I would go ahead and do the unified drivers before the AMD drivers. It's been a while since I last installed Windows on this board, but pretty much as long as you install the AGP drivers (included in the Unified) before your video card drivers (assuming it's an AGP card; otherwise it doesn't matter), everything should work fine.
I would go ahead and do the unified drivers before the AMD drivers. It's been a while since I last installed Windows on this board, but pretty much as long as you install the AGP drivers (included in the Unified) before your video card drivers (assuming it's an AGP card; otherwise it doesn't matter), everything should work fine.
There were 3 settings on the drive. Slave, Master with slave present, and Single Drive. I changed it from Master with slave present to Single Drive and the slow boots stopped. I simply overlooked this when I put the new system together.
I installed the ULi integrated drivers 2.20 and it seems the installation process exited before it was done. It installs really quickly but I think it got to approx. 75% and then it just exits suddenly without any message that says it installed successfully. Is this normal? There's no install directory so I can't even check if the install completed or not.
I could really use some help in how to proceed now, if anyone could help. I've installed everything, but 9 times out of 10 I don't even get to the POST screen, the fans all come on and I hear the harddrive make some noise and the CD tray light lights up, but no POST and no BIOS.
That rare 1 time out of 10 when I do post it gives me a message like, "This PC has failed to boot a number of times, press F2 . . ." etc. So I do and get into the BIOS. I'm able to move around and mess with it as usual. However, when I save and exit it fails to POST again when it resets itself.
Here's the config:
Athlon64 X2 4200+
Arctic Cooling Freezer 64 Pro
OCZ PC3200 512MBx2 (in slots 3 and 4, the rightmost slots, would this cause an issue?)
Sapphire X800GTO w/ modded BIOS to unlock extra pipeline quad
450W Coolermaster Real Power PSU
Seagate 7200.9 on SATA 1
DVD Burner on a Serriell IDE->SATA adapter on SATA 2 (yes, this is the Seriell2 and so it supports optical drives, I was using the drive this way on my NF7-S)
Not using the SATAII controller at all
The only "weird" thing in the system that I thought may cause issue is my X800GTO AGP that is modded to unlock an extra pipeline quad. Initially I thought it might have issues with this (can't see what it could be, but it was my only thought), but after replacing it with my old Radeon 9500Pro it still behaves the same (except strangely it takes forever to draw the BIOS screens with this card in it, weird).
In any case, I noticed that if I turn off the power supply the BIOS gets reset to default values. This seems odd, perhaps the battery for the BIOS is bad? Should I replace it, might this cause this behavior?
I also noticed that with everything set to auto it's setting my PC3200 at DDR266 instead of DDR400. Maybe I need to try my RAM in the other two slots instead of the two right ones? One more bit of strange behavior, I have yet to hear a single peep from the case speaker. I've tried reversing the plug, but no go, so I'm assuming the mobo actually isn't signalling some failure condition via beeps, since I seem unable to hear anything.
In any case, I'd be grateful if anyone has any suggestions. Of course, it's possible the mobo is just DOA and I'll need to RMA, though I've already written in the board manual some info, hehe.
That rare 1 time out of 10 when I do post it gives me a message like, "This PC has failed to boot a number of times, press F2 . . ." etc. So I do and get into the BIOS. I'm able to move around and mess with it as usual. However, when I save and exit it fails to POST again when it resets itself.
Here's the config:
Athlon64 X2 4200+
Arctic Cooling Freezer 64 Pro
OCZ PC3200 512MBx2 (in slots 3 and 4, the rightmost slots, would this cause an issue?)
Sapphire X800GTO w/ modded BIOS to unlock extra pipeline quad
450W Coolermaster Real Power PSU
Seagate 7200.9 on SATA 1
DVD Burner on a Serriell IDE->SATA adapter on SATA 2 (yes, this is the Seriell2 and so it supports optical drives, I was using the drive this way on my NF7-S)
Not using the SATAII controller at all
The only "weird" thing in the system that I thought may cause issue is my X800GTO AGP that is modded to unlock an extra pipeline quad. Initially I thought it might have issues with this (can't see what it could be, but it was my only thought), but after replacing it with my old Radeon 9500Pro it still behaves the same (except strangely it takes forever to draw the BIOS screens with this card in it, weird).
