Printable View
Got bored today so I took the optional nb cooler from the rampage 3 extreme and got it to fit on the UD7, just not sure how I am going to plug that tiny fan in. This is now my Gigapage 7 Extreme
http://img709.imageshack.us/img709/8971/gigapage7.png
Uploaded with ImageShack.us
the only thing i don't like about the X58 waterblock is that it sits on fins. instead of directly on the chip.
Alright I moved the card a couple of days ago. I left the ATI card in the primary slot and moved the GTX 260+ card into the 3rd x16 slot and the Nvidia card now works and is detected. The cards are right next to each other now, which I didn't want as I was trying to keep them cooler by having them further away. I wonder if there is an issue with the bottom slot as I want to go Crossfire again at some point.Quote:
Originally Posted by SteveRo
btw Can anyone tell me the speed of each x16 slot? When I look at GPU-Z the Nvidia card is only running at x8, but I am pretty sure that 3rd slot is x16. The second x16 slot you can't use because it's always blocked if you have a dual slot cooler on your vid card, which 95% of high end cards do these days, so I assume that second slot is x8?
Just wanted to add to my previous post about the BIOS. Could this still be a BIOS issue with the Nvidia card not working in the 4th x16 (x8) slot? I did some googling and someone else had the problem with his Nvidia card not being detected (different brand MB though) when used with an ATI card as the primary video card and a BIOS update on his MB fixed the issue.Quote:
Originally Posted by Ice009
Hello,
after having used F7 now for a while I can safely say that it is no improvement on my D0 920.
1) It requires 1.325V on the core for 4 GHz. F6 works fine with 1.29
2) F7 requires 1.18V on the QPI PLL, F6 only 1.14. Too low voltage makes the POST a lot longer before the GPU BIOS is displayed.
3) Despite relative stability at 4 GHz, I am getting 'esoteric' problems on F7 despite not showing issues in Linx or any other stability tests. I had a BSOD 50 once on a shutdown, occasional vanishing of part of my systray icons in Win7-x64, and maybe other more subtle issues.
I somehow suspect that part of the problem is the changed LLC.
Any suggestions or confirmations on what I saw here?
I finally got 4.8ghz stable here are my highest benchmark so far.
http://i753.photobucket.com/albums/x...4/DSC_0662.jpg
http://i753.photobucket.com/albums/x...4/DSC_0665.jpg
http://i753.photobucket.com/albums/x...4/DSC_0643.jpg
http://i51.tinypic.com/oaarys.png
I reached the limit values of the voltages of the options in the BIOS @ 4.0GHz i7-975.
VTT 1.335v
IOH 1.18v
DRAM 1.64v 1600MHz 6-8-6-20 1T
PLL 1.86v
IOH I/O 1.58v
vCore 1.30000v (CPU-Z 1.28v)
http://img201.imageshack.us/img201/1...128v686208.png
i7 975 is not the best for overclocking, I'm currently testing with 4.2GHz 1.40000v BIOS (1.392v CPU-Z) less than that, blue screen occurs in the system.
REAPER, Very nice looking rig :up:, nice oc also.
I seem to have a problem with my usb 3.0 every time i connect my hdd dock to my usb 3.0 port on the m/b it pops up in windows saying drive can perform fastest if plugged into a super speed 3.0 usb it is isnt it? the other problem i have is when i restart my pc i lose the external drive and have to turn in off and on again to get it to reconnect
After played with my lovely Q9550 and UD3P, I'm planning to upgrade with i7 1366 (still can't decide which one, maybe i7 950). Please give me suggestions which mobo that I should buy, still can't decide:
1. UD7 rev.1.0 -> I found the second one with very good conditions
2. UD7 rev 2.0 -> brand new in a box
3. UD5 rev 2.0 -> brand new in a box
4. UD3R rev. 2.0 -> brand new in a box
5. Asus Rampage III Gene -> brand new in a box
In my country, price range for UD3R rev 2.0 and R3 Gene is closely match, even you guys in US/Europe said on the web review that UD3R is bang for the bucks, but the fact in my country, UD3R is way too pricey. Speaking of the performance for oc, is it any difference between rev.1.0 and 2.0 for UD7?
