Phenom 9500 w/ MSI K9A2 Platinum

[jesteronyer6]

I tried the same ^^^^^ settings and all is well at 2.5 gonna try 2.7 after you report the results..

[Daveburt714]

Plus 9x NB and 11.5x CPU on Phenom 9600 doesn't work, you're right.

I'm sure you meant 5x NB and 11.5x CPU..

Nice settings there. For NB VID, you're looking at the wrong tool.

That'll change the RD790 voltage insteead, which is labeled NB and doesn't need to be touched.

Check in AMD Power Monitor: http://www.amd.com/us-en/assets/cont...ties/setup.exe

It'll show you the current operating VIDs for CPU and NB, as below:

WooHoo!! Your absolutely right (like I would ever doubt ya.. )
I have a new setting to play with! I've changed it before, but actually set it back to 36 thinking that it wasn't working because AOD didn't show a change..

Anyway... I changed it to 28 (1.20v) and the NB still won't run at 9x250...
Oh well, at least I know NB VID actually does something now! Maybe I can raise HT Ref a little more... I'll keep you folks posted...

[KTE]

Thanks guys.

System is now on drugs.

Ran 250HT fine: http://valid.x86-secret.com/show_oc.php?id=305718

Check these for 1:2 divider EVEREST (single thread) bandwidth results.



You should check out 499 1T (high IMC clock cannot manage high RAM MHz 1T with low volts though)





And WinRAR at those speeds:



I benched this: http://valid.x86-secret.com/show_oc.php?id=305721
..and this: http://valid.x86-secret.com/show_oc.php?id=305722
...fine today.

ok, Settings for 2.5Ghz 2.0nb:
Core FID 09
VID 024
DID Divide by 1
NB FID 06
VID 036
DID Divide by 1

Settings for 2.7Ghz
Core FID 11
VID 023
DID 1
Same as before on the NB.

According to AMD power monitor, at 2.5Ghz VID 024, the core voltage was 1.25 on all cores. That was rock solid stable, even with C&C enabled, did gaming, and benching no problems at all. Nb voltage reading on AMD pow monitor is 1.1v

Yep, that's true. But like I said, those aren't voltages, they are Voltage Identities.
Real voltage is lower than the identity, as 1.25VID=1.240V and 1.1VID=~1.09V.

With my 2.7Ghz settings AMD power monitor shows 1.2625v per core, currently running with C&Q disabled, so that may help.

That's roughly 1.255V. I'll check the mapping out and edit this figure in a moment.

But, then again, the RAM and PSU are more or less hold overs until I can afford to get a decent unit, though going by the price I could of ended up getting a decent 600w psu from a better brand. Didn't know how infamous Raidmax PSU's were till after I sent in for the rebate.

And intersting results, I installed WinXP SP3 and performance on most benchies in Everest jumped by between 500 and 1000 points, just while running at 2.5Ghz.

SP3 is better, much better IME, yup.

As for the PSU, I hope it doesn't blow your system as it usually does, i.e. CPU/MOSFETs on board.

Would you try and jump off Eiffel Tower with a handmade box containing many wires carrying two 1m cardboard wings on either side, even for one second and expect to live?

Exactly. Depending on your GPU, you don't really need much power. Phenom with HD 3870 gets around 220-230W AC full load with two HDs and two 2.2V RAMs (etc). For safety a 500W will be more than perfect. Try for the Enhance ENP-5150GH, it's excellent and usually on for cheap around ($70 last I saw it). Some others are Antec TP3/Earthwatts, Corsair VX/HX, Seasonic S12 I/IIs and Enermax Infinitys in that wattage range, from 450-550W. There was an Ultra X3 that ran pretty good too, not the X-Finity though since I suffered bad experience with that.

I'm sure you meant 5x NB and 11.5x CPU..

You sure?

NB: 05=9x
CPU: 07=11.5x

Phenom 9600 =>

WooHoo!! Your absolutely right (like I would ever doubt ya.. )
I have a new setting to play with! I've changed it before, but actually set it back to 36 thinking that it wasn't working because AOD didn't show a change..

36 is too low. Stock is 24=1.250VID. Put it to 15 if you go plus 2.1G NB and see it cruise.

