DFI NF4 Thread

[xgman]

Originally Posted by Sierra

The 2.4-2.6GHz range doesn't sound very impressive. Someone else can be the guinea pig for M2. I'll wait for the smoke to clear.

I agree with that. It is taking AMD far to long to come over the 3ghz stock fence. Are any of these M2's dual core?

[CrowdAds]

[Derf_abx]

Hi,
New to posting but have been around quite sometime gathering info for my latest build as per my sig. It runs great while everything is stock but as soon as I overclock 3D Mark 03 and 05 stop working. Locks up shortly after the opening screens.

I runPrime 95 to check if the cpu is at fault but prime runs all night long with no errors. I was wondering if my Enermax PSU is not cutting it. It is not a version 2.0 therefore I am running a 20-24pin adapter. From reading the threads I know I will be chastized for running this but there were no decent PSU's in my local area for immmediate purchase so I went this route for now.

I was running the same cpu on an A8V-D at 11x255. vcore of 1.575. Could run 3dmark 03 and 05 with no problems. Mind you I was using my 6800GT with it.

Thanks

Derf

[Sierra_abx]

I believe there will be dual-core M2 processors. Dual-core seems to be the direction everyone is headed. What AMD needs to do is lower the price so that more people would be willing to try them.

[gocchin]

Originally Posted by Sierra

I got my Mushkins (2X512) up to 277 (SuperPi 1M stable) with 704-2BTA. May just be your sticks. How much voltage are you giving them? I needed 3.8v. I've run them as high as 4v. :eek:

Wow 277!!! I was trying 255 @ 3.5v and 3.6v and blue screening with 704-2BTA.

I just completed 32m @ 255 @ 3.4v with 623-3. Time was around a second better than using 704-2BTA (but it's probably the tweaks you gave me )

Anyway the stable lower voltage is a better sign for me anyway

[Sierra_abx]

Originally Posted by Derf

Hi,
New to posting but have been around quite sometime gathering info for my latest build as per my sig. It runs great while everything is stock but as soon as I overclock 3D Mark 03 and 05 stop working. Locks up shortly after the opening screens.

I runPrime 95 to check if the cpu is at fault but prime runs all night long with no errors. I was wondering if my Enermax PSU is not cutting it. It is not a version 2.0 therefore I am running a 20-24pin adapter. From reading the threads I know I will be chastized for running this but there were no decent PSU's in my local area for immmediate purchase so I went this route for now.

I was running the same cpu on an A8V-D at 11x255. vcore of 1.575. Could run 3dmark 03 and 05 with no problems. Mind you I was using my 6800GT with it.

Thanks

Derf

Hi and welcome to ABX. Rather than using an adapter, try plugging the 20-pin connector into the board. As long as you have strong rails, you should be fine. Not recommended though if you're running SLI.

[Derf_abx]

Hi Sierra,

Thanks I will give that a try. I looked at the 20 pin plug earlier and was not sure it would work with the board hence the adapter. I am not running SLI and the PSU puts out +3.3@35A, +5@35A and +12@33A.

Derf

[gocchin]

Can someone remind me how the Tref Refresh number in the bios relates to the Tref in A64 Tweaker?

Thanks!

[Sierra_abx]

Originally Posted by Derf

Hi Sierra,

Thanks I will give that a try. I looked at the 20 pin plug earlier and was not sure it would work with the board hence the adapter. I am not running SLI and the PSU puts out +3.3@35A, +5@35A and +12@33A.

Derf

Your rails have plenty of current. It's worth a try. The 4-pins that need to be left empty are the ones closest to the corner of the board, near the 4v jumper. Also, if you haven't done so already, plug a Molex connector into the board for some added power.

[Sierra_abx]

Originally Posted by gocchin

Can someone remind me how the Tref Refresh number in the bios relates to the Tref in A64 Tweaker?

Thanks!

Refresh Period(tREF)

Settings = Auto, 0032-4708 in variable increments.

1552= 100mhz(?.?us)
2064= 133mhz(?.?us)
2592= 166mhz(?.?us)
3120= 200mhz(?.?us)(seems to be a/ Bh-5,6 sweet spot at 250+mhz)
---------------------
3632= 100mhz(?.?us)
4128= 133mhz(?.?us)
4672= 166mhz(?.?us)
0064= 200mhz(?.?us)
---------------------
0776= 100mhz(?.?us)
1032= 133mhz(?.?us)
1296= 166mhz(?.?us)
1560= 200mhz(?.?us)
---------------------
1816= 100mhz(?.?us)
2064= 133mhz(?.?us)
2336= 166mhz(?.?us)
0032= 200mhz(?.?us)
---------------------
0388= 100mhz(15.6us)
0516= 133mhz(15.6us)
0648= 166mhz(15.6us)
0780= 200mhz(15.6us)
---------------------
0908= 100mhz(7.8us)
1032= 133mhz(7.8us)
1168= 166mhz(7.8us)
0016= 200mhz(7.8us)
---------------------
1536= 100mhz(3.9us)
2048= 133mhz(3.9us)
2560= 166mhz(3.9us)
3072= 200mhz(3.9us)
---------------------
3684= 100mhz(1.95us)
4196= 133mhz(1.95us)
4708= 166mhz(1.95us)
0128= 200mhz(1.95us)

