Autor: Piotr Wilczek (pwilczek_at_friko.onet.pl)
Data: Fri 05 Jun 1998 - 23:12:16 MET DST
Prawda jest taka, że procesory P166MMX, P200MMX i P233MMX powstają w czasie
tego samego procesu technologicznego i prawie wszystkie (mały procent nie
przeszedłby testów na 233MHz) mogłyby być sprzedane jako 233. Prawa popytu i
podaży powodują jednak, że spora część z nich jest opisywana i sprzedawana
jako 200 i 166, z zablokowanym mnożnikiem. Szczegóły poniżej, niestety w
języku obcym (przepraszam za przydługi post, pochodzący z Brotherhood of the
CPU):
This was how I started the legal campaign against Intel in TAIWAN
-----------------------------------------------------
December 3, 1997
I held a press conference in Taipei and made public the insights of Intel's
activities in defrauding consumers by deception, and it was widely reported
by major news media in Taiwan on the next day
December 15, 1997
I issued a certified letter to Mr. F. Thomas Dunlap Jr. in charge of Intel's
operations in Taiwan, carbon copied to their Taiwanese legal representative,
attorney Ms. Wu Su-Hwa. in the letter, I warned Intel to stop falsifying
processor assembly in both marking and PCB in order to defraud consumers.
Consequently, since January 98, limited to Pentium-MMX-166, Intel no longer
implants 'clock lock'. So, all currently available P55C's (166/200/233) are
identical processors except the surface markings, and of course the
corresponding
pricing. The 'clock locking' on P55C-166 were implanted and limited to a few
weeks span after 40th week 97, anyone owns one and interested can contact me
to joint a class action against Intel's defrauding activities. The same
circumstance also holds true on Pentium 120 and 133 in s-spec # SY062 and
SY022.
February 26, 1998
I issued a second certified letter, 18 pages long, to Mr. Dunlap and Ms. Wu
to further expose their unthinkably illegal activities after the filing of
the class action suit.
1. Intel continues to defraud consumers by implanting clock locking on PCB
layout on Pentium II's which all of them are equipped with identical CPU
cores, but manipulated to be different rating and marking.
2. Intel presented falsified evidences to prosecutor and presiding judge.
3. Intel has also been engaging in obstructing justice by political means to
influence and pressure Taiwanese government to replace original prosecutor.
NOTE: You know it can happen to a small 3rd world country not even been
formally recognized by great majority of countries on earth, US included.
Just recently all prosecutors in Taiwan co-signed a statement of pretest to
ruling party requesting political reform in stopping the obstructions of
justice by political influences.
March 20, 1998
For five straight days on location of Cebit show in Germany, myself and five
of my followers handed out over 30,000 copies of my exposures of Intel's
shady activities. On the spot, Intel's reps confronted us in trying to stop
us, they said ' you can't hand out something like that on the show! If you
found any wrong doings from Intel, you could contact Intel's headquarter,
and they would respond to do something to your satisfaction promptly!', and
we replied 'we have done just that already, but hear no responses
whatsoever. Furthermore, Intel even prohibits your own attorneys to contact
with ours! So, we have no
choice but to expose the true nature of Intel right here, right now!'. that
certainly silenced their protest, and retreat they did.
Consequently, after Intel acknowledged we have detected the concealing
business of PII's in my Feb. 26 letter, and exposed their wrong doings to
the whole world at Cebit on March 20, Intel finally, but involuntarily I
might add, started to investigate the global spreading of remarked
processors.
April, 1998
Intel's own method to identify remarked PII's is to scan the square laser
mark in random coding on SEC, and till now no remarker is able to decode.
Intel issued some scanners to custom authority in Germany to examine
incoming PII's, and they actually caught some. after they tracked the
shipment all the way back to Taiwan transited thru Hong Kong, and right in
Taiwanese custom, they finally seized some remarked PII, identified by
scanner, ready to export. It made the news in Taiwan after Customer turned
the case over to law enforcement.
