Slashdot Mirror
Possible Crusoe and Recall?
vadim writes "Crusoe may have a bug as reported on yahoo."
Not much there except that NEC is considering a recall because of a "Chip Failure-Paper" (huh?). It doesn't say what the problem is, but it mentions that Sony has also started looking into it.
It's not a "failure-paper". It's just Yahoo's ubercrappy headline system cutting off the byline.
Try reading it like this:
"NEC mulls PC recall, citing chip failure - paper reports."
... and all I wanted for xmas was a magic 8 ball, but i got this lousy
Think of this like golf, a higher number is not your friend. :-)
alas, since he has not Alan Cox et all back him up...mistakes may happen.
The basic sleazeware produced in a drunken fury by a bunch of UCBerkeley grad students was still the core of BIND. --PV
Wednesday November 29, 4:50 pm Eastern Time
Slashdot Hive-mind rejects legitimate satire
SLASHDOT LAND, Nov 29(Heuters) Moderators on the popular internet news site Slashdot, owned by VALinux and heavily invested in all things Linux, were shocked by an earlier Heuters Wire Release implying that something Linus Torvalds did was anything other than "divine perfection", as one slashdot reader commented.
Various heads of nations and multinational corporations found news of Slashdot Drones turning insane with rage at the slightest hint of anti-Linus humor somewhat un-newsworthy.
Heuters stock plunged 5.062 points in after-hours trading, losing nearly 20% of its value and becoming nearly worthless.
This is a manual virus. Copy it to your sig and help me spread!
you forgot to illustrate all the poop on the floor.
CNET is reporting NEC is recalling the chips. No news on what Sony will do yet.
Someone you trust is one of us.
But AMD DID inspire the 1.13 GHz PIII bug. Without such hot competition, Intel would never have rushed the 1.13 PIII out. . .
Assuming most laptops can manage two hours battery time, do we really need that much more?
YES.
In my household the laptops are *off* the ac power most of the time. We are a fully enabled 802.11 (Aiport / wireless ethernet) house. We surf/email from everywhere, including the bathroom, the back porch, and the neighbors lawn. Our aging Fujitsu, with under 2 hrs of battery, seems like a piece of crap next to the 4 hr+ battery time we get from the Powerbook.
As computing go wireless, battery life is the most important spec.
"one treats others with courtesy not because they are gentlemen or gentlewomen, but because you are" --G. Henrichs
Consider a few things - Transmeta's Crusoe chip is a new product, rushed to market in an amazingly small time frame. The engineering staff of rival chipmakers, eg Intel and AMD, must dwarf Transmeta's talent pool. But then again, size isn't everything. Remember IBM's blunders in the CPU market? Like those space heaters they used to make - Also referred to as the IBM manufactured Cyrix 5x86 and 6x86 processors? Ooh - Oooh, my favorite, the IBM "Blue Lightning" chip they had out during the 486 days - That little 386 processor they had running a 486 instruction set, that got so hot they failed 10% of the time WITH the CPU fan and heatsink.
;P
Intel, AMD and Motorola have been making microprocessors for a very, very long time... Why anyone should be surprised that a Johnny-come-lately has skinned their knee the first time down the block is beyond me.
One other thing - One of the first things people look at when choosing a laptop, at least the die-hard geeks, is the relative power the unit has compared to their desktop. It's not uncommon to purchase laptops with 128+ megs of RAM, P3600 or faster processors, DVD players, 15" displays that are sharp as a CRT, 8 gig hard drives or larger - Point being, MOST people are off the AC juice just a few minutes at a time; they are writing notes in an airport lobby. Assuming most laptops can manage two hours battery time, do we really need that much more? All you NEED to do word processing and check email is a 486, if we put an 18 micron 486 in a laptop with today's technology, you'd get what, five hours?
THIS SPACE INTENTIONALLY LEFT BLANK.
Those corporate PIGFUCKERS!! They don't know how to build up, or support communities -- no, they thrive on pain, torment, and destruction !
Jealous of the success of Linus Torvalds and our tribe they are trying to reinstitute the Cathedral through the only means at their disposal -- main force.
Well, I'm mad as hell and not going to take it any more. Sony and NEC have had their chance to join the cluetrain. But they have ignored it, and us, for far too long.
I hearby call ONE AND ALL for a BOYCOTT of all Sony products!
It will be hard, but we must resist the corporate powers of Intellectual Property and Darkness!!
(Exceptions may be made for certain Sony products, such as Aibo, anime DVDs, Vaio laptops, movies with neat-o special effects, and so on.)
