Milo, re: aceshardware article
The article has several weak arguments. Few remarks
follow:
The article says:
"The 1 GHz Athlon runs at 1.8V, much more than the normal 1.65-1.7V. Almost every Athlon 850 out there can reach 900 with 1.8V. In fact, most manufacturers of overclocking devices (Golden Finger Devices or GFD) advise that the overclocker not apply more than 1.8V. In other words, 1.8V is really the limit, and the 900 MHz Athlon doesn't have much margin left."
Obviously the conclusion is wrong. First, if every
850-Athlon can reach only 900 at 1.8V, it means that
the 1GHz Athlons are DIFFERENT if they got to 1000!
Second, the 1.8V is not the limit, see my post:
Message 12851877
(I think Aceshardware should do better searches on SI
and better understand the meaning of posts).
Aceshardware partially contradict themselves by
mentioning:
"...and the Spitfire will use 1.5V.", which means
that there is no such limit for AMD technology.
About the power consumption:
"The K7-1000 typically draws 60W and a maximum of 65W"
This is BS. Did Aceshardware ever try to measure
the current? Obviously AMD is too "technical" and
quotes those values for WORST CASE BIN and for
WORST CASE TEMPERATURE. More, if
the maximum is "65W", the "typical" consumption
(DOS, regular office apps) would be no more than 40W.
And this is at 1.8V. At 1.6 it could be less than
32W.
Again, to determine the "maximum" power, AMD uses
(and distributes under NDA) some synthetical DOS
program. It draws the power that has never been
observed in any avalable application yet, or at least
it never is sustained. In contrast, Intel has never
defined what their "typical" and "maximum"
power is. At least for Intel to say that the "typical"
is 32 and "maximum" is 33 is a total, outrageous
bullsh*t.
Ace's is worried:
"It remains to be seen if the Athlon systems that are out there will be able to upgrade to these new power hungry monsters."
Baloney again. For the off-shelf users there is no
worry about, and the main power supply is not an issue
(Gateway, AFAIK, uses a ridiculous 200W P/S).
For the home-grown builders-"upgraders"
I would suggest the following: if you want to upgrade
to 1000GHz and have concerns about your mainboard
capabilities, use "goldenfingers" and try it at 1.7,
1.65, 1.6, etc. core voltage. You may have a double
gain: reducing the voltage will reduce power, it
will reduce the junction temperature, which in turn
will give you room for functionality and stability.
I believe that 80% of all 1000GHz athlons will
be fairly happy at 1.65V and with decent fan-cooled
heat sink in a normally-ventillated box.
Finally, the picture of 1GHZ Coppermine #44 at
Anand website, with HUGE EXPENSIVE ALL-COPPER
heat sink, is a clear evidence that something
is still wrong with Intel's ability to handle
standard junction temperatures. Remember, the
first Coppermines were limited to 533MHz since
they could not handle higher case temperatures?
It was a year ago, Paul must has all references...
It looks like the problem is still there...he-he.
- Ali, from the "screwdriver shop" |
