[how to] Get the best SuperPi efficiency out of the K10 platform

The aim:
As the K10 platform has matured more people start benching their Phenom II cpu's in all known bechmarks.
As SuperPi has become a standard here more or less it's good to know how to get the best result out of your K10 system knowing this platform isn't as strong as others in SuperPi.
I tried a lot of relevant settings to see what works best and what not, in other words, use it to your advantage. ;)

The method:
The effficiency of the system is gauged by the so called Performance Product (PP) and is basically the CPU speed in MHz x the output result of SuperPi.

The hardware:
Motherboard- Asus M4A79 deluxe bios 0903
CPU- Phenom II 940 black edition running on all 4 cores
RAM- Cellshock Black PC6400 2x1GB

The sofware:
Winows XP Professional SP2, no tweaks, just a clean install.

The testresults:
Let's start off pretty basic, i tried to have all the results at around 3.6GHz or at least close to that as possible.

Here is 18x200 RAM at DDR1070 with 5-5-5-15 2t timings, NorthBridge at 2GHz:

http://img.photobucket.com/albums/v6...05-5-520NB.jpg

PP=3612x19.203=69361 This should be easy to improve.
Let's just raise the NB to 2.6GHz :

http://img.photobucket.com/albums/v6...05-5-526NB.jpg

PP=3612x18,906=68288. That's a lot better, only by just raising the NB frequency.

Now let's see what tighter timings will do.
18x200 RAM at DDR 800 4-4-4 2t NB at 2.6GHz:

http://img.photobucket.com/albums/v6...-4-42t26NB.jpg

As you can see, exactly the same as with DDR 1070 5-5-5.


PP=3612x18,906=68288

Now using the 1:2 divider gives the advantage of using tighter timings and also allows 1t timing, i was really curious how much 1t would bring on this platform.

18x200 RAM at DDR 800 4-4-4 1t NB at 2.6GHz:

http://img.photobucket.com/albums/v6...-4-41t26NB.jpg

PP=3612x18.922=68346
WTF? It's worse than 2t! I tried several times but this was the best i could get.
To the best of my knowledge command rate is the timing of the bus between the cpu and the memory, at 1t the bus gets data at every clockcycle, at 2t every other clockcycle.
Appearently due to the integrated memory controller the latency is that low that 1t makes zero impact on the performance or makes it even worse.

So, basically forget 1t for the K10 platform if you're planning to run SuperPi, DDR3 might differ here.

So what else do we have? Well even 4-3-3 1t timings were possible at the 1:2 divider so why not?

18x200 RAM at DDR 800 4-3-3 1t NB at 2.6GHz:

http://img.photobucket.com/albums/v6...-3-31t26NB.jpg

PP=3612x18,906=68288

No real improvement here, now let's raise the NB frequency.

18x200 RAM at DDR1070 with 5-5-5-15 2t timings, NorthBridge at 2.8GHz:

http://img.photobucket.com/albums/v6...05-5-528NB.jpg

PP=3612x18.828=68006. Now that's better! Still not sub 68K but we're getting close now.

How about raising the busspeed?

15x240 RAM at DDR 800 4-4-4 1t NB at 2.6GHz

http://img.photobucket.com/albums/v6...04-4-426NB.jpg

PP=3600x18.985=68346. No, that's not the way. We could use higher busspeed to our advantage cause with raising the busspeed the memoryspeed goes up as well.:)

Now 14x260 RAM at DDR 866 4-4-4 NB at 2.6GHz

http://img.photobucket.com/albums/v6...-4-41t26NB.jpg

PP= 3640x18.812=68475. Not really what i was looking for but maybe some more speed on the RAM?

14x260 RAM at DDR 1040 5-5-5 NB at 2.6GHz

http://img.photobucket.com/albums/v6...05-5-526NB.jpg

PP=3640x18.765=68304. Getting better now. Unfortunately the board refused to boot or run stable above 260 bus so i couldn't get more RAM speed, i'm sure more speed on the RAM would have improved results.

The last thing i could try was running more NB speed.

Here is 14x260 RAM at DDR 1040 5-5-5 NB at 2.86GHz:

http://img.photobucket.com/albums/v6...05-5-528NB.jpg

PP=3640x18.688=68024 Still no sub 68K but there's one final trick up my sleeve...