In any case, I noticed that if I turn off the power supply the BIOS gets reset to default values. This seems odd, perhaps the battery for the BIOS is bad? Should I replace it, might this cause this behavior?
I also noticed that with everything set to auto it's setting my PC3200 at DDR266 instead of DDR400. Maybe I need to try my RAM in the other two slots instead of the two right ones? One more bit of strange behavior, I have yet to hear a single peep from the case speaker. I've tried reversing the plug, but no go, so I'm assuming the mobo actually isn't signalling some failure condition via beeps, since I seem unable to hear anything.
In any case, I'd be grateful if anyone has any suggestions. Of course, it's possible the mobo is just DOA and I'll need to RMA, though I've already written in the board manual some info, hehe.
The message you're seeing ("The PC has failed to boot....") is usually seen when overclocking and you've pushed it too far, to a speed that won't boot. I've never heard of that appearing in normal (stock) usage, and to me that would seem to indicate something running out of spec (or at least the board thinking that it is). I'd probably start by moving your RAM to slots 1 and 2, and possibly even just booting with one stick at a time. Is this RAM from an older build (so you know it works)? The fact that Auto sets your RAM to DDR266 does seem fishy...if your RAM can't run at DDR400 for some reason, that would explain the message you're seeing (and the failed boots/no POST/etc.). Maybe run Memtest86 if you can get that far (or with the RAM in another PC...)
Hi Guys, i just got this board and on XP its been rock stable, but i bought it so i could test out Vista as well.
My main problem is getting my 9600XT working in Vista, as both Beta 2 and the latest Vista builds give me a Error 43 and no drivers or anything make any difference.
Are there any people running this motherboard with an AGP card successfully in Vista, and if so, are there any special steps to take.
Thanks guys
My main problem is getting my 9600XT working in Vista, as both Beta 2 and the latest Vista builds give me a Error 43 and no drivers or anything make any difference.
Are there any people running this motherboard with an AGP card successfully in Vista, and if so, are there any special steps to take.
Thanks guys
Yes, the RAM is from an old build so it should work fine I'd think. It's OCZ PC3200, admittedly one is Rev. 2 and one is Rev. 3, but they both have the same timing rating and played nicely without incident in my other PC. I'll try messing with it and seeing what happens.
Well, the memory angle seems to be yielding results. I took both sticks out of slots 3 & 4 and put just one stick in slot 1, it boots fine and I'm able to run memtest. I ran it for a little over an hour with no errors. I've just put the other stick in slot 1 to test it as well, at which point I'll try to dual channel them both in slots 1 and 2.
The Auto still detects only DDR266 if everything is set to auto, but I can manually set it to DDR400 2-3-3-6 and it runs fine. I think this was happening on my old NF7-Sv2 as well, so maybe the spd in these OCZ sticks are weird or something.
If I have issue with them in dual channel I might mess with the TRAS timing and move it up to 7. When I initially bought the new stick of RAM and noticed it was a Rev. 3 (since they'd stopped doing Rev. 2) I asked them if they should still work fine together, they said yes, but I may have to bump TRAS to 7. I didn't have to on my NF7-Sv2, but maybe I'll need to on this Asrock. Well, off for some more troubleshooting - wish I were playing Oblivion though, lol!!
[EDIT / UPDATE] : OK, well, both sticks tested out fine in Slot 1 with memtest, I didn't really expect otherwise. However, it seems the board doesn't like having more than one stick in it at all. Anytime I put more than one stick in it it reverts back to the sporadic "will it boot or won't it" behavior - this is even when the sticks are on seperate channels (ie. one in slot 1 and one in slot 3).
However, it seems to work fine with more than one stick - even in dual channel configuration - if I turn on that "Flexibility Option" that's in the BIOS. Does anyone know what this actually does?
