ASUS Maximus II Formula - new P45 king?

[Simps]

Jor3lBR,

Wow, very nice numbers, but yes, prime is needed. No prime, the comparison is even less apples to apples as it already is.
I also see that your CPU is a 45nm, it will by definition need less volts to it its business on the same clock a 65nm will. That is independent on boards or anything. Its just a better and newer cpu. That is the major thing making the comparison even less fair.

I see the vPLL and vFSB you needed was lower then mine. But those are very cpu related, and this is a problem when comparing 45nm to 65nm cpus.
The vNB I used was lower then yours. I was on 1.23 on bios and 1.24 reported on everest, as you can see on pics. This voltage is much less cpu dependend that the others, but still, you are on 2x2gb mem, and I am on 2x1GB mem, which could explain it too.

I guess the only way to compare it, would be the same CPU running on each board. That would be the best. Or, second way, if I had a 45nm or you a 65nm. I don't have a 45nm. Do you have a 65nm quad by a chance?

Anyway, the numbers are good. Congrats. But as a comparison, it won't do, because the difference on the CPU's are much more then I imagined.

Anyway, this is what I get on the everest, for the mems clocked as yours. My write and copy values, are lower because I have less cache on the CPU (65nm vs 45nm). Every 65nm cpu will have lower write / copy. But the read / latency are unafected by the less cache and are fair comparison.



[]'s
Simps

[Grnfinger]

Just to reply Simps on Gigabyte Thread, I won't be answering any questions here only there to avoid people getting upset. This is Asus thread so discussions should be Asus only...

More results just click on my signature down there



I know it's not a competition there is nothing to compete for, but sometimes it sounds like you underestimate engineers, if they build something to work with 6 Phase it's because they know what they are doing. Just because Asus has 16 Phase doesn't mean they can hold anything better. In fact as far as I'm concern the smallest route between to points is still a straight line...

Here:

[System Setup]
CPU: Q9650@3500Mhz 500x7
Voltages:
VCORE=1.29V
VTT=1.20V
VPLL=1.5V
VNB=1.26V
VSB=1.1V
VDIMM=2.24V
Mobo: Gigabyte EP45-UD3P
Cooler: Sunbeam Core Contact Freezer
Ram: 2 x 2Gb - 4gb G.Skill Pi 9600 @1200 5-5-4-15
Video: X1300
PSU: OCZ Powerstream 520w Adj


[Stability and Memory Benchmark Testing]
I can't show Prime95 now as I don't have 7hrs of the day, but testing was done using IBT Max and HCi Memtest 3.8 150%. If my systems passes those 2they can Prime95 all year. But I can Prime95 during the week if you insist.

Another note is that Everest OSD doesn't show much info on this board only the stuff is on screen so you have to see all voltages on ET6 (Easy Tuner 6).


IBT 35 Loops Maximum Stress starting - LOAD (I can only see pics of your system on Idle)


This board is pretty much a joy to use, so easy you can hit and tweak and you can get the job done.

No matter how low volts you use this board will use less and remain stable and no matter how high fsb you want this board will keep breaking records with Quads.

Hope that settles the friendly inter-thread challenge

Good looking clocks there, I had a chance to play with one and wish I had bought it , it would have been a great benching board.

Jor3lBR,

Wow, very nice numbers, but yes, prime is needed. No prime, the comparison is even less apples to apples as it already is.
I also see that your CPU is a 45nm, it will by definition need less volts to it its business on the same clock a 65nm will. That is independent on boards or anything. Its just a better and newer cpu. That is the major thing making the comparison even less fair.

I see the vPLL and vFSB you needed was lower then mine. But those are very cpu related, and this is a problem when comparing 45nm to 65nm cpus.
The vNB I used was lower then yours. I was on 1.23 on bios and 1.24 reported on everest, as you can see on pics. This voltage is much less cpu dependend that the others, but still, you are on 2x2gb mem, and I am on 2x1GB mem, which could explain it too.

I guess the only way to compare it, would be the same CPU running on each board. That would be the best. Or, second way, if I had a 45nm or you a 65nm. I don't have a 45nm. Do you have a 65nm quad by a chance?

Anyway, the numbers are good. Congrats. But as a comparison, it won't do, because the difference on the CPU's are much more then I imagined.

Anyway, this is what I get on the everest, for the mems clocked as yours. My write and copy values, are lower because I have less cache on the CPU (65nm vs 45nm). Every 65nm cpu will have lower write / copy. But the read / latency are unafected by the less cache and are fair comparison.



[]'s
Simps

What ram you running there?
Everest scores look a little low...

