PDA

View Full Version : Re: geforce 3 ti


Beowulf
June 28th 03, 04:48 AM
Water,or any liquid, is not an optimal cooling solution. Too complex,
prone to leaks and an enginering nightmare. I am surprised that
overclockers and computer cooling freaks haven't yet discovered Peltier
thermoelectric heat pump chips. Stop living in the 20th century!

raj wrote:

>can water cooling cause corrosion, cause I will be water cooling my gf3 card
>as a test, then i will use my radeon 9700 pro AIW if the out come is good
>
>
>
>

Strontium
June 28th 03, 05:53 AM
Unless you have a very large heatsink, peltiers are useless for CPU cooling.
You have to DO something with all of the heat that is being pumped from the
other side of the element..... Last time I messed around with that, a small
30-40Watt peltier required a heatsink the size of small novel to even be
effective and keep the heat from going back over to the other side of the
element. This was back when CPU's only pumped about 50Watts. Nowadays,
it's just impractical. So, basically, peltiers came and went. Old-school.
Yesterday's news. Etc..

-
Beowulf stood up, at show-n-tell, and said:

> Water,or any liquid, is not an optimal cooling solution. Too complex,
> prone to leaks and an enginering nightmare. I am surprised that
> overclockers and computer cooling freaks haven't yet discovered
> Peltier thermoelectric heat pump chips. Stop living in the 20th
> century!
>
> raj wrote:
>
>> can water cooling cause corrosion, cause I will be water cooling my
>> gf3 card as a test, then i will use my radeon 9700 pro AIW if the
>> out come is good

--
You, probably, thought you weren't going to die, today...SURPRISE!"

-The main character in Postal 2

J.Clarke
June 28th 03, 05:59 AM
On Fri, 27 Jun 2003 21:48:43 -0600
Beowulf > wrote:

> Water,or any liquid, is not an optimal cooling solution. Too complex,
>
> prone to leaks and an enginering nightmare. I am surprised that
> overclockers and computer cooling freaks haven't yet discovered
> Peltier thermoelectric heat pump chips. Stop living in the 20th
> century!

I think if you investigate you will find that most overclockers who use
Peltiers these days (there is nothing new about their use in
overclocking) water-cool them. With a Peltier you're dissipating the
heat generated by the CPU and the additional waste heat generated by the
Peltier. If you can't cool the CPU without the Peltier then you don't
have a chance in Hell of cooling the CPU _with_ the Peltier. And
Peltiers introduce their own complications.

> raj wrote:
>
> >can water cooling cause corrosion, cause I will be water cooling my
> >gf3 card as a test, then i will use my radeon 9700 pro AIW if the out
> >come is good
> >
> >
> >
> >
>


--
--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)

RaiderFan
June 28th 03, 03:42 PM
No corrosion if you are using distilled, de-ionized water.
H2O is not corrosive, just some of the dissolved and suspended solids are.


"J.Clarke" > wrote in message
...
> On Fri, 27 Jun 2003 20:02:41 -0700
> "raj" > wrote:
>
> > can water cooling cause corrosion, cause I will be water cooling my
> > gf3 card as a test, then i will use my radeon 9700 pro AIW if the out
> > come is good
>
> Water cooling will cause corrosion only if it leaks.
>
> --
> --
> --John
> Reply to jclarke at ae tee tee global dot net
> (was jclarke at eye bee em dot net)

pete
June 28th 03, 04:14 PM
Didn't I read somewhere that peltiers cause condensation around the CPU? I'm
pretty sure I did.

Pete

"J.Clarke" > wrote in message
...
> On Fri, 27 Jun 2003 21:48:43 -0600
> Beowulf > wrote:
>
> > Water,or any liquid, is not an optimal cooling solution. Too complex,
> >
> > prone to leaks and an enginering nightmare. I am surprised that
> > overclockers and computer cooling freaks haven't yet discovered
> > Peltier thermoelectric heat pump chips. Stop living in the 20th
> > century!
>
> I think if you investigate you will find that most overclockers who use
> Peltiers these days (there is nothing new about their use in
> overclocking) water-cool them. With a Peltier you're dissipating the
> heat generated by the CPU and the additional waste heat generated by the
> Peltier. If you can't cool the CPU without the Peltier then you don't
> have a chance in Hell of cooling the CPU _with_ the Peltier. And
> Peltiers introduce their own complications.
>
> > raj wrote:
> >
> > >can water cooling cause corrosion, cause I will be water cooling my
> > >gf3 card as a test, then i will use my radeon 9700 pro AIW if the out
> > >come is good
> > >
> > >
> > >
> > >
> >
>
>
> --
> --
> --John
> Reply to jclarke at ae tee tee global dot net
> (was jclarke at eye bee em dot net)

