AMD Athlon 64FX first impressions

[Chris]

Well today we received 2 NEW AMD CPU's, the Athlon 64FX and Athlon 64 3200+.
The mainboards were pre-production Gigabyte, GA-K8NNXP and GA-K8NNXP-940. A
good start CPU coolers fit like the P4 so the risk of damage when fitting is
much lower. We found that the N-vidia FX5900 (Asus) video cards did not
perform that well and that the ATI Radeon 9800 pro (Sapphire) worked much
better and had an easier install on both systems. We monitored both CPU's
with a NAMAS calibrated thermocouple 'K' type recording 48c idle for the
Athlon 64 and 56c under load, the FX chip was hotter by 4c at idle and under
load but neither caused a problem, though this is on a bench not in a case.
We loaded Windows XP some updates and patches both ran stable on the 3 day
burn in test, though the 32bit OS on the Athlon 64 3200+ showed results no
better than the Athlon XP3200+, the FX did how ever but at double the price
you would expect this. Then came out our BETA Windows XP64 OS, loaded well
on both machines and seemed stable, then we loaded the drivers and bang,
blue screens, rebooting and all sorts of driver conflicts, we have narrowed
it down to the USB2 driver the IEEE1394 and onboard sound, if we disable
any one of these the other 2 work but not all 3 together. So at the moment
we have a cheap soundcard in which has helped but not entirely cured the
problem. Both systems ran quite a bit faster using the 64bit OS though in
real terms it was not something you would see by eye. We used 1Gb (2x512Mb)
of OCZ EL DDR RAM PC3200 Dual Channel Platinum memory CL 2-2-3-5. At
present cost of the chips and the 63bit OS make this an expensive option and
in comparison to a 3.2GHz P4 system the Athlon 64 3200+ was 20% more
expensive
for no gain if you used a 32bit OS, the Athlon XP3200+ would be a cheaper
and better bet. The price of XP64 will determine the success of this chip,
chip prices may fall in the future but Microsoft is not known for dropping
the price of it's OS and XP64 bit edition is likely to be 50% more than the
32bit version.
--
Chris
Technical director CKCCOMPUSCRIPT
Apple Computers, Intel, Roland audio, ATI, Microsoft, Sun Solaris, Cisco and
Silicone Graphics.
Wholesale distributor and specialist audio visual computers and servers
FREE SUPPORT @,
http://www.ckccomp.plus.com/site/page.HTM
(e-mail address removed)

 

[Morten Holberg Nielsen]

Well today we received 2 NEW AMD CPU's, the Athlon 64FX and Athlon 64 3200+.
The mainboards were pre-production Gigabyte, GA-K8NNXP and GA-K8NNXP-940. A
good start CPU coolers fit like the P4 so the risk of damage when fitting is
much lower. We found that the N-vidia FX5900 (Asus) video cards did not
perform that well and that the ATI Radeon 9800 pro (Sapphire) worked much
better and had an easier install on both systems. We monitored both CPU's
with a NAMAS calibrated thermocouple 'K' type recording 48c idle for the
Athlon 64 and 56c under load, the FX chip was hotter by 4c at idle and under
load but neither caused a problem, though this is on a bench not in a case.
We loaded Windows XP some updates and patches both ran stable on the 3 day
burn in test, though the 32bit OS on the Athlon 64 3200+ showed results no
better than the Athlon XP3200+, the FX did how ever but at double the price
you would expect this. Then came out our BETA Windows XP64 OS, loaded well
on both machines and seemed stable, then we loaded the drivers and bang,
blue screens, rebooting and all sorts of driver conflicts, we have narrowed
it down to the USB2 driver the IEEE1394 and onboard sound, if we disable
any one of these the other 2 work but not all 3 together. So at the moment
we have a cheap soundcard in which has helped but not entirely cured the
problem. Both systems ran quite a bit faster using the 64bit OS though in
real terms it was not something you would see by eye. We used 1Gb (2x512Mb)
of OCZ EL DDR RAM PC3200 Dual Channel Platinum memory CL 2-2-3-5. At
present cost of the chips and the 63bit OS make this an expensive option and
in comparison to a 3.2GHz P4 system the Athlon 64 3200+ was 20% more
expensive
for no gain if you used a 32bit OS, the Athlon XP3200+ would be a cheaper
and better bet. The price of XP64 will determine the success of this chip,
chip prices may fall in the future but Microsoft is not known for dropping
the price of it's OS and XP64 bit edition is likely to be 50% more than the
32bit version.

Nice.

Hovever, I saw some guys at an AMD forum, who were pleased about the results
they got by testing the 64. This was in terms of overclocking abilities.

 

[CBFalconer]