[gocchin]

Originally Posted by Sierra

Refresh Period(tREF)

Settings = Auto, 0032-4708 in variable increments.

1552= 100mhz(?.?us)
2064= 133mhz(?.?us)
2592= 166mhz(?.?us)
3120= 200mhz(?.?us)(seems to be a/ Bh-5,6 sweet spot at 250+mhz)
---------------------
3632= 100mhz(?.?us)
4128= 133mhz(?.?us)
4672= 166mhz(?.?us)
0064= 200mhz(?.?us)
---------------------
0776= 100mhz(?.?us)
1032= 133mhz(?.?us)
1296= 166mhz(?.?us)
1560= 200mhz(?.?us)
---------------------
1816= 100mhz(?.?us)
2064= 133mhz(?.?us)
2336= 166mhz(?.?us)
0032= 200mhz(?.?us)
---------------------
0388= 100mhz(15.6us)
0516= 133mhz(15.6us)
0648= 166mhz(15.6us)
0780= 200mhz(15.6us)
---------------------
0908= 100mhz(7.8us)
1032= 133mhz(7.8us)
1168= 166mhz(7.8us)
0016= 200mhz(7.8us)
---------------------
1536= 100mhz(3.9us)
2048= 133mhz(3.9us)
2560= 166mhz(3.9us)
3072= 200mhz(3.9us)
---------------------
3684= 100mhz(1.95us)
4196= 133mhz(1.95us)
4708= 166mhz(1.95us)
0128= 200mhz(1.95us)

Thanks Sierra, I needed that .

[Sierra_abx]

Try using 4708 or 3684. See if that helps. Also loosen Max Async Latency to 7ns.

[Derf_abx]

Well I have the board running on 20 pin ATX power and it doesn't seem to make a difference. The 4 pin molex and the floppy pwr cables are connected to the board. I did that right from the beginning after reading the threads concerning stability. I can run 11@225 without 3dmark crashing but as soon as I tried to raise the core of the GPU 3dmark crashed again. Even though the GPU worked at 450 core while the CPU was at stock settings.

The GPU as well is running on the adapter from BFG. TOOOOOOOOO many adapters me thinks. Will look for a new PSU and leave the machine at stock timings till I do.

The box has 2x 80 mm fans, 2x120mm fans, fan controller, water pump plus the rest of the computer. I have tried to estimate the wattage required but the calculators I have found online seem a little dated.

Thanks for your help.

Derf

[Eldonko]

Originally Posted by Derf

I can run 11@225 without 3dmark crashing but as soon as I tried to raise the core of the GPU 3dmark crashed again. Even though the GPU worked at 450 core while the CPU was at stock settings.

I am confused. Are you OCing the GPU and CPU at the same time? If your CPU OC is stable, it may be heat issues from the GPU OC.

[Derf_abx]

I have had the 3700 SD in my A8V-D stable at 2.8(11@255). While trying some benchmarks with my new rig 3DMark would lock up. Thats was with the CPU at 11*235 and the 7800GT at stock sttings. All temps seemed good albeit I am using speedfan to monitor those.

CPU temp 11*200 = 38c while priming
CPU temp 11*225= 39 c while priming
CPU temp 11*255= 46 c while priming...will run all night long without errors

7800GT temp immediately after running a successfull 3DMark was 47 c.


It just seems that as soon as I overclock the CPU a little and try to run 3Dmark 03/05 the vid card craps out. Screen is all dark though you can barely see the scene. But yet I can prime no problem. I thought it was a power issue with not enough juice to go around after I start to bump the CPU.

[Eldonko]

Could be that yeah. Make sure your CPU/RAM OC is 100% stable before OCing the GPU. Which Prime test have you been using, small FFT? Give blend a try if you haven’t already.

Kinda OT but I thought this post was interesting:

Originally Posted by RyderOCZ @ XS

TCCD when performing certain tasks (memtest #5 and #8 for instance) actually run's hotter than BH/CH chips with 3.5V running through them in my experience. They pull more amperage than many IC's do.

I noticed that my Plat Rev 2 ran really hot and the UTT I have is much cooler. Just shows that you should active cool all RAM.