PDA

View Full Version : ROG Team Super Pi Challenge Oct 18 - Nov 17 2012



Pages : [1] 2 3

HiVizMan
10-17-2012, 05:35 PM
To celebrate the formation of the new ROG Supreme OC HWBOT (http://hwbot.org/team/rog_supreme_oc/) team. I think it is only appropriate that we have a little benching challenge to get things really started. This challenge is only open to ROG Supreme OC members.

Not a member of ROG Supreme yet, join the team now. Register Here (http://hwbot.org/profile/register/)

Submissions open - 18 October 2012
Submissions close - 17 November 2012 12:00 GMT

And what better benchmark than Super Pi.

Download here (http://91.121.148.119/downloads/benchmarks/super_pi_mod.exe)

Rules here (http://hwbot.org/news/879_application_7_rules/)

Challange wallpaper downloaded here (http://www.mediafire.com/view/?p86fnmf9y9y2m54)

The Challenge - ROG Super Pi efficiency

Post your best 32 mil and 1 mil times.
Both submissions must be run at the same CPU frequency.
Memory frequency is capped at 2400 MHz. Not one MHz more. :)
Open to AMD and Intel users.
Both submissions are rated for efficiency and the efficiency rating for each benchmark combined.
The winner has the best efficiency total.

Requirements
Air and Water cooling only, for memory and CPU - No sub zero cooling allowed.
A photo of your benching rig must accompany your submission.
Contest is only open to ROG Supreme OC team members - your entry must include your HWBOT profile link. :)
Each submission must have the wallpaper that Chris designed for this challenge.
Each submission must comply with the HWBOT rules for submission.
Each submission must have the motherboard tab of CPUz open.


Prize

There is a CPU or kit of Ram up for grabs.

The CPU will be either a:

Intel® Core™ i7-990X Processor Extreme Edition (Brand New and never used)
or
Intel® Core™ i5-2550K Processor (binned 5.5GHz) not used for HWBOT submission will be certain silver cups in all 2D benchmarks and CPUz

I will let the winner decide which prize they need or want the most.

HiVizMan
10-17-2012, 05:37 PM
12802

Leader Board Intel

1 Alex-Ro 53896.10
2 Kr0nz3 54863.30
3 Chino 55538.11
4 henkenator68NL 55565.23
5 flexnl 55575.53
6 Squallalaha 56320.69
7 Arne Saknussemm 56977.70
8 Sinebrychof 57055.29
9 THE SE7EN SINS 57470.19
10 Nodens 57518.80
11 Dartmaul 57647.74
12 asipamn 59220.53
13 smokeintheeye 59276.83
14 RonanC 59797.54
15 Skull6 62000.00







https://encrypted-tbn2.gstatic.com/images?q=tbn:ANd9GcQUF-pch2BhEp_1cBJV1OGAGfj9f_0u5LK1POndumKvYLLdgEdD

Leader Board AMD



1 Myk SilentShadow 110390.78
2 Henkenator68NL 113693.55
3 chrsplmr 114071.80
4
5
6
7
8
9
10

Want an example of a submission clickHERE. (http://rog.asus.com/forum/showthread.php?24118-ROG-Team-Super-Pi-Challenge-Oct-18-Nov-17-2012&p=172659&viewfull=1#post172659)



Important Disclaimer. Neither ASUS, ROG or any affiliate of ASUS is in any way responsible for the prizes that I am giving away. The prizes are from my personal collection of hardware. They will all be in 100% working condition at the time of shipping. There is no warranty and no recourse to RMA from the original supplier in the future, and definitely no recourse to any ASUS representative if something happens to the prizes in the future. If you guys are not happy or unwilling to live with this condition I am instructed to remove the prizes, as ASUS can not and should not be held accountable for any support for and warranty off these prizes. By participating you agree to his condition. Please pm me if you do not and I will remove your submission, or the prizes if that is most suitable to all.

Arne Saknussemm
10-17-2012, 05:44 PM
Hehe excellent! Should be fun!:)

EDIT: Chrs HiVizMan a big thanks for the set-up very generous of both of you... time and effort!

Zka17
10-17-2012, 06:49 PM
This is very exciting! :cool: Can not wait to get home... :) It will be definitely a lot of fun! :)

Just to get it correctly, there are no CPU limitations? Do we have to submit the results to HWBOT before posting here?

That's a cool wallpaper, Chris - congrats for doing it!

HiVizMan
10-17-2012, 07:04 PM
This is very exciting! :cool: Can not wait to get home... :) It will be definitely a lot of fun! :)

Just to get it correctly, there are no CPU limitations? Do we have to submit the results to HWBOT before posting here?

That's a cool wallpaper, Chris - congrats for doing it!

No CPU limitations besides it has to be water cooled or air cooled. Oh and both submissions have to be at the same frequency.

Result does not have to be submitted to hwbot, it simply has to meet all the hwbot requirements for validity.

And yeah Chris is the man, big thanks to him for doing the wallpaper of this contest, very talented dude.

Trying to decide on the prize for this challenge. Was thinking of putting either a CPU up for grabs or a CPU LN2 pot, but not sure how the pot would get through customs. So might have to go with either CPU or a kit of benching ram. Lets see how it goes and maybe give the winner the choice to choose what they most want or need.

Henkenator68NL
10-17-2012, 08:48 PM
Looks good cant wait to try it out. First I have to get familiar with the test... Newby on this part. But exciting though! Thanks for the opportunity! Can I compete with 2 systems? AMD and Intel Right?

HiVizMan
10-17-2012, 09:52 PM
Yeah sure can, which ever is the more efficient of your submissions is the one that goes onto the leader board. Sadly can not spring for a prize for Intel and one for AMD. :p

_
10-18-2012, 08:04 AM
Oooh nice!! Gonna write this up :D

chrsplmr
10-18-2012, 11:52 AM
HiVizMan, this is awesome .. theArt is in the masterpieces all you guys paint with this equipment everyday.
I am humbled by your kind words and honored to offer this tiny piece. (It helps 'timestamp' the submission)

Best of Luck to all that Enter .. and those that just think about it as well.
Super Congrats on the new ROG Supreme OC team .. outstanding.

btw..
I read the announcement thread @ http://rog.asus.com/forum/showthread.php?23270-A-new-ROG-Team-on-HWBOT&highlight=supreme
I see the password is <republic> but ...
How does one join and where do you put the password <republic> ???

Salute to all involved.c.

Chino
10-18-2012, 02:48 PM
I'll start this off with my modest scores. lol

SuperPi score: 7sec 166ms (http://hwbot.org/submission/2321802_chino_superpi_core_i7_3770k_7sec_166ms)

12681

12682

SuperPi 32m score: 6min 39sec 244ms (http://hwbot.org/submission/2321803_chino_superpi_32m_core_i7_3770k_6min_39sec _244ms)

12683

http://hwbot.org/user/chino/

HiVizMan
10-18-2012, 03:47 PM
Well done to Chino our early leader. :)

Mate I hope you do not mind but I have used your images as the template for how a submission should look. Oh and I edited your post a bit. :p

Chino
10-18-2012, 04:36 PM
Well done to Chino our early leader. :)

Mate I hope you do not mind but I have used your images as the template for how a submission should look. Oh and I edited your post a bit. :p

Not at all. lol :p Your format definately looks better than what I posted. :D

I like the sound of that "early leader" haha.

Zka17
10-18-2012, 04:41 PM
Good job, Chino! Now at least we have a landmark where we're heading too... :cool:

Chino
10-18-2012, 04:47 PM
Curious about how HiVizMan calculated the efficiency score though. :D

HiVizMan
10-18-2012, 05:20 PM
Can't tell you all my secrets. :p

Speed of CPU vs time taken.
Then simply align the two benchmarks ratings using a fixed constant and add them together.

Henkenator68NL
10-18-2012, 05:26 PM
Ha Guys check this score ... almost exactly the same as Chino's 1M score. But mine is al on stock air coolers ;)

12687

http://hwbot.org/submission/2321823_henkenator68nl_superpi_core_i7_3770k_7sec_ 167ms?recalculate=true

Hope I can get Water Cooled before end of the contest ... LOL

Zka17
10-18-2012, 05:31 PM
Henk, can you do the SuperPi 32M too?

Henkenator68NL
10-18-2012, 05:34 PM
Yes I will do it tomorrow (promised my wife I would spend some time on her instead of my beloved computers LOL).

To be continued ...

Henkenator68NL
10-18-2012, 05:42 PM
LOL check the World leader List I am at position 263, Chino on 262. This is hilarious

Chino
10-18-2012, 05:43 PM
You have a nice Chip there, Henk. Mines wouldn't even let me get into Windows @ 1.6v. Much less being stable lol.

How are your temps with the stock fan?

HiVizMan
10-18-2012, 05:48 PM
Just a word to the wise. You have as much chance to win on air as you do with water. :)

Zka17
10-18-2012, 05:50 PM
LOL check the World leader List I am at position 263, Chino on 262. This is hilarious

Well, just keep the comparison among us... have you seen at what frequencies they're getting those low times? And what cooling?

Anyway, as I understood, here we're competing in efficiency scores, not timings... :)

HiVizMan
10-18-2012, 05:55 PM
Anyway, as I understood, here we're competing in efficiency scores, not times... :)

The key phrase. :)

Zka17
10-18-2012, 06:00 PM
Yeah, wanted to type "times"... - but I think because of the "efficiency score" this competition is really exciting! That's why the CPU doesn't really matter... actually getting lower SuperPi times with lower CPU frequency should be more efficient... - and that's already about fine tuning! That's how we can learn from this competition... :cool:

Very well thought, HiVizMan!

Chino
10-18-2012, 06:09 PM
lower SuperPi times with lower CPU frequency

Interesting... :eek:

Henkenator68NL
10-18-2012, 07:36 PM
@ Chino : I am not sure about the temp it was 67 degrees celcius i believe

@HiVizMan:
are you going to keep a "running" score board, so we can see who is in the lead? Might give me some more insight in the score you are going to calculate and hopefully can do something with that to try to improve..

Zka17
10-18-2012, 07:40 PM
I think, he does, Henke... the thing is that right now Chino's submission is the only valid one... - that's why I was curious about your 32M results...

Henkenator68NL
10-18-2012, 08:50 PM
Oké on it.tomorrow hope to get nice score.there. Later mates. I am glad I joined I like this!

Henkenator68NL
10-19-2012, 10:03 AM
Hey guys,

Couldn't help myself to give the SuperPi 1M another try .. I improved the score just a little bit !!!

To 7.161s !!

:D

12708

This scores some nice points for the Team !!

http://hwbot.org/submission/2321975_henkenator68nl_superpi_core_i7_3770k_7sec_ 161ms?recalculate=true

Later on I will have a go at the 32M

Tada

Zka17
10-19-2012, 10:13 AM
Good job, mate! Just don't forget that you will need to run the 32M with the same CPU frequency to have a valid submission for this competition...

Henkenator68NL
10-19-2012, 10:22 AM
Wow thats good that you mention that .. I did forget about the fact that it has to be the same cpu freq. !

Thanks mate!

HiVizMan
10-19-2012, 10:56 AM
I do have one request. Please guys I will be very disappointed if folks sand bag. If you are not sure what that is let google help you. I find the practice to be despicable.

