Gigabyte EX58-UD5/Extreme Discussion Thread

[rge]

F5b same problem as F4 and F5a on mine. Any bclk above 170, and I can no longer reboot, just get blank screen with computer running. Have to hit the reset key to get into windows. I wrote GB and asked if they would pass on to bios manufacturers about the high bclk bug that was introduced between F4m and F4, and persists in F5a that affected a lot of people (and now F5b). They wrote back saying that OCing is not guaranteed, but would pass the info along. If GB Extreme (and UD5) bios testers are not OCing these things and testing them that way, and they dont support Ocing, then maybe GB could change the name of the mobo to GB Stock instead of GB Extreme, so the public would know.

It is really going to annoy me, if that bug persists in all future bioses, and F4m is as far as it goes with OCing.

[safan80]

here's a quick linpack run (only 10) on bios F5c with the same settings used in my f4kr/f4r/f5a bioses

note: the cold boot issue seems to take less time to boot than any of the other bioses

[vega22]

F5b same problem as F4 and F5a on mine. Any bclk above 170, and I can no longer reboot, just get blank screen with computer running. Have to hit the reset key to get into windows. I wrote GB and asked if they would pass on to bios manufacturers about the high bclk bug that was introduced between F4m and F4, and persists in F5a that affected a lot of people (and now F5b). They wrote back saying that OCing is not guaranteed, but would pass the info along. If GB Extreme (and UD5) bios testers are not OCing these things and testing them that way, and they dont support Ocing, then maybe GB could change the name of the mobo to GB Stock instead of GB Extreme, so the public would know.

It´s really going to annoy me, if that bug persists in all future bioses, and F4m is as far as it goes with OCing.

I said above that this problem can have other side. I had exactly the same problems as you when i upgraded from F4j to F4, but yesterday when I upgraded from F4m to F5a I had no problem in reaching the 211 bclk, and none of the serious stability problems that I had before (only worse performance now). It´s very difficult for me to explain this in English, but I think it is something related to how the board adjusts its internal values, specially for memory, after a flashing or a re-setting. For example, working in the same Bios, I have found that is much more stable and safe set the voltage first and then go up the bclk than set the bclk first and then try to adjust the voltage (normal procedure).
At the bottom, I can only say that with this new procedure the board is fully Linpack stable @ 1.34v Qpi, while in the "old fashion" way is impossible to descend under 1.40v Qpi, and now when upgrading my bios i not have none serious problem.
I hope that at least some of what Im trying to say is understandable...

[burningrave101]

So noone else is currently experiencing this problem I have?? It all started out when I had a failed overclock. Now no matter how many times I reboot, shutdown, clear the CMOS for 10s, boot into safe mode and reboot, ect. it blue screens and restarts when it starts to load Vista Ultimate x64. It gets to the part where the green bar is running across the screen and as soon as it finishes that it acts like it's going to go and then blue screen.

I'm also having a problem where at times it's restarting twice during POST instead of only once. Starts up, shuts down, starts up, away we go.

What's the deal with this board? I know I seen at least one other guy having some serious issues like this. Is it corrupting the OS install on the failed OC and then from then on it's just an endless cycle till you reinstall the OS again? I've already reinstalled it once when I had a similar issue yesterday. Do I need to make sure I bump up the PCI-E speed to 101Mhz and use a specific SATA port for my hard drive so that this doesn't happen?

[rge]

I said above that this problem can have other side. I had exactly the same problems as you when i upgraded from F4j to F4, but yesterday when I upgraded from F4m to F5a I had no problem in reaching the 211 bclk, and none of the serious stability problems that I had before (only worse performance now). It´s very difficult for me to explain this in English, but I think it is something related to how the board adjusts its internal values, specially for memory, after a flashing or a re-setting. For example, working in the same Bios, I have found that is much more stable and safe set the voltage first and then go up the bclk than set the bclk first and then try to adjust the voltage (normal procedure).
At the bottom, I can only say that with this new procedure the board is fully Linpack stable @ 1.34v Qpi, while in the "old fashion" way is impossible to descend under 1.40v Qpi, and now when upgrading my bios i not have none serious problem.
I hope that at least some of what Im trying to say is understandable...