14x260 RAM at DDR 1040 5-4-4 NB at 2.86GHz:

http://img.photobucket.com/albums/v6...05-4-428NB.jpg

PP=3640x18.671=67962. So there you have it, sub 68K

For those who think Windows7 is faster:

http://img.photobucket.com/albums/v6...y/Windows7.jpg

PP=3612x19.031=68739.

I'd rather stick to XP.;)

I hope you guys enjoyed this little write up.

thanks for the in-depth research in making K10 run faster on SuperPI

Thanks for the work in investigating this. Perhaps a chart would make the presentation & comparison between the various settings easier for the viewers.

Quote:

---

Originally Posted by Zeus

WTF? It's worse than 2t! I tried several times but this was the best i could get.
To the best of my knowledge command rate is the timing of the bus between the cpu and the memory, at 1t the bus gets data at every clockcycle, at 2t every other clockcycle.
Appearently due to the integrated memory controller the latency is that low that 1t makes zero impact on the performance or makes it even worse.

So, basically forget 1t for the K10 platform if you're planning to run SuperPi, DDR3 might differ here.

---

Command rate has no direct influence on the databus or data transfer rate. What it changes is the number of clock cycles the controller uses to qualify new commands and their corresponding addresses. As you add more memory sticks, the capacitive load on the controller's address unit increases, which in turn reduces the slew rate of the signals. Since the signals switch slower from 0 to 1 (and vice versa), the controller might need an additional clock cycle to generate the correct combination of 0s and 1s. The biggest drawback of 2T is probably not the delay itself, but rather that the address bus gets more "crowded". Each entry takes up two timeslots, so there would simply be less space (time) to squeeze in new commands.

The difference was rather big in the 939 days, but small for AM2 (although 1T was still slightly faster).

In K10's unganged mode, the difference should be even smaller, because there are two independent controllers, which gives potentially longer read/write bursts per command. This lowers the risk of collisions on the address bus. A transfer of a given data chunk would simply require fewer commands (per controller), so fewer entries would have to "compete" for the same timeslots.

As a result of the "new" Fly-By address bus topology and the on-DIMM termination, DDR3 systems can handle 1T at reasonably high loads and clockrates.


Anyway, nice work! I enjoyed analyzing the results :up:

Nice work there :up:

Quote:

---

Originally Posted by Particle

Thanks for the work in investigating this. Perhaps a chart would make the presentation & comparison between the various settings easier for the viewers.

---

I might just do that, thanks.

Quintero,

Thanks for shedding a light onto the command rate part, very interesting and much appreciated.

This is sweet! Too bad you didn't get 520 4-4-4:p: That would get your time down a bit, I think. What about ganged vs unganged, btw?

Thanks. :) I would have loved to run 520 4-4-4 but the 3:8 divider won't let me boot at cas level 4.

Ganged vs unganged never crossed my mind to be honest, i might as well give that a try later, this testing is very time consuming stuff you know.

I'm working on this atm, trying to get this gigabyte to cooperate at 300fsb....

Here we go, I would not reccomend this to anyone for everyday use, this particular board is a PITA to get 300 fsb stable on and the voltage I am feeding everything is :shocked:, I could probably get the efficiency better with higher ram speed but there is fsb wall at 300+, I will try a diff configuration later...

The times don't look good but the efficiency is just over 67K,

http://members.cox.net/bostonreefer3/efficiency.JPG

@ Zeus

realy good job! thank you!


@ chew

i have the same problem with my asus MVP

if i would make 301 in the bios the mainboard wants to boot an CPU Clock auf 4200mhz with every CPu multiply

but if i boot with 300 and pust it in windows with AOD over 300 it is no problem

http://www.abload.de/img/superpi4fertigjfiq.jpg

yah tried aod, even with max voltage everywhere on the board it doesn't want to here it.

nice, very informative. a little off topic but has anyone found a time where tightening the timings and changing to 1T has actually had a benefit to performance? i have changed my timings all around and never once have i seen a difference in performance.

@ chew*

did you have Micron D9GMH to make a test with them?

maybe your 2 x 2gb memmory does not work with more than 600

I tried a 2x2g set of corsair dominators v1.1 and it puked them up, hosed my OS twice, oddly they passed memtest fine.