Then you can lower the VID to find lowest stable it when you reach your CPU/HT limit.

[Rammsteiner]

I would have thought Phenom had a bigger Bandwith? Im running 251Mhz on RAM 4-4-4-10 T2 (and subtimings tuned with Memset) and I get almost 11GB/s with Read and Write. Copy is AFAIK about 10GB/s.

[Daveburt714]

You sure?

NB: 05=9x
CPU: 07=11.5x

Phenom 9600 =>

Excuse me Master...
I think we're mixing VID's and Multi's here....
NB Vid = 1 / CPU Vid =7 (NB 5x / CPU 11.5x), is Bad Karma..
NB Vid = 5 / CPU Vid =7 (NB 9x /CPU 11.5x), is Zen...

36 is too low. Stock is 24=1.250VID. Put it to 15 if you go plus 2.1G NB and see it cruise.

Then you can lower the VID to find lowest stable it when you reach your CPU/HT limit.

My CPU VID is at 24 (1.25v)... Are you saying I should set my NB VID to 15 (1.3625v)? That seems awfully high to me!!

I can almost see why the new BIOS removed the P-States option. It leads to alot of confusion!! I'm sure it was the easiest way, but if they just gave us Multi's and Voltage options for CPU, NB and HTT it would make things alot more user friendly...

You've done alot of testing KTE, is there really any advantage to going over 1.20v (VID 28) on the NB?

[KTE]

I would have thought Phenom had a bigger Bandwith? Im running 251Mhz on RAM 4-4-4-10 T2 (and subtimings tuned with Memset) and I get almost 11GB/s with Read and Write. Copy is AFAIK about 10GB/s.

Phenom does have higher bandwidth but only with software which supports it and is multi-threaded. Single threaded like EVEREST+Phenom, it will show low bandwidth. Sandra loads all cores for memory testing and in Sandra it gets plus 12.8k at those settings, maybe even plus 13.2k. Should be fair high. A very loose run of mine got >12.8k last I tried with low IMC clocks.
In EVEREST, Phenom has around 7700MB/s Ganged and Unganged 7000MB/s stock bandwidth at 800 4-4-4-4-11 1T on the 790FX with TLB Caching boost (Red).

I think we're mixing VID's and Multi's here....
NB Vid = 1 / CPU Vid =7 (NB 5x / CPU 11.5x), is Bad Karma..
NB Vid = 5 / CPU Vid =7 (NB 9x /CPU 11.5x), is Zen...

I think we are too.
I stated NB FID 05 in BIOS = 9x multi and CPU FID 07 in BIOS = 11.5x multi.
I've not mentioned VIDs yet, they're much lower and in regards to voltages instead of the multi.

FID=>Multi
VID=>Voltage ID

Plus 9x NB and 11.5x CPU on Phenom 9600 doesn't work, you're right.

This is true ya know. Phenom 9500/9600 [locked] don't allow plus 9x NB multi and plus 11x/11.5x CPU multi, although you can downclock the multiplier and you can't downclock the NB multi on the BE's according to my testing.

NB/CPU VID 001=1.5375VID -> ~1.53V

My CPU VID is at 24 (1.25v)... Are you saying I should set my NB VID to 15 (1.3625v)? That seems awfully high to me!!

If you want plus 2.16G on IMC you need to give it extra volts. 15 is good for 2.4G IMC at least IME and would give around 1.35V to the IMC.

I can almost see why the new BIOS removed the P-States option. It leads to alot of confusion!! I'm sure it was the easiest way, but if they just gave us Multi's and Voltage options for CPU, NB and HTT it would make things alot more user friendly...

To most users, yes, it would cause many problems because it's that powerful an area. For me, it was the highlight and what oc was. Without it, it's the same as having no oc on the chip to me.

You've done alot of testing KTE, is there really any advantage to going over 1.20v (VID 28) on the NB?

Firstly, like I stated earlier, stock Phenom is 1.250VID on the IMC which is NB VID 024. If you want higher MHz than stock then obviously you'll need more voltage for the IMC soon after 1.8G. Typically you can have the system fully stable with 1.25VID @ 2.1G IMC Unganged and 2.2G Ganged IME but more MHz will require higher volts than stock. Up-to 2.6G on the IMC I've seen possible with ~1.37VID.