Many thanks for give me the enlightment :yepp:
Well, to my experiences:Quote:
Originally Posted by Sadhana
1) There is no real difference between a 920 or 930 and a much more expensive 940/950/960. They all hit 3.8-4 Ghz at 1.3V, get equally hot, and require watercooling beyond 4 GHz. Even a 975X usually doesn't go much further, if at all. (same production process. 45 nm, go figger)
2) UD7 V1 vs. v2: I haven't seen comments here on Rev. 2 clocking better than V1 overall. It is more of a factor if your particular CPU and board combination work well together but that's nothing you can predict. The V2 has less power phases for the CPU which is theoretically not as good, yet even 16 phases is kinda beyond good and eveil making it questionable if there is a difference at all. Other modifications may affect the 980X (V2 has the 6 Ghz mod AFAIK, not sure)
3) UD5 or UD3R: no personal experiences but as far as I know even the UD3R has the same BIOS overclocking options as the UD7. If you are happy with BCLK 200 or a bit above any of the boards will do, also the ASUS ones. Extreme overclocking above 210 maybe easier on the UD7 or even UD9....it doesn't look feasible to me tho for 24/7 use.
As for ASUS boards, I had a P6T Deluxe V2, it was going to BCLK 200 too without much hassle. What took me voer to Gigabyte is that the GA boards are having less or no stupid design kinks.
- enough SATA ports on GA. P6T D V2 had enough theoretically, jsut some of the JMB383 controller plainly had no plug, just one esata.
- PCI Express slot arrangement is stupid on many ASUS Gamer boards unless you pay beyond 300 Eur. PCI slot often unusable.
- less BIOS bugs
And a few more. I don't want those if I am paying 250 Euros already.
Got my UD7 rev.2 today, but didn't test it yet.The only thing , there is only 1 fanconnector around the cpu.
Well, I managed to stabilize the i7-975 4.0GHz with voltage to 1.28V (CPU-Z) 1.30000V (BIOS) running in the past 20 LinX using all the memory without problems. Now I try to run past the 50 LinX with the same settings, but I can not in any way, always giving errors between the fifth tenth ~ twentieth in the last LinX. I did some different combinations in the BIOS and the problem persists.
I wonder if anyone has any tips for me to get those 50 run in the past LinX. Thanks!
vCore 1.30000v
VTT 1.335v
IOH 1.18v
ICH I/O 1.58v
PLL 1.86v
DRAM 1600MHz 6-8-6-20 1.64v (Uncore 3200)
http://img843.imageshack.us/img843/1...128v686208.png
Looking to get a UD7 rev. 2 but had a few questions:
Will a w3680 work on it? (I think so, just want to make sure)
Which works better - the Silent pipe or waterblock setup on the NB? (or both at the same time?)
Has anyone tried the new OC bios out for the rev. 2 board (GOOC2010.W08) to see if it is better (for OC'ing) than the F6/7 bios?
Thx
Dr_DX
Sure working well for W3680;) and i suggest GO OC BIOS for overclocking.Quote:
Originally Posted by dr_dx
Thanks hicookie!
Looking forward to "updating" my primary system.
One more thing: I noticed in some earlier posts that there seemed to be an issue with SSD's w/ Sandforce controllers. Was this in a RAID or non-RAID configuration? If in a RAID config, was the issue fixed?
I notice on Gigabyte's site they say not to run RAID SATA3 SSD's on the Marvel SATA3 ports, which I wasn't going to anyway. I have 3 sata2 Sandforce SSD's in RAID0 and want to run them on the sata2 ports. Any problem with this?
THX
Dr_DX
I have 2 x HD5970 wich means 4 gpu processors what should I set voltage to according to this ->
Quote:
bclock voltages
For this step, there are only two voltages you should play with; VTT, and IOH Core. IOH Core voltage is easy, if you are running a single PCIe card (graphics card), give the IOH Core 1.1V, two graphics cards 1.15V, three graphics cards 1.2V, or four graphics cards 1.25V. If you plan to run bclock frequencies far beyond 200MHz, sometimes it may help for the IOH core to be even higher.
Got problems installing xp32bit, after first reboot , I get a drive failure.. Someone also has the problem?
W7 64bit is no problem for install.
What kind of drive failure? Yellow bubble on an optical drive?Quote:
Originally Posted by Bullu$
I had that often in XP/Server 2003 after installing intel's AHCI/RAID drivers. Just install the generic optical drive driver and you'll never see it again....
Here is a thread i found in OCZ Forums, maybe helps you.Quote:
Originally Posted by dr_dx
http://www.ocztechnologyforum.com/fo......look-inside
Do i understand it right ? The F7 final bios should have been fixed and there istn't a sandforce error anymore ?
It appears that the SMART issue was solved for both rev. 1 & rev. 2 boards, but it didn't say which ports they were using. Based on what Gigabyte is saying (see my earlier post) it sounds like there is still an issue with them on the Marvel sata3 ports.
No, I can't even get into windows. After the post I get that failure.Quote:
Originally Posted by Amurtigress
I will than oc in W7 for superpi...:mad:
Tried 2 mobo for high blck,
rexII; max blck 215
UD7; max blck 221, set everything to low, exept qpi-link freq.