[Simps]

Grnfinger,

Everest "write" and "copy" bandwidths, are affected directly by the CPU amount of cache. The 45nm CPU's have 6MB L2 (x2) and 65nm CPU's have 4MB L2 (x2).
So a 45nm CPU will have a much much higher "write" and "copy" bandwidth, then a 65nm CPU.

The other values, "read" and "latency" aren't affected my the CPU cache. And they are the only ones comparable between a 65nm CPU and 45nm CPU.

That is why those "write" and "copy" values are a bit low, because of my 65nm CPU. These days people are used to 45nm results and so.

[]'s
Simps

[radaja]

well im on the asus merry-go round once again.i call and ask to talk to the manager and the phone jockeys take my info and give me a new support ticket and they say someone will call,and of course if its like the last time no one does,so ill have to call again and do it all over again with someone else and the will tell me something similar about someone calling me but they never do so i call again this time im a little upset,cant explain myself clearly then they get some inkling of what im talking about and take some more info and tell me that they will have someone call me in the next hour,but no one does,and then guess what happens?nothing..........that is untill i call them again and again and again.

[Simps]

Grnfinger,

I just did a search on this thread, and found your old 65nm results. As you can see, your results showed the low "write" and "copy" values on 65nm too. And much lower then mine (that you tought was low), and that was on your x38. See for yourself, might refresh your memory.

Here is some old 65nm quad runs











45nm C2D












Will post more when I get home from work and will match cable tie guys settings for a comparison.
CryptiK maybe you could run some tests with your MIIF and E8400 that would be a great head to head

[]'s
Simps

[Grnfinger]

Grnfinger,

Everest "write" and "copy" bandwidths, are affected directly by the CPU amount of cache. The 45nm CPU's have 6MB L2 (x2) and 65nm CPU's have 4MB L2 (x2).
So a 45nm CPU will have a much much higher "write" and "copy" bandwidth, then a 65nm CPU.

The other values, "read" and "latency" aren't affected my the CPU cache. And they are the only ones comparable between a 65nm CPU and 45nm CPU.

That is why those "write" and "copy" values are a bit low, because of my 65nm CPU. These days people are used to 45nm results and so.

[]'s
Simps

Thanks I'm well a where what the differences are in 65nm and 45nm cpu's and how they affect bandwidth.
So again I'll ask you what ram is that?

well im on the asus merry-go round once again.i call and ask to talk to the manager and the phone jockeys take my info and give me a new support ticket and they say someone will call,and of course if its like the last time no one does,so ill have to call again and do it all over again with someone else and the will tell me something similar about someone calling me but they never do so i call again this time im a little upset,cant explain myself clearly then they get some inkling of what im talking about and take some more info and tell me that they will have someone call me in the next hour,but no one does,and then guess what happens?nothing..........that is untill i call them again and again and again.

Perhaps your being alittle too polite, maybe take a royal dump all over someone, then say this is unacceptable you would like to speak with a supervisor.
Once he/her gets on the phone you will have to tone it down anotch or 2 or they will hang up on you....explain the problem and ask to be elevated to level 3 tech, once there remind the guy how much the board cost's, that your not some douche bag with more money than brains and refuse anything less than an upgrade, it worked for me and my Max X38 and it will work for you if your just irritating enough without crossing the line.

[CryptiK]

I got an RMA approved for my MIIF today, they're going to do the cross-ship method. I'm not sure if I'm going to keep the replacement or not as I'm having a lot of fun with my DFI LT P35 and my Gigabyte EP45 UD3P...

IMHO I think the LT P35 and the UD3P are much more capable ram clocking boards than the M2F. I still can't get 541 FSB & 650MHz to pass SPI32M, nor get 434 FSB & 650 MHz to pass. It's beyond me why, that isn't even pushing it frequency or FSB wise. My P5K could do that even, so could my mates LT P35.

I can get 534 FSB & 640 MHz to pass, but as soon as I crank up the multi at the same settings it fails straight away and no amount of skewing will make it stable. This board is a nightmare to bench on.

If I were you I'd sell it as new when you get it back (after testing it to make sure it works of course), they are still really expensive boards so if you took ~$100 off retail price you'd sell it in no time.

I can give it try for you...
You mean 1:2 divider, as a 400FSB giving DDR2-1600?
Is that right?

Yes, that's the one. 333 FSB & 666 MHz would be easier to test though, however my board wont boot no matter what I set on the 1:2 divider. I'm quite sure it's broken and Asus just never expected anyone to try and use it and find out.

Just to reply Simps on Gigabyte Thread, I won't be answering any questions here only there to avoid people getting upset. This is Asus thread so discussions should be Asus only...