The score board will be a live score board, and what the real aim of this thread is meant to do, is provide a platform for all ROG Supreme OC members to learn from each other how to bench 32 Mil. And I fully expect that the first score that Chino posted will be improved by him quite considerably before the end of the contest. As I expect all who participate to improve hugely during the contest submission period. Now the only way to do that is by submitting so we can have a base to work from. Submit and improve. :)

Henkenator68NL
10-19-2012, 11:19 AM
I do have one request. Please guys I will be very disappointed if folks sand bag. If you are not sure what that is let google help you. I find the practice to be despicable.

Well I googled ... but the meaning if that expression eludes me, ofcourse my English is not as good as yours, being you're native language.. Please explain

HiVizMan
10-19-2012, 12:10 PM
LOL - sorry about that. And you might be surprised about what some folks have as a first language. :)

There are many very sad people in the world. For them it is so very very important to win. Makes no difference what it is but they feel they have to win. So when there are on-line contests like this some folks do not want to put up their scores until the very last minute. They are scared that people will see what they get, and will then try harder to beat it. Those kinds of people think that if they show what they have done it will reduce the chance they have of winning.

What that does is make contests very boring, that is why this is a challenge. Not a contest. The challenge is to learn how to run Super PI and gain knowledge in the tweaking, set up and management of a 2D benching system. You compete against yourself as much as against anyone else.

Once we have some entries on the leader board, I will start putting out some tweaks and best practices for 2D benching. And we will see how peoples score will improve. It is as I always say about the journey and not the outcome.

Hope that helps.

Henkenator68NL
10-19-2012, 12:35 PM
Oke here are my first submits for the Challenge.

Super Pi 1M: 7.232s
12711
Super Pi 32M: 6m36.742s
12712

My Rigg:
12713

My HWBOT profile: http://hwbot.org/user/henkenator68nl/

I hope I have every point checked if not let me know !

Greetings
Henk

HiVizMan
10-19-2012, 12:45 PM
Thank you. Will update the leader board in the next two minutes :)

Updated and you are the new leader my friend.


It would be nice if you listed your hardware, make and model and stuff like that. Have a look at the second post of the thread for an example. But it is not the end of the world if you don't want to. :)

Zka17
10-19-2012, 12:47 PM
Here is my first attempt:

Hardware:
CPU - Intel Core i7-3930k @ 5000MHz
Mobo - Asus Rampage IV Extreme
Cooling - Water cooled: Swiftech Apogee HD, XSPC RX360, 6xCougar push-pull
Memory - Corsair Dominator Platinum @ 2133MHz

Picture of the system:

12714

SuperPi 1M score: 7sec 737ms

12715

SuperPi 32M score: 7min 43sec 196msec

12716

Note: This is just my first attempt... like a starting point... - I'm about to start tweaking the memories! :cool:

HWBOT profile: http://hwbot.org/user/zka17/

HiVizMan
10-19-2012, 01:03 PM
Thank you Leader board updated.

Arne Saknussemm
10-19-2012, 06:03 PM
Not teabagging here just no time until now and contemplation of meaning of efficiency.....

http://hwbot.org/user/arne_saknussemm/

i73930K @ 4600MHz
Rampage IV Extreme
Cooling H2O Custom loop
RAM Gskill Z series 1600MHz

Pi1M
12727
Pi32M
12728
12729

HiVizMan
10-19-2012, 06:22 PM
I truly hope no-one searches google images for teabagging. :D

Updated and Arne you make top 3 mate.

Henkenator68NL
10-19-2012, 06:32 PM
My system:
Asus Maximus V Extreme, Intel 3770K, G.Skill RipjawX 2 x 4 Gb CL9 PC17000, Asus GTX 660 Direct CU II OC, CM Silent Pro M2 1000W, Case: CM Storm Stryker

Zka17
10-19-2012, 06:40 PM
Good job, Arne! And you already have the two GTX680s in there... :cool: I guess, soon we will see some nice scores with those too... :)

Arne Saknussemm
10-19-2012, 06:46 PM
I truly hope no-one searches google images for teabagging. :D

Yes, Henk don't do it!

Hmmm....have to think some more then! lol

Yes! The second 680 is here and OCs a hair better than the other card so I will work on it.....but I think watercooling is necessary....for real scores......project in progress

HiVizMan
10-19-2012, 06:59 PM
In that case next month we should have a little 3 D challenge here on ROG hmm

Zka17
10-19-2012, 07:47 PM
In that case next month we should have a little 3 D challenge here on ROG hmm

Oh, that would be wonderful! :cool: But let's finish this one first...

Arne Saknussemm
10-19-2012, 08:07 PM
In that case next month we should have a little 3 D challenge here on ROG hmm

HaHaHa! :D

http://hwbot.org/user/arne_saknussemm/

i73930K @ 4830MHz
Rampage IV Extreme
Cooling H2O Custom loop
RAM Gskill Z series 1680MHz

Pi1M
12733
Pi32M
12734
12735

HiVizMan
10-19-2012, 08:29 PM
Arne cheers mate, there is an efficiency gain :) Not enough to get you into second place but gains were made.

Just a quick update, I used Henks individual submission super Pi time for the calculation. Not seeing he had run the benchmarks at different frequency for his later 32 mil and 1 mil submission, a minor adjustment was needed of course, what it does is put Chino back in top spot. For now... :)

Arne Saknussemm
10-20-2012, 10:25 AM
Cheers HiVizMan! This has me intrigued...thought voltage might play a part in the formula

Anyhow this comp has taught me a few things alright!!! Got 4.8Ghz out of 1.344v which makes me think I have a pretty good chip and RAM overclocking, which I have never tried, I get 1904 out of a 1600 MHz kit!! But sub timings are Greek to me at the moment so probably a dirty overclock and may not have helped....more research needed.

Does it help efficiency....times are lower but...:confused:;)


http://hwbot.org/user/arne_saknussemm/

i73930K @ 5100MHz
Rampage IV Extreme
Cooling H2O Custom loop
RAM Gskill Z series 1600MHz @ 1904MHz

Pi1M
12755
Pi32M
12756
12757

HiVizMan
10-20-2012, 11:00 AM
Nope way worse from an efficiency point of view.

Speed of CPU is not the determining factor here. Things like the number of processes that are active and using resources, the memory timings are key. Again the highest frequency on your memory might not be the most efficient, it is the relationship between frequency and latency that counts.

Arne Saknussemm
10-20-2012, 11:37 AM
Yeah suspected as much.....tweaking to continue:D

HiVizMan
10-20-2012, 12:44 PM
Test your tweaking skills at 4.4GHz. That is a nice place to work from. :)

Nodens
10-21-2012, 02:15 AM
Ok here goes my first submission so let's see how it scores:P

HWBOT: http://hwbot.org/user/nodens_rog/

Hardware:
Intel 3930K C2 @ 4576Mhz
RIVE
Corsair H100
Corsair Vengeance @ 1664Mhz

1M Score: 8sec 250ms

12776

32M Score: 7m 35s 184ms

12777

Rig photo (yes I know the photo is crap but I currently don't have my DSLR here:))

12778

_
10-21-2012, 06:51 AM
Now front page article!: http://rog.asus.com/170612012/overclocking/rog-team-super-pi-challenge-oct-18-nov-17-2012/

HiVis: Can we separate air cooling and LLC/watercooling leagues? Make it a bit easier for the air cooled guys (me!) :)

I've gotta drop in some 2133 mem and push my 2500K now. Need to wait until Tuesday when winter comes, cause it's still 30C here. Pi is single threaded right? Will help us i5 guys ;)

I still have a BF3 code knocking about for the winner!

THE SE7EN SINS
10-21-2012, 07:57 AM
Specs:
CPU: Intel 3770k @4.7Ghz
RAM: G.Skill Trident X 16Gbz (2x 8Gbs) @2400Mhz
Mainboard: Asus Maximus V Extreme
127801278212781
Also I know the picture sucks... but I invested in a computer not camera. lol

8 Pack
10-21-2012, 08:23 AM
Great idea this HivizMan watching this thread with interest.

HiVizMan
10-21-2012, 10:44 AM
Now front page article!: http://rog.asus.com/170612012/overclocking/rog-team-super-pi-challenge-oct-18-nov-17-2012/

HiVis: Can we separate air cooling and LLC/watercooling leagues? Make it a bit easier for the air cooled guys (me!) :)

I've gotta drop in some 2133 mem and push my 2500K now. Need to wait until Tuesday when winter comes, cause it's still 30C here. Pi is single threaded right? Will help us i5 guys ;)

I still have a BF3 code knocking about for the winner!


Rich it makes no difference. The air cooling guys are not in any way shape or form disadvantaged. You can win this challenge at stock CPU speed. It has nothing to do with frequency of the CPU. :D It has all to do with the tweaking and set up with 32 MIL

There should be a separate table and challenge for the AMD guys. And that is a sad thing to have to say as AMD was my first love. But there you have it.

HiVizMan
10-21-2012, 11:11 AM
Specs:
CPU: Intel 3770k @4.7Ghz
RAM: G.Skill Trident X 16Gbz (2x 8Gbs) @2400Mhz
Mainboard: Asus Maximus V Extreme
127801278212781
Also I know the picture sucks... but I invested in a computer not camera. lol


You did do good, but you do not have a link to your ROG supreme OC HWBOT profile :)

57871.88 is your efficiency rating. Next time please try and get your CPU frequency to show in both benchmarks, the one has the down-clocked frequency.

Post your HWBOT profile in the submission page and get onto the leader board. :p

HiVizMan
10-21-2012, 11:11 AM
Depending on the number of folks who participate I will have a CPU or Ram up for grabs.

The CPU will be either a:

Intel® Core™ i7-990X Processor Extreme Edition (Brand New and never used)
or
Intel® Core™ i5-2550K Processor (binned 5.5GHz) not used for HWBOT submission will be certain silver cups in all 2D benchmarks and CPUz

HiVizMan
10-21-2012, 11:25 AM
Rich I have put up some prizes for the challenge. I will supply the first place prize a CPU or kit of ram up to the winner to decide what they want. If you have some goodies like lanyards or BF3 stuff then great we can do them as runner up stuff.

Have updated the front page.

chrsplmr
10-21-2012, 03:20 PM
Well .. This is simply fantastic. Although I disqualify myself because I 'supplied' the wallpaper .. (and being an AMDguy .. OCnoob) .. I
could not resist the fun, community and a chance to learn.
I have no clue as to what these numbers mean. (a little explaination, if someone has the time, would be nice.) I was a little suprised
at the difference in numbers of the submissions and my own .. and am very please to find I have so much to learn.
(As I see that 'Team' averages count on HWbot .. and not wanting to bring the 'team' down in the future I will withdraw from
the team .. after a 'we' bit more fun .. Learning is afterAll FunDaMental.)
Below is the result of the first effort .. ideas and suggestions for improving this are More than Welcome. Thank You All.c.

Specs:
cpu: AMD 1100T @ 3950.9
ram: Corsair Dominator GT 1866 X 8gb
mainboard: CrossHair V Formula bios 506

12794

12792
12793

Chino
10-21-2012, 04:18 PM
Just to keep this challenge competitive. :)

Hardware
CPU - Intel i7 3770K @ 4400.2 MHz
Mobo - Asus Maximus V Extreme
Cooling - Antec KUHLER H2O 920
Memory - Corsair Dominator Platinum @ 1866 MHz

SuperPi score: 8sec 369ms
12796

SuperPi 32m score: 7min 37sec 242ms
12795


Chino @ HWBot (http://hwbot.org/user/chino/)

Zka17
10-21-2012, 04:26 PM
Chris, welcome in the benching world!

