I have a Gateway P-75 and just read the article about overclocking in
Sim Racing News. All I can say is Hallelujah!! I want to overclock to
100 Mhz, but I wanted some input from anyone who has already done this
as far as success and improved performance. Any info. will be GREATLY
appreciated.
Keith Branch
>I have a Gateway P-75 and just read the article about overclocking in
>Sim Racing News. All I can say is Hallelujah!! I want to overclock
to
>100 Mhz, but I wanted some input from anyone who has already done this
>as far as success and improved performance. Any info. will be GREATLY
>appreciated.
>Keith Branch
I have done this in the past, for lab testing purposes, and they will
not last real long. I had a P60 last ten minutes at 66. They are made
to run at the printed speed, no faster. Do you think if a P75 would run
at 90-120 Mhz, Intel would sell it as a 75 for less money? I don't
think so.
Mike
If I understand correctly, the chips are only tested to a set speed, but
are actually the same as ones rated faster (assuming the same design is
used)? Doesn't the 486SX cpu still contain the math co-processor
(disabled, of course)?
People have been overclocking since motherboards came out with jumpers to
adjust the bus speed. When I had my 486-25, I was able to successfully
clock it to 33. When I finally upgraded to a dx2-50, I tried clocking it
up to 66MHz. It lasted for 1/2 an hour, and then fried the chip AND my
upgrade socket. Thank god that was all it fried!
Although my motherboard can be upped to 166MHz now, I believe I'll stay at
100MHz until I buy a new chip. With the 200MHz chips coming out, I'm sure
the 166MHz chips will drop down in price pretty fast. And whatever I pay
for the chip will be less than a new motherboard if mine blows up from
overclocking.
And one last thought before I jump off the soap box. Would you put racing
fuel (higher octane) into your family sedan and hope that your engine
doesn't go boom? Isn't that essentially the same thing you do when you
overclock?
Well, specifically regarding P-75s, I've just sold one that I've had
running at 100Mhz for about 9 months. The new owner also has it running
at 100Mhz in a new PC, and it's zipping along fine - no problems.
Cheers!
John
_________________________________
__ _____| |_____ __
_________| |__| :| John Wallace | |__| |_________
\ :| |::| :| Team WW Racing TSW | |::| | /
> :| |::| :|_________________________________| |::| | <
/ :|__|::|____:/ Sim Racing News \.____|::|__| \
/_______:/ \::/ http://sneezy.dcn.ed.ac.uk/simnews/index.htm\::/ \._______\
Mike,
I don't mean to be insulting, but it's patently obvious you know very
little about processor tolerances, nor about the marketing devices
employed by today's semiconductor companies.
If what you say is true, then someone needs to send some statisticians
over to Edinburgh, because the laws of probability just aren't operating
here. I personally have overclocked every processor I've ever had
(LOTS!), as have all of my friends - we have NEVER, repeat NEVER, had a
processor fail on us. Some haven't worked at the higher speed and have
been put back to the original without damage, but NONE have failed.
Learn a bit about the subject before simply scaremongering based upon
ONE instance of a VERY old technology CPU.
Cheers!
John
_________________________________
__ _____| |_____ __
_________| |__| :| John Wallace | |__| |_________
\ :| |::| :| Team WW Racing TSW | |::| | /
> :| |::| :|_________________________________| |::| | <
/ :|__|::|____:/ Sim Racing News \.____|::|__| \
/_______:/ \::/ http://sneezy.dcn.ed.ac.uk/simnews/index.htm\::/ \._______\
>> TO EVERYBODY WANTING TO DO THIS. DON'T. YOU WILL BE BUYING A NEW
>> PROCESSOR VERY SOON IF YOU DO.
>> I have done this in the past, for lab testing purposes, and they
will
>> not last real long. I had a P60 last ten minutes at 66. They are
made
>> to run at the printed speed, no faster. Do you think if a P75 would
run
>> at 90-120 Mhz, Intel would sell it as a 75 for less money? I don't
>> think so.