Well today we received 2 NEW AMD CPU's, the Athlon 64FX and Athlon 64 3200+.
The mainboards were pre-production Gigabyte, GA-K8NNXP and GA-K8NNXP-940. A
good start CPU coolers fit like the P4 so the risk of damage when fitting is
much lower. We found that the N-vidia FX5900 (Asus) video cards did not
perform that well and that the ATI Radeon 9800 pro (Sapphire) worked much
better and had an easier install on both systems. We monitored both CPU's
with a NAMAS calibrated thermocouple 'K' type recording 48c idle for the
Athlon 64 and 56c under load, the FX chip was hotter by 4c at idle and under
load but neither caused a problem, though this is on a bench not in a case.
We loaded Windows XP some updates and patches both ran stable on the 3 day
burn in test, though the 32bit OS on the Athlon 64 3200+ showed results no
better than the Athlon XP3200+, the FX did how ever but at double the price
you would expect this. Then came out our BETA Windows XP64 OS, loaded well
on both machines and seemed stable, then we loaded the drivers and bang,
blue screens, rebooting and all sorts of driver conflicts, we have narrowed
it down to the USB2 driver the IEEE1394 and onboard sound, if we disable
any one of these the other 2 work but not all 3 together. So at the moment
we have a cheap soundcard in which has helped but not entirely cured the
problem. Both systems ran quite a bit faster using the 64bit OS though in
real terms it was not something you would see by eye. We used 1Gb (2x512Mb)
of OCZ EL DDR RAM PC3200 Dual Channel Platinum memory CL 2-2-3-5. At
present cost of the chips and the 63bit OS make this an expensive option and
in comparison to a 3.2GHz P4 system the Athlon 64 3200+ was 20% more
expensive
for no gain if you used a 32bit OS, the Athlon XP3200+ would be a cheaper
and better bet. The price of XP64 will determine the success of this chip,
chip prices may fall in the future but Microsoft is not known for dropping
the price of it's OS and XP64 bit edition is likely to be 50% more than the
32bit version.

I suggest you look over the above monolythic paragraph and
consider how you expect people to read it. It simply rambles.
Breaking it up into paragraphs, each dealing with a single
thought, would make a major difference. I, for one, won't bother.

 

[Stacey]

Nice.

Hovever, I saw some guys at an AMD forum, who were pleased about the
results they got by testing the 64. This was in terms of overclocking
abilities.

And you'd expect 'some guys at an AMD forum' to form a different opinion of
AMD's newest chip?

 

[Dave]

I suggest you look over the above monolythic paragraph and
consider how you expect people to read it. It simply rambles.
Breaking it up into paragraphs, each dealing with a single
thought, would make a major difference. I, for one, won't bother.

Seconded!

 

[Morten Holberg Nielsen]

And you'd expect 'some guys at an AMD forum' to form a different opinion of
AMD's newest chip?

No, but the results i saw was really good.

You missed the actual point. I was just writing that for overclocking, its
doing quite well. Dont know the price for it yet, but if its too high, the
64 isnt really interesting. Not even in terms of oc'ing.

 

[Stephen Austin]

Well today we received 2 NEW AMD CPU's, the Athlon 64FX and Athlon 64
3200+.

I suggest you look over the above monolythic paragraph and
consider how you expect people to read it. It simply rambles.
Breaking it up into paragraphs, each dealing with a single
thought, would make a major difference. I, for one, won't bother.

I found it an informative and interesting read, don't bother reading if you
don't want to, but don't bother critisizing either in that case.

Steve

 

[Martin]

I found it an informative and interesting read, don't bother reading if you
don't want to, but don't bother critisizing either in that case.

Steve

Spot on.

 

[kony]

No, but the results i saw was really good.

You missed the actual point. I was just writing that for overclocking, its
doing quite well. Dont know the price for it yet, but if its too high, the
64 isnt really interesting. Not even in terms of oc'ing.

.... and it's an important point, means that they shouldn't have any
problems ramping up the speeds released, if the tested CPUs are a fair
representation.

Dave

 

[kony]

I found it an informative and interesting read, don't bother reading if you
don't want to, but don't bother critisizing either in that case.

Steve

It's a fair criticism when a writing is meant for public consumption,
rather than private scribble. Minor readability problems are easily
overlooked, but whose keyboard lacks an <Enter> key?

Futher it's rather unnecessary for YOU to tell others not to
criticize, that they could not "bother reading", when this is exactly
what you did to the above post.... the pot calling the kettle black.


Dave

 

[Stephen Austin]

I suggest you look over the above monolythic paragraph and

Futher it's rather unnecessary for YOU to tell others not to
criticize, that they could not "bother reading", when this is exactly
what you did to the above post.... the pot calling the kettle black.

Excuse me? Fancy explaining that one?

 

[kony]

Excuse me? Fancy explaining that one?

You feel he shouldn't have made the comment, should just "not read
it". Likewise, you then should also refrain from making comment,
should just "not read" the post you replied to.

The difference is that the initial post WAS a mess, and it was helpful
to explain to the OP why the post won't be read by some people. It's
the OP's choice how to post, but unlike a person-to-person
conversation, it's often helpful to give written feedback.

They can't observe your reaction in-person, may not realize why they
don't get the feedback they (might) expect. They can't tell what your
reaction is over usenet without a reply.

What cbfalconer wrote, is a close approximation of what I was also
thinking when I saw that post.



Dave

 

[CBFalconer]

.... snip ...

I found it an informative and interesting read, don't bother reading
if you don't want to, but don't bother critisizing either in that case.

My objective was to give some feedback, and possibly improve his
postings. If everybody simply ignored such things the noise level
would be even higher.

 

[Stephen Austin]

I suggest you look over the above monolythic paragraph and

My objective was to give some feedback, and possibly improve his
postings. If everybody simply ignored such things the noise level
would be even higher.

But there is a difference between constructive critisizm and being rude. In
my personal opinion, I found your comment to be more rude than anything
else. Maybe I'm alone in this, if so I apologise.

Steve

 

[Stephen Austin]

I suggest you look over the above monolythic paragraph and

You feel he shouldn't have made the comment, should just "not read
it". Likewise, you then should also refrain from making comment,
should just "not read" the post you replied to.

...snip..

I *did* read the post, and that is not what I said. I said that, *IF* he
didn't want to read it (he implied that he hadn't - "I for one won't
bother"), then he shouldn't critisize. I have read the entire thread,
completely, and I found that the reply was unnecessarily rude. That was why
I replied as I did. As I said in the reply to CBFalconer, if I'm alone in
this point of view, I apologise. I'm just giving an opinion.

Steve