While I have CAS-TRCD-TRP-TRAS set manually to 2-3-3-6, the other timing options are all set to [AUTO], MA Timing, etc. I wasn't sure what these should be if I were to set them manually for these particular sticks, so if anyone has any advice on that it would be great.
Otherwise, can anyone enlighten me on this "Flexibility Option"? It doesn't change the timing settings in the BIOS as far as I can tell, so what is it actually doing?
Ah, apparently "Flexibility Option" seems to set a 5/6 divider, which isn't really acceptable. Dammit, I wish I could figure out why this board isn't playing nice with this RAM!!
The Auto still detects only DDR266 if everything is set to auto, but I can manually set it to DDR400 2-3-3-6 and it runs fine. I think this was happening on my old NF7-Sv2 as well, so maybe the spd in these OCZ sticks are weird or something.
If I have issue with them in dual channel I might mess with the TRAS timing and move it up to 7. When I initially bought the new stick of RAM and noticed it was a Rev. 3 (since they'd stopped doing Rev. 2) I asked them if they should still work fine together, they said yes, but I may have to bump TRAS to 7. I didn't have to on my NF7-Sv2, but maybe I'll need to on this Asrock. Well, off for some more troubleshooting - wish I were playing Oblivion though, lol!!
[EDIT / UPDATE] : OK, well, both sticks tested out fine in Slot 1 with memtest, I didn't really expect otherwise. However, it seems the board doesn't like having more than one stick in it at all. Anytime I put more than one stick in it it reverts back to the sporadic "will it boot or won't it" behavior - this is even when the sticks are on seperate channels (ie. one in slot 1 and one in slot 3).
However, it seems to work fine with more than one stick - even in dual channel configuration - if I turn on that "Flexibility Option" that's in the BIOS. Does anyone know what this actually does?
While I have CAS-TRCD-TRP-TRAS set manually to 2-3-3-6, the other timing options are all set to [AUTO], MA Timing, etc. I wasn't sure what these should be if I were to set them manually for these particular sticks, so if anyone has any advice on that it would be great.
Otherwise, can anyone enlighten me on this "Flexibility Option"? It doesn't change the timing settings in the BIOS as far as I can tell, so what is it actually doing?
Ah, apparently "Flexibility Option" seems to set a 5/6 divider, which isn't really acceptable. Dammit, I wish I could figure out why this board isn't playing nice with this RAM!!
Originally posted by: deadlast
Hi Guys, i just got this board and on XP its been rock stable, but i bought it so i could test out Vista as well.
My main problem is getting my 9600XT working in Vista, as both Beta 2 and the latest Vista builds give me a Error 43 and no drivers or anything make any difference.
Are there any people running this motherboard with an AGP card successfully in Vista, and if so, are there any special steps to take.
Thanks guys
I only still had my AGP 6800GT for a little while after Vista beta 2 was out (and before I upgraded to an X1900XT), but I seem to remember it working fine...have you installed the AGP drivers beforehand? Are you trying to use 32-bit or 64-bit Vista? I believe the latest ULI Integrated drivers (which include AGP drivers) will work in Vista, but failing that you should be able to install the AGP drivers for Windows XP (either 32-bit or 64-bit as appropriate).
If you give a bit more info on what you're installing and in what order, we might be able to figure out the cause of your error...(I'm not familiar with that specific error code right now)
It sounds like you're heading in the right direction with the troubleshooting...Flexibility does make the RAM run at a 133MHz divider I believe (2/3), so you'll generally want to avoid that unless all else fails.
Have you tried increasing the TRAS to 7 yet? Also, some people have trouble running 1T on this board (even with 2 sticks of RAM; pretty much no one can run 1T if you have 4 sticks like I do). So you might try changing MA Timing from Auto (defaults to 1T) to 2T. The performance loss will be completely insignificant, and I've been running at 2T with my 4x512MB for 4-5 months now with no issues. And if you are having issues with running at 1T, that would certainly cause your no-boot issue. I forgot to switch from Auto to 2T before installing my second gigabyte (2x512MB) of RAM, and then the computer didn't boot since 4 sticks won't run at 1T...