>> Mike
>If I understand correctly, the chips are only tested to a set speed,
but
>are actually the same as ones rated faster (assuming the same design
is
>used)? Doesn't the 486SX cpu still contain the math co-processor
>(disabled, of course)?
Thus my point from above.
Mike
I don't know about my mom, but my wife would probably kick my ass out the
door if I came to her and said, "Honey, I need a new computer because I
was stupid."
--
Well, it's more or less true... Overclocking a P75 to 100mhz is very
drastic on the processor... more recent Intel P75 can't even be
overclocked anymore, they just will refuse to start!!
Running a P90 at 100, or a 120 at 133, is perfectly possible... IF you
have a good cooler on your pentium (not the ones that you can 'bounce
around') and also use a heat-conducting fluid between processor and
cooler... My P90 runs at 100mhz for more than a year, several hours a
day... The thing gets damn hot, its lifespan WILL be shorter, but hey,
who uses his/her processor for more than 2 years anyway?
---
Yves Smolders - 2 by 2
(almost) always on 'UNDERNET' IRC as TONIO... channels #babbel and
#jongeren
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
MP
Scurvydog #00 Simpson-Cowlings Motorsports
http://www.netcom.com/~svydog/scms/index.html
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
underclocked cpu than to race your local dirt track a few week ends.
Bear in mind that if you blow up your cpu, that would be a good excuse
to go out and get that faster unit
CT
Hawaii: Cole
"If you dont like the way I drive, stay off the sidewalk"
In my experience of overclocking 486's and pentiums you can
* sometimes * get away with it, consider the following :-
486DX33 Overclocked to 50Mhz A difference of 50% and 17 Mhz
Pentium 90Mhz overclocked to 100Mhz A difference of 11% and 10Mhz
Both had heatsinks and fans.
The 486 is now 2 years old and works faultlesly
(Except with some VESA cards which is normal for a DX50 anyway)
The Pentium works fine idling but locks up when pushed hard (by NASCAR for
example)
This is for the simple reason that more of the pentium's logic is being
used = more heat. The Processor will be prone to errors (i.e Crashing)
at higher clock rates.
If you insist on overclocking any processor follow my advice :-
* DON'T go any more than 1 step faster on a Pentium (i.e 75 to 90Mhz)
* You MUST fit a heatsink and a fan.
* Fit a second case fan sucking into the case (as opposed to the PSU's
which blows out). Do not allow this fan to work against the processor
fan, though.
* Use a proper heat transfer compound (e.g silicone) between the processor
case and the heatsink.
* Blow all the dust and ***from the PC's innards every 6 - 12 months
Short of fitting a refridgeration unit into your PC (has anybody tried
this - just a thought) this is about all you can do to minimize the risk
of processor burnout.
Just one more thought - that guy who oc'd a P60 to P66, didn't he think
that maybe a processor that runs too hot anyway (the P60 REQUIRES a heat
sink to be reliable) might just stretch thing a LITTLE too far?
>>TO EVERYBODY WANTING TO DO THIS. DON'T. YOU WILL BE BUYING A NEW
>>PROCESSOR VERY SOON IF YOU DO.
>>I have done this in the past, for lab testing purposes, and they will
>>not last real long. I had a P60 last ten minutes at 66. They are made
>>to run at the printed speed, no faster. Do you think if a P75 would
run
>>at 90-120 Mhz, Intel would sell it as a 75 for less money? I don't
>>think so.
>Mike,
>I don't mean to be insulting, but it's patently obvious you know very
>little about processor tolerances, nor about the marketing devices
>employed by today's semiconductor companies.
>If what you say is true, then someone needs to send some statisticians
>over to Edinburgh, because the laws of probability just aren't
operating
>here. I personally have overclocked every processor I've ever had
>(LOTS!), as have all of my friends - we have NEVER, repeat NEVER, had
a
>processor fail on us. Some haven't worked at the higher speed and have
>been put back to the original without damage, but NONE have failed.
>Learn a bit about the subject before simply scaremongering based upon
>ONE instance of a VERY old technology CPU.
>Cheers!