I set all my settings, except bclk, reboot, and then set my bclk, but still does not help. I have manually locked all my secondary settings to bios F4m/f4j values both, but no help. I spent hours and hours trying different ways, but I can boot the first time at any bclk, including I can run stable at 200x21 in F5a or F5b. The problem is, whenever I reboot or shutdown the computer, after having "stable settings of 21x200", it will not reboot, unless I shut it down completely, hit reset, and then it will boot ok at 21x200. There is some bug that everytime I reboot, I have to hit reset.

I take it you can reboot without hitting reset button? If I left my computer on all the time, and never rebooted, I could use bios F5a or F5b. But having to do a hard shutdown, reset everytime is ridiculous.

I think it has something to do like you said with settings we can not manipulate or a bug introduced by them trying to fix resume from sleep mode. But problem isnt getting stable settings, it is having to hit reset button on every reboot.

If you can reboot yours, can you list exactly all your settings you changed...and I will try to duplicate, even secondary mem timings.

[lowdog]

So noone else is currently experiencing this problem I have?? It all started out when I had a failed overclock. Now no matter how many times I reboot, shutdown, clear the CMOS for 10s, boot into safe mode and reboot, ect. it blue screens and restarts when it starts to load Vista Ultimate x64. It gets to the part where the green bar is running across the screen and as soon as it finishes that it acts like it's going to go and then blue screen.

I'm also having a problem where at times it's restarting twice during POST instead of only once. Starts up, shuts down, starts up, away we go.

What's the deal with this board? I know I seen at least one other guy having some serious issues like this. Is it corrupting the OS install on the failed OC and then from then on it's just an endless cycle till you reinstall the OS again? I've already reinstalled it once when I had a similar issue yesterday. Do I need to make sure I bump up the PCI-E speed to 101Mhz and use a specific SATA port for my hard drive so that this doesn't happen?

Try reflashing your bios, you may have a corrupted boot block.

[burningrave101]

Try reflashing your bios, you may have a corrupted boot block.

I tried reflashing a couple of times between BIOS F4m and F4j and that didn't seem to help any. All I can think of is that it's corrupting the OS in some way on a failed OC. I need to test my memory good I guess and make sure they're performing like they should be. I have 2x3GB Corsair 1600Mhz Dominators installed.

I did a clean install again and I think I finally got it up and running fresh. Only thing I did notice though is that I got a blue screen when installing Vista which I believe was due to me enabling AHCI in the BIOS but I don't know. In a couple of parts of the Vista install it also seemed like it would take a lot longer than it should to go on to the next screen or start the install. I think the SATA ports were giving me issue since I was trying to switch from the Gigabyte ones to the Intel southbridge ones. Anyway it's running now so fingers crossed that it won't do it again.

I need to get my Thermalright Ultra 120 Extreme w/ 1366 bolt-through so I can start actually pushing the board. The stock Intel heatsink/fan doesn't allow much headroom past 3.6Ghz.

[ReverendMaynard]

I tried reflashing a couple of times between BIOS F4m and F4j and that didn't seem to help any. All I can think of is that it's corrupting the OS in some way on a failed OC. I need to test my memory good I guess and make sure they're performing like they should be. I have 2x3GB Corsair 1600Mhz Dominators installed.

I did a clean install again and I think I finally got it up and running fresh. Only thing I did notice though is that I got a blue screen when installing Vista which I believe was due to me enabling AHCI in the BIOS but I don't know. In a couple of parts of the Vista install it also seemed like it would take a lot longer than it should to go on to the next screen or start the install. I think the SATA ports were giving me issue since I was trying to switch from the Gigabyte ones to the Intel southbridge ones. Anyway it's running now so fingers crossed that it won't do it again.