Yes, I've tried changing timings to stuff completely absurd. There seems to be no rhyme nor reason. I have MA Timing set to 2T manually for all of my trials. I had timing at like 3-4-4-8 and still nothing, not at DDR400 or DDR333. The only way I seem to be able to run with both sticks in the mobo is to have one in slot 1 and the other in slot 3 (or slot 2 and slot 4) and have it no higher than DDR266 (I had all other times set to AUTO for this test). Ugh, frustrating. At this point I'm not sure whether to blame the RAM or the mobo - leaning toward the mobo, but not sure whether this is something an RMA would fix.
I don't suppose going to another BIOS revision could help could it? My board shipped with 1.9.
I don't suppose going to another BIOS revision could help could it? My board shipped with 1.9.
I seem to remember on the OCW Forums some posts about people having trouble with memory compatibility in post-1.70 BIOS releases. I'm using 2.10 myself at the moment, 2.20 is the most recent, but honestly as far as I can tell 1.80 and newer are all pretty much the same for Socket 939 users (meaning, all of the improvements after 1.80 seem to have been for the Socket AM2 board).
I would recommend perhaps trying 1.60 or 1.70, but if you still have the same issues then maybe you just have a bad board and need to RMA...you know the RAM itself works fine...
Edit: Something else I forgot to mention in my last post, is that have you increased the RAM voltage? The range is really limited, "High" being only 2.7-2.8V, but I always run mine at High to avoid any issues. Just another thought...
I just today bought this Motherboard and it will be here soon. I have a question about the RAM. I have 2 sticks of 512 PC2700 DDR333 and they are not the same brand. It is my understanding that the board will run but not in Dual mode. Is that right? Also, there are 4 slots DDR1,2,3,4. Can I put the unmatched RAM in 1 and 2 or do I need to use 1 and 3? Thanks much.
The board might not boot initially because the default is DDR400 (if SPD is not read correctly, etc.). But after a few failed boots, it *should* lower the speed and allow you to boot and get into the BIOS and change the speed. If the sticks are unmatched you may or may not be able to run in dual channel. You can always try (slots 1 and 2), and if that doesn't work 1 and 3 should be non-dual-channel, I believe.
Where can I change the RAM voltage? I was actually looking for that and didn't see it.
Also, I've just been running memtest again with a stick on slot 1 and a stick in slot 3 (so no dual channel) and I've gotten about five errors I didn't previously when running each stick separately. Of course, at this point I'm more inclined to blame the mobo for this rather than the RAM.
The RAM voltage is under the Chipset Settings section, not the one where you change CPU/RAM speed and timings.
Thanks. I seem to be getting weird results from memtest. At DDR400, with only one stick in the mobo, I sometimes get errors with one particular stick, the Rev.3. I might just have to run it some more. Of course, running them one at a time and testing them is fine, but if it goes to hell when they're dual channel it doesn't really matter.
System is as follows:
Athlon X2 4200+ 939
Asrock 939Dual-Sata2
2gb Corsair C2 Ram
2 x 200gb WD Hdds in Raid0
1 x 400gb via e-Sata in via PCI 3112 Controller
9600XT Video Card
Pretty much i install Vista, the XP AGP Drivers will not install, and from what i've read there is usually a driver by default.
Problem is, the Video card always has an exclamation mark next to it and cause "code 43" there is a problem with this device and it is not using any resources.
Works fine in XP, and detects in Vista, but doesn't function, and there is no display listed under display properties.
Athlon X2 4200+ 939
Asrock 939Dual-Sata2
2gb Corsair C2 Ram
2 x 200gb WD Hdds in Raid0
1 x 400gb via e-Sata in via PCI 3112 Controller
9600XT Video Card
Pretty much i install Vista, the XP AGP Drivers will not install, and from what i've read there is usually a driver by default.
Problem is, the Video card always has an exclamation mark next to it and cause "code 43" there is a problem with this device and it is not using any resources.
Works fine in XP, and detects in Vista, but doesn't function, and there is no display listed under display properties.