I don't think that having an AMD system would be negative in any way! In contrary, I really think that it's making both this challenge and our team more colorful. Regarding the scores/points, no worries there! Every point counts in a positive way, there is no such thing as "to bring the 'team' down"... If you fall in love with benching, you're very welcome to join and stay with us! :)

Unfortunately, I'm not experienced with the AMD system... can not be of too much help... but I'm sure HiVizMan has some very nice inputs on that! Maybe, others with AMD systems will join too...

Thank you, my friend! And please... enjoy!

HiVizMan
10-21-2012, 05:06 PM
Well .. This is simply fantastic. Although I disqualify myself because I 'supplied' the wallpaper .. (and being an AMDguy .. OCnoob) .. I
could not resist the fun, community and a chance to learn.
I have no clue as to what these numbers mean. (a little explaination, if someone has the time, would be nice.) I was a little suprised
at the difference in numbers of the submissions and my own .. and am very please to find I have so much to learn.
(As I see that 'Team' averages count on HWbot .. and not wanting to bring the 'team' down in the future I will withdraw from
the team .. after a 'we' bit more fun .. Learning is afterAll FunDaMental.)
Below is the result of the first effort .. ideas and suggestions for improving this are More than Welcome. Thank You All.c.

Specs:
cpu: AMD 1100T @ 3950.9
ram: Corsair Dominator GT 1866 X 8gb
mainboard: CrossHair V Formula bios 506

12794

12792
12793

Chris Team ROG does not have that many AMD folks. Now there are two ways to get team points, and as you are pretty much the only AMD guy you will be bringing in many team hardware points for the guys. So stay part of the team and have fun with the guys. I really do not want the quest for HWBOT points to get so all consuming, the whole thing of benching is about having fun. The fun is the point, not the points. :D

HiVizMan
10-21-2012, 05:07 PM
Chris, welcome in the benching world!

I don't think that having an AMD system would be negative in any way! In contrary, I really think that it's making both this challenge and our team more colorful. Regarding the scores/points, no worries there! Every point counts in a positive way, there is no such thing as "to bring the 'team' down"... If you fall in love with benching, you're very welcome to join and stay with us! :)

Unfortunately, I'm not experienced with the AMD system... can not be of too much help... but I'm sure HiVizMan has some very nice inputs on that! Maybe, others with AMD systems will join too...

Thank you, my friend! And please... enjoy!

I would be very keen for some of the AMD crew to join in. That way we have a AMD and a Intel Leader board. :)

kkn
10-21-2012, 05:14 PM
ive converted mine to a server whit stock cooler :s
and a old 8800 ultra card whit the corsair ram

Chino
10-21-2012, 05:16 PM
I would be very keen for some of the AMD crew to join in. That way we have a AMD and a Intel Leader board. :)

Submitted some benchmarks from my parent's AMD HTPC today actually. lol :p

HiVizMan
10-21-2012, 05:16 PM
Chino you have upped your game mate. 56944.84

Zka17
10-21-2012, 05:32 PM
Chino, you're doing very good, mate! Both here and there... :cool:

HiVizMan, I would really love to have some AMD guys in our team!

Where are you, guys? When it's about arguing AMD vs Intel, there are so many comments... and now? Please, join us - no fights here, just fun! :)

Zka17
10-21-2012, 05:57 PM
OK, now I'm back in business! :) Have done some changes on my system, but more details about it in my TestBench thread...

Here is my first run with the new system:

Hardware:
CPU - Intel Core i7-2700k @ 5000MHz
Mobo - Asus Maximus V Gene
Cooling - Water cooled: EK Supreme HF, XSPC RX360, Swiftech MCP655
Memory - Corsair Dominator GT @ 2133MHz

Picture of the system:

12798

SuperPi 1M score: 7sec 582msec

12799

SuperPi 32M: 6min 54sec 244msec

12800

HWBOT link: http://hwbot.org/user/zka17/

HiVizMan
10-21-2012, 06:14 PM
Zoltan new efficiency is 58622.20 you have gone ahead of Arne but still to break into the Top 3

Chino clear leader with some daylight between the pack.

AND

I have started an AMD Leader Board.

Chris you are the clear leader at present. :)

Zka17
10-21-2012, 06:17 PM
Thanks! No worries, this was just a test run... while was leak-testing the loop... :cool: Soon will be back with some better results... at least, I hope so...

HiVizMan
10-21-2012, 06:29 PM
Just want to repeat that there is a AMD Leader Board now so we have two separate challenges running here.

Chris is the early leader of the pack. :D

Zka17
10-21-2012, 06:32 PM
Excellent! I think, I will send some PMs around... :)

chrsplmr
10-21-2012, 07:23 PM
Well .. this is a pleasant suprise. Leading by default. hahahhaa ... is still leading. For the moment. I shall enoy the one at a time.

As to AMD/vs/Intel. ... hahhahaha I hug mine every night. .. she never refuses me or disappoints.
Let the fun ROGon ... kool.c.

HiVizMan
10-21-2012, 07:35 PM
Thing is Chris, AMD and Intel have followed two different design trajectories. So Intel have focused on and is very strong on single thread applications, and Super Pi is a single threaded application. Whereas AND have followed a very different path. For example I love my 8150 for my media rendering. And as a 24/7 work station you can not beat the Llano CPU. Well maybe the next gen will be even better. :)

So the efficiency ratings are very far apart and it is for that reason that we needed to have a separate leader board.

THE SE7EN SINS
10-21-2012, 07:54 PM
Specs:
CPU: Intel 3770k @4.7Ghz & @4.8Ghz
RAM: G.Skill Trident X 16Gbz (2x 8Gbs) @2400Mhz
Mainboard: Asus Maximus V Extreme
1M @ 4.8Ghz
12803
32M @ 4.7Ghz *4.8Ghz causes bluescreen around loop 12/13 :c
12804
Hyperion (crap picture, I know)
12805

HWBOT Profile: http://hwbot.org/user/the_se7en_sins/

chrsplmr
10-21-2012, 07:58 PM
Well then I am really pleased goodSir.
I would suggest the hesitation that we see is probably the same as my first considerations,
the comparison on single issues to 'prove' or 'disprove' .... "Better/Best" .. which like all else
in life can not be weighed by single or grouped 'presumed' tests but in SumTotalities.
No matter the sport or group endeavor the victorious are Never of one single strength
but united.
Like I said, I am really pleased. ROG.

Thanks 7 .. I don't mind if I do. Any suggestion for improvement?
What would be the first 3 things you would do if you were me ? Anybody?
(and..I don't want to hear noth'n bout .. 'get an Intel'..hehe..cuz..well, no.haha).c.

Chino
10-21-2012, 08:04 PM
Chino, you're doing very good, mate! Both here and there... :cool:

HiVizMan, I would really love to have some AMD guys in our team!

Where are you, guys? When it's about arguing AMD vs Intel, there are so many comments... and now? Please, join us - no fights here, just fun! :)

Thanks, mate! :D I've finished my benchmark runs lol. I've ran out of hardware to bench :(. Such a drag. Currently 2nd in the National League, just 2 points shy of being in first. :(

HiVizMan
10-21-2012, 08:17 PM
The Se7en Sins - both submissions have to be run at the same speed. Sorry mate, but the upside of this challenge is that I fully expect the winner to have their CPU at the 4.4GHz range.

Could I please ask that you have another go with both runs completed at the same CPU frequency.

But to give you an idea your efficiency was 57387.20 - however both runs need to be at the same frequency.

RonanC
10-21-2012, 08:39 PM
here's mine :)

Zka17
10-21-2012, 09:05 PM
I've finished my benchmark runs (

There is no such thing! He-he... once you got the bug, you got it... - sit back a little bit, take a deep breath and think about what you've learned so far... find your weak points and start over! :cool:

Nodens
10-21-2012, 09:37 PM
Ok let's see if I can improve.

HWBOT: http://hwbot.org/user/nodens_rog/

Hardware:
Intel 3930K C2 @ 4240Mhz
RIVE
Corsair H100
Corsair Vengeance @ 1979Mhz

1M: 8s 906ms

12808

32M: 8m 1s 899ms

12809

HiVizMan
10-21-2012, 10:24 PM
Oh yes indeed sir. 58193.96

To those who have submitted a score and have not yet received a copy of the 32 mil notes. Please check your in boxes or send me a pm.

HiVizMan
10-21-2012, 10:28 PM
here's mine :)

Mate could I ask you to have a look see at the way the other guys are presenting their submissions and ask you to use the same format. But that is looking pretty good by the way.

RonanC
10-21-2012, 10:57 PM
no problem m8 ;)

RonanC
10-21-2012, 11:18 PM
http://www.hwbot.org/user/ronanc_irl/

Hardware
CPU - Intel i7 3960 ES @ 4800MHz
Mobo - Asus Rampage IV Extreme
Cooling - Custom WC
Memory - Corsair Vengence @ 1600 MHz

SuperPi score 1m: 7.865

SuperPi 32M : 7m 39.334

Myk SilentShadow
10-22-2012, 01:29 AM
Ooook, decided to see how badly I would do lol

1MPass
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/1MPiPass.jpg

32MPass
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/32MPiPass.jpg

Shot of Rig
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/MykSilentShadowSystem.jpg

www.hwbot.com/user/myksilentshadow

Good luck C, there's no chance I can take you on in this Bro lol :cool:

Zka17
10-22-2012, 01:37 AM
It's not bad at all! As a second subscriber with an AMD system, you're automatically on the second place! :)

Myk SilentShadow
10-22-2012, 01:48 AM
It's not bad at all! As a second subscriber with an AMD system, you're automatically on the second place! :)

And miles behind lmao

Zka17
10-22-2012, 01:53 AM
Come on, Myk! That's just the basic run... as for Chris... - now, both of you start tweaking! Please, pay attention to HiVizMan's post #83 in this thread... :)

Myk SilentShadow
10-22-2012, 02:03 AM
Come on, Myk! That's just the basic run... as for Chris... - now, both of you start tweaking! Please, pay attention to HiVizMan's post #83 in this thread... :)

basic run? lol I just clocked my CPU to 4.6 before I ran the tests as for the "tweaking" I got no clue what i'm doing...I used Turbo V Evo again to OC my CPU lmao :cool: :p

Zka17
10-22-2012, 02:10 AM
As I said before, I'm not expert in AMD tweaking... but what I've got about SuperPI is that the memory tweaking is more important... - at least, in this challenge it certainly is! (remember, here we're chasing efficiency, not low times)

So, if I would be you, I would let the CPU at 4.6GHz for now and would start playing with the memories... just checked the Crosshair mobo section, Raja has an OC guide for that too... focus on the memory section...

Chino
10-22-2012, 03:09 AM
There is no such thing! He-he... once you got the bug, you got it... - sit back a little bit, take a deep breath and think about what you've learned so far... find your weak points and start over! :cool:

Or send me your old hardware you no longer use and I'll bench them rofl :p

Zka17
10-22-2012, 03:14 AM
Ha-ha... :D Not saying no... but those things I'm not using right now are really old... - I'm following the advice I gave you... after a certain time I'm starting over with every hardware... and usually there is place for better... :)

THE SE7EN SINS
10-22-2012, 03:20 AM
The Se7en Sins - both submissions have to be run at the same speed. Sorry mate, but the upside of this challenge is that I fully expect the winner to have their CPU at the 4.4GHz range.

Could I please ask that you have another go with both runs completed at the same CPU frequency.