I need to get my Thermalright Ultra 120 Extreme w/ 1366 bolt-through so I can start actually pushing the board. The stock Intel heatsink/fan doesn't allow much headroom past 3.6Ghz.

ACHI fault at OS install = 7B

I just set up a bench drive with XP and had a BSOD with ACHI enabled.

[tipes]

Can someone post a link to the F4J bios for the UD5 as many are saying this is the best bios, but I can't find a download link.

Many thanks

Mark

[Splave]

http://www.xtremesystems.org/forums/...postcount=1228

[vega22]

If you can reboot yours, can you list exactly all your settings you changed...and I will try to duplicate, even secondary mem timings.

Well...yes, i can reboot. The behavior is similar to F4m with the same settings, except that sometimes in the restart, when i change some settings, the system hangs and the black screen before entering the logo is frozen. Then push the reset buttom, and go into windows.

My procedure & settings:
From F4m:
1-Load optimal defaults (excessive prudence) -> save and exit (reboot)
2-Qflash -> flashing to F5a (power off)
3-Power on -> enter Bios -> Load optimal defaults -> save and exit (reboot)
4-Enter Bios -> save an exit (reboot) again (to ensure that optimal defaults are loaded)
5-Power off before enter in Bios again
6-Unplug PSU power cord
7-60seg push clear cmos buttom
8-Plug PSU cord -> Power on
9-Enter Bios F5a

10-Setting peripherals, advanced cmos, etc. -> Save and Exit (reboot)
11-Enter Bios
FIRST SETTINGS:
CPU: Ht off/Turbo off/Virtualization tec. off (everything else default -Eist/C1e enabled-)
19x205
PCIe 100 (Cpu clock and Pcie clock are in defaults 700/700)
Qpi/uncore Auto (36x/16x)
Mem 8x
8-8-8-24 (default spd)
Command Rate = 1
Subtimings A, B and C: Auto
Turnarounds A, B and C:
Writes followed by reads:
7
7
6
(Everything else in turnarounds A, B and C: Auto)
LLC: Enabled
Vcore: 1.38750v
Qpi/Vtt: 1.340v
IOH: 1.280v
Vdimm: 1.66v
12-Save an Exit (reboot) -> Windows XP x64 sp2
13-5 passes Linpack test (so that the board be "accommodating" internally)
14-Reboot -> Enter Bios
SECOND SETTINGS:
Everything as before, except:
Bclk: 211 (19x211)
Subtimings A, B and C:
tRRD: 7
tWR: 8
tRFC: 90
(Everything else in Auto)
15-Save and exit (reboot) -> Go SO again
16-Ready for heavy test

...I dont think i forgot anything ...

[burningrave101]

Vega22, don't you need to disable EIST and C1e when overclocking or will those Intel power/clock management features not affect an overclock? Just wondering since I disabled them all.

[burningrave101]

ACHI fault at OS install = 7B

I just set up a bench drive with XP and had a BSOD with ACHI enabled.

Yeah I figured that was what it was. It BSOD'd somewhere during the Vista install because I came back to the blue screen. I went back and disabled AHCI and tried again and it didn't do it. I wonder what the performance difference is running in IDE mode instead of AHCI for the enhanced SATA features. I don't care much about having the drives hot swappable but NCQ is helpful.

[vega22]

Vega22, don't you need to disable EIST and C1e when overclocking or will those Intel power/clock management features not affect an overclock? Just wondering since I disabled them all.

It is not recommended. But my goal is to use DES once it reaches an optimal 24/7 OC.

Yeah I figured that was what it was. It BSOD'd somewhere during the Vista install because I came back to the blue screen. I went back and disabled AHCI and tried again and it didn't do it. I wonder what the performance difference is running in IDE mode instead of AHCI for the enhanced SATA features. I don't care much about having the drives hot swappable but NCQ is helpful.

Earlier in this thread you can read that after a failed OC my 750Gb AHCI ICH10R HDD was dead.