Beowulf
June 28th 03, 07:38 PM
No Peltiers don't need big heatsinks - that's a misconconception and a
design flaw. They aren't popular since they were initially introduced
prior to the Pentium1 (several patents were filed); however, passive
cooling in those days was more than adequate. Passive cooling needs big
heatsinks - even with big fans - with today's CPUs. You will still
need to put in enough case fans to exhaust the heat your pulling off.
The biggest mistrake poeple make is not setting the heat pump voltage
properly or choosing the wrong heat pump. Many people erroneously think
that in order to cool the CPU (overclocked or not) you need to chill it.
You don't . You just need to drop and regulate the temp to 15-20
degrees celsius - and thus no condensation. These devices are extremely
efficient and effective at cooling. I have had great success with them
in the past. - not just for CPUs, but also for a variety of scientific
instrumentation.

I don't use them anymore simply because it is easy and relatively
inexpensive to pick up a faster CPU, mobo, and/or graphics card.
Although I do a bit of gaming, mostly my PC runs office apps, www
browsing, a few DVDs, and some video/image editing, I only see a small
difference real world performance in overclocking. With 3GHz CPUs, most
tasks are done before I can reach for the mouse - anything that may take
a few minutes just means that I can check my email or go take a ****
while I'm waiting. Going faster the 100-150fps in an OPENGL shooter at
1600x1200x32 full quality, FSAA 4x, etc, is just a bunch of weenies
saying "mine's faster!" So what? I used to have the same discussions
with motor heads over whose car was faster. Big whoop.

pete wrote:

>Didn't I read somewhere that peltiers cause condensation around the CPU? I'm
>pretty sure I did.
>
>Pete
>
>"J.Clarke" > wrote in message
...
>
>
>>On Fri, 27 Jun 2003 21:48:43 -0600
>>Beowulf > wrote:
>>
>>
>>
>>>Water,or any liquid, is not an optimal cooling solution. Too complex,
>>>
>>>prone to leaks and an enginering nightmare. I am surprised that
>>>overclockers and computer cooling freaks haven't yet discovered
>>>Peltier thermoelectric heat pump chips. Stop living in the 20th
>>>century!
>>>
>>>
>>I think if you investigate you will find that most overclockers who use
>>Peltiers these days (there is nothing new about their use in
>>overclocking) water-cool them. With a Peltier you're dissipating the
>>heat generated by the CPU and the additional waste heat generated by the
>>Peltier. If you can't cool the CPU without the Peltier then you don't
>>have a chance in Hell of cooling the CPU _with_ the Peltier. And
>>Peltiers introduce their own complications.
>>
>>
>>
>>>raj wrote:
>>>
>>>
>>>
>>>>can water cooling cause corrosion, cause I will be water cooling my
>>>>gf3 card as a test, then i will use my radeon 9700 pro AIW if the out
>>>>come is good
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>--
>>--
>>--John
>>Reply to jclarke at ae tee tee global dot net
>>(was jclarke at eye bee em dot net)
>>
>>
>
>
>
>

J.Clarke
June 28th 03, 10:59 PM
On Sat, 28 Jun 2003 15:14:00 GMT
"pete" > wrote:

> Didn't I read somewhere that peltiers cause condensation around the
> CPU? I'm pretty sure I did.

Depends on how cold you're running. Any time you're running below
ambient temperature there is a possibility of condensation. There are
ways to deal with it, generally involving an impermeable barrier of some
sort.

> Pete
>
> "J.Clarke" > wrote in message
> ...
> > On Fri, 27 Jun 2003 21:48:43 -0600
> > Beowulf > wrote:
> >
> > > Water,or any liquid, is not an optimal cooling solution. Too
> > > complex,
> > >
> > > prone to leaks and an enginering nightmare. I am surprised that
> > > overclockers and computer cooling freaks haven't yet discovered
> > > Peltier thermoelectric heat pump chips. Stop living in the 20th
> > > century!
> >
> > I think if you investigate you will find that most overclockers who
> > use Peltiers these days (there is nothing new about their use in
> > overclocking) water-cool them. With a Peltier you're dissipating
> > the heat generated by the CPU and the additional waste heat
> > generated by the Peltier. If you can't cool the CPU without the
> > Peltier then you don't have a chance in Hell of cooling the CPU
> > _with_ the Peltier. And Peltiers introduce their own complications.
> >
> > > raj wrote:
> > >
> > > >can water cooling cause corrosion, cause I will be water cooling
> > > >my gf3 card as a test, then i will use my radeon 9700 pro AIW if
> > > >the out come is good
> > > >
> > > >
> > > >
> > > >
> > >
> >
> >
> > --
> > --
> > --John
> > Reply to jclarke at ae tee tee global dot net
> > (was jclarke at eye bee em dot net)
>
>