Hey guys,
Hopefully this thread is still very much alive, unlike some other boards (like extreme oc asrock board thread which seems dead) because I need all the help I can get.
1. Does the easy Vcore mod work with all/any BIOS revision? I'm at OCWBeta2 BIOS and it doesnt work for me (could be something else - the type of conductive grease I'm using).
I have a 3500+ venice E6 LBBWE currently overclocked to 2.55GHz as I'm typing this with the Vcore max at 1.45V (neeed more but having problems getting vcore mod to work), max ram voltage, memory at 1:1 and NB/SB link at 600MHz (3x).
Edit: Maybe I should introduce my setup...
Here will be my general setup.
ASROCK 939Dual-SATA2
AMD Athlon 64 3500+ Venice
2x256MB of Corsair TWINX512-3200LL memory (2-3-2-6)
2x80GB Hitachi Deskstar HD in RAID_0
ATI Radeon 9700PRO AGP - IS the AGP bus locked?
Seasonic S12 500W PSU
Hopefully this thread is still very much alive, unlike some other boards (like extreme oc asrock board thread which seems dead) because I need all the help I can get.
1. Does the easy Vcore mod work with all/any BIOS revision? I'm at OCWBeta2 BIOS and it doesnt work for me (could be something else - the type of conductive grease I'm using).
I have a 3500+ venice E6 LBBWE currently overclocked to 2.55GHz as I'm typing this with the Vcore max at 1.45V (neeed more but having problems getting vcore mod to work), max ram voltage, memory at 1:1 and NB/SB link at 600MHz (3x).
Edit: Maybe I should introduce my setup...
Here will be my general setup.
ASROCK 939Dual-SATA2
AMD Athlon 64 3500+ Venice
2x256MB of Corsair TWINX512-3200LL memory (2-3-2-6)
2x80GB Hitachi Deskstar HD in RAID_0
ATI Radeon 9700PRO AGP - IS the AGP bus locked?
Seasonic S12 500W PSU
I have the Asrock 939 dual sata mobo and am running it in dual channel mode. I have 4 sticks of Corsair ram 512mb's PC 2700. Post says running in dual channel. So in other words, if slots 1&2 or all slots are filled with matching sticks it will run in dual mode. If one or more sticks is not matched, it will not run in dual mode.
Also, it is my understanding that "matched" sticks is sort of misleading. There is nothing special about memory sticks that make them matched. If the sticks have the same capacity and timing perameters they will be recognized as matched by the board even if they are from different mfgrs. Others may know more though.
Also, it is my understanding that "matched" sticks is sort of misleading. There is nothing special about memory sticks that make them matched. If the sticks have the same capacity and timing perameters they will be recognized as matched by the board even if they are from different mfgrs. Others may know more though.
I just put together a system based on this board and I'm impressed as to how well it performs. The CPU is a 3500 Venice, along with 2 sticks of Corsair Vaue Select 400 DDR. The reason for the choice of this board was that almost all my other systems are Vencie based AMD 64 bits, except for 3 systems that are still older Barton 32 bit Athlons (130 nm dies). These are in service because they have AGP ATI All-In-Wonder video capture cards. All I had found originally for 939 boards were pci video versions, unfortunately. With this new board functioning this well I plan on putting 2 more together so as to get away from the older 32 bit based systems. Much less heat, even under the strain of video editing than the older 32 bitters.
Glad I ran accross this thread to get pointed in the right direction. I had almost given up on finding an AGP 939 board...
Glad I ran accross this thread to get pointed in the right direction. I had almost given up on finding an AGP 939 board...
The "conductive pen" Vcore mod should work with any BIOS version as far as I'm aware...I've briefly run OCWbeta2 with the mod and it worked fine....Which variation did you do? I did the VID4 (I think) one where you have to select 1.15V in the BIOS to get 1.55V actual. I had to do it two or three times to get it to "take"; I probably didn't actually connect the traces/solder pads on the first try (the area is really tiny and my vision isn't the greatest
). And what type of conductive material are you using?As for your question about the AGