**Unified** DFI LanParty UT 790FX-M2R-M2RS / Phenom /Data Base Thread**

[justapost]

If i just up the HTT i can only do 203.
If i up the HTT+BHTT then things proceed as expected but i never used BHTT before you mentioned it as there was no need with the 6400BE.

Glad it helped. Good to know i can skip BHTT with the 5000BE next time.

@Oese: Thank you for your watercooling tips. I ordered such an mosfet cooler. Should arrive tomorrow, i'll report here how it works.

[Oese]

yeah i'm curious as well and pics please

and could you confirm please that htt oc would work with bhtt set? or is it like may or may not...

[Final8ty]

so if setting the bhtt right there is no problem running high htt? then i'd test some 9550 or so...

No idea as that was only about getting the MY 9950 to post at a given FSB.
But so far my 9950 don't like high FSB.

[Oese]

203mhz is also not what i consider "high" of course..

something around 260mhz htt would be nice with some cheap 9550 phenom.. but phenoms in general dont seem to like high htt..

[justapost]

Hmm I found a gap ~10MHz is max betwen HTT and BHTT i'll test that now.
I use an HT speed of 1.4GHz a lower NB multi and high tRFC's (195ns) and have NP running 250-260MHz ref HT.
I'll test how far i can get and will post sucess later.

@Final8ty have you played with HTT Gap? Someone reported it helped using a smaller gap than 5. Maybe your triple crossfire setup needs that too.
Beside that have you other ram to test with or tried DDR800/DDR667 mode? DDR800 worked best here dunno about DDR1066 my Tracers are completely unusable atm.

[Oese]

sounds nice.. tweaking is always good if it yields success thats what dfi boards are for..

[justapost]

First i tried was 200MHz BHTT and 210MHz HTT. System stuck at 25.

I used those settings and upped both ref HT's in steps of 10MHz. 260MHz did not post with those settings, i'll see if i can go further anyhow. I use 4x1GB Dimms atm maybe with two i'll get better results.

>HT Link Control:
 
HT Link Width......................... - 16 bit
HT Frequency.......................... - 1.6GHZ MHz
IH Flow Control Mode.................. - Disabled
HT Link Tristate...................... - Disabled
2x LCLK Mode.......................... - Disabled
Unit Id Clumping...................... - Disabled
 
>CPU Feature:
 
CPU-NB DID........................... - 1
CPU-NB FID........................... - 04
AMD CPU stepping..................... - P0

>DRAM Configuration:

MCT Memory Timing.................... - Manual
2T Mode.............................. - Auto
Cas Latency Time..................... - 5
Pre Charge Delay (tRAS).............. - 15
DRAM RAS# Precharge.................. - 5
DRAN RAS# to CAS# Delay.............. - 5
Bank to Bank Cmd (TRRD).............. - Auto
TRC.................................. - Auto
TRTP................................. - Auto
Write Recovery (TWR)................. - Auto
Write-Read Command (TWTR)............ - Auto
TRFC0................................ - 195ns
TRFC1................................ - 195ns
TRFC2................................ - 195ns
TRFC3................................ - 195ns
Trdrd................................ - Auto
TrwtT0............................... - Auto
Twrrd................................ - Auto
Twrwr................................ - Auto
CS/ODT Pin Fine Delay................ - Enabled
Bank Swizzle Mode.................... - Enabled
Auto Tweak Performance............... - Disabled
Optimal Performance Mode............. - Disabled

>ODC Control Mode:

DRAM Drivers Weak Mode.............. - Normal drive
Output Drivers Comp................. - Auto

>PWM Settings:
 
PWM Control.......................... - Enabled
CPU VID Control...................... - 1.30000v
CPU Vid Special Add.................. - Auto
CPU NB Vlotage Control............... - 1.30000v
CPU NB Vid Special Add............... - Auto
DRAM Voltage Control................. - 2.07v
SB PLL 1.2V Voltage.................. - 1.20v
NB Core Voltage...................... - 1.160v
NB PCI-E Voltage..................... - 1.160v
NB HT Voltage........................ - 1.130v

 
Genie BIOS Settings:
 