More results just click on my signature down there



I know it's not a competition there is nothing to compete for, but sometimes it sounds like you underestimate engineers, if they build something to work with 6 Phase it's because they know what they are doing. Just because Asus has 16 Phase doesn't mean they can hold anything better. In fact as far as I'm concern the smallest route between to points is still a straight line...

Here:

[System Setup]
CPU: Q9650@3500Mhz 500x7
Voltages:
VCORE=1.29V
VTT=1.20V
VPLL=1.5V
VNB=1.26V
VSB=1.1V
VDIMM=2.24V
Mobo: Gigabyte EP45-UD3P
Cooler: Sunbeam Core Contact Freezer
Ram: 2 x 2Gb - 4gb G.Skill Pi 9600 @1200 5-5-4-15
Video: X1300
PSU: OCZ Powerstream 520w Adj


[Stability and Memory Benchmark Testing]
I can't show Prime95 now as I don't have 7hrs of the day, but testing was done using IBT Max and HCi Memtest 3.8 150%. If my systems passes those 2they can Prime95 all year. But I can Prime95 during the week if you insist.

Another note is that Everest OSD doesn't show much info on this board only the stuff is on screen so you have to see all voltages on ET6 (Easy Tuner 6).


IBT 35 Loops Maximum Stress starting - LOAD (I can only see pics of your system on Idle)
http://i566.photobucket.com/albums/s...R/a47f50c9.jpg


IBT 35 Loops Maximum Stress ending -IDLE
http://i566.photobucket.com/albums/s...R/67fcc5b3.jpg


HCI Memtest 3.8
http://i566.photobucket.com/albums/s...R/fff6555e.jpg


DDR2 500x7 FSB 1200Mhz Bandwith
http://i566.photobucket.com/albums/s...R/248a7998.jpg

This board is pretty much a joy to use, so easy you can hit and tweak and you can get the job done.

No matter how low volts you use this board will use less and remain stable and no matter how high fsb you want this board will keep breaking records with Quads.

Hope that settles the friendly inter-thread challenge

Nice work, that board is great and your quad is a monster, I saw some of your results with it (4.4GHz stable) and was very impressed. You have got the hang of the Gigabyte well for a DFI aficionado.

well im on the asus merry-go round once again.i call and ask to talk to the manager and the phone jockeys take my info and give me a new support ticket and they say someone will call,and of course if its like the last time no one does,so ill have to call again and do it all over again with someone else and the will tell me something similar about someone calling me but they never do so i call again this time im a little upset,cant explain myself clearly then they get some inkling of what im talking about and take some more info and tell me that they will have someone call me in the next hour,but no one does,and then guess what happens?nothing..........that is untill i call them again and again and again.

That's appalling service, I had no idea Asus were so bad. Heh, now I think back, Grnfinger actually convinced them to upgrade him after the nightmare he went through, and they sent him the wrong board. They can't even make up for their bad service & faulty hardware properly

Hope you have better luck with them.

[seban]

I got this strange issue with MF2 and downloading.

When I start a file with IDM it goes let's say 2.5mb/sec then next file from list would go 2.3mb/sec and after 1h file will go 1.5mb/sec and the decrease rate will gradualy slow down so it is less noticeable but still present.

Probably you would say that my provider is stuffed and that can happen but when I put some speed meter or I watch firewall activity it will state all the time 2.5mb/sec and in fact I got very stable connection and 95% of time it is 2.5mb/sec.

Also when I stop download the meters and firewall will still be showing full activity and it will not drop down - and those are also not fake values because when I play some on line games pings r affected as with 100% loaded connection.

I already tested both LAN ports and in addition USB Wifi Stick - all the same. On my previous NF4 board all was fine and whenever I dont use network I download 24/7 so I am sure there was no issue like this.

The only solution I discovered "for hanging full load" is to disable Network adapter which is currently in use and re-enable it again (all from system lvl). Then it is downloading again full speed and slowly slowing down.

Very irritating.

Anybody saw something like this on their systems? Can you test it?

Cheers

------------------

...
I just bought a new HDTV and have a blu-ray drive in the mail to me that I want to output to the TV.
I have a 4870x2, so I'm HDCP compliant there. However, I want to output the audio via the onboard Optical SPDIF to my 7.1 receiver.

Hi m8,

I dont have ATI card but as far as I know they got DVI to HDMI adapter and build in audio chip to passthrough PCM sound up to 5.1 (DTS and DD5.1) so I would do it this way on your place and forget onboard spdif. With HDMI it is cheaper and less cables also it is less stress and tuning then with onboard spdif.

