Enthusiast fixes what AMD won't (Phenom II + XP)

[Silent_Buddha]

I've been going through some non-POSTing joy with a friend's Abit IP35 and 4 DIMMs. Whether or not you have problems really comes down to the quality of the DIMMs. Stick with the 1.8v DIMMs for the best luck IMO.

My confusion with Socket 939-related AMD hate is caused by my having moved on from it in 2006. AMD wasn't going to port Phenom over to 939, so K8x2 was all it would ever have. Once Core 2 Duo came out, getting more K8x2 chips was anything but exciting for such an expensive platform. RAM costs a bucketload for it if you want lots of it. 939 was only interesting for a short while IMO (a year or two) and that's not really entirely related to AMD moves. Right now, AM2+ is actually way more interesting and carries more value than 939 ever did IMO, with access to $40 dual cores and $40 4gig kits. It's a bitchin' low cost platform today.

939 was actually pretty good and long lasting. It supported 1 series Opterons, Athlon 64 (FX), and Athlon 64x2. Socket 754 was the one I didn't really like, I think they would have been much better off either not bothering with Socket 754 or not bothering with Socket 939.

It still lasted 2 years which is more than can be said of many other sockets. Socket 423 (first P4 socket) for example lasted less than a year. I don't think the replacement Socket 478 lasted much longer as it was subsequently replaced by LGA 775. Which was then revised after about a year or two making it newer boards with Socket 775 incompatable with older 775 CPUs and vice versa.

Yeah a right mess. At least with Socket-F and Am2+/3 things have been fairly stable on the AMD front. And Socket 775 (after the midlife revision) has lasted a fair while. It'll be interesting to see if Socket 1366 will last as long.

Then again processor advancements have slowed quite a lot since the heady days of the p3->p4 and Athlon XP->Athlon64 days. Meaning there's been less need for rapidly evolving sockets.

Regards,
SB

 

[Bludd]

Socket 478 lasted from August 2001 to August 2004 when the LGA775 version of the Prescott core was released.

At least according to this list.

 

[Davros]

the only real downside to 478 i had was trying to get a pci-e 478 board
thankfully asrock made one

 

[Kaotik]

Because that just happened yesterday. PIIs have been out for more than just one day. Plus, XP is still mega popular, especially with all the Vista FUD out. Not only that, but people supported Windows 98SE/ME in 2003, two years after XP came out. Why shouldn't companies support XP two years after Vista came out?

Your logic is at best flawed.

Because they follow MS's lifecycle policy, each OS has full support minimum of 5 years, but at least 2 years over the release of successor.
In case of XP, it was that 2 years over the release of successor, in case of earlier OSes, it was that 5 years because the successors came on so fast

 

[Sxotty]

I have vista x64 and PII x3 cpu. With CnQ enabled I had all kinds of stability issues stock with no overclock. After disabling CnQ everything is rock solid. I wish it actually worked well for me, but I gave up on it some time ago.

 

[entity279]

Still, you can try to undervolt manually.. Power savings will still be there in full load as well this way.

 

[homerdog]

Yep, I disabled SpeedStep on my E6750 since it runs at 3.2GHz all the time, but I was still able to undervolt it to 1.26V. Idle temps went down a lot (~11C) and I still maintain 24/7 Prime stability.

 

[Sxotty]

Still, you can try to undervolt manually.. Power savings will still be there in full load as well this way.

I did this in my car as well (on car computer) and it has worked well, but I am not that worried yet on the main PC as heat is not as big an issue with good ventilation.

I may try it as summer come though.

 

[Silent_Buddha]

I did this in my car as well (on car computer) and it has worked well, but I am not that worried yet on the main PC as heat is not as big an issue with good ventilation.

I may try it as summer come though.

Well there is also a nice side benefit of lower electricity costs if you use the computer a lot.

Regards,
SB

 

[Mariner]

Undervolting is always a good idea, IMO.

I currently use CrystalCPUID to both undervolt and underclock but prior to that merely undervolted my AthlonX2 down to 0.975V (stock voltage 1.3V) in BIOS. The upsides? Much lower power consumption, a cooler running CPU & PC. The downside? It took me about two days to discover the lowest achievable undervolt. Hardly an onerous task!

 

[swaaye]

I usually either undervolt or I overclock as far as I can on stock voltage.

With a Phenom BE and those utilities I posted in the first post here, you can do whatever you want. Even have CnQ functioning as well.

 

[hoom]

Isn't this whole CnQ thing a problem with the Windows scheduler?
It switches affinity for no particularly good reason, preventing cores from properly shutting down/de-clocking or something.
I seem to recall that W7 scheduler has a better algorithm for that?

 

[Sxotty]

Well I d/led the MSRtweaker and installed. I tipped the multipliers up to 15 so it is running at 3Ghz, then I dled CPUz to see if it was working as stated.

Strange thing is MSRtweaker says voltage is 1.325, CPUz says voltage is 0.936. Both say multi is 15 and running at 3GHz. So it should be stock voltage whatever it is though.

Went to 17x it crashed, 16.5 for an hour then set at 16, but no power saving features.

I guess I need to try k10 stat if I want to use the low power setting as well, but to be honest I am leery of it as it has a pretty sketchy site, the readme is full of gibberish in the download. Doesn't inspire confidence.

Edit:
Does anyone know a listing of the hex like codes to decimal? They are not true hex though, my multiplier was set at 0C which =14, I changed it to 0F in bios=15.5 it seems a bit random, but it would be nice to know what the heck they stand for. I was hoping to get 16 in bios

 

[Sxotty]

Well nevermind
http://ocia.net/aimages/309_5_full.jpg

There it is.

 

[entity279]

I seem to recall that W7 scheduler has a better algorithm for that?

That would be a very good thing.. Will p2 re-enable the original Phenom's CnQ behaviour? I mean, would it be possible?

 

[hoom]

Can't immediately find what I thought I'd seen but from Engineering Windows 7 blog thingie

To further help reduce idle power, we are focusing on core processor power management improvements. Windows scales processor performance based on the current amount of utilization, and making sure Windows only increases processor performance when absolutely required can have a big impact on power consumption.