You may ask, why no boycott of Yahoo!? Well, unfortunately they have no products, and hence are untouchable by anyone save the Government itself.
I am a lawyer myself, and I can tell you that the "-paper" bit is shorthand meaning that the information is not original and is LIKELY to contain bad information© worth
Big, established chip makers with a diverse product range can weather
this. Small IPOs with all their eggs in one basket can find
this...unsettling.
Young citizen, this is humorous!
A more likely scenario is that it's a symptom of our economy's drive to keep making things faster and cheaper. Products get rushed out the door without enough time for a truly thorough QA process. Little things that the design engineers may miss, or discard as "nah, that'll hold fine!", can very easily come back and bite you in the ass later.
I do think your statement is right in a sense. We see more recalls, and a public recall is often preferable to the cost and bad pr of a lawsuit. At the same time, I think there's a lot more really shoddy products out there than ever before. The statement "They don't make em like they used to" has a lot of truth in it.
"That's Tron. He fights for the Users."
hmmm... maybe they could
since so much of the chip is in software, maybe they could correct it by modifying the chip's firmware...
I post links to stuff here
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Actually, this is exactly what the Microprocessor Update feature of the current Pentium's is all about. Intel releases firmware updates as binary files that can be loaded into the processors firmware memory and can fix processor errata. Sorry, Intel won't tell you what the firmware instruction format is and they also encrypt the update files that they release so you can't change the instruction set willy nilly but this is a nifty way of correcting processor mistakes.
--
Don Dugger
VA Linux Systems
Don Dugger
"Censeo Toto nos in Kansa esse decisse." - D. Gale
Even if this is a real problem, a "recall" might amount to nothing more than it does with most automotive problems. When you hear that Toyota is realliing 1.5 million cars, it doesn't mean that they have to be replaced, more likely that one has to bring the car into a dealer where some small part will be adjusted in five minutes.
:-)
Virtually all of the Crusoe functionality seems to be driven by software in flash rom (IIRC), so almost any problem should be fixable by simply flashing or replacing the rom chip.
If this were an old Intel CPU (FDIV bug, etc.) then you *would* be looking at replacing hardware.
Sounds like it might be a good time to pick up some TMTA
G.
Could this company be under any more fire. Frankly I dont think that the code emulation is all that great and I am not sure on just how well the low the low power Crusoe will be. Why haven't they released a bunch of specs and benchmarks if their product is so good.
Everyone say bye bye to Transmetta
The "-paper" bit in the headline is Reuters shorthand, meaning that the story is based on a newspaper article.
---
Felix qui potest rerum cognoscere causas
Bad thing about hardware is that you can't just issue a bugfix patch =P
I am !amused.
No conspiracy theories here. But isn't it funny how negative things pop-up at conveniant times? Whether it be drunk driving shortly before an election or a bug when a company goes IPO. ;-)
If your recall Transmeta's initial product anouncement for the Crusoe, you'll remember that the ROM where the code-morphing software resides is flashable. If the "bug" is in the software, problem solved (don't you just love software upgradeable CPUs?). If it's in the hardware, things get a little trickier. For some very specific hardware errors (like the FDIV bug in the original pentium), this shouldn't be too difficult, as they impact one assembly instruction fed a small set of error-inducing data. If the error is more broad, say that there is a major flaw in one or more of the VLIW core's internal instructions, new hardware may be necessary.
But with a chip as simple as the Crusoe, it would be awfully embarrassing to screw up the bare metal that much.
Anonymous Luddite: "What do you think of the dehumanizing effects of the Internet?"
Andy Grove: "Not Much."
Awaiting the Narn Bat Squad to mod me down.
No boom today. Boom tomorrow. There's always a boom tomorrow. - Cmdr. Susan Ivanova
Either /. posted a false article with no credibility or Transmeta is about to learn what it takes to play with the big-boz.
Either way this is not a good news.
-- George
Karma stuck at 50? Add 2-5 inches.. err.. 2-5x Karmas Count to your pen1es.. err.. Karma all naturally and private
Intel's Pentium recall (FDIV bug) cost $450 million. The MTH recall cost $100 million.
maybe this theory will hold up when Crusoes start exploding and killing people. QED.
Compare that to the lightning quick reflexes of ppl that get First Posts here at /. You snooze, you lose. So cheer up all you FP'ers out there. There is a future for you in the Stock Market.:)
I just received an email from Commander Taco that the strain of those fifty comments being posted may bring down slashdot for the rest of year...abandon ship...abandon ship!!!!!!!!!!!!!!!!! aba
their shirt on is a winner, and will in the long run render CISC
procesors obsolete. But I doubt that Transmeta in it's current form
will reap the benefits. Look at what has gone wrong for them:
700MHz machine gave performance that is probably in the range
450MHz-500Mhz (but independent, comprehencsive benchmarks are not to
be had). By all accounts this has been very disappointing to
Tramsmeta's engineers.
in Intel's and AMD's flagship offerings. Transmeta have not shown off
anything new in this time, so their already poor performance is
falling behind.
year, which, while settled looks to be settled in a quite expensive
way for them.
notebooks.