--
--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)

J.Clarke
June 28th 03, 11:15 PM
On Sat, 28 Jun 2003 12:38:50 -0600
Beowulf > wrote:

> No Peltiers don't need big heatsinks - that's a misconconception and a
>
> design flaw. They aren't popular since they were initially introduced
>
> prior to the Pentium1 (several patents were filed);

Peltiers were _introduced_ in 1834 and the modern semiconductor version
in the 1950s. They predate the microprocessor by many years.

> however, passive
> cooling in those days was more than adequate. Passive cooling needs
> big heatsinks - even with big fans - with today's CPUs. You will
> still need to put in enough case fans to exhaust the heat your pulling
> off. The biggest mistrake poeple make is not setting the heat pump
> voltage properly or choosing the wrong heat pump. Many people
> erroneously think that in order to cool the CPU (overclocked or not)
> you need to chill it.
> You don't . You just need to drop and regulate the temp to 15-20
> degrees celsius - and thus no condensation.

What does that gain you over 30C?

> These devices are
> extremely efficient and effective at cooling. I have had great
> success with them in the past. - not just for CPUs, but also for a
> variety of scientific instrumentation.

Yes, they are efficient and effective but they still must dissipate not
only the thermal energy produced by the CPU but also the thermal energy
of their own power consumption. If the CPU needs a large heat sink with
a fan the Peltier also needs a large heat sink with a fan.

> I don't use them anymore simply because it is easy and relatively
> inexpensive to pick up a faster CPU, mobo, and/or graphics card.
> Although I do a bit of gaming, mostly my PC runs office apps, www
> browsing, a few DVDs, and some video/image editing, I only see a small
>
> difference real world performance in overclocking. With 3GHz CPUs,
> most tasks are done before I can reach for the mouse

Overclocking is not about "real world performance".

> - anything that
> may take a few minutes just means that I can check my email or go take
> a **** while I'm waiting. Going faster the 100-150fps in an OPENGL
> shooter at 1600x1200x32 full quality, FSAA 4x, etc, is just a bunch of
> weenies
> saying "mine's faster!" So what? I used to have the same discussions
>
> with motor heads over whose car was faster. Big whoop.

If you don't want to participate in the hot rodding or overclocking
hobbies, then don't. But the only "weenie" I see here is the one who is
putting down others because of the hobbies they choose.

> pete wrote:
>
> >Didn't I read somewhere that peltiers cause condensation around the
> >CPU? I'm pretty sure I did.
> >
> >Pete
> >
> >"J.Clarke" > wrote in message
> ...
> >
> >
> >>On Fri, 27 Jun 2003 21:48:43 -0600
> >>Beowulf > wrote:
> >>
> >>
> >>
> >>>Water,or any liquid, is not an optimal cooling solution. Too
> >complex,>>
> >>>prone to leaks and an enginering nightmare. I am surprised that
> >>>overclockers and computer cooling freaks haven't yet discovered
> >>>Peltier thermoelectric heat pump chips. Stop living in the 20th
> >>>century!
> >>>
> >>>
> >>I think if you investigate you will find that most overclockers who
> >use>Peltiers these days (there is nothing new about their use in
> >>overclocking) water-cool them. With a Peltier you're dissipating
> >the>heat generated by the CPU and the additional waste heat generated
> >by the>Peltier. If you can't cool the CPU without the Peltier then
> >you don't>have a chance in Hell of cooling the CPU _with_ the
> >Peltier. And>Peltiers introduce their own complications.
> >>
> >>
> >>
> >>>raj wrote:
> >>>
> >>>
> >>>
> >>>>can water cooling cause corrosion, cause I will be water cooling
> >my>>>gf3 card as a test, then i will use my radeon 9700 pro AIW if
> >the out>>>come is good
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>--
> >>--
> >>--John
> >>Reply to jclarke at ae tee tee global dot net
> >>(was jclarke at eye bee em dot net)
> >>
> >>
> >
> >
> >
> >
>


--
--
--John
Reply to jclarke at ae tee tee global dot net
(was jclarke at eye bee em dot net)