Timing Mode.......................... - Manual
Memory Clock value or Limit.......... - DDR 800
PCIE 3,4 State....................... - Auto
PCIE 2 State......................... - x2
CPU TLB Cache Fix.................... - Auto
AMD K8 Cool & Quiet control.......... - Disabled
CPU Clock Ratio...................... - Auto
Spread Spectrum...................... - Disabled
CPU Host/HTT Clock................... - 250mhz
CPU Host/HTT Boot Up Clock........... - 250mhz
CPU Host/HTT Adjustment Gap.......... - 5mhz
PCIE Clock........................... - 100mhz

[informal]

Hi justapost,thanks for the 9950 results
Did you try it in your M3A motherboard?Any luck with OC attempts?I wonder how will 5-phase pwm react to this chip

[justapost]

Hi justapost,thanks for the 9950 results
Did you try it in your M3A motherboard?Any luck with OC attempts?I wonder how will 5-phase pwm react to this chip

Hi informal, i'm also curious if and how that chip will work in the M3A. I have not yet tried it. I'm waiting for new watercooling parts and want to sort out that +8°C issue i have since a few days first. I fear the chip did not like my 1.6V+ testings and degraded abit.Before that 3GHz at 1.3625V and 3.1GHz at 1.425V was stable under winxp 32bit and linux 64bit. 3.2GHz at 1.475V was winxp 32bit stable but not under linux 64bit even at 1.5V. 3.4GHz at 1.625V was benchable under winxp 32bit.

[Final8ty]

Hmm I found a gap ~10MHz is max betwen HTT and BHTT i'll test that now.
I use an HT speed of 1.4GHz a lower NB multi and high tRFC's (195ns) and have NP running 250-260MHz ref HT.
I'll test how far i can get and will post sucess later.

@Final8ty have you played with HTT Gap? Someone reported it helped using a smaller gap than 5. Maybe your triple crossfire setup needs that too.
Beside that have you other ram to test with or tried DDR800/DDR667 mode? DDR800 worked best here dunno about DDR1066 my Tracers are completely unusable atm.

Yep i dropped the ram to 800 & that did not help at all & would go against what i was trying for as i have my max CPU over clock & wanted to over clock the ram past its default 1066

[informal]

Hi informal, i'm also curious if and how that chip will work in the M3A. I have not yet tried it. I'm waiting for new watercooling parts and want to sort out that +8°C issue i have since a few days first. I fear the chip did not like my 1.6V+ testings and degraded abit.Before that 3GHz at 1.3625V and 3.1GHz at 1.425V was stable under winxp 32bit and linux 64bit. 3.2GHz at 1.475V was winxp 32bit stable but not under linux 64bit even at 1.5V. 3.4GHz at 1.625V was benchable under winxp 32bit.

Thanks man.When you resolve the cooling issues and have some spare time,test it in M3A if possible (and cool those mosfets! )
Are you sure about the chip degrading?Could it be a lower efficiency of your water cooling system-hence higher cpu temps and lower OCs,as you mentioned you had some issues with it?

Cheers!

[Oese]

@ justapost: looks nice.. thank you as well..

[justapost]

I ran the universe suite from the phoronix-test-suite with 10x250MHz vs. 12.5x200MHz yesterday. Means the CPU ran at 2.5GHz, NB and HT at 2GHz and only the mem at 400Mhz vs 480MHz with the same timings.

Here are the results.
http://global.phoronix-test-suite.co...4326-730-12625

@Final8ty: How far did the ref HT go at 1066mode, 203MHz? With two 1GB Tracers I maxed out at ~630MHz and ~615MHz stable. I'd be interested if it works better with just two 2GB dimms (assuming you use four).

Thanks man.When you resolve the cooling issues and have some spare time,test it in M3A if possible (and cool those mosfets! )
Are you sure about the chip degrading?Could it be a lower efficiency of your water cooling system-hence higher cpu temps and lower OCs,as you mentioned you had some issues with it?

Cheers!

Temps where high the first time i mounted the chip. I guess thats why my power consumption comparison results where that high. I thought i fixed it by reseating the D-Tek but it seems after a few hours of usage the temps went up again. Can be the chip does fluctate that much. At 2.6GHz only CoreDamage pushes the cpu temp to 60°C with prime95 i'm at ~50°C most of the time with air cooling (BigTyphoon).