Unfortunately neither your ATI or onbaord will be able to passthrough any 7.1 standard to your amplituner (so no MHD DTS and or HD DD 7.1 for you my friend). If you would like to pass through any of HD Audio formats you would need to buy ASUS 1.3 Audio card sorry cant remember full name at the moment.

As an alternative you can decode all formats on your PC and send them via analog cables but then your room with all cabling all over the place, would look more like rain forest rather than cozy little cinema (also if you would like to go this route then again you need good audio card for nice sound quality as all sounds would be generated by it - and MF2 free addition is kind of joke - for pass through you need anything even 5e card would do as all processing would be done by amplituner).

There are nice forums focused on HTCP and HDTV like http://www.avforums.com/ or http://www.avsforum.com/ you can get all knowledge and answerer you need there.

[ndGreg]

As stated in my last post, after testing to get NB GTL I had 3 values that stopped Prime @ 9 minutes. -40, -60, & -65.
Not sure if this would make any difference or be correct, but what I did was, Ran Everest Cache and Memory Benchmark, and found that -60 had the better latency.
So thats what I went with.
-40= 54.1ns
-60= 53.9ns
-65= 54.2ns

[Darknezz]

Hello CryptiK would you mind posting your bios settings so i can compare them with my own for i am having some problems getting it stable. 9x500 with Bios 1901 I can prime for about a 1½ hour and thats it....

[WaterFlex]

Jor3lBR

[CryptiK]

Hello CryptiK would you mind posting your bios settings so i can compare them with my own for i am having some problems getting it stable. 9x500 with Bios 1901 I can prime for about a 1½ hour and thats it....

Sure here it is.

CPU Clock: 4.23GHz
Processor: E8400 E0
BIOS: 1307

Ai Overclock Tuner: Manual
OC From CPU Level Up: Auto
Ratio CMOS Setting: 9
FSB Frequency: 470
CPU Clock Skew: Normal
NB Clock Skew: Normal
FSB Strap to North Bridge: 333
DRAM Frequency: DDR2-1129MHz

DRAM CLK Skew on Channel A1: Normal
DRAM CLK Skew on Channel A2: Normal
DRAM CLK Skew on Channel B1: Normal
DRAM CLK Skew on Channel B2: Normal
DRAM Timing Control: Manual

1st Information: 

CAS# Latency: 5 DRAM Clocks
DRAM RAS# to CAS# Delay: 5 DRAM Clocks
DRAM RAS# Precharge: 5 DRAM Clocks
DRAM RAS# Activate to Precharge: 15 DRAM Clocks
RAS# to RAS# Delay:  3
Row Refresh Recycle Time: 40 DRAM Clocks
Write Recovery Time: 6
Read to Precharge Time: 3

2nd Information:

Read to Write Delay (S/D): 8
Write to Read Delay (S): 3
Write to Read Delay (D): 5
Read to Read Delay (S): 4
Read to Read Delay (D): 6
Write to Write Delay (S): 4
Write to Write Delay (D): 6

3rd Information:

Write to PRE Delay: 14
Read to PRE Delay: 5
PRE to PRE Delay: 1
All PRE to ACT Delay: 5
All PRE to REF Delay: 5
DRAM Static Read Control: Enabled
DRAM Read Training: Enabled
MEM. OC Charger: Disabled
Ai Clock Twister: Moderate
Ai Transaction Booster: Manual

Common Performance Level: 8
Pull-in of CHA PH1: Disabled
Pull-in of CHA PH2: Disabled
Pull-in of CHA PH3: Disabled
Pull-in of CHA PH4: Disabled
Pull-in of CHA PH5: Disabled
Pull-in of CHB PH1: Disabled
Pull-in of CHB PH2: Disabled
Pull-in of CHB PH3: Disabled
Pull-in of CHB PH4: Disabled
Pull-in of CHB PH5: Disabled

PCIE Frequency: 100

CPU Voltage: 1.33750 (1.272v load in cpuz, 1.263v load in LCD POSTER)
CPU PLL Voltage: 1.50000
FSB Termination Voltage: 1.25900
DRAM Voltage: 1.95900
North Bridge Voltage: 1.27225
South Bridge 1.5 Voltage: 1.50000
South Bridge 1.1 Voltage: 1.10000

CPU GTL Reference (0): +60mv
CPU GTL Reference (1): +20mv
CPU GTL Reference (2): +60mv
CPU GTL Reference (3): +20mv
NB GTL Reference: -60mv
DDR2 ChA Reference Voltage: Auto
DDR2 ChB Reference Voltage: Auto
North Bridge DDR Reference: Auto

CPU Configuration:

Ratio CMOS Setting: 9
C1E Support: Disabled
Max CPUID Value Limit: Disabled
Intel Virtualization Tech: Enabled
CPU TM Function: Enabled
Execute Disable Bit: Enabled

Load-Line Calibration: Disabled
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled

[Darknezz]

CryptiK Thanks you. I think i have found the problem... Are you running your mobo with raid???