But to give you an idea your efficiency was 57387.20 - however both runs need to be at the same frequency.

Thanks for the tips, I'm kinda new at this kinda thing .-.
I am learning a lot, I've made the changes.

Specs:
CPU: Intel 3770k @4.7Ghz
RAM: G.Skill Trident X 16Gbz (2x 8Gbs) @2400Mhz
Mainboard: Asus Maximus V Extreme
1M
12825
32M
12826
Hyperion (best camera ever am I right?)
12827
HWBOT Profile: http://hwbot.org/user/the_se7en_sins/

Zka17
10-22-2012, 04:00 AM
I start tweaking the memories...

Hardware:
CPU - Intel Core i7-2700k @ 5000MHz
Mobo - Asus Maximus V Gene
Cooling - Water cooled: EK Supreme HF, XSPC RX360, Swiftech MCP655
Memory - Corsair Dominator GT @ 2133MHz

System:

12828

SuperPi 1M: 7sec 520msec

12829

SuperPi 32M: 6min 52sec 886msec

12830

HWBOT link: http://hwbot.org/user/zka17/

speed
10-22-2012, 04:03 AM
Thing is Chris, AMD and Intel have followed two different design trajectories. So Intel have focused on and is very strong on single thread applications, and Super Pi is a single threaded application. Whereas AND have followed a very different path. For example I love my 8150 for my media rendering. And as a 24/7 work station you can not beat the Llano CPU. Well maybe the next gen will be even better. :)

So the efficiency ratings are very far apart and it is for that reason that we needed to have a separate leader board.

That is why I don't bench. It is all designed around and for intel.

Zka17
10-22-2012, 04:13 AM
Come on, speed! There are two separate challenges here... one for AMD and one for Intel... with an AMD system your results would be compared only with other AMDs...

This is more about learning and having fun... - you won't experience any of these if you won't try... :)

You have a great experience with AMD-based systems, why you're just not giving a shot?

_
10-22-2012, 06:30 AM
I'm nipping that in the bud right now. If you want to join, join, otherwise don't. Simple as that.

HiVizMan
10-22-2012, 08:50 AM
This thread is only for our ROG forum guys, and I am even thinking about removing the need to belong to HWBOT at all as a prerequisite. Want ROG members to just have a bit of fun. So send me your views about that folks.

However please keep this thread only about the Super Pi challenge. I will be cleaning up the thread later today and all posts not about the challenge will be removed in the interest of keeping this thread tidy just how I like it. :)

Leader board has been updated and I can say that it is hotting up at the top. Some movement in the top 3 I am pleased to say.

Also I have decided I needed to clarify a point about the prize. Namely they are not ASUS endorsed, sanctioned or sourced from. Therefore ASUS has nothing to do with the support or warranty of the prize. In short there is none. There is no transferable warranty for any of kit listed as a possible prize, selectable by the winner of this challenge, as all Intel items are OEM. If you are doing this for the prize then you should not be in this thread to be blunt. :)

Arne Saknussemm
10-22-2012, 10:56 AM
Lol....well said..... yeah sure...the more the merrier...I think we should particularly encourage entries from anyone who is going to score less than myself.....and yes please, tidy away!

chrsplmr
10-22-2012, 11:40 AM
This is fun and frustraiting at the same time.

Question:
I tried to submit SuperPi32 .. to 'get on the board' .. sortof speak. (sorry for the english .. hahaha)
and in the drop down box there is no 6X for the 1100T. Am I missing something ?? or do'n it wrong ??
This is a pic of it ..
Should I be turning off 2 or more cores for this ??? Noobs !!!! hahahhaha

12834

On a happy note .. I went looking to see what to shoot past on air with this cpu .. and was super
pleasantly suprised to find our goodFriend Flank3r on the HardOCzech Team just tearing it up with some
outrageous speeds on some AMD's .. cudos Flank .. When your done practicing and fooling around how
bout come on over .. with all due respect to HardOCzech Team.

EDIT *****
Duh ... NooBs .. thnxViz.c.

HiVizMan
10-22-2012, 11:49 AM
That is number of CPU's (sockets) present on a single motherboard.

Lets keep ROG for the family, eh.

HiVizMan
10-22-2012, 12:44 PM
I just want to show folks that CPU speed is not what it is all about.

CPU @ 4GHz
Ram @ 2200MHz

Efficiency 56076.52

Super Pi 1 Mil - 9.031
12842

Super Pi 32 Mil - 8,18.813

12841

I fully expect folks to break the 55500 barrier in this contest. :)

Just some food for thought.


Edit: Added image of rig

12843

Please note that I have not even fastened the cooler down onto the CPU and motherboard. And that the cooler is without a fan attached. Point being it is not about how high or how far you over-clock your system, but how best you manage your system.

Arne Saknussemm
10-22-2012, 05:50 PM
http://hwbot.org/user/arne_saknussemm/

i73930K @ 4400MHz
Rampage IV Extreme
Cooling H2O Custom loop
RAM Gskill Z series 1600MHz

Pi1M
12864
Pi32M
12865
12866

HiVizMan
10-22-2012, 06:07 PM
Arne 59177.50

Arne Saknussemm
10-22-2012, 06:17 PM
:eek: oops headed the wrong way lol ;)

Zka17
10-22-2012, 06:44 PM
I've been playing around with the memory's primary timing... Also reduced the CPU speed... - so, here is an another shot:

Hardware:
CPU - Intel Core i7-2700k @ 4200.1MHz
Mobo - Asus Maximus V Gene
Cooling - Water cooled: EK Supreme HF, XSPC RX360, Swiftech MCP655
Memory - Corsair Dominator GT @ 1866MHz

System:

12867

SuperPi 1M: 8sec 643msec

12868

SuperPi 32M: 7min 45sec 989msec

12869

HWBOT link: http://hwbot.org/user/zka17/

HiVizMan
10-22-2012, 06:49 PM
Now we are talking. 55873.47

The top 3 on the Leader Board Intel has been rearranged. :)

Zka17
10-22-2012, 07:04 PM
Ooops, I headed in the right direction... :)

Sincerely, the system is not entirely stable yet... passes the SuperPis, but it still gives some errors sometimes... in AIDA64 stops the stress test saying "Hardware failure" once in 3 runs... - maybe, some voltages needs to be adjusted...

Edit: I would never thought that will say this... but I really wish to give me a BSOD with a code...

I also tried prime95 with "blend" settings and sometimes one of the workers fails/shows an error... but no BSOD... and it doesn't fail after each restart (30 min runs) and doesn't fails always on the same core...

How to guess where is the trouble? VCCIO or VDIMM? Vcore or VCCSA? or any mix of these... - it will take a lot of time to experiment each variant... :( But, no hurry... soon or later I will find it! :cool:

Raja@ASUS
10-22-2012, 07:21 PM
32M efficiency comes from DRAM and the frequency of the memory controller. Memory controller frequency is usually 1:1 DDR3. So use highest memory ratio possible and make sure you don't need to slacken 1-2 clocks on any primaries or third timings in order to run a higher ratio. EG running DDR3-2133 at CAS 7 with tight third timings, can be faster than running DDR3-2400 with CAS 10 and slack third timings.

If the efficiency is calculated properly, CPU frequency is only a factor in terms of stability when looking at clock-for-clock efficiency in this benchmark. You could effectively leave the CPU core frequency at stock if you wish. Depending on how good your CPU memory controller is, it will likely have an easier time running higher memory frequencies at lower CPU frequency (possibly toy with various multipliers once maxxed out). Just be sure to turn off any power saving features that don't help the benchmark. Also disable Execture Disable Bit in EFI. After that lot you can get into Maxmem and Waza (if you want to get serious about things).

Zka17
10-22-2012, 07:30 PM
Maxmem?! Is there a correlation between the Maxmem score and SuperPi efficiency?

HiVizMan
10-22-2012, 07:34 PM
Raja is not talking about the benching application called Maxmem. Specifically an undocumented /maxmem switch is mentioned on page 160 of Volume 3 of the Microsoft Windows NT Resource Kit. This switch enables you to specify the maximum amount of random access memory (RAM) memory that Windows NT may use.

There you have it. :D

Raja@ASUS
10-22-2012, 07:35 PM
We are talking about two different things.


This is the "maxmem" I am refering to: http://support.microsoft.com/kb/108393


This will limit the memory available to the OS. You need to limit OS memory to a manageable size for Waza. Google 32M and Waza for some background info. Most of the theory is out there and accessible. Anything you get confused on just ask one of us.


Before you get to that, work on your memory timings (use the UEFI guides to work out minimum possible timings). EG minimum tFAW is 16 at tRRD 4 etc. Then you can start getting advanced. Those of you using X79, use three channels only - not four (for 32 bit XP).

Zka17
10-22-2012, 07:38 PM
Oh, I got it now! And also got a BSOD... the 0x3B one... - hah, that's supposed to be the Vcore! :D

Raja@ASUS
10-22-2012, 07:47 PM
FWIW Waza and tight memory timings can shave 5-10 seconds off an unoptimized run. Just depends how much time you want to spend on 32M to shave time...hehe

HiVizMan
10-22-2012, 07:53 PM
FWIW Waza and tight memory timings can shave 5-10 seconds off an unoptimized run. Just depends how much time you want to spend on 32M to shave time...hehe

Not that we have ever spent any time working on 32 Mil now have we... :D :p


Raja thanks for pitching in here. Most of the guys are first time 32 mil bencher's, and we will be having a more detailed exploration of the benchmark in the next couple of weeks. They have been promised a clinic but for now it is just great to see folks having fun with their hardware and running 32 mil. And if you are willing to share your considerable knowledge about DDR3 tuning I would be very very grateful indeed.

ZKA17 here are the next set of results I did. All that is different is one tweak. OS based, and it shaved 8 seconds off my base line time.

Efficiency here is now down to 55744.52

12872

12873

Arne Saknussemm
10-22-2012, 07:57 PM
I fail to see how this can help 32M runs....


http://youtu.be/UDTZCgsZGeA

HiVizMan....you'll have to tidy this post away too:o

Myk SilentShadow
10-22-2012, 08:01 PM
Ok so all this "efficiency" stuff is RAM based correct? ok...so what the heck do I have to do for better efficiency? I just read all that stuff Raja posted at the top of this page and i'm completely confused lol.

G.Skill 1600MHZ running @ 1766 currently.
DRAM running on stock Voltage of 1.5V
FSB: DRAM 1:4
CAS# Latency 9
RAS# to CAS# Delay 9
RAS# Precharge 9
Cycle Time 24
Bank Cycle Time 39 clocks
NB Frequency 2439 MHZ

I've tried reading through the OCing thread Raja posted in the CHVF Sub Forum...but yeah, more confusion lol

Zka17
10-22-2012, 08:02 PM
I didn't got yet to the point to start playing with the OS... first I would like stability... :)

HiVizMan
10-22-2012, 08:03 PM
Now you play that vid three times from you D drive to your C drive and your 32mil is 10 seconds faster. :D :D

Arne Saknussemm
10-22-2012, 08:04 PM
hahaha exactly!:D

HiVizMan
10-22-2012, 08:07 PM
Works better in HD mind you.

HiVizMan
10-22-2012, 08:07 PM
Ok so all this "efficiency" stuff is RAM based correct? ok...so what the heck do I have to do for better efficiency? I just read all that stuff Raja posted at the top of this page and i'm completely confused lol.


Yeah welcome to the world or Raja. You get used to it after a couple of years. :)

Efficiency in Super Pi32 is the synthesis of your OS, your CPU and your RAM. Each one needs to be just perfect for the Super Pi efficiency to be good.

And the only instability you will have in this challenge (well besides mental instability I am starting to realise based on the direction this thread is going) is going to be your ram. CPU can be anything you like. Stock is good too for this challenge.

Myk for you I would work on my ram frequency and play both up and down the frequency scale and see what give you better 32 Mil times you might be surprised at the out comes. Just keep your CPU at one frequency.

Myk SilentShadow
10-22-2012, 08:15 PM
Yeah welcome to the world or Raja. You get used to it after a couple of years. :)

Efficiency in Super Pi32 is the synthesis of your OS, your CPU and your RAM. Each one needs to be just perfect for the Super Pi efficiency to be good.

And the only instability you will have in this challenge (well besides mental instability I am starting to realise based on the direction this thread is going) is going to be your ram. CPU can be anything you like. Stock is good too for this challenge.

Myk for you I would work on my ram frequency and play both up and down the frequency scale and see what give you better 32 Mil times you might be surprised at the out comes. Just keep your CPU at one frequency.

Thanks mate, but what should I keep in mind...i'm guessing I can't just choose a Frequency and expect it to work, those timings I posted are the stock ones for my RAM...so should I slacken them for a better frequency? or go the opposite? lol can't get more n00b than me right now :p

HiVizMan
10-22-2012, 08:19 PM
We were all n00bs at one point or other, some folks just have a hard time remembering that. See what you can do to increase your DDR3 frequency at the current timings. Do not damage your ram, but most of the time it requires a little more voltage either to the memory or to the controller.

Once you feel you have reached your peak frequency then you start to play with the timings. But for now leave your primary timings as is.

Myk SilentShadow
10-22-2012, 09:31 PM
Ok mate thanks for the tips, i'm gonna leave my current CPU clock as is..it's only a 400MHZ boost but it will have to do for now, though i'll mess with my RAM frequency a bit later as i'm supposed to be studying right now lol. i'll mess later and post if I can get better scores :D

speed
10-22-2012, 09:54 PM
I'm nipping that in the bud right now. If you want to join, join, otherwise don't. Simple as that. That's cool I understand.

asipamn
10-23-2012, 12:36 AM
I dont own any rog only normal asus motherboard. can i still enter?

Nodens
10-23-2012, 01:02 AM
One more. (I should get extra points for using Vengeance RAM with X79 :p)

HWBOT: http://hwbot.org/user/nodens_rog/

Hardware:
Intel 3930K C2 @ 4133Mhz
RIVE
Corsair H100
Corsair Vengeance @ 1978Mhz

1M: 9s 78ms

12875

32M: 8m 12s 791ms

12876

Luumi
10-23-2012, 02:45 AM
I had to try myself aswell :).

3930K C2 @ 4759
RIVE
32gb G.Skill TridentX @ 2400
I need a better kit for this bench and a clean OS :D

Superpi 1M
12877

Superpi 32M

12878

I have to say the pic of my rig is kinda crappy since it was taken just before a leaking test but it should be ok enough :).

12880


And my HWBOT profile is: http://hwbot.org/user/luumi/

Raja@ASUS
10-23-2012, 06:29 AM
I didn't got yet to the point to start playing with the OS... first I would like stability... :)

You will have to find stability again after you change the OS and perform Waza. If you're game, might as well do it all at once. FWIW for you guys, I'd recommend not stripping the OS. Just use XP SP3. Good ole Fredyama (32M legend) uses a basic install with GPU drivers and it works well enough for him. The stripping stuff is very complex, most of the time it gives rise to wild swings in efficiency due to page file usage.

Monteboy's tool here will do most of the Waza stuff for you:

http://www.ocxtreme.org/forumenus/showthread.php?t=3469

And be sure to set large system cache in performance options or the registry:

http://www.overclock.net/t/1249432/super-pi-32m-tweaking-guide


All you need is two partitions (a single drive will do). Run an SSD in IDE mode. Any questions on that or memory timings and just ask.

12887





(http://www.ocxtreme.org/forumenus/showthread.php?t=3469)

Raja@ASUS
10-23-2012, 08:21 AM
Ok so all this "efficiency" stuff is RAM based correct? ok...so what the heck do I have to do for better efficiency? I just read all that stuff Raja posted at the top of this page and i'm completely confused lol.

G.Skill 1600MHZ running @ 1766 currently.
DRAM running on stock Voltage of 1.5V
FSB: DRAM 1:4
CAS# Latency 9
RAS# to CAS# Delay 9
RAS# Precharge 9
Cycle Time 24
Bank Cycle Time 39 clocks
NB Frequency 2439 MHZ

I've tried reading through the OCing thread Raja posted in the CHVF Sub Forum...but yeah, more confusion lol


Go back to the guide and read the part about CAS and what the figure denotes in time at a given frequency. A smaller number denotes better efficiency.

You can use the formula tCAS= (CAS / 2000) / Memory Frequency to show the time in nano seconds.

For example:

If running CAS 9 at DDR3 1866

tCAS=(9 X 2000) / 1866

Which is 9.64ns


If instead you ran CAS 9 at DDR3-2133

tCAS=(9 X 2000) / 2133

Which is 8.43 ns.


So, CAS 9 at DDR3-2133 is faster. Of course, this assumes that most other memory timings remain unchanged, tighter, or +1 clock maximum, between the two points.


Plug in your numbers and start working from there.


You can apply that formula to ALL timings that are shown in units of DRAM clock cycles, unless otherwise noted in the guides. You will find an interesting pattern every 266 MHz (DDR3), in that adding +1 to CAS results in the same time. It is however still faster because it takes less time to make a data burst at higher bus frequency.


There are some complexities to the rules depending upon the platform. On AMD, you get to choose the memory controller bus frequency, so the rules above are "more" applicable. On current Intel platforms, the memory controller frequency is tied to the memory ratio - it cannot be controlled independently (for good reason before anyone starts cussing). That means you can sometimes add +2 to CAS per 266 MHz (DDR3) and see no loss in time, or even an improvement.

HiVizMan
10-23-2012, 08:22 AM
I dont own any rog only normal asus motherboard. can i still enter?

yeah I see no reason why not.

HiVizMan
10-23-2012, 08:29 AM
I had to try myself aswell :).

3930K C2 @ 4759
RIVE
32gb G.Skill TridentX @ 2400
I need a better kit for this bench and a clean OS :D

Superpi 1M
12877

Superpi 32M

12878

I have to say the pic of my rig is kinda crappy since it was taken just before a leaking test but it should be ok enough :).

12880



And my HWBOT profile is: http://hwbot.org/user/luumi/


The mysterious Luumi has been uncovered. :D

58540.36

Will update leader board once I have done the other scores. Good first go mate. Your 32 mil time was very good efficiency wise, but the 1 mil seemed way off the pace for some reason.

HiVizMan
10-23-2012, 08:31 AM
Nodens 60787.79 No progress there.

Leader Board is now updated.

Zka17
10-23-2012, 11:38 AM
Indeed, finally the mystery was solved! Welcome onboard Sinebrychof/Luumi! :cool:

Zka17
10-23-2012, 12:21 PM
You will have to find stability again after you change the OS and perform Waza. If you're game, might as well do it all at once. FWIW for you guys, I'd recommend not stripping the OS. Just use XP SP3. Good ole Fredyama (32M legend) uses a basic install with GPU drivers and it works well enough for him. The stripping stuff is very complex, most of the time it gives rise to wild swings in efficiency due to page file usage.

Monteboy's tool here will do most of the Waza stuff for you:

http://www.ocxtreme.org/forumenus/showthread.php?t=3469

And be sure to set large system cache in performance options or the registry:

http://www.overclock.net/t/1249432/super-pi-32m-tweaking-guide


All you need is two partitions (a single drive will do). Run an SSD in IDE mode. Any questions on that or memory timings and just ask.

12887 (http://www.ocxtreme.org/forumenus/showthread.php?t=3469)

Thank you, Raja! Yes, I'm aware that will need to find stability after I will change the OS... I'm just looking for a nice Windows XP SP3... - I thought I have one, but for some reason none of my optical drives recognizes the disk anymore... :(

What I really meant by "first I would like stability" is to have a functional OS after tweaking the memory's timings... I didn't wanted to start with SuperPi immediately after I made changes, so used first other softwares to test...

I was considering my system enough stable for SuperPi if it passed at least 5 minutes of AIDA64 stress testing. So, stressed the system, and if passed, restarted and run SuperPi. But now I run into a problem...

I'm down with the CAS at 7 and trying to adjust the other primary timings... When I'm trying to stress with AIDA, immediately crashes... - but if I restart the stressing (not touching any other things) will go as long as I let it... The strange thing is that if I try to run SuperPi immediately after I got into OS, will stop showing some error, but if I'm trying it after the AIDA run (with the first crash and restart), it will complete...

Here is a picture for better understanding:

12889

I also tried prime95 for 30 min runs.... - and it's the same thing! If it's the first to start after booting into OS, then on one of the cores will show an error... not always on the same core, it seems random... If I restart it or running it after the AIDA as it's shown above, will go with no problems...

Note, both AIDA and prime95 are running perfectly and with no errors at all at stock and at XMP settings.

Any thoughts?

HiVizMan
10-23-2012, 12:41 PM
What BIOS version are you using on your board, XP is not installing on some motherboards with the latest cap bios. :(

Zka17
10-23-2012, 12:46 PM
Oh, my disk is not readable - I'm pretty sure of that... It is very old and vent through a lot since I have it... I'm just looking for a new disk...

HiVizMan
10-23-2012, 12:48 PM
you have pm

Raja@ASUS
10-23-2012, 01:35 PM
Thank you, Raja! Yes, I'm aware that will need to find stability after I will change the OS... I'm just looking for a nice Windows XP SP3... - I thought I have one, but for some reason none of my optical drives recognizes the disk anymore... :(

What I really meant by "first I would like stability" is to have a functional OS after tweaking the memory's timings... I didn't wanted to start with SuperPi immediately after I made changes, so used first other softwares to test...

I was considering my system enough stable for SuperPi if it passed at least 5 minutes of AIDA64 stress testing. So, stressed the system, and if passed, restarted and run SuperPi. But now I run into a problem...

I'm down with the CAS at 7 and trying to adjust the other primary timings... When I'm trying to stress with AIDA, immediately crashes... - but if I restart the stressing (not touching any other things) will go as long as I let it... The strange thing is that if I try to run SuperPi immediately after I got into OS, will stop showing some error, but if I'm trying it after the AIDA run (with the first crash and restart), it will complete...

Here is a picture for better understanding:

12889

I also tried prime95 for 30 min runs.... - and it's the same thing! If it's the first to start after booting into OS, then on one of the cores will show an error... not always on the same core, it seems random... If I restart it or running it after the AIDA as it's shown above, will go with no problems...

Note, both AIDA and prime95 are running perfectly and with no errors at all at stock and at XMP settings.

Any thoughts?

Yes, I have some thoughts :)

Forget about AIDA and Prime95 stability for running 32M. Wasting time, as the stability of one of those may not correlate to 32M stability at all. Don't waste your time with those two for running this benchmark. :)

Simply apply primary memory timings in UEFI and the memory frequency. Go into OS, run 32M. If it passes, you have two options:

1) Use MemTweakit on the secondary and third timings. Pull them in and check for impact on time and stability.

