Positive experience with ACC on Phenom II

[mattkosem]

So, I've read over and over again that ACC doesn't do anything on Phenom II chips. Blocking out all of those bad thoughts, I decided to give it a try...

The problem: Core 3 on my PII 940 does not clock over 3.5Ghz

Research to discover the problem: Setting each core to 3.7Ghz via AOD, one by one then running Prime95 64-bit set to 8k FFTs with 1600MB of ram allocated.

Core 0: No problem
Core 1: No problem
Core 2: No problem
Core 3: Cannot even get into prime before the system reboots.

Attempted solutions:
1) more vcore - no luck
2) lower memory ratio - no luck
3) lower overall clock - lame, but works
4) *ACC

I started by setting ACC to 0% on all but core 3, which I set to +2%. I rebooted the system, and set the clock. Hmm...Interesting...I can set that clock without almost instantly rebooting. So I thinks to myself, this seems promising. I kick off prime, it runs for a few seconds then reboots. Feeling a bit more lucky, I decided to kick it up another notch to 4%. Whadya know, I've gotten through 10 passes of this test without any issues so far.

Did the optimizations for ACC in recent bios revisions do something for ACC that does actually make it effective?

--Matt

[leoy]

Maybe your chip needed the acc.
If it works for you , then you are ok, don't mind for others.

[crazydiamond]

acc worked wonders on my 9850 but my board wont post w/ acc on auto with my 940

[mattkosem]

Maybe your chip needed the acc.
If it works for you , then you are ok, don't mind for others.

Yes, you really seem to believe me . It certainly makes a marked improvement from what I'm seeing now. There's, obviously, quite a big difference between instantly rebooting and being able to prime with the most stressful test that I've been able to muster so far at a given clock. I feel like this mobo is junk, and have a M4A39 Deluxe on the way to replace it. Perhaps the new mobo won't need it, but this one most certainly benefits from it.

--Matt

[leoy]

Yes, you really seem to believe me . It certainly makes a marked improvement from what I'm seeing now. There's, obviously, quite a big difference between instantly rebooting and being able to prime with the most stressful test that I've been able to muster so far at a given clock. I feel like this mobo is junk, and have a M4A39 Deluxe on the way to replace it. Perhaps the new mobo won't need it, but this one most certainly benefits from it.

--Matt

I would certainly bet on a new mobo....

[richierich]

For some strange reason when I turn on ACC CPUID Hardware Monitor can't detect my CPU temps

[mattkosem]

Believe what you like. I'll enjoy the benefits. It looks like 2% and 1.55v is the best combo for me. 4% and 1.55v doesn't get as far.
--Matt

[Veedo]

i was going to try the same thing on my x3 720. i have one core that will not boot into windows at stock clock, but with a lower multi it will np. this might help!

[mattkosem]

It couldn't hurt to give it a shot. What have you got to lose?

--Matt

[i found nemo]

i was going to try the same thing on my x3 720. i have one core that will not boot into windows at stock clock, but with a lower multi it will np. this might help!

then you should return it to amd ....

[Oese]

then you should return it to amd ....

i think he is meaning the fourth

[Veedo]

yes i mean the freebie core . so far i have the other 3 at 3.5ghz running occt. hope to get a bit more, but we shall see!

[i found nemo]

he said stock so i assumed he was using it as x3 still, but idk wich one he means.

edit =

ahhh i c. well in that case more voltagess! lol

[G0ldBr1ck]

ACC does some tricky things with the voltages that may be the key to what your getting.

[mattkosem]

Hopefully I won't have to deal with it when the M4A79 Deluxe comes. It looks like a much better board overall. I only have this one because it was the only one locally available when my other board died.

--Matt

[Throwed]

I have a Foxconn board that has SB750, a Biostar that has SB600, and a ASUS that has SB700. Out of three the ASUs overclocks the best. ACC does nada for my PII 940. In fact, when enabled it makes things screwy so I don't even deal with it. The SB700 suits my taste and is ALOT more stable. Some love ACC and some don't. IIRC, ACC was really just ment for the 1st Phenoms and I am beginning to believe it because I fair no better/worse on the SB750 vs SB700.

[mattkosem]

In my situation, where one core lags terribly behind the other three, ACC on the slow core seems to have bridged the gap. I obviously can't prove to any of you non-believers that the system crashes immediately after setting 3.7Ghz without it enabled on that core, but I've shown that it runs under Prime with it enabled.

--Matt

[AliG]

just a word of advice, don't bother running prime for more than half an hour. I say this because all prime does is torture the cpu, how many times a day do you think you'll be literally torturing your cpu with average load? Priming too long will infact cause some unnecessary damage to the cpu, unless you have some killer cooling, but I wouldn't trust a TRUE for that.

I would use OCCT and if it passes more than ten minutes, you're set (people have found that OCCT will cause errors in just a few minutes while it takes prime95 hours to do so). Just a word of advice, since I was totally into prime95 stability testing myself and now my cpu doesn't oc as well.

anyways, good luck, you seem to have a fairly good chip, I would consider investing in watercooling if you want to get the full potential out of it

[Final8ty]

just a word of advice, don't bother running prime for more than half an hour. I say this because all prime does is torture the cpu, how many times a day do you think you'll be literally torturing your cpu with average load? Priming too long will infact cause some unnecessary damage to the cpu, unless you have some killer cooling, but I wouldn't trust a TRUE for that.