[Grnfinger]

Sure here it is.

CPU Clock: 4.23GHz
Processor: E8400 E0
BIOS: 1307

I see you run Load-Line Calibration: Disabled, clearly you are having alot of success running disabled. How much do you need to overvolt to compensate for vdroop?

[CryptiK]

CryptiK Thanks you. I think i have found the problem... Are you running your mobo with raid???

No I dont run raid, just a single drive at this stage. What do you think your issue is?

I see you run Load-Line Calibration: Disabled, clearly you are having alot of success running disabled. How much do you need to overvolt to compensate for vdroop?

I have 1.312v idle, 1.272v load (according to cpuz which reads slightly high) so 0.04v droop. Its fine for now, but I need around 1.32 - 1.336v for 4.3GHz stable so if I move up to that speed I will have to enable LLC to avoid going over spec at idle.

[xoldboyx]

Last run this time

grnfinger could you please post your settings for that 495 fsb?
I m trying to reach with no chance ,and still havent read the whole thread!
thx in advance!

[CryptiK]

Well I can bench 3D at 4500MHz at 1.336v. Don't laugh at the score it's just a 8800GT, I got a lowly 57th on HWBOT with this score. Better than with my old cpu though.

[radaja]

that is a really nice chip cryptik.very nice vcore.my e8500 needs at least 1.39 just to get into windows at anything above 4400ghz.i hope my new 9650 will be able to do great things at low volts

[Celcius]

I just got this mobo and noticed that it comes with an x-fi card. I've already got an x-fi xtreme music and I'm wondering if I should use the mobo's card and sell the xtrememusic or is the xtrememusic a better card?

thanks

[CryptiK]

that is a really nice chip cryptik.very nice vcore.my e8500 needs at least 1.39 just to get into windows at anything above 4400ghz.i hope my new 9650 will be able to do great things at low volts

Thanks I'm heat limited in going any higher than 4.23GHz for 24/7 it hits 68*C @ 1.268v at that clock and going to 4.3GHz pushes temps to 75*C+ which to me is unacceptable. Short benches like the one above are ok though it doesn't heat up too much.

Congrats on your Q9650, hope it goes well for you. There are some real gems out there.

[Evil_AnAnAs]

Hi, just bought this card for 100€ to break some WR with some p4 cpus. Can I use setFSB with this card? have been looking a long time for the clockgenerator but I can't find it. And If someone could give me a link to a page that could help med understand the settings with skew, like the cpu clock skew.

[radaja]

thanks cryptik,i hope i get one of those gems.what was the vid on your chip?is it one of the lower vids(the ones that run hot)?

[JoeBar]

I just got this mobo and noticed that it comes with an x-fi card. I've already got an x-fi xtreme music and I'm wondering if I should use the mobo's card and sell the xtrememusic or is the xtrememusic a better card?

thanks

Don't even bother with the included "X-Fi" card. Keep your XtremeMusic...

[Grnfinger]

Hi, just bought this card for 100€ to break some WR with some p4 cpus. Can I use setFSB with this card? have been looking a long time for the clockgenerator but I can't find it. And If someone could give me a link to a page that could help med understand the settings with skew, like the cpu clock skew.

http://www.ocztechnologyforum.com/fo...ad.php?t=40747

[CryptiK]

Hi, just bought this card for 100€ to break some WR with some p4 cpus. Can I use setFSB with this card? have been looking a long time for the clockgenerator but I can't find it. And If someone could give me a link to a page that could help med understand the settings with skew, like the cpu clock skew.

The clock gen is either ICS9LPRS918BKL or ICS9LPRS918JKLF, have a look at the chip that is half covered by the NB heatsink to the left of the NB itself, if you get the right angle you can read the number off it without having to remove anything.

thanks cryptik,i hope i get one of those gems.what was the vid on your chip?is it one of the lower vids(the ones that run hot)?

No strangely my chip is 1.25 VID but runs way hotter than other 1.25 VID chips I have seen, in fact its significantly hotter than my 1.225 VID chip so not sure what's going on there.

It's a bit irritating when a lapped true with mx-2, a solid mount and good fan can't keep up with cooling a dual core.