BTW I'm not a Master of anyone that I know... apart from my wife to be.

[Brother Esau]

Where the Beef? Errrr I mean DFI 790FX-M2R

@Tye, testing new 1/30/08 Beta find very good worth in New Bios

[Daveburt714]

Geez, your right Man... Most of what I called VID, should be FID...

You get the idea though... Hic/Cough..

I'll shaddup up now!!

[jonspd]

Geez, your right Man... Most of what I called VID, should be FID...

You get the idea though... Hic/Cough..

I'll shaddup up now!!

LOL this is what happens when your testing at 2 in the morning LOL hehe

[DaMulta]

So has anyone put the black under dry ice yet?

[KTE]

So has anyone put the black under dry ice yet?

You were going to try it, weren't you?

Tried them TECs on it?

I've tried up-to -30C on the 9500 and on my last BE from my work place, but that was 10 minutes, no clampdown, no insulation and I was just testing coldbug and IHS temps. 3G failed bootup, 2.8G failed, 2.7G failed and 2.63G was max on it with stock cooler anyway, so I didn't bother more.

[DaMulta]

I sold mine

BUT

I think I will have a black next week sometime.



So no cold bug down to -30 at least.

[Brother Esau]

@KTE......Hey have you had any issues with AMD overdrive if the Bios itself is configured for you're overclock? After installing new Bios last night I wated to see if there was any AMD Overdrive support added into the Bios for it so I installed it with optimized defaults set rebooted and upon boot Desktop froze and had to do a manual restart so after about 3 reboots and opening and closing of the program I set the Bios to my O.C Settings and when reaching desktop after a few minutes I open AMD Overdrive and its fine so I start Priming with it open and am on the forum with that in the Back ground and it Blue Screens on me this sh@t has been consistent with this board and this program Lock ups . BSOD while in Windows and just flaky behavior in general

So the Million dollar question is do any of you have a problem using this program at anything other than optimized defaults set in Bios?

BTW at no time have I ever used this thing other than to monitor temps or view settings never used it to make changes to Bios what so ever!

Please give me some feedback on this as I wish to submit info to DFI for Feedback on issues with Bios and I would say this is one of them and its been present since I have owned the Board and I have tried all versions with no success and was currently using Latest Beta.

[Hosser]

@KTE......Hey have you had any issues with AMD overdrive if the Bios itself is configured for you're overclock? After installing new Bios last night I wated to see if there was any AMD Overdrive support added into the Bios for it so I installed it with optimized defaults set rebooted and upon boot Desktop froze and had to do a manual restart so after about 3 reboots and opening and closing of the program I set the Bios to my O.C Settings and when reaching desktop after a few minutes I open AMD Overdrive and its fine so I start Priming with it open and am on the forum with that in the Back ground and it Blue Screens on me this sh@t has been consistent with this board and this program Lock ups . BSOD while in Windows and just flaky behavior in general

So the Million dollar question is do any of you have a problem using this program at anything other than optimized defaults set in Bios?

BTW at no time have I ever used this thing other than to monitor temps or view settings never used it to make changes to Bios what so ever!

Please give me some feedback on this as I wish to submit info to DFI for Feedback on issues with Bios and I would say this is one of them and its been present since I have owned the Board and I have tried all versions with no success and was currently using Latest Beta.

Whenever I do anything in AOD except adjust PCI-e Speed, the system hangs, and an ear-splitting high-pitch tone screams out of my speakers. It's a pile of *ahem* if you ask me.

[justapost]

Phenom no POST after clear CMOS

Well, well. My Phenom did not POST for over 24hrs. This is very abnormal, I was about to return it because this is the exact problem my older BE had and I really do get sick of it. I wonder why this occurs?
None of the 9500/9600s had it though on the same board with the same BIOS. I had phoned my uncle to drop it off Saturday after trying everything which usually worked with the last BE.
The error started when I chose 1066 mode. It failed POST, so I had to clear CMOS. That's it, after that it was stuck in a reboot loop.