>John
Eight years in the PC industry, four as an engineer, I think I know how
these things work. If you have never had one fail, good for you and
your friends. PC companies are always trying to get better benchmarks
than everyone else, which sells systems. This sometimes includes trying
to "push" things. Don't even get me started on marketing.
New technology processors are not much more reliable than the older
ones, when you push the limit.
Learn a little about people before you put you foot in your mouth and
accuse them of not knowing anything.
Mike
Emmmmm yes :)
A peltier cooler is essentially a small "fridge" for your CPU and is
very efficient. A liquid cooling system was considered also, but
anything that needs cooling THAT badly is a bad idea to clock
Cheers!
John
_________________________________
__ _____| |_____ __
_________| |__| :| John Wallace | |__| |_________
\ :| |::| :| Team WW Racing TSW | |::| | /
> :| |::| :|_________________________________| |::| | <
/ :|__|::|____:/ Sim Racing News \.____|::|__| \
/_______:/ \::/ http://sneezy.dcn.ed.ac.uk/simnews/index.htm\::/ \._______\
Overclocking to P100 is _NOT_ drastic on a P75. The reason the latest
P75 chips "refuse to start" is because Intel realised that every chip
could be overclocked and many people were doing this - they put
circuitry to STOP people doing that.
Try UNDERclocking your P75 to 66Mhz - wonder why it doesn't work even at
a SLOWER speed? That's not because of strain....
_________________________________
__ _____| |_____ __
_________| |__| :| John Wallace | |__| |_________
\ :| |::| :| Team WW Racing TSW | |::| | /
> :| |::| :|_________________________________| |::| | <
/ :|__|::|____:/ Sim Racing News \.____|::|__| \
/_______:/ \::/ http://sneezy.dcn.ed.ac.uk/simnews/index.htm\::/ \._______\
Mike,
Sorry to have upset you, but I am/was going on the information I have -
what you posted. And what you posted is the scaremongering propaganda of
the manufacturers and it's wrong.
I too have worked in the semiconductor industry for several (12) years,
specifying these products, and developing the raw material to make these
products alongside the manufacturers - as such we know a lot about how
they are made and manufactured. I'm not talking about the rubbish
claimed by PC sellers, I'm talking of the info of the people who make
them - Intel etc.
To say that "They are made to run at the printed speed, no faster" is
just plain not true as you will know if you have the experience you
said. Intel make processors to run as fast as they possibly can - call
it "continuous improvement", or "product development", if they can make
a P-75 that runs at 300Mhz they'll do it - that's the processor that
will beat Cyrix/STM/IBM/AMD-NexGEN in the coming months. As for why they
don't sell it as a "P-300", that's marketing. As for why they don't put
overclock protection, that's product yield - you can't do it until
sufficient quantity of the line's product meets that standard.
You had a bad experience with a P-60, a 0.8um technology early Pentium.
P-75s were made on 0.6um and latterly 0.35um lines - these can turn out
P-200 chips, so overclocking a P-75 to 100Mhz or higher is WAY short of
a risk, and new technology processors are much more reliable than older
ones since they're being produced on lines which are designed to be
ready for P-Pro and even P7 production. Intel's switch from 0.6um to
0.35um and below has gone much better than anticipated, and that is why
you're seeing a glut of good quality processors, price cuts and the P-
200 which wasn't supposed to be released (although there are other
reasons for that too). Perhaps your knowledge just needs to be more up
to date?
I've seen a lot of your posts here, and you do talk a lot of sense, but
on this issue I had to say you were/are incorrect. I'm not accusing you
of "not knowing anything", just not being very "au-fait" with this
subject.
Cheers!
John
_________________________________
__ _____| |_____ __
_________| |__| :| John Wallace | |__| |_________
\ :| |::| :| Team WW Racing TSW | |::| | /
> :| |::| :|_________________________________| |::| | <
/ :|__|::|____:/ Sim Racing News \.____|::|__| \
/_______:/ \::/ http://sneezy.dcn.ed.ac.uk/simnews/index.htm\::/ \._______\