I would use OCCT and if it passes more than ten minutes, you're set (people have found that OCCT will cause errors in just a few minutes while it takes prime95 hours to do so). Just a word of advice, since I was totally into prime95 stability testing myself and now my cpu doesn't oc as well.

anyways, good luck, you seem to have a fairly good chip, I would consider investing in watercooling if you want to get the full potential out of it

OCCT3 error-ed at 8 minutes remaining on a 1 hour test Large Data Set. So a few minutes are not enough & now i set it for 2 hours.

[mattkosem]

just a word of advice, don't bother running prime for more than half an hour. I say this because all prime does is torture the cpu, how many times a day do you think you'll be literally torturing your cpu with average load? Priming too long will infact cause some unnecessary damage to the cpu, unless you have some killer cooling, but I wouldn't trust a TRUE for that.

I would use OCCT and if it passes more than ten minutes, you're set (people have found that OCCT will cause errors in just a few minutes while it takes prime95 hours to do so). Just a word of advice, since I was totally into prime95 stability testing myself and now my cpu doesn't oc as well.

anyways, good luck, you seem to have a fairly good chip, I would consider investing in watercooling if you want to get the full potential out of it

Thanks for the advice. I'm just trying to get the setup stable, and prime seems to be the best tool for the job so far. OCCT PT's linpak test doesn't seem to get results as quickly and prime has a pretty wide range of possible tests. So far I'm considering my CPU stable if it passes:

1. one hour of prime small FFTs (in place)
2. one hour of prime large FFTs with 1600mb allocated
3. one hour of prime small FFTs with 1600mb allocated

I'm not the kind of person that runs prime 24/7 and isn't content until it runs for a week without an error. I'm just trying to find a completely stable OC for this cpu that I just bought.

I'm planning to re-install the WC system when the new board arrives on Tuesday. I didn't install it initially because the GTZ that I have didn't include mounting hardware for AM2. I'm going to stick a Fuzion V1 that I picked up from another forum member on it along with the D5/DetroitAC top and TFC360 that I've already got.

--Matt

[AliG]

cool that should hold down the fort, I would probably stop a bit over 4ghz or whatever your max with 1.55v is. Hz won't deteriorate your cpu, but volts will, so try to keep the voltage to a minimum (ironic that cpus can't run without volts and yet too many kills them). Also I would seriously consider using OCCT over prime from now on, as I said, its much more efficient

[chew*]

My 2 cents on ACC.

It is not necesary with PH II, yes I know it enables a core that was disabled because it was "BAD" maybe the NB is weak on it, or maybe it doesn't scale well or maybe it runs really hot with all 4 who knows........

What ACC does is adds voltage to the chip, note the fact when enabled temps dissapear from AOD.......probably either shuts the thermal safeties off or AMD did not want you to know they were increasing voltage etc creating more heat............If it helps your PH II then maybe you need to fine tune your voltages a tad more

The only way ACC did anything for me was on a bios that though my chip was a PH I etc, did not support PH II. It clocked higher but at the cost of performance with a bios not supporting the chip.....

[[XC] Oj101]

just a word of advice, don't bother running prime for more than half an hour. I say this because all prime does is torture the cpu, how many times a day do you think you'll be literally torturing your cpu with average load? Priming too long will infact cause some unnecessary damage to the cpu, unless you have some killer cooling, but I wouldn't trust a TRUE for that.

I would use OCCT and if it passes more than ten minutes, you're set (people have found that OCCT will cause errors in just a few minutes while it takes prime95 hours to do so). Just a word of advice, since I was totally into prime95 stability testing myself and now my cpu doesn't oc as well.

anyways, good luck, you seem to have a fairly good chip, I would consider investing in watercooling if you want to get the full potential out of it

I can run OCCT for an hour without a hitch but can't run F@H for three days. I think the ultimate stability test is the one that literally takes the longest - normal daily use

[AliG]

I can run OCCT for an hour without a hitch but can't run F@H for three days. I think the ultimate stability test is the one that literally takes the longest - normal daily use

folding is a different story. That probably is the hardest legit task since it goes on non-stop doing as many calculations as the cpu can handle (not even gaming will be close to the stress). However, for a person that does not fold, anything that lasts an hour in a stability test is more than enough

[mattkosem]

My 2 cents on ACC.

It is not necesary with PH II, yes I know it enables a core that was disabled because it was "BAD" maybe the NB is weak on it, or maybe it doesn't scale well or maybe it runs really hot with all 4 who knows........

What ACC does is adds voltage to the chip, note the fact when enabled temps dissapear from AOD.......probably either shuts the thermal safeties off or AMD did not want you to know they were increasing voltage etc creating more heat............If it helps your PH II then maybe you need to fine tune your voltages a tad more

The only way ACC did anything for me was on a bios that though my chip was a PH I etc, did not support PH II. It clocked higher but at the cost of performance with a bios not supporting the chip.....

I still see temps in AOD and and CoreTemp with it enabled. I'm not sure why everyone else doesn't...

I'm running the latest bios for the board, version 802 - released in January. The release notes for it state "01. Support ACC for AM2+ 45nm CPU". Could something have been done in that bios that makes it more usable? I cannot report anything other than what I've seen on my PC. I have nothing to gain or lose by reporting my findings. I'm just reporting what I'm seeing. As I've mentioned, setting 3.7Ghz on core 3 or all 4 cores in AOD previously resulted in an almost instant reboot when under no load. This occurs at 1.40v or 1.55v and with any other variations of NB, HT, and memory voltages and timings. With ACC enabled, this does not happen and it is 3.7 is completely stable. I'm not sure why this is so difficult to believe, but I encourage others that have a weak core or weak cores on their cpu to try it. There's nothing to lose, and it provided sizable gains for me.

--Matt