So how was it fixed? Ironically, you had to go subzero AND take the chip out of the socket and reinsert it to fix it. Stuck the chip under -20C ice in a plastic bag inside a tub within a -31C refrigeration unit. Booted without HSF on, and here is the IHS temp after the first successful bootup at ambient 27C (HSF still disconnected):



Yep, it took roughly -10C - 2C BIOS temps to boot it up and that's when it was coming back to ambient. After that I could reinsert everything but using AMD stock cooler, I haven't put any more paste on it and now the stock temperatures are 8-10C higher.

Lol, how did you get the idea putting the cpu in the refrigerator?

Achim: Ganged mode is fastest for single threaded applications and unganged mode is fastest for multi-threaded applications.

Here's Ganged Vs. Unganged mode at stock (little tweaked timings for you to compare). AOD button = Red.

[center]Unganged DCT Mode Vs. Ganged DCT Mode
(Click for full detail screenshots)

Thx, but I can not compare my max 11x multi cpu results with those 11,5 ones.

Here are my results at 2,2GHz with identical memory timings.

[KTE]

@KTE......Hey have you had any issues with AMD overdrive if the Bios itself is configured for you're overclock? After installing new Bios last night I wated to see if there was any AMD Overdrive support added into the Bios for it so I installed it with optimized defaults set rebooted and upon boot Desktop froze and had to do a manual restart so after about 3 reboots and opening and closing of the program I set the Bios to my O.C Settings and when reaching desktop after a few minutes I open AMD Overdrive and its fine so I start Priming with it open and am on the forum with that in the Back ground and it Blue Screens on me this sh@t has been consistent with this board and this program Lock ups . BSOD while in Windows and just flaky behavior in general

So the Million dollar question is do any of you have a problem using this program at anything other than optimized defaults set in Bios?

BTW at no time have I ever used this thing other than to monitor temps or view settings never used it to make changes to Bios what so ever!

Please give me some feedback on this as I wish to submit info to DFI for Feedback on issues with Bios and I would say this is one of them and its been present since I have owned the Board and I have tried all versions with no success and was currently using Latest Beta.

I don't open AOD, hardly, because it calls the PLL and RD790 chipset and those will and can lockup, freeze, reboot your system. They do this a lot, so I avoid it because I absolutely loathe it happening!

That's one of the bad things because I'd like to change AOD TLB caching button from green to red, but I can never open it most of the time which gives low memory performance.

However, I can oc and change memory/voltage timings with it quite OK many times with oc'd BIOS settings. Most of the time though, AOD starts at stock and then it has to load your current clocks... that's when it'll fail. Not with low MHz, with only high MHz.

Try it-> set like 100MHz above stock CPU speed from the BIOS and bootup. Start AOD... does it work?
Then increase the MHz at bootup... the higher you go, the buggier it gets and the freeze/lockup is when the PLLs are fluctuating highly actually.

Lol, how did you get the idea putting the cpu in the refrigerator?

Do it all the time in some sealing and insulation.

Thx, but I can not compare my max 11x multi cpu results with those 11,5 ones.

Here are my results at 2,2GHz with identical memory timings.

Those RAM/L3 are directly comparable, they're the same NB/RAM.
The L1/L2 caches aren't comparable because my CPU is 100MHz faster.
I'll run some 200x11 today and post here.

However you only ran one single threaded bench which doesn't support Phenom fully yet. Try and run Sandra/WinRAR as it supports Phenom multi-threaded memory access and that'll show the real bandwidth of Phenom better. Unganged will perform better than Ganged in multi-thread, EVEREST can't be used to show the real bandwidth but can be used to show the single-threaded memory bandwidth.

[Brother Esau]

So basically AMD Overdrive is just a Piece of Sh**t then?

[gOtVoltage]

Yup Burrrrrrp...many beers later i use AOD for just viewing my Bios stats and nothing more!

[KTE]

Its very good actually unless you're looking for high end oc's. Too many haven't even tried it past one or two times, so they have no clue and are making misleading statements. A bit like with Vista when it was released.

It's not a clockgen/setfsb by any means, its too heavy an application at loadup. With Intel systems now, most oc's/benches are done through using the above tools or scores/max oc's would be not even 100MHz near with bootup MHz. AMD RD790 lacks anything as such.

[gOtVoltage]

Its very good actually unless you're looking for high end oc's. Too many haven't even tried it past one or two times, so they have no clue and are making misleading statements. A bit like with Vista when it was released.

It's not a clockgen/setfsb by any means, its too heavy an application at loadup. With Intel systems now, most oc's/benches are done through using the above tools or scores/max oc's would be not even 100MHz near with bootup MHz. AMD RD790 lacks anything as such.

Exactly...Also Bioses need to be fully reconized by AOD .A properly written bios is a must have for AOD. I believe this is why people get a mojority of lock ups , unoptimized Oem Bioses!.

[KTE]

Here ya go Achim, everything's matched now even the CPU speed. Only thing different is you have the theme service switched off, I'm on XP Pro SP3 and RD790 chipset.


Ganged DCT Mode




Unganged DCT Mode



Interesting thing to note-> people really haven't worked out this platform properly yet, let alone a week or a few days for a review. Look below

Power Consumption:

CPU 11.5x 2.3GHz @ 1.160V, 1.175VID
NB 9x 1.8GHz @ 1.050VID
HT 9x 1.8GHz @ 1.1V
2x 1GB RAM 800 @ 2.2V
HSF 1290RPM @ 5V - 400mA
Sapphire HD 2600 XT 800/700 @ 1.26V/DK
2x SATA II WD Caviar SE 80GB
Antec Earthwatts 430W

Hibernate: 9W AC
Standby: 55W AC
Idle: 94W AC
Load: 161W AC


Compare with stock system findings 15th Jan: Power Draw

CPU 11x 2.2GHz @ 1.160V, 1.175VID
NB 9x 1.8GHz @ 1.050VID
HT 9x 1.8GHz @ 1.1V
2x 1GB RAM 800 @ 2.2V
HSF 1290RPM @ 5V - 400mA
Sapphire HD 2600 XT 800/700 @ 1.26V/DK
2x SATA II WD Caviar SE 80GB
Antec Earthwatts 430W

Ganged DCT Mode:-
Idle AOD Green: 92W AC
Idle AOD Yellow: 95W AC
Idle AOD Red: 119W AC
Load: 156W AC

Unganged DCT Mode:-
Idle AOD Green: 94W AC
Idle AOD Yellow: 97W AC
Idle AOD Red: 118W AC
Load: 158W AC

[Daveburt714]

Ok, Now that my head has cleared, I figured out what was screwing me up last night (besides Beer )

You said "Plus 9x NB and 11.5x CPU on Phenom 9600 doesn't work."
And I read it as "Plus , 9x NB and 11.5x CPU on Phenom 9600 doesn't work."
If you would only have said "Over 9x NB and 11.5x CPU on Phenom 9600 doesn't work."

I actually do have a legit question though.. You said these were stock VID settings for the 9600 (non BE)?
CPU VID = 24
NB VID = 24

When I first opened P-States after flashing back to 1.13, these were the Pre-Sets...
CPU VID = 28
NB VID = 36
Which I assumed to be Stock...

I'm currently running @ CPU=24 / NB=36... If stock NB is 24 then I'll set it to that....

[KTE]

Yep, that's how they load from disabled->enabled... but, why not see what FID/DID are presets.

Stock is 1.2000-1.250/1.250 for both parameters. You can disable P-States and bootup at stock all auto and check in AMD PowerMonitor.

On a side, you don't need high MHz for high bandwidth on Phenom...




Tried CAS3 1000 today, my sticks can do CAS3 1066 on P35, but with this system = no go. The system will boot but it will boot with bad CMOS: reset and everything reset.

Also, I've been experimenting. Two things to note, two very important things:

* Phenom is starved for memory bandwidth. Yes, damn, I've been playing with high clocks and low clocks and Phenom performance is low at low RAM clocks (same CPU/NB) compared to at higher RAM clocks, say 400 3-3-3-3 compared to 500 4-4-4-4.