Try to use higher multi, but I'm stuck at 210x23...
I'll post the voltages soon.
edit
At these settings I get max 221 @ multi 19+21Code:Advanced CPU Features:
CPU Clock Ratio ................................ [19 and 21 ]
Intel(R) Turbo Boost Tech ...................... [dis]
CPU Cores Enabled .............................. [2]
CPU Multi Threading .............................[dis]
CPU Enhanced Halt (C1E) ........................ [dis]
C3/C6/C7 State Support ......................... [dis]
CPU Thermal Monitor ............................ [dis]
CPU EIST Function .............................. [dis]
Virtualization Technology ...................... [dis]
Bi-Directional PROCHOT ......................... [en]
Uncore & QPI Features:
QPI Link Speed ..............................[7956]
Uncore Frequency ............................[3757]
Isonchronous Frequency ......................[en]
Standard Clock Control:
Base Clock (BCLK) Control .......................... [en]
BCLK Frequency (MHz) ................................[221]
PCI Express Frequency (MHz) ........................ [108]
C.I.A.2........................................... .. [-]
Advanced Clock Control:
CPU Clock Drive ..............................[800]
PCI Express Clock Drive ................... [900]
CPU Clock Skew ............................. [0]
IOH Clock Skew ............................. [0]
Advanced DRAM Features:
Performance Enhance ...................... []
Extreme Memory Profile (X.M.P) ........... []
System Memory Multiplier (SPD) ........... [ ]
DRAM Timing Selectable (SPD) ............. []
Channel A + B + C
Channel A Timing Settings:
##Channel A Standard Timing Control##
CAS Latency Time ......................[9]
tRCD ..................................[9]
tRP ...................................[9]
tRAS ..................................[24]
CR.....................................[]
B2B.....................................[]
Advanced Voltage Control:
CPU Intel Core i7 920 D0 Stepping
Load Line Calibration ................. [level2]
CPU Vcore ..............................[1.55]
QPI/VTT Voltage 1.150v .................[1.5]
CPU PLL 1.800v .........................[1.4]
PCIE 1.500v ...........................[1.5]
QPI PLL 1.100v ........................[1.14]
IOH Core 1.100v .......................[1.10]
ICH I/O 1.500v ........................[1.5]
ICH Core 1.1v .........................[1.1]
DRAM
DRAM Voltage 1.500v .............. [1.66]
DRAM Termination 0.750v.......... [AUTO]
Ch-A Data VRef. 0.750v ...........[AUTO]
Ch-B Data VRef. 0.750v ...........[AUTO]
Ch-C Data VRef. 0.750v ...........[AUTO]
Ch-A Address VRef. 0.750v ........[AUTO]
Ch-B Address VRef. 0.750v ........[AUTO]
Ch-C Address VRef. 0.750v ........[AUTO]
But when I use multi 23 ,I only get max. 216. Tried to raise Vcore, qpi vtt and pci-e but it doesn't help.:shrug:
Temps are ok, max. 45degrees.
Hi,Quote:
Originally Posted by dr_dx
I read a long article on SSDs on onboard controllers. They used a huge RAID made out of 8 Intel SSDs a couple of months ago on Tom's hardware (I think, might also have been on Anandtech), including our Marvell chip.
The result was that the Intel ICH10R was the clearly best performing onboard controller they used despite not supporting SATA 3. The only superior devices were expensive RAID cards.
The marvell chip was clearly behind the ICH10R (Maybe due to being connected through PCI Express? Not sure) along with the AMD chipset controllers on Socket AM2+/AM3 systems.
You might be better off using the ICH10R controller despite losing the SATA 3 throughput.
Just a thought...sorry for not having a link to that test, but you might want to look into that option.
Sorry about that, I was answering the poster before me and forgot to quote him. I am planning on using the ICH10R ports for my Sandforce sata2 SSD's - see post #2016.Quote:
Originally Posted by Amurtigress
On several forums, I saw where Gigabyte MB's were getting a S.M.A.R.T error on raided SSD's that use the Sandforce controller. Since other MB makers weren't, the problem seemed to be with Gigabyte. It appears they realized it around early Aug. and by early Sept. they had issued new bioses to correct the problem. In post #2016, we see that Gigabyte doesn't recommend raided SSD's on the Marvel ports - most likely, as you say, for performance reasons.
I am still a little bummed, because I wanted to to put my 2 2TB Sata3 Seagates on them for a performance boost, but that doesn't look very promising. :shrug:
So it's only with Sandforce based drives in RAID that there is an issue?