[rge]

Vega22, thanks for the detailed response

The hanging you sometimes notice, I never get in bios F4m or F4j, but always get once overclock past 3.8 in F4, F5a, F5b. And I have to hit reset to reboot then as well. For me having to do it once is not acceptable, but for some reason I never have that issue in F4m and always on any bios after that.

I will have to play with that tomorrow or next day when have time. I can already tell you something is different. Whenever I hit load optimal defaults it changes cpu/pcie clock drive to 800/900. The only time I get 700/700 is when I dont enter optimal defaults.

maybe when you clear cmos it resets them to 700/700? So something is different. Can you double check in your bios, if the clock drives are 700/700?

Edit: yep clearing cmos resets clock drives to 700/700. load optimal defaults sets them back to 800/900. So clearing cmos permanently changes settings, until you reload optimal defaults...that is screwy.

I will have to try both resetting cmos, and 700/700 clock drives. But if you are getting the need to hit reset sometimes on reboot, that is exactly what I am complaining about.

Edit 2: well tried it your way, and ended up with 700/700 clock drives, but still have to hit reset button to reboot.

[yenniedn]

So noone else is currently experiencing this problem I have?? It all started out when I had a failed overclock. Now no matter how many times I reboot, shutdown, clear the CMOS for 10s, boot into safe mode and reboot, ect. it blue screens and restarts when it starts to load Vista Ultimate x64. It gets to the part where the green bar is running across the screen and as soon as it finishes that it acts like it's going to go and then blue screen.

I'm also having a problem where at times it's restarting twice during POST instead of only once. Starts up, shuts down, starts up, away we go.

What's the deal with this board? I know I seen at least one other guy having some serious issues like this. Is it corrupting the OS install on the failed OC and then from then on it's just an endless cycle till you reinstall the OS again? I've already reinstalled it once when I had a similar issue yesterday. Do I need to make sure I bump up the PCI-E speed to 101Mhz and use a specific SATA port for my hard drive so that this doesn't happen?

I had those exact same problems - to a tee. On the first instance, I had to completely re-install Vista 64bit. A week after that fresh OS install (and with no change in my OC settings that I've been using for weeks) it happened again. Luckily, I was able to just do the repair option on the Vista installation DVD and it worked again.

Now, my SATA drives just show up on boot whenever they want to. I don't even expect to see them anymore - if they're there, great - if not, it's typical. Luckily my P6T Deluxe will be here tomorrow and I can finally RMA both of these Extremes back to MWave.

[godsfist]

Anyone know if the TRUE 120 can be mounted vertically? (fan pointing toward back out take fan)

[vega22]

rge: The hanging you sometimes notice, I never get in bios F4m or F4j, but always get once overclock past 3.8 in F4, F5a, F5b. And I have to hit reset to reboot then as well. For me having to do it once is not acceptable, but for some reason I never have that issue in F4m and always on any bios after that.

Of course, for me is too unacceptable, this is why I returned quickly to F4m. The difference is that with my "new procedure" such problems occur to me less and less serious than before. Before my system was hanging at every time, in BIOS or OS, and was unable to finish the smallest test @ 19x211. (EDIT: in Bioses after F4m, of course)

Whenever I hit load optimal defaults it changes cpu/pcie clock drive to 800/900. The only time I get 700/700 is when I dont enter optimal defaults.

This is why I do clear cmos after load optimal settings. On my system the true defaults of the bioses after F3 (specially 700/700) are more stable for Vcore than "optimal default" (800/900). But I have not investigated whether 700/700 is more unstable or not for the memory subsystem. Depending on your results, it appears that 800/900 does not help either in this matter ...

[Xello]

You don't need to disable power saving features to overclock, i don't

They only take effect while idle, so stress testing is no different. If you don't like the downclocking / volting @ idle, you can disable it in windows power management.

[tonytiger]

Anyone know if the TRUE 120 can be mounted vertically? (fan pointing toward back out take fan)

My True Black is mounted vertically on the UD5

[Lukee]

Anyone know if the TRUE 120 can be mounted vertically? (fan pointing toward back out take fan)

When I had my True, it blocked the first memory slot so you had to put the memory in first. If you use "tall" memory, then you may have a problem, but you should be able to mount it vertically.