2) Go back to UEFI and change memory timings to tighter.


Get as close to this as you can with your DIMMs

tRRD = 4
tFAW = 16
tRTP = 4
tWTR = 4


Third timings get all to 4 to start if possible. Add 1-2 clocks if your DIMMs can't handle it and tighten from there.

Zka17
10-23-2012, 01:50 PM
OK, will try to forget about AIDA and prime95... will be hard though, as I really love AIDA (with prime95 will be no problem) :)

But now I started over, preparing a special HDD with WinXP... :cool: then I will follow your instruction with the memory settings - thanks again, Raja!

Luumi
10-23-2012, 01:53 PM
Heh you didn't like my surname as my nickname on HWBOT :D. I need to do some more runs and see how much better score I can get :). I'd like to try with some better kit for this since 8gb modules don't like super tight thirds as we all know :D.

Zka17
10-23-2012, 02:18 PM
No, it's not that! Just wanted to know who Luumi is... - now it's all set! As I said: Welcome onboard!

8 Pack
10-23-2012, 02:58 PM
Enjoying following this thread guys. Some good stuff coming out.

Keep tweaking those systems.

asipamn
10-23-2012, 03:52 PM
i tried to register the rog supreme oc but it require password to register. can mod help?
i mean the hwbot team registration.

chrsplmr
10-23-2012, 04:20 PM
The password is: republic

Welcome asipamn to ROG.c.

Nodens
10-23-2012, 07:05 PM
This one should be better...and probably the limit of my RAM...

Hardware:
Intel 3930K C2 @ 3498Mhz
RIVE
Corsair H100
Corsair Vengeance @ 2260Mhz


1M: 10s 719ms

12894

32M: 9m 32s 434ms

12895

HiVizMan
10-23-2012, 07:26 PM
Bravo sir, really well done. 57518.80

Do you mind if I update the leader board after the UEFA football?

Nodens
10-23-2012, 07:28 PM
No worries mate, anytime you wish! :)

HiVizMan
10-23-2012, 09:54 PM
Leader board updated.

Chino
10-24-2012, 02:22 AM
More of a personal achievement than an attempt to catch Zka lol. He's still light years away.


Hardware
CPU - Intel i7 3770K @ 3920.5 MHz
Mobo - Asus Maximus V Extreme
Cooling - Antec KUHLER H2O 920
Memory - Corsair Dominator Platinum @ 2000 MHz

SuperPi score: 9sec 311ms
12909

SuperPi 32m score: 8min 22sec 382ms
12908


Chino @ HWBot (http://hwbot.org/user/chino/)

HiVizMan
10-24-2012, 07:34 AM
Chino that is a huge leap in progress, 56199.66 and mate you still have more to go I am sure.

Dartmaul
10-24-2012, 06:18 PM
1M - 08.589s
12919
32M - 08m 39.299s
1292012921
http://hwbot.org/user/dartmaul/

HiVizMan
10-24-2012, 10:14 PM
Hey dartmaul good to see you join us. Here is the efficiency rating of your first submission. I will update the leader board tomorrow
60640.76

Myk SilentShadow
10-24-2012, 11:22 PM
OK I got some slightly better times!! :D

1Mil Pass: 19.018s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/1MPiPass2ndAttempt.jpg

32Mil Pass: 17m58.212s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/32MPiPass2ndAttempt.jpg

http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/MykSilentShadowSystem.jpg

hwbot.org/user/MykSilentShadow

smokeintheeye
10-25-2012, 10:27 AM
1292912930

all on stock speeds. will try with an overclock later today.

http://hwbot.org/user/smokeintheeye/

HiVizMan
10-25-2012, 10:35 AM
Myk SilentShadow 138782.14

Leader boards updated.

Myk SilentShadow
10-25-2012, 10:46 AM
Myk SilentShadow 138782.14

Leader boards updated.

Awesome!!! almost a 1000 point drop, now to consider using that stripped OS....

HiVizMan
10-25-2012, 10:47 AM
Welcome smokeintheeye and nice submissions. I have a small problem in that I am not certain your system ran both instances at the max single core turbo mode. You see the efficiency ratio between the two results is quite off. So that tells me that the 32mil was run at the max single thread turbo multiplier, but the 1mil did not.

Would you mind selecting a fixed multi for all cores, and if possible for the submission disable the Speed step too.

Thanks.

Zka17
10-25-2012, 10:50 AM
Awesome!!! almost a 1000 point drop, now to consider using that stripped OS....

Myk, just an idea... if I would be you, I would wait a little bit with the stripped OS... I would rather try to tighten the memory timings first... - as I said, just an idea... but HiVizMan certainly can point you to the right direction... :)

HiVizMan
10-25-2012, 11:07 AM
LOL Myk has pushed his ram I think, but it might be worth your while to see just what is left in the timings of your ram. The Frequency may be tapped out, but there might be a little more in the second or third tier timings.

Myk SilentShadow
10-25-2012, 11:17 AM
LOL Myk has pushed his ram I think, but it might be worth your while to see just what is left in the timings of your ram. The Frequency may be tapped out, but there might be a little more in the second or third tier timings.

Actually I haven't mate, I haven't touched the RAM at all. I tried messing with it the other morning, but I couldn't get anywhere...mostly due to just trying to adjust the frequency to see where I could get before playing with the timings...ended up having to do a CLRCMOS to POST and I redid used Turbo V Evo to get myself a slightly better OC...last time was a 10% increase, this one was 11 lmao.

HiVizMan
10-25-2012, 11:37 AM
Ah sorry I misunderstood Myk. Well there you have it a whole new vista of tweaking improvement awaits, and the seriously the good thing about this all is that you and many of us will learn heaps about our systems. All win as far as I am concerned.

smokeintheeye
10-25-2012, 12:07 PM
me second go hope it worked this time lol.

129351293612937

http://hwbot.org/user/smokeintheeye/

HiVizMan
10-25-2012, 12:13 PM
Nice one mate, this time it is much better.



Here is your Efficiency rating for this submission. 59276.83

Please do post a picture of your system at some time please. And guys once you have posted your system photo once, and you are using the same hardware do not worry about posting again.

Myk SilentShadow
10-25-2012, 12:16 PM
Ah sorry I misunderstood Myk. Well there you have it a whole new vista of tweaking improvement awaits, and the seriously the good thing about this all is that you and many of us will learn heaps about our systems. All win as far as I am concerned.

I agree mate and this weekend i'm gonna try and read through Raja's OC thread and see if I can understand it and try get my RAM to a good point where I can at least come close to Chris' score lol.

Zka17
10-25-2012, 12:19 PM
Please do post a picture of your system at some time please. And guys once you have posted your system photo once, and you are using the same hardware do not worry about posting again.

I was actually about to ask regarding this... sounds logic...

smokeintheeye
10-25-2012, 12:26 PM
thanks HiVizMan.... so is that a good score? is there anything you recommend trying to get a better one iv not done anything to the ram yet its still at stock speed.

Raja@ASUS
10-25-2012, 12:31 PM
You will get more from learning Waza on a basic XP Sp3 or Server OS install than fooling with stripped OSes too early.


Prioritize like this:

1) Memory timings.

2) Waza (includes making the 32M OS changes to LSC and Maxmem switch), then tweak memory timings again. Use Xp 32 bit SP3 or 2003 Server.


3) OS stripping. But you can leave this out for now. Too many variables and far less to gain than the two above. Further, most of the stripped OSes out there have issues with Waza making things difficult to pin down.


And don't worry about understanding memory timings at the electrical level, very few people do. I have not met a single extreme overclocker that truly does, but they still get results from graft. So, it's not mandatory. The only real advantage is that you can tweak a system faster than other people if you do know what the timings mean. It's a slow learning process for all of us.

HiVizMan
10-25-2012, 01:07 PM
thanks HiVizMan.... so is that a good score? is there anything you recommend trying to get a better one iv not done anything to the ram yet its still at stock speed.

Start playing with your ram frequency and timings for now. Get a feel of what makes a difference to the times. Worry about the other stuff down the road, this challenge is in its infancy still, we have heaps of time to get through all or at least most of the major 32 Mil tweaks. Now just focus on getting the most out of your ram. Find your kit of rams sweet spot.

HiVizMan
10-25-2012, 04:24 PM
Zoltan could you please put the team password in your sig, that way new members can see it easily and not have to pm to get it. Thank you.

Dartmaul
10-25-2012, 04:29 PM
1M - 08.808s
12943
32M - 7m 54.788s
12944
12945
http://hwbot.org/user/dartmaul/

My 1st run has actually on my 24/7 settings without any serious tweaking.
But the last place makes me sad so I decided to spend some time and ask some people (Thanks Nodens) to get better results.
(I've cut off 45sec from my precious 32M result, even with lower frequencies, hope it'll have some good impact)

flexnl
10-25-2012, 05:28 PM
1m run : 7.457s
https://www.dropbox.com/s/2gy2ntjswggttfu/rogpikort.png?m

32m run : 6m50.265s
https://www.dropbox.com/s/n6n8ccxp3xhhrff/rogpilang.png

myrig : https://www.dropbox.com/s/5q3euf8a34cp1l9/2012-10-25%2019.20.44.jpg

hwbot: http://hwbot.org/user/flexjenl/

and a big fat HELLO to you all:):):):D:D:D

Zka17
10-25-2012, 06:00 PM
Zoltan could you please put the team password in your sig, that way new members can see it easily and not have to pm to get it. Thank you.

How about now? Is it something like you were thinking about?

HiVizMan
10-25-2012, 06:03 PM
Welcome my friend.

Your ER is 56642.29

Outstanding first submission. I will update the leader board later tonight.

Dartmaul I hear you my friend and a good improvment. 57647.74

Dartmaul
10-25-2012, 06:14 PM
I will update the leader board later tonight.

Dartmaul I hear you my friend and a good improvment. 57647.74
TY)
BTW Please correct my nickname in ladder, you've made 3 mistakes in it))

Also better IMC gives some advantage to Z77 vs X79(

HiVizMan
10-25-2012, 07:12 PM
I sincerely apologise, when I did the last update I was rather tired. I shall not mess it up again I promise.

HiVizMan
10-25-2012, 07:18 PM
I sincerely apologise, when I did the last update I was rather tired. I shall not mess it up again I promise.

HiVizMan
10-25-2012, 07:20 PM
How about now? Is it something like you were thinking about?

Thank you perfect.

Leader board has been updated and there is movement in the Intel Top 3.

Ladies and gentlemen I introduce flexnl who has made a rather sparkling first submission and has shot up into third place.

Zka17
10-25-2012, 09:21 PM
Congratulations, flexnl - and welcome! :cool:

Dartmaul
10-25-2012, 09:26 PM
I sincerely apologise, when I did the last update I was rather tired. I shall not mess it up again I promise.
The only one who does nothing isn't mistaken, so it's no problem my friend, newermind.

What do you think about both X79 and Z77 systems being in the same ladder? Did your score system has some balancing multiplier?)

Menthol
10-26-2012, 02:38 AM
Zka,
Why am I not surprised your in the lead, I never seen the point in 2D benching, no pretty graphics, but may give it a shot this weekend, have to read up on it a little first.

Rockford
10-26-2012, 02:44 AM
yes..

i mean.. return to proxycon.. When he is blasting with the bfg..........

or the tranquility of the fireflies..

Does Jane have a girlfriend?, if not.. why

Not to forget the excitement in New Calico...

many fine entries in here...

HiVizMan
10-26-2012, 08:06 AM
The only one who does nothing isn't mistaken, so it's no problem my friend, newermind.

What do you think about both X79 and Z77 systems being in the same ladder? Did your score system has some balancing multiplier?)

That is a very good point, and one I have been looking at these last few days. The key consideration that has held me back in factoring in an adjustment is that the X79 processors have a larger L3 cache. And theoretically that should make 32mil more effective on that platform. However the IMC strength of the Z77 processors is overpowering that hoped for balancing factor. But for now I need more testing and then we shall see.

This challenge is certainly a work in progress and as such is no where near perfect, or the finished article. I am making most of the stuff up as I go along, so with the help and input of all the guys we can just get better.

Dartmaul
10-26-2012, 11:13 AM
For some time ago I've played WoW. (In few words, it's a game that cannot be balanced perfectly, but developers are always missing the nearest spot between classes.)
To find out the actual "balance" state I always visited worldoflogs website with players submissions. The class that took more 1st places than others is owerpowered at the most.

Replace "wow" with "overclosking", "classes" with "platforms", and "worldoflogs" with "hwbot" and you'll understand what I mean))

Raja@ASUS
10-26-2012, 11:27 AM
That is a very good point, and one I have been looking at these last few days. The key consideration that has held me back in factoring in an adjustment is that the X79 processors have a larger L3 cache. And theoretically that should make 32mil more effective on that platform. However the IMC strength of the Z77 processors is overpowering that hoped for balancing factor. But for now I need more testing and then we shall see.

This challenge is certainly a work in progress and as such is no where near perfect, or the finished article. I am making most of the stuff up as I go along, so with the help and input of all the guys we can just get better.


X79 is slower. The internal architecture is two separate memory controllers strung on the ring bus. Even if you go dual channel on one controller, the impact of the additional logic remains.

HiVizMan
10-26-2012, 11:29 AM
Yeah I know what you mean, it is such a difficult thing to balance, and not favour one class (processor type).

Henkenator68NL
10-26-2012, 12:21 PM
Hi Guys,

Since there is a separate leaderbord for AMD and Intel, I now submit my first scores for my AMD system.

HWBOT link: http://hwbot.org/user/henkenator68nl/


1M Score: 17.472s
12980
32M Score: 17m30.584s
12981

My Rigg:
12983

My AMD Rigg:
Crosshair V Formula, Phenom II 1100T BE, 4 x 2 Gb ddr 1333 GEiL, 6990, CPU+GPU+MEM water Cooled
see my signature for further details and for pictures my Album on my ROG profile Page

Now I am curious so see the score for my AMD system

HiVizMan
10-26-2012, 12:40 PM
Here you are 113693.55


Think that puts you in first place. :)

Henkenator68NL
10-26-2012, 01:21 PM
Alright thats not bad, now trying to improve.

My HWBOT link: http://hwbot.org/user/henkenator68nl/

New run on higher frequency, Improvement on 32M ! on 1M same score. Very interesting result, now I have to figure out what this means...

Here is Super Pi 1M at 4176MHz: 17.472s
12996
Here is improved Super Pi 32M score on 4176 MHz: 17m21.287s
12998
My Rigg:
13000

My AMD Rigg:
Crosshair V Formula, Phenom II 1100T BE, 4 x 2 Gb ddr 1333 GEiL, 6990, CPU+GPU+MEM water Cooled
see my signature for further details and for pictures my Album on my ROG profile Page

What will the Impact be on my overall score??

Thanks for updates HiVizMan

HiVizMan
10-26-2012, 01:42 PM
Nope the 1mil hurt you big time and the efficiency of the 32 mil was a little higher too.

116447.22

Myk SilentShadow
10-26-2012, 01:55 PM
Nope the 1mil hurt you big time and the efficiency of the 32 mil was a little higher too.

116447.22

Didn't hurt him enough to knock Chris and myself one rank down lol

HiVizMan
10-26-2012, 02:02 PM
hehe well I only take the best ER not the last submission.

Actually there is a trick to the AMD benching but I need to do more testing before I post up here.

My best is 109068.30 but I did not take screen shot :(

Henkenator68NL
10-26-2012, 11:07 PM
hehe well I only take the best ER not the last submission.

Actually there is a trick to the AMD benching but I need to do more testing before I post up here.

My best is 109068.30 but I did not take screen shot :(

I sure would like to know you're ideas/findings on that !

flexnl
10-26-2012, 11:37 PM
superpi : https://www.dropbox.com/s/1vqs6jrrs2b3ttm/rogpikortfinall.png

superpi 32m :https://www.dropbox.com/s/2418vojjmravqqu/rogpilangfinal.png

http://hwbot.org/user/flexjenl :cool:

myrigpic 13033

Henkenator68NL
10-27-2012, 02:50 AM
Oke I had another run on my INTEL system, lower frq, bumbed up the mem freq.

Curious to see how this results holds up in end score. Major tweaks are still on hold ... until I decided what to do about cooling. The stock Intel Cooler is not the best hehehe

Super Pi 1M: 7.302S
13016

Super Pi 32M: 6m42.730s

13017

My rigg:

13018

HiVizMan
10-27-2012, 10:12 AM
56646.84 for Henk. And it blows my mind that you are getting 5GHz on a Intel stock cooler. You do know that you can run your system at way lower clocks and it will not impact negatively at all on ER.


56397.15 for flexnl. Mate could you work out how to use the image upload button. Our server is way quicker for images that dropbox and it is more mobile device friendly. As I tend to use a mobile device most of the time. And BIG PLUS it means folks can actually see at a glance what your bench was. :)

Nice gains both of you. Not enough to change the leader board ranking but it is certainly getting tighter at the top. I will update later today, hope you both do not mind, but since both scores do not change the order it should be fine for a while. :)

Zka17
10-27-2012, 11:50 AM
And it blows my mind that you are getting 5GHz on a Intel stock cooler. You do know that you can run your system at way lower clocks and it will not impact negatively at all on ER.

Actually, while I was stress testing before run SuperPI (I know, I won't do that from now on), I did observe something... sometimes I left open the AIDA's stress window, with the actual stress test stopped, and I saw that SuperPi is actually means a very low load on the CPU, around 25% - and the temps were also just a little bit over the idle ones... of course, in my case there is the big rad, but I could imagine with the stock cooler to have the idle temps around 50C with Henk's settings and then to increase to 70-80 while SuperPi is running... of course, no serious stress testing is advised...

But I agree, in this challenge the point is about efficiency... so, better play safe and reduce the CPU frequency, instead, tweak the memories! When you find a nice, tight spot for the memories, then you can up the CPU again to have a run for HWBOT too...

Arne Saknussemm
10-27-2012, 12:05 PM
http://hwbot.org/user/arne_saknussemm/

i73930K @ 4200MHz
Rampage IV Extreme
Cooling H2O Custom loop
RAM Gskill Z series 1600MHz

Pi1M
13030

Pi32M
13031
13032

Myk SilentShadow
10-27-2012, 12:19 PM
OK here's run 3: a little disappointing that my 32M was a little slower after a really good start.

1MIL: 18.796s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/1MPiPass3rdAttempt.jpg

32MIL: 17m59.269s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/32MPiPass3rdAttempt.jpg

Though, on a personal note...I managed to take out #2 ranking for my 4170!! :D

Zka17
10-27-2012, 12:36 PM
Though, on a personal note...I managed to take out #2 ranking for my 4170!! :D

I saw that - your doing good, keep going! :cool:


Arne, I see that you have new setup... Do you have a thread with more details about it?

Myk SilentShadow
10-27-2012, 12:47 PM
I saw that - your doing good, keep going! :cool:


Thanks mate, i'm hoping the "efficiency" is good enough to close the gap to Chris lol

Arne Saknussemm
10-27-2012, 12:53 PM
Arne, I see that you have new setup... Do you have a thread with more details about it?

Yeah, I was thinking of calling it "The chronicle...." LOL only kidding ;)....no no thread just got it to drop temps and make changes easier....

Zka17
10-27-2012, 01:05 PM
Arne, if you want your own chronicle, just go ahead! I think it would be interesting... :cool:

HiVizMan
10-27-2012, 01:10 PM
57544.42 Arne Should now be just off the top 5.

139596.48 Myk still not as efficient as your first run.

Will do the Leader Board after my afternoon nap.

HiVizMan
10-27-2012, 01:12 PM
Arne's adventure - now that has a nice ring to it.

flexnl
10-27-2012, 01:22 PM
56646.84 for Henk. And it blows my mind that you are getting 5GHz on a Intel stock cooler. You do know that you can run your system at way lower clocks and it will not impact negatively at all on ER.


56397.15 for flexnl. Mate could you work out how to use the image upload button. Our server is way quicker for images that dropbox and it is more mobile device friendly. As I tend to use a mobile device most of the time. And BIG PLUS it means folks can actually see at a glance what your bench was. :)

Nice gains both of you. Not enough to change the leader board ranking but it is certainly getting tighter at the top. I will update later today, hope you both do not mind, but since both scores do not change the order it should be fine for a while. :)

ii know how to do it, its just that my png files are invaled it says
i take the shots with printscreen and then paint
am i doing something wrong?
ooh and eehhh it willl get tighterrrrr :)

Henkenator68NL
10-27-2012, 01:23 PM
Oke now at much lower frequency, I am curious about the ER now:

Super Pi 1M: 7.956s
13035

Super Pi 32M: 7m13.180s

13036

13037

Henkenator68NL
10-27-2012, 01:28 PM
ii know how to do it, its just that my png files are invaled it says
i take the shots with printscreen and then paint
am i doing something wrong?
ooh and eehhh it willl get tighterrrrr :)

@Flex:
Flex you make a screenshot by pressing Prt Scr, than go to paint, press paste. Than save as ----> choose JPG !!
Than it will work!
Good Luck Mate!

kkn
10-27-2012, 01:36 PM
ii know how to do it, its just that my png files are invaled it says
i take the shots with printscreen and then paint
am i doing something wrong?
try and keep the file size under 1mb, and post as jpg mabye.

Myk SilentShadow
10-27-2012, 01:45 PM
OK!!! I finally managed to crack 2000MHZ on my RAM

1MIL: 18.742s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/1MPiPass4thAttempt.jpg

32MIL: 17m50.742s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/32MPiPass4thAttempt.jpg

HiVizMan
10-27-2012, 01:56 PM
134254.36 OK now you are heading in the right direction. :)

HiVizMan
10-27-2012, 01:59 PM
56523.88 Henk there you go not much in it.

Myk SilentShadow
10-27-2012, 02:06 PM
134254.36 OK now you are heading in the right direction. :)

FINALLY!!! lol, though admittedly. I did push the RAM voltage up a couple notches and loosened the timings from 9-9-9-24 to 10-10-10-25. Everytime I use stock 9-9-9-24 I cannot for the life of me push the RAM really beyond 1700MHZ...though, it may have something to do with me not adjusting the NB Frequency beyond the 2247MHZ mark...lol :p

Chino
10-27-2012, 02:09 PM
ii know how to do it, its just that my png files are invaled it says
i take the shots with printscreen and then paint
am i doing something wrong?
ooh and eehhh it willl get tighterrrrr :)

Try saving them as JPG.

chrsplmr
10-27-2012, 02:09 PM
OK .. 2nd shot at this. (I must confess, I still have no idea how to read these or where you come up with the scores .. et.)