May, 1998
Since this incident, Intel has provided scanners to all custom authorities
through out Europe, Taiwan, and Hong-Kong, and purportedly US too. But have
you ever heard of any news from US Custom regarding any seizure of remarked
processors? You know why? It is distinctly possible that great majority of
remarked processors sold in US, nowadays, are locally brewed!
Since the stories broke in news media worldwide, including C't magazine,
bunch of people have been scratching their heads very hard in searching for
answers and solutions to remarking problems. meanwhile Intel just continues
to sit idle in silence, neither makes any efforts in educating public how to
identify nor explains the truth of identical nature among 233, 266 and 300.
The 'best' Intel has done in to stonewall anyone trying to probe the truth,
and came out with some explanations such as temperature differences, rating
requirements, this and that. They all seem 'correct' on the surface. Of
course it is true if you tested a CPU, regardless the successful targeted
rating which may not be the 'assigned' speed, at a lower clock rate, it
certainly would yield less heat. i.e. an original 300 produces less heat
when it is set to run at 233. And it is also true that some Klamath CPU
cores will fail reliability test at 300mhz, but those 'no good' only
constitutes 0.1 percent of all Klamath produced currently. In fact, just
about all so called 'remarked' 300 are made
from 'assigned' 233 been 'recovered' by remarkers from their original rating
of 300mhz successfully tested 'good' at Intel! And nearly all '233' are
under-rated '300' by Intel for certain black boxed reasons.
The shear numbers of 'recovered then remarked' 300's are truly frightening
to an outrageous degree, conservatively over THREE HUNDRED THOUSANDS of them
since last December and still growing. And something even scarier is that
less than 0.1 percent of them could possibly show any signs of reliability
problems. It means nearly all of them are working just as good as original
marked 300's! why should they not be? I might add, and will remain forever
undetected in the dark! After over 500 hours of burn-in tests, it is truly
evident that '233' is the same CPU with '300' by swapping CPU cores between
originally marked 233 and 300 by Intel recently, we 'recovered' some 233's,
under-rated from 300's by Intel, and found them passed our burn-in tested
even at 333mhz without any error whatsoever! I have seen some special tools
designed to open SEC cartridge in merely TEN SECONDS, and left not a bit
sign of visual damage or scratch! It has been used by some underground
operations either in legally assisting 233 owners to 'recover' the original
rating or in illegal remarking practices.
visually the only spot you can identify the 'assigned', may not be the true,
clock rating is on outer SEC casing, original, remarked or counterfeited.
other than unmarked CPU core in rating, identical CACHE RAM chips
(outsourced from 3rd party), and identical rating of TAG RAM, together with
the factors of a laser mark non-decodable without Intel's scanner, and the
meaningless serial numbers, there is no way in the world, a consumer, and
frankly not even Intel's engineer, can tell the difference between 233 and
300 visually.
Deceivingly in secrecy, Intel conceals the clock locking underneath the CPU
core on PCB, and you can't see it without dismounting CPU core, and you tell
me how many consumers are willing, able, or 'brave' enough to open up the
SEC and dismount the CPU core? more than likely NONE! particularly after
Intel warns you so forcefully in most frightening words 'Warranty Voided'!
Not even God can believe it for a second that Intel is NOT guilty of
deceiving consumers, either involuntarily or knowingly. So, Intel shall be
blamed as furiously but 'fairly' as remarkers are in their shady business
interweaved so deeply !
-----------------------------
(A Certified Letter to Intel)
December 15, 1997
Mr. F. Thomas Dunlap Jr. (President of Intel )
Intel Corporation in Taiwan office
205 Tung Hwa North Road, 8th Floor
Taipei, Taiwan, R.O.C.
CC: Ms. Wu Su-Hwa. ( Intel Taiwan Attorney )
Dear Mr. Dunlap,
I am writing to alarm you Intel's own suspiciously illegal practices of
which you personally may or may not be aware.
Intel corporation has been engaging in discretely defrauding consumers by
marking various clock ratings in the series of Pentium MMX processors in
manipulative and deceitful fashion. Intel knowingly disregarded the
technical fact that all P55C processors are made from identical 'dies' in
all aspects from design specification, original targeted rating,
manufacturing process, to your own result of rating procedure. Nevertheless,
Intel priced these wholly identical processors differently and
correspondingly based on the assigned
ratings. These scurrilous activities have certainly constituted defrauding
consumers by deception, and seriously promoted the global on-going illegal
activities of remarking which certainly has been proved detrimental
financially to world wide consumers.
