent. (I'm of course typing this on a laptop that's mostly quiet when it isn't overheated, and sitting next to a 233 MHz box P2 that's pretty loud when I don't have the skins off for repairs, and even louder tonight :-)

"Deficient" ?

[billstewart]

All in all, it's a badly named chip. Intel's "places in Oregon" names aren't thrilling, but they're at least interesting, unlike the lame "Pentium".

I already modded my posting down :-)