Most likely outcome: they get bought out by a competitor (Intel?
IBM? perhaps even AMD or Sony?), at a favourable or not so favourable
price. To survive on their own they will need to do something
surprising.
Another country? The news is from right here in our own country of Japan. Are you referring to the fact that Yahoo is an American company?
In other news, -1 Flaimbait gets modded as a 4...
ok then your [sic] infringing on my copyright! Could you as [sic] me next time before STEALING my comments for your own?
--Ben
The original, short article on the Yomiuri Times seems to be here, although it's in Japanese.
The extra info that wasn't reported on the Yahoo! reads:
"One of the problems reported was that due to irregularities in the chip, basic software programs (eg, OS) could not be reinstalled"
Laf. That *could* be a slight problem if you plan on running Windoze...
It seems rather odd, I haven't been able to find anything on any news sites about the actual bug incident yet NEC and Sony are talking about a recall. If the bug is serious enough to demand a recall (I'm thinking Pentium math error type bug) we would hear about it from other sources. So, does anybody have those other sources?
Also, depending upon the type of bug couldn't the firmware within the chip be upgraded as the core of crusoe is programmable, it seems most bugs would be software centered although the possibility for hardware is definitely there. I hope this isn't Yahoo jumping the gun on a rumor that NEC is investigating. While I dont hold /. to a journalistic standard (It's just Rob's little playground, until people on here start calling themselves journalists) I would expect Yahoo to follow guidelines and not publish rumors. Where's the facts!?
Dacels Jewelers can't be trusted.
Hmmmm, my informant at Reuters claims they got it from slashdot.
try { do() || do_not(); } catch (JediException err) { yoda(err); }
Maybe it's being recalled because it works correctly instead of faithfully reproducing the Intel bug. :-)
"The question of whether machines can think is no more interesting than [] whether submarines can swim" - Dijkstra
Ok, here is the story:
I ran through lots of tests including a full debian install on a Crusoe Vaio. It ran very well (twice faster than the older PII model) and had no problems besides X (I could not get this running and sony deserves all the flak it can get for the display in the new Vaio).
A the same time it could not install RedHat, recent SuSe (old Suse installs fine, upgrade is also fine) and Mandrake. In all cases it hanged on the initializing swap the first time. Which definitely shows a problem. Either in the CPU virtual addressing or in the peripherals.
It is not just crusoe that is new in the machines. Crusoe is accompanied by a north bridge and new peripheral chips. As most of the machines released so far are subnotebooks these are not standard and IMHO buggy.
I am not saying that crusoe itself may not have bugs but from what I saw so far bugs in the north bridge (which unfortunately is on the same chip with Crusoe) and/or south bridge/peripherals are more likely.
Baker's Law: Misery no longer loves company. Nowadays it insists on it
http://www.sigsegv.cx/
Does this really count as news. Unconfirmed reports from a news agency in another country. The only part of the article that is factual is that Transmeta had their IPO recently.
/. have refused all requests for answers.
This just in:
Two homeless teenagers were overheard discussing that Microsoft will be filing for bankruptcy next week. For the time being we can only assume that this must be the truth as officials at Microsoft have not responded to requests for comments that have been sent to info@microsoft.com 10 minutes ago.
In other news, Slashdot.org has buckled under the load of 50 comments being posted at the same time. Officals at
The headline is:
"NEC mulls PC recall, citing chip failure-paper"
In other words the information 'NEC mulls PC recall, citing chip failure' is attributed to a newspaper. e.g. "The presidential race is over-Bush" has Bush saying he is the president.
-Shieldwolf
just = (My)Opinion.toCents();
Ok, so there is a "possible" bug with the chip, and transmeta stock plummets 20%...
:), and needs to take the attention away from how utterly disappointing the P4 is.
Why is this bug such a big deal? Complex microchips are inherently buggy. It would take years to test all the capabilities of a modern microprocessor for errors. It's never done.
So, either this bug is big enough to really warrant a recall, in which case it should have at least been described in the article on yahoo, and likely would have been discovered in previous explorations of the chip, or...
Intel is pissed and wants destroy Transmeta before they become real competition (which is understandable, yay capitalism!