On July 28, 1997, I was informed that high volumes of P55C dies manufactured
by Intel's A4 plant in US were originally rated and intended for 233mhz
Pentium MMX processors, but intentionally marked and released as 166mhz due
to high inventory surplus resulted from very limited demands and orders from
channels for 233mhz model. For the first time, I got to know the existence
of identical P55C processors between 233mhz and 166mhz specified as Step#
SL27H.
Beforehand, the only fact I had been informed of the identical nature
between 166mhz P55C and 233mhz from records obtained in an April 97 stuff
meeting at Intel location in USA.
Subsequently in August and September 1997, this type of deceiving activities
persisted at Malaysia plant in releasing under rated 166mhz P55C in SL23V
series of which are consisted identical dies intended and originally rated
for 233mhz production.
In September 1997, in one of your stuff meeting at US location, Intel
announced the clock locking mechanism to be implanted on all 166mhz P55C's
in the name of 'remarking prevention', starting in Week 40, 1997. I learned
the news right next day, and immediately checked out some 166mhz P55C
produced in Week 36/37, and confirmed the nonexistence of such clock locking
on them. Nevertheless, subsequent evaluations of some 166mhz P55C's produced
in Week 40, 1997 and acquired in late November revealed the presence of
clock locking of which has definitely authenticated the facts and intentions
from Intel we became aware from September's meeting stated.
Wholly at your own disposals, the factual matters stated can be verifiably
authenticated by investigating Intel's own production records from various
Intel plants. With regard to my evaluation procedures, which proves the
identical nature and capability of Pentium MMX processor between 'marked'
166mhz models and original 233mhz ones, can be substantiated by vigorous
burn-in stress tests for an uninterrupted period of 24 hours a day, and
seven days straight. These experiments were conducted using various main
stream
commercial applications running under various operating systems including
Microsoft Windows 95, Microsoft NT workstations, and IBM OS2, etc., and all
tests were thermally monitored by industry level electronic thermometers at
45 degrees Celsius plus minus 2 degrees.
As an electronic engineer, I have been monitoring Intel's activities in all
aspects for last 20 years, andhave personally evaluated as many as one
hundred thousands plus PC systems. It would be very foolish of your company
if Intel were attempting to stonewall my technical findings by technical
means.
Any contradictory factual matters Intel can possibly present are wholly
warranted and expected in one week time. Meanwhile, at the same time this
certified letter is in your hand, I am ready to file legal actions against
Intel Corporation in the matter of defrauding consumers by deception in
state of California, USA.
(signed & sealed)
Certified and delivered by Post Office in Taipei, Taiwan.
Chicken napisał(a) w wiadomości: ...
>Mam kilka możliwości, czy ktoś to próbował z P200MMX:
>
>66 * 3,5 = 233 MHz (wg mnie za wolna magistrala)
>83,3 * 2,5 = 208,3 MHz
>90 * 2,5 = 225 MHz (trzeba mieć świeżą płytę)
>100 * 2 = 200 MHz (j.w. - pozorny brak dopału, jednak transfer na
magistrali
>:-))))
>
>z P233MMX:
>
>83,3 * 3 = 250 MHz ( sądze że wielu ludzi tak ma)
>83,3 * 3,5 = 291 Mhz ( bez komentarza..)
>100 * 2,5 = 250 MHz (byłoby ekstra )
>90 * 3 = 270 MHz ( coolowo )
>
>a może można jeszcze szybciej ?
>Ja mam P166MMX - działa od samego początku na 200 MHz ( 3 * 66 ) bez
żadnych
>problemów. Dałbym go na 208,3 MHz ale trzeba najpierw zmienić płytę, bo
>stara ma tylko max. 66 MHz magistrali.
>
>Chicken
>
>
To archiwum zostało wygenerowane przez hypermail 2.1.7 : Tue 18 May 2004 - 17:19:54 MET DST