Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: a?n?g?e?l@xxxxxxxxxxxxxxxxxxxxxxxxxxx (The little lost angel)
- Date: Fri, 28 Jul 2006 04:58:35 GMT
On Thu, 27 Jul 2006 15:34:35 -0400, George Macdonald
<fammacd=!SPAM^nothanks@xxxxxxxxxxxxx> wrote:
No, what I meant was you appear to be making assumptions about what other
people are doing and experiencing, with *their* equipment, which is
different from yours and your friend's... based on what I read elsewhere.
In that case I guess we just have to agree to disagree since I prefer
to trust the figures from myself and friends whom I know at least know
something of what they are doing compared to MSI forums which would
likely include fudge agent from various companies with vested
interests.
I don't see what months of intensive use has to do with it - a fraction of
a second of transient overload is what counts... and ruins a gamer's
My point is that, if the most demanding of games only needs this much
power, it's highly unlikely the PSU will have to deal with surges much
above that. Are there any documented proof that system transient can
surge more than 2x the average maximum load?
day.:-) I suggest a visit to nVidia's SLIZone and check out the Certified
Power Supplies - nothing below 500W that I see and damn few of them; 600W
*cough* *cough* Now we want to bring in commercial and corporate
considerations? That's going to really muck up the muddy waters even
more :ppPp
seems to be the average. You want "proof" - go read the MSI forum for a
start... and figure that Asus has a video card which comes with an external
80W power supply.
Two cards = 160W, base system 160W, that's about 320W for ya, which is
about the same as the figured obtained by Anandtech and what not for
total system draw.
So you think your meter is telling you that the board is using the 4-pin
12V for things other than the CPU? So why are they adding a 4-pin molex
too? Surely this is also to do with how the P/S maker has distributed the
wiring inside the P/S. Apparently it's quite common now for the PCI-E
connector(s) to come off the same 12V side as the CPU... if there are
indeed separate "sides" at all.
Well, I doubt there's different sides for most cases, just which
current limiter path. So the only difference is which wire the system
is drawing power from to the mainboard, which as I mentioned, is
minimal on the actual main ATX connector.
You're measuring what comes out of the wall? I think you mean *average*
peak and I think you may be missing something here - internal transient
surges which are smoothe out by the P/S.
Again, if the average is only about 200 peak on very specific
subtests, what's the chance internal transients would even reach 400W?
Furthermore, most of the better PSU if stating 400W as their max
capacity, tend to also have a peak capacity between 450W to 500W to
deal with such spikes. Please bear in mind I'm not talking about crap
$25 "400W" PSU which are really 200W in disguise.
That must be an awfully unusual 300W P/S. I know that a "good" but not
brand new 400W I used, which had 18A on 12V, was not satisfactory for a
system with just a A64 3500+ and a couple of hard drives. I was wary when
I first built it from an exisitng case+P/S since I'd read that 18A 12V was
the absolute minimum for an A64 system - sure enough it was not enough.
I don't think it's very unusual, just a standard no frills 300W PSU.
But it is a newer model with "dual" +12V. According to him, the
manufacturer's spec *** on it gives a chart showing that it's
capable of delivering almost entirely +12V output with minimum 5V/3.3V
so I think the key difference is that, the +12V output more so than
the actual total capacity of the PSU. A 450W PSU with only 18A of 12V
isn't going to be as suitable for today's systems compared to a 350W
with 25A of +12V.
At one time, I was usinng a P4-HT 3Ghz and about 6 drives without
problem with a 350W 18A (or less, can't remember exactly) +12V PSU. I
seriously think that the quality of the PSU matters much more than
what's on the label.
Then again, another possibility is the AC in your area is bad so
causing problems for the PSU on the input end?
--
A Lost Angel, fallen from heaven
Lost in dreams, Lost in aspirations,
Lost to the world, Lost to myself
.
- Follow-Ups:
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: George Macdonald
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- References:
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: George Macdonald
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: The little lost angel
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: George Macdonald
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: The little lost angel
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: George Macdonald
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: The little lost angel
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: George Macdonald
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: The little lost angel
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: George Macdonald
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: The little lost angel
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- From: George Macdonald
- Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- Prev by Date: Re: Merged AMD-ATI monster embarks on monopoly-busting
- Next by Date: Re: AMD to produce ATI GPUs in Dresden
- Previous by thread: Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- Next by thread: Re: AMD getting rid of all 1MB L2 Athlons (except FX)
- Index(es):