question on +3.3V current requirements

[Guest]

Archived from groups: alt.comp.hardware.overclocking.amd (More info?)

My question is general but in my case based on the
MSI Neo2 Platinum (Socket 939) board with Athlon 3000+
Winchester CPU.

The board takes +3.3V power from the PSU and supplies
the CPU with Vcore which is in the range 1.4 - 1.6 V,
in my config.

Question: does the DC-DC conversion on the motherboard
involve any "current gain", as would be the case with an
ac transformer? Or to put it another way, if my CPU were
to use up to (for example) 60A at 1.5V, does that mean
the PSU should be rated for (at least) 60A on the +3.3V
rail?

If so, that would mean that at least half the power goes
in ohmic losses on the conversion, right? So the MB would
be dissipating at least as much as the CPU?

If not so, can one estimate the efficiency of the conversion
somehow, and hence the PSU requirements?

TIA
David

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking.amd (More info?)

According to this

http://takaman.jp/D/?english

the A-64's seem to be just using the 12v line.

--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
uce@ftc.gov
Thanks, robots.

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking.amd (More info?)

Hi Ed, very interesting, thanks.
That would explain the extra +12V connectors most newer
boards have next to the CPU socket and power supplies.
But if the CPU takes it's Vcore from +12V, what the hell
needs 30A and more at 3.3V then?!

"Ed Light" <nobody@nobody.there> wrote in message news:r_bCd.6592$232.2378@fed1read05...
> According to this
>
> http://takaman.jp/D/?english
>
> the A-64's seem to be just using the 12v line.
>

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking.amd (More info?)

"David Johnstone" <david_ffm@mail.com> wrote
> But if the CPU takes it's Vcore from +12V, what the hell
> needs 30A and more at 3.3V then?!

Some agp video cards, pci cards. They differ alot.
These articles test cards for what amps they're pulling at each voltage.
http://www.xbitlabs.com/articles/video/display/ati-powercons.html
http://www.xbitlabs.com/articles/video/display/ati-vs-nv-power.html
--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
uce@ftc.gov
Thanks, robots.

 

[chip]

Archived from groups: alt.comp.hardware.overclocking.amd (More info?)

"David Johnstone" <david_ffm@mail.com> wrote in message
news:crcj6j$nq$00$1@news.t-online.com...
> Hi Ed, very interesting, thanks.
> That would explain the extra +12V connectors most newer
> boards have next to the CPU socket and power supplies.
> But if the CPU takes it's Vcore from +12V, what the hell
> needs 30A and more at 3.3V then?!

Older motherboards drew more current from the 5v and 3.3v lines. PSU's are
simply trying to cover as many possible requirements as they can - to offer
the greatest compability.

And remember the "30A at 3.3V" spec (where did you get that from???) that a
PSU may quote, will be its *maximum* available current at that voltage,
assuming no load on the 5V (and also possibly the 12V) lines. In practice a
PSU rated at 30A at 3.3V will not deliver 30A. And apart from anything
else, 30x3.3 is only 99w. Its not a massive amount.

Chip