13038

13039

HiVizMan
10-27-2012, 02:17 PM
114455.79 Still way ahead of the pack Chris.


The formula is a derivative of time divided by frequency. But it has a added constant that makes it easier to work with for the contest.

chrsplmr
10-27-2012, 02:28 PM
hahhahaahha ....... ooooooooooohh .. that's much clearer bro.. hahahaa ..

This is my favorite thing about ROG .. I worry not of not knowing this or that ..
It is said one truely knows All that knows how and where to look it up ... ROG adds another Layer ..
-- 'Who to ask.' ... What one does not know, the next knows and shares Levels above. cudos and thankYou.c.

So, if I read this right though .. Henkenator .. still has the lead/AMD ?? awww .. Heck. hehe
1. Henkenator68NL 113693.55 (http://rog.asus.com/forum/showthread.php?24118-ROG-Team-Super-Pi-Challenge-Oct-18-Nov-17-2012&p=174897&viewfull=1#post174897)

HiVizMan
10-27-2012, 02:48 PM
Yeah but his best submission is with an Intel system. So you are the actual AMD leader. Only the best submission will count. :)

Arne Saknussemm
10-27-2012, 03:37 PM
http://hwbot.org/user/arne_saknussemm/

i73930K @ 3800MHz
Rampage IV Extreme
Cooling H2O Custom loop
RAM Gskill Z series 1600MHz

1M
13042
32M
13043
13044


Wondering which way this is going to go....:)

Myk SilentShadow
10-27-2012, 04:06 PM
OK!!! got my RAM to 2133MHZ @ 10-10-10-26 and these times are terrible!!

1Mil: 20.989s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/1MPiPass5thAttempt.jpg

32Mil: 19m55.538s
http://i483.photobucket.com/albums/rr194/TheVaempyrMyk/32MPiPass5thAttempt.jpg

chrsplmr
10-27-2012, 04:36 PM
This was/is my confusion.(post192)
13053

And .. ok .. I'm addicted .. I was even over in my AMDversion XP64 .. but it updated drivers on boot after a year .. and
now is unstable .. more fun to go.. hahhaha .... Shot 3 !!! 1134xxx ??? am I adding correctly ???
13051
13052

HiVizMan
10-27-2012, 06:40 PM
Close Chris 114071.80

Sorry for confusion meds are way to strong. :D

Myk yes they are. I am not even going to put that score in a thread. LOL

Here is a tip, look at the 1866MHz range. That is the sweet spot I found.


Arne 57295.11

flexnl
10-27-2012, 08:47 PM
Congratulations, flexnl - and welcome! :cool:

tnx m8
:o

Henkenator68NL
10-27-2012, 09:20 PM
56523.88 Henk there you go not much in it.

HiVizMan, question are you going to add this score in the Leaderboard? :cool:

HiVizMan
10-27-2012, 11:05 PM
Did I miss one, my apologies. Indeed I shall add it right now.

HiVizMan
10-27-2012, 11:13 PM
Must just say it is really tight at the top. Well done to the four or five guys who are currently at the top end of the leader board. I still think that the winning ER will be under 55500. And that it will be achieved at a low processor frequency. :)

HiVizMan
10-27-2012, 11:13 PM
Not in the TOP 5

Please do not be disheartened if you are not in the top two or top three, or feel you will not be able to challenge for the top place. You might just surprise yourself. The result I posted which no one has equaled yet was completed with a air cooler and ram that cost Ł49.

However, I have decided to include a lucky prize draw, and will be open to all who have submitted and taken part in the thread. I have sent the qualification number of submissions to be included in the lucky draw to a moderator. All I can say is it is more than 2 and less than 10 submissions that will qualify you for inclusion into the draw.


If you have submitted and have not received your Super Pi 32 Mil notes, or a link to a pretty nifty 32 Mil Benching OS please PM me. One submission gets you the notes, and two + access to the OS.

chrsplmr
10-28-2012, 12:29 AM
by ROG .. I think I'm inching forward .. look Ma .. I'm learn'n ...
1306113062

Zka17
10-28-2012, 01:22 AM
Chris, you're doing very good! :cool:

chrsplmr
10-28-2012, 01:41 AM
Im trying to get one with no OC on the cpu .. it's not hav'n it.
I cant wrap my head around the formula to come up with the result to know .. whether I did better or not ??? hahhaha a... well, everybody's got some kinda dumb in them .. hahahhaa .. this is that algebra stuff aint it ??? hahhahaa

[[If your afraid to look stupid for asking .. you always will be for not...besides I have forgotten
more than I ever thought I knew .. I think.]]

(I had two concerns in school, Not going to Vietnam, if I quit and CheerLeaders .. and I am sorry for that.. hahahhaa
Which is funny to me now because I would love to go to Vietnam and ... CheerLeaders .. hahahhaha..no.

Sorry for OffTopic .. a little.... ThanksZ..

Henkenator68NL
10-28-2012, 02:13 AM
Oke finally try for this day (hmm or is it already the other day??? hmmm don't know anymore :cool:)

Super Pi 1M Score: 8.073S
13071

Super Pi 32M Score: 7m18.206s

13072

My Rigg:
13073

Good night or day cya all later !!

Myk SilentShadow
10-28-2012, 02:37 AM
Im trying to get one with no OC on the cpu .. it's not hav'n it.
I cant wrap my head around the formula to come up with the result to know ..

You and me both Bro, i've been pushing hard to try and get my RAM to run as fast as I could, now having 2133 and I got told to throttle back on it lol...and i'm also currently @ stock too. I did, however manage to push my CPU OC to 5GHZ :D

Henkenator68NL
10-28-2012, 02:43 AM
I had a little insight so this is REALLY the finall try for to day (and I believe a good one !!)

Super P1 1m score: 8.859s @ 4123MHz
13078
Super Pi 32M score: 7m55.314s @ 4123MHz
13077

My Rigg:
13076

NOW I keep my fingers crossed ... LOL

Myk SilentShadow
10-28-2012, 06:22 AM
This is starting to frustrate me now somewhat. Running CPU @ stock 4.2GHz, RAM @ 1866 with the timings 9-11-11-28-45(tRC). NB Frequency @ 2400 and the FSB: DRAM is @ 3:14 and got these terrible times 1MIL: 21.282s 32MIL: 20m14.107s. Everytime I try to "tighten" the first 3 by 1 I always get an Overclocking has failed! message at startup.

The annoying thing is that when I had my CPU sitting @5GHz, my 1MIL was the best time i'd gotten @ 17.778s...though then SuperPI kept throwing out errors while attempting a 32MIL run lol.

Henkenator68NL
10-28-2012, 08:10 AM
This is starting to frustrate me now somewhat. Running CPU @ stock 4.2GHz, RAM @ 1866 with the timings 9-11-11-28-45(tRC). NB Frequency @ 2400 and the FSB: DRAM is @ 3:14 and got these terrible times 1MIL: 21.282s 32MIL: 20m14.107s. Everytime I try to "tighten" the first 3 by 1 I always get an Overclocking has failed! message at startup.The annoying thing is that when I had my CPU sitting @5GHz, my 1MIL was the best time i'd gotten @ 17.778s...though then SuperPI kept throwing out errors while attempting a 32MIL run lol.Myk, It happend to me too. The only way I found sofar is adjusting the freq of your ram to a stable OC. Try te test than....I think it depends on the ram tyoe and series... But hey just like you... I doing it by trial and error too...

Chino
10-28-2012, 08:53 AM
This is starting to frustrate me now somewhat. Running CPU @ stock 4.2GHz, RAM @ 1866 with the timings 9-11-11-28-45(tRC). NB Frequency @ 2400 and the FSB: DRAM is @ 3:14 and got these terrible times 1MIL: 21.282s 32MIL: 20m14.107s. Everytime I try to "tighten" the first 3 by 1 I always get an Overclocking has failed! message at startup.

The annoying thing is that when I had my CPU sitting @5GHz, my 1MIL was the best time i'd gotten @ 17.778s...though then SuperPI kept throwing out errors while attempting a 32MIL run lol.

Somtimes RAM can really be a pain in the ass. If you've tried the impossible and it won't get higher than 1866MHZ, then maybe that's your max OC, If that's the case, you can start tweaking the second and third timings. Or your OS. xD

HiVizMan
10-28-2012, 11:41 AM
Right the Leader Board has been updated.

Apologies to those who I missed out, it will happen from time to time, simple remind me in the thread and post a link to your submission that makes it easier for me to find.

Chris 114384.10

Henk you have moved up the Leader Board with 56122.85 Well done mate.

Dartmaul
10-28-2012, 11:45 AM
Now, when score calculation system is known and Arne brings some news in X79 standing, I want to make my next attempt.

But there's one more question to ask:
There's a rule about the CPU frequency - must be the same on both 1 and 32 runs. What's about mem frequency?

Alex-Ro
10-28-2012, 12:09 PM
hey hivizman,check your pm please :)

HiVizMan
10-28-2012, 12:09 PM
Memory frequency has only one restriction, may not be above 2400MHz.

The formula is very simple and is - time taken divided by frequency. I then apply a constant value to make it simple to combine the two and have one ER value for both benchmarks.

I am not sharing the formula because it does have one very key anti-cheat effect, that I am sure will not be needed but you have to build these things into the equation.

Now as to X79 vs Z77 - I think that more results need to be submitted by X79 boards before I can see what if anything needs to be adjusted.

Dartmaul
10-28-2012, 12:23 PM
1M - 09.664s CPU - 3904 MHz Mem - 1138.8 MHz (2277.6 MHz DDR)
13088
32M - 08m 41.974s CPU - 3904 MHz Mem - 813.4 MHz (1626.8 MHz DDR)
13089
13090
http://hwbot.org/user/dartmaul/

HiVizMan
10-28-2012, 12:41 PM
Dartmaul 58106.12

Dartmaul
10-28-2012, 12:50 PM
Seems that I need some software improvements to continue. (I'm to lazy to install XP with all that stuff Raja told about)

HiVizMan
10-28-2012, 01:01 PM
Seems that I need some software improvements to continue. (I'm to lazy to install XP with all that stuff Raja told about)

LOL I do understand.

Myk SilentShadow
10-28-2012, 01:39 PM
Hey aren't the memory frequencies supposed to be the same for both 1 and 32M passes?

HiVizMan
10-28-2012, 02:14 PM
Hey aren't the memory frequencies supposed to be the same for both 1 and 32M passes?

Nope only the CPU must be at the same frequency, the memory must be 2400MHz or lower. :)


Due to the amount of cache on modern CPU's 1 mil will hardly use the system memory at all. 32 Mil will however test your ram big time, and be impacted by your ram big time. So no, only the CPU frequency needs to be the same.

Myk SilentShadow
10-28-2012, 02:37 PM
Nope only the CPU must be at the same frequency, the memory must be 2400MHz or lower. :)


Due to the amount of cache on modern CPU's 1 mil will hardly use the system memory at all. 32 Mil will however test your ram big time, and be impacted by your ram big time. So no, only the CPU frequency needs to be the same.

Ok, so i've been messing with the Bus speed as you suggested and it's made my times even worse. Currently my CPU is @ 4.4GHz, Bus is @ 266, Multiplier @ 16, RAM @ 1772MHz timings of 9-9-9-24-33. if I adjust the 2nd Tier of the timings, loosen them by one times are crap, if I tighten them by 1, my System refuses to boot and gives me an OC has failed! message....even using the stripped down OS has made no real difference...might be because i'm using an extra SSD to hold the XP install lol