[Oese]

and those 4x1gb did no adverse effect? i use 4x1gb too and luckily the 6400+ handles it extremely well, the 3800+ (which is crap anyway) doesnt like them..

[justapost]

and those 4x1gb did no adverse effect? i use 4x1gb too and luckily the 6400+ handles it extremely well, the 3800+ (which is crap anyway) doesnt like them..

Using two instead of four dimm's give me ~10MHz better ref HT results.

[Final8ty]

I ran the universe suite from the phoronix-test-suite with 10x250MHz vs. 12.5x200MHz yesterday. Means the CPU ran at 2.5GHz, NB and HT at 2GHz and only the mem at 400Mhz vs 480MHz with the same timings.

Here are the results.
http://global.phoronix-test-suite.co...4326-730-12625

@Final8ty: How far did the ref HT go at 1066mode, 203MHz? With two 1GB Tracers I maxed out at ~630MHz and ~615MHz stable. I'd be interested if it works better with just two 2GB dimms (assuming you use four).

I have done tests with only 2 sticks & it makes no difference for my setup.
My 6400BE with 4x2GB sticks & mem setting at 800 would do 275-280 FSB.

[justapost]

I have done tests with only 2 sticks & it makes no difference for my setup.
My 6400BE with 4x2GB sticks & mem setting at 800 would do 275-280 FSB.

So what is the max you can get with DDR1066, 9950BE and 8GB? I'm curious because I plan to buy 2GB dimms as a replacement for the tracers.

[Final8ty]

So what is the max you can get with DDR1066, 9950BE and 8GB? I'm curious because I plan to buy 2GB dimms as a replacement for the tracers.

I don't think you can judge what it would be like for you because the Phenom chips are so inconsistent & the fact that my cpu will not do anything near your
Mhz or FSB no matter the Vcore or the amount of dimms in the slots.

[justapost]

I don't think you can judge what it would be like for you because the Phenom chips are so inconsistent & the fact that my cpu will not do anything near your
Mhz or FSB no matter the Vcore or the amount of dimms in the slots.

This is my fifth phenom and all the previous ones where medium quality clockers. But they all reached at least 225MHz with ram at DDR1066 with low tRFC's and high HT/NB multis.
Any chance your share your results?

[Final8ty]

This is my fifth phenom and all the previous ones where medium quality clockers. But they all reached at least 225MHz with ram at DDR1066 with low tRFC's and high HT/NB multis.
Any chance your share your results?

I have been upto FSB216 i was aiming for 1150-1200

[Final8ty]

im going to give your settings a try.

[Oese]

Using two instead of four dimm's give me ~10MHz better ref HT results.

Ah thx..

[justapost]

HD995ZFAJ4BGH JAAHB AA 0820APMW

just arrived.

For comparison 9850 power consumption and temps at stock:

Idle: 153,7W CPU 21°C CORES 33°C
Load: 243 W (everest julia)
Load: CPU 42°C CORES 45°C (coredamage 5mins)

9950BE at 2.5GHz 1.3V

Idle: 169W CPU 27°C CORES 26°C
Load: 267W (everest julia)
Load: CPU 52°C CORES 45°C

So the 9950BE requires ~15W more than the 9850BE here, temp reading seems more correct on the 9950BE.

With my temp issues fixed i reran the power test

Idle: 152,5W CPU 22° CORES 23°C (lower core temps than the 9850 and equal cpu temps now).
Load: 244W (everest julia)
Load: CPU 47°C CORES 37°C

So my results are opar with most reviews showing an nearly equal power consumption.

[justapost]

I have been upto FSB216 i was aiming for 1150-1200

Hmm 533MHz * 216/200 = 1151,28MHz

[Oese]

With my temp issues fixed i reran the power test

Idle: 152,5W CPU 22° CORES 23°C (lower core temps than the 9850 and equal cpu temps now).
Load: 244W (everest julia)
Load: CPU 47°C CORES 37°C

So my results are opar with most reviews showing an nearly equal power consumption.

should be same consumption i think?

btw, did the pwm-watercooler arrive?