I had it going out the top, but I can't see a reason why you can't have it going out the back.

[yenniedn]

Anyone know if the TRUE 120 can be mounted vertically? (fan pointing toward back out take fan)

Yes, most of the TRUE setups I've seen on most boards (including this one) are vertical, exhausting towards the back. I had mine set up this way. And yes, it also blocked the first DIMM slot when using my Dominator DDR3's.

[JWMc]

Edit: Had issues with multi setting in BIOS getting stuck with F4n so I went to F5e, stable with all the same settings across various OC targets so far:

Vista Ultimate SP1 64-bit
Gigabyte GA-EX58-UD5 @ F5e
G.Skill F3-12800C8-2GBPI-B x 2 @ 1600mhz 2:8, 8-8-8-21-1T
Ci7-920 (3841A383 from Newegg)
TRUE 120 Extreme with push-pull Panaflo 120x38mm M1-BX fans @ 12v @ ~2100 rpm
4000mhz Cores, 3200mhz Uncore, 3600mhz QPI

Temps are from Real Temp 2.90
vCore is from CPU-Z 1.49
Stability tested with LinX x 20 and Prime Blend for 6 hours

Relevant BIOS settings:
Turbo: off
QPI link: x36
Uncore: x16
Load-Line Calibration: enabled
Everything else not mentioned: Auto/Default

4.0ghz (20x200) HT enabled @ 1.360v load, 1.376v idle
vCore: 1.40625v
vDIMM: 1.660v (1.65v in HW Monitor)
vQPI: 1.375v
LinX max temps: 83, 83, 81, 82

4.0ghz (20x200) Hyperthreading DISABLED @ 1.264v load, 1.280 idle
Timings: 8-8-8-21-1T
vCore: 1.30000
vDIMM: 1.660v (1.65v in HW Monitor)
vQPI: 1.335v
LinX max temps: 67 66 66 66 (edit: my temps increased with higher ambient)

Edit: Finally found stability with one more voltage step in BIOS:
3.8ghz (19x200 turbo off) @ 1.23125v bios, 1.184v load, 1.200v idle, with HT on, 1.275v QPI, 8-8-8-21-1T, 1.65v DRAM (1.66 BIOS), max LinX temp 65c. LinX x 20 and Prime blend 11 hours stable.

New: Same as above @ 3.8ghz but with all 3 x 2gig modules @ 8-8-8-21-2T.

Final stable air OC I'm going to try: 4.2ghz (21x200 turbo on) @ 1.45625v bios, 1.408v load, 1.424v idle, with HT on, 1.395v QPI, 9-9-9-24-2T, 1.68v DRAM, 6gigs, 89c max temp @ 18c ambient.
http://valid.canardpc.com/show_oc.php?id=488873
http://img222.imageshack.us/my.php?i...6gblinxts1.jpg

Idle temps with HT on @ 1.376v: 37 35 38 37
Ambient temp: ~20C
I did all my early testing without the 3rd stick of RAM because my second fan gets in the way of the heatsink on the PI Black stick. To install the third module I moved the fan to blow through the ram sticks and at the lower portion of the CPU heatsink and am getting about the same temps.

I have the motherboard mounted to a removable motherboard tray laying flat on a table. My TRUE 120 extreme is not lapped and has a very uneven surface. I used a big blob of the paste that came with the TRUE 120 extreme in the middle of the IHS and screwed her down evenly.

Is this chip worthy of liquid cooling?

It seems to me that I'm ready to go for gaming at the moment with no cooling upgrades.

I'm sure the cooling will still be adequate for HT disabled even in a hot room with one less fan based on the LinX temps I am getting.

[Splave]

looks great

[godsfist]

Yes, most of the TRUE setups I've seen on most boards (including this one) are vertical, exhausting towards the back. I had mine set up this way. And yes, it also blocked the first DIMM slot when using my Dominator DDR3's.

But you can still get 3 sticks in right???