Had to run 5-4-4-10 with the 4gig kit though........didn't want to hear CL 4.

Someone donated me a benching set 2x1g dominator v1.2, I will give them a shot.....

My board and mem can clock decent, no clue what these sticks are but they crank at rated timings.....they are preety much dead now though.

http://members.cox.net/bostonreefer3/massmanPI3.JPG

http://members.cox.net/bostonreefer3/massman4.JPG

my corsair dominators always passed mem test but they would crash in prime and had to be overvolted to run stock clocks and timings.

@ chew*

forget memtest for ramtest
did you know S&M ?
it is the best ram test i ever use! much better than memtest ore prime

here is the DLL http://rapidshare.com/files/198699292/S_M.exe

here are the best settings for the ram test

http://www.abload.de/img/smram8r64.jpg

if you search for e good benchmark DDR2 ram, you can have some from me. but i am in austria in europ!

to bad S&M doesnt work for 64bit...... :(

I have 2 sets of 2x2 dominators V1.1 and V1.2 and they both do 4.4.4.10 at 1000Mhz. I have faith that those 2x1g set will do the same easy Chew*. Sadly one of my 2x2 sets died so I RMA'd them and got back a set of V4.2 And they wont do any better than 5.5.5.12. I have some Old (at least a year) Gskill ram 2x2 im testing right now at CAS 4, gonna start memtest and go to bed.

Quote:

---

Originally Posted by Slagathor

to bad S&M doesnt work for 64bit...... :(

---

i use XP32 and XP64 and there it works. oese use Vista 64 and there it also runs.

you must unpack it first

It's all in the CPU/NB frequency balance: if the NB is clocked a lot lower than the cpu, your efficiency will drastically decrease (performance product will increase)

Quote:

---

Originally Posted by chew*

I'm working on this atm, trying to get this gigabyte to cooperate at 300fsb....

Here we go, I would not reccomend this to anyone for everyday use, this particular board is a PITA to get 300 fsb stable on and the voltage I am feeding everything is :shocked:, I could probably get the efficiency better with higher ram speed but there is fsb wall at 300+, I will try a diff configuration later...

The times don't look good but the efficiency is just over 67K,

---

Would you like to share which voltages you raised to get 300fsb?

Your efficiency is good due to high ram speed and high NB clock not because of 300 fsb but i think you know that.

Keep them coming and show me sub 68K chew*! ;)

Quote:

---

Originally Posted by techtrancer

@ Zeus

realy good job! thank you!


@ chew

i have the same problem with my asus MVP

if i would make 301 in the bios the mainboard wants to boot an CPU Clock auf 4200mhz with every CPu multiply

but if i boot with 300 and pust it in windows with AOD over 300 it is no problem

---

Thanks. :) Very nice efficiency there, 600+ on the ram seems to do a great job, i might give that a try too.

Quote:

---

Originally Posted by massman

It's all in the CPU/NB frequency balance: if the NB is clocked a lot lower than the cpu, your efficiency will drastically
decrease (performance product will increase)

---

Very true my friend, NB clock seems to have the most impact on the score with the memoryspeed in second place. Timings seemed to matter a lot less.

i think much memmory is also good for superpi!

4gb are better than 2 gb

Quote:

---

Originally Posted by Zeus

Would you like to share which voltages you raised to get 300fsb?

Your efficiency is good due to high ram speed and high NB clock not because of 300 fsb but i think you know that.

Keep them coming and show me sub 68K chew*! ;)

---

Zeus If you want it via PM no problem however I'd like to refrain from posting them for all to see, not for the faint of heart and people will be blowing up cpu's/mobos left and right.

Currently working on the second part of my Phenom performance scaling analysis (first part was about the Ph-I) and i just hit 67947 - no tweaks whatsoever.

Will upload the screenie later ;)

//edit: 67763

67579, this is easy. I added task manager.

http://i154.photobucket.com/albums/s...man-/22516.png

I must have a good chip after all: 2.8GHz NB without touching the voltages (except from the cpu, just to be safe when testing)

Mine could also do 2.8 NB at stock volt (auto) but after i raised the CPU-NB voltage to 1.45V it now refuses to boot at 2.8GHz, strange huh?

Running NB clock higher than CPU clock has a negative effect on the efficiency: not equal, but worse than 1:1 :(

Quote:

---

Originally Posted by Zeus

Mine could also do 2.8 NB at stock volt (auto) but after i raised the CPU-NB voltage to 1.45V it now refuses to boot at 2.8GHz, strange huh?

---

That voltage for chipset may need to be adjusted for the NB @ that speed .If you Push the NB Higher you should Raise the NBchipset volts a little for your Ram. Usually its under CORE PCI-E volts for this. But there are many factors beside just raising the NB volts for CPU. Such As Ram Divider CPUmulti and timing all play a part. Most IMC's for PHENOM and PHENOMII can handle 1.55volts easy @ High NB with good cooling.

Ive also found since the NB is on CPU you have to balance the Volts and Heat for it, as the CPU /NB clocks go up. When you cant go Higher than say2800+NB with less than 1.5v its usually heat related or Chipset volts need a bump up for the ram stability.


Example,,
9950 @ 2500nb i use 1.45volts,, @ 2700nb i need all 1.68v my mobo has to offer with Great watercooling . Both require me to Raise the Core PCI-E for my ram when i run 1100+mhz for the Ram.

For 940 PHII the IMC is stronger allowing less volts for High NB. Once you go above a certain limit HEAT/Volt wise ,,you must have good cooling to run the Higher Volt range and keep things stable. This dragon platform requires lots of time to find that nice Max OC.


Sometimes the IMC/NB can run with more/less volts and Higher NB using different CPU/Ram mulis.

One key is to keep HT close to stock with using High NB for stability. Sometimes a Lower CPU multi will work better with a hIhger NB multi.

PHenom ,,PhenomII,, Again i must say take your time and thus require lots of tweaking for Max NB Cpu performance along with Ram timing.

Note:,, Higher VDD can help with the CPU/NB by limiting vdroop as well.
Just my 2cents:D

Quote:

---

Originally Posted by techtrancer

i use XP32 and XP64 and there it works. oese use Vista 64 and there it also runs.

you must unpack it first

---

It wont let the driver start.
A warning comes up and whines about drivers needing to be signed.....

did you have all SP installed?

Quote:

---

Originally Posted by massman

Running NB clock higher than CPU clock has a negative effect on the efficiency: not equal, but worse than 1:1 :(

---

really?!I thought, the highest NB = the best scores..Or what do u mean exactly?

edit:ouch:am:...u mean CPU clock and NB clock, not HT clock and NB clock :D.Im sorry

Quote:

---

Originally Posted by FlanK3r

really?!I thought, the highest NB = the best scores..Or what do u mean exactly?

edit:ouch:am:...u mean CPU clock and NB clock, not HT clock and NB clock :D.Im sorry

---

your Not ,, StuPid,, The Brain just suffering from lack of Nb and ram timing:rofl:

Zues and massman thx for giving me a challenge :up: this board is very very hard to get PI to perform, etc it sux with memory performance however on the brightside just about no one can touch me in cinebench with it 32bit OS etc the cpu scores are out of this world........

Anyway you guys made me strive to squeeze every last bit of performance out of this giga GX board.

If anyone feels the need to bang there head against a wall or just like a challenge try the giga 790GX some crappy ram and try to break 68K efficiency :p:

Yes that is 3144nb frequency on air ;) anyone want to see 3200 :p:
http://members.cox.net/bostonreefer3...encysub68k.JPG

Here we go sub 67,900 efficiency 3200+ nb ;) this is air cooled

I have been making this poor motherboard scream for mercy the past couple of days, lets see what else I can force feed it :D

http://members.cox.net/bostonreefer3...ncysub68k2.JPG

Outstanding result there chew! Your result emphasizes the importance of high NB clocks to get every last bit of performance out of this platform. Great work, how much volts on the CPU-NB for that?

Quote:

---

Originally Posted by Zeus

Outstanding result there chew! Your result emphasizes the importance of high NB clocks to get every last bit of performance out of this platform. Great work, how much volts on the CPU-NB for that?

---

:rofl: really want to know? :eek: IIRC +.50 vcore at default 1.250 to aid in keeping heat down. Bumped vcore .05 once I felt comfortable with temps for second run, also needed it to be stable enough above 3.3 .

Massman I tried 1/1= horrible performance @ 3.3 cpu/3.3 NB

So that is 1.75v CPU/NB? (thanks god it's up again-- that was a long 15 minutes for rebooting!)

no clue what default is :shrug: board just has + voltage not current, isn't default 1.35v? thats what aod says at least.

I would highly not reccomend this on air, I have no aversion to breaking hardware.......

Wow, long 15 minutes!

Nice results Chew*

Excellent thread :clap:

I've had better results than you it seems with Window7 though. Here are 2 quick tests I did a while ago.

The first result I was running less northbridge than you (I haven't gone past 2.7Ghz but the test was only at 2.4Ghz) and the second was just processor speed alone when checking what my rough max clock was.

I'm sure I could get better times with tweaking everything, but these were just quick boot up runs to see if it was stable enough to keep going at all.

Kei

nice 3.9 on air

Quote:

---

Originally Posted by roofsniper

nice 3.9 on air

---

Thank you, but please note that was running in dual core settings only (2 cores active) not in the quad core setting.

The highest I was able to go on all 4 cores was a shade over 3.8Ghz, but I didn't put in a huge amount of effort trying.

I was just posting that to show that Windows 7 does work very quickly as well as XP. :)

Kei

Quote:

---

Originally Posted by Kei

Thank you, but please note that was running in dual core settings only (2 cores active) not in the quad core setting.

The highest I was able to go on all 4 cores was a shade over 3.8Ghz, but I didn't put in a huge amount of effort trying.

I was just posting that to show that Windows 7 does work very quickly as well as XP. :)

Kei

---

I wouldn't go as far to say that, I can hammer those times with this board at same speeds in xp and this board has horrible PI performance....efficiency not to be confused with OS performance ;)

We wont even go there if I load sp1 XP....

I was just saying that Windows 7 performs very well is all, I'm not sure which one is faster because I only use W7 now. I haven't used XP in years I just wanted to show that the numbers in the first post for W7 won't be the same for everyone of course. :)

I do LOVE W7 though more than any OS I've used so far (i've been around...a lot) it's been excellent and I'll definitely buy the retail version when it hits. :up:

Kei

Yah I need to try it because I hate vista, but would like to step away from xp as a everyday OS, mostly use XP to bench lately.

Kei, first of, thanks for your contibution to this thread. Looking at your Performance Product a quick math shows a PP of 68875 which might be better than my run in W7 but it's still much worse than XP. That's why i advised to stay away from W7 if you want to get everything out of your system in SuperPi. ;)

Quote:

---

Originally Posted by chew*

Yah I need to try it because I hate vista, but would like to step away from xp as a everyday OS, mostly use XP to bench lately.

---

I used Vista for a long time until Windows 7 became available and I'm one of the ones that really liked Vista. Once you took the time to learn it and set it up the way you wanted it to run it was great for me. I liked Vista more than XP for overall use.

Windows 7 is about a billion miles better than Vista in my opinion from the couple months I've used it now. I really am extremely impressed with how good it is even in the Beta stages that, like I said I'm buying it as soon as it's released and my Beta runs out no hesitations.

Quote:

---

Originally Posted by Zeus

Kei, first of, thanks for your contibution to this thread. Looking at your Performance Product a quick math shows a PP of 68875 which might be better than my run in W7 but it's still much worse than XP. That's why i advised to stay away from W7 if you want to get everything out of your system in SuperPi. ;)

---

Thanks for the breakdown on those runs Zeus. I'm not sure if they were the best runs as like I said I just booted ran one took a shot and went back to the bios to keep clocking. I've never really dug deep into getting the best SuperPi time before though I think it's pretty interesting. :)

I do believe that XP could produce a faster time than W7 could in this benchmark especially since it's a single threaded benchmark. I'm not sure really how fast I could actually get a run out of W7 with max speeds though I know it could go faster than the 17.9 run I posted. No idea on what it could do though if everything was turned up along with the processor.

Thanks again, and very cool indepth thread on SuperPi efficiency. :up:

Kei

@ chew

realy nice

how much CPU-NB did you give?

Quote:

---

Originally Posted by techtrancer

@ chew

realy nice

how much CPU-NB did you give?

---

IIRC +.50 no clue what default is board just has + voltage not current, isn't default 1.35v?

i think @ default it is a litle bit higher than 1.0