* Phenom is very IMC dependent. As you may know, running Super Pi at 70k efficiency when the best efficiency is around 75k is impossible. On an Intel chip, best you can manage is 49k when you can get 50k very easily without any tweak. But with Phenom I've tried this and you gain quite a bit of performance in applications through just IMC overclocking. Number calcs will show this up very quickly, like wPrime, Pi apps and so on. You'll be quite shocked at what you observe if you do it the right way.

Check this. My system runs 1-core 2.8G bootup fine-> http://valid.x86-secret.com/show_oc.php?id=306352

1-core 2.9GHz fine but with high volts. Minimum needed to run 2.9G is 1.45V-> (click for full image)



3GHz was max, it froze soon after-> (click for full image)




BUT... I met very bad performance going from 2.75G to 2.8G. 2.8G was 2.2s slower than 2.75G in SuperPi 1M and 2.6G was 0.3s faster than 2.8G



Yeah, it's probably called instability, I gather. I better stability test my RAM to see if it's okay, this is very typical of dying RAM.

[justapost]

Hmm..

^ That was your comment about MaxVid/MinVid.
Tried it here on the m2a-vm with 1,325V vcore and MaxVid is 1,3V here. Gonna inspect that register further.

Those RAM/L3 are directly comparable, they're the same NB/RAM.
The L1/L2 caches aren't comparable because my CPU is 100MHz faster.
I'll run some 200x11 today and post here.

However you only ran one single threaded bench which doesn't support Phenom fully yet. Try and run Sandra/WinRAR as it supports Phenom multi-threaded memory access and that'll show the real bandwidth of Phenom better. Unganged will perform better than Ganged in multi-thread, EVEREST can't be used to show the real bandwidth but can be used to show the single-threaded memory bandwidth.

Yep, I tried RightMark's multi threaded memory benchmark whom SocketMan recommended.
Unganged mode is faster here. Even more faster (ganged vs unganged relation) with ram at 1066 than at 800.

Here ya go Achim, everything's matched now even the CPU speed. Only thing different is you have the theme service switched off, I'm on XP Pro SP3 and RD790 chipset.

Yep, theme service and some other useless services are deactivated by default. No international SP3 available I guess.
Beside that I see the biggest differences in latencies. MSI seems to be a tight better in unaganged, Asus in ganged mode.
Thank you for running those benchmarks, I expected bigger differences.

Interesting thing to note-> people really haven't worked out this platform properly yet, let alone a week or a few days for a review. Look below

Power Consumption:

CPU 11.5x 2.3GHz @ 1.160V, 1.175VID
NB 9x 1.8GHz @ 1.050VID
HT 9x 1.8GHz @ 1.1V
2x 1GB RAM 800 @ 2.2V
HSF 1290RPM @ 5V - 400mA
Sapphire HD 2600 XT 800/700 @ 1.26V/DK
2x SATA II WD Caviar SE 80GB
Antec Earthwatts 430W

Hibernate: 9W AC
Standby: 55W AC
Idle: 94W AC
Load: 161W AC


Compare with stock system findings 15th Jan: Power Draw

CPU 11x 2.2GHz @ 1.160V, 1.175VID
NB 9x 1.8GHz @ 1.050VID
HT 9x 1.8GHz @ 1.1V
2x 1GB RAM 800 @ 2.2V
HSF 1290RPM @ 5V - 400mA
Sapphire HD 2600 XT 800/700 @ 1.26V/DK
2x SATA II WD Caviar SE 80GB
Antec Earthwatts 430W

Ganged DCT Mode:-
Idle AOD Green: 92W AC
Idle AOD Yellow: 95W AC
Idle AOD Red: 119W AC
Load: 156W AC

Unganged DCT Mode:-
Idle AOD Green: 94W AC
Idle AOD Yellow: 97W AC
Idle AOD Red: 118W AC
Load: 158W AC

AMD should have merchendised the tlb-fix as a power saving feature.
Is the load consumption also affected by the tlb-fix?

[KTE]

System CPU load power consumption stays the same, hence why I only left one value for both conditions.

Importantly, if any reviewer clicked that or opened AOD without changing it back to yellow/green, they will give higher power figures, 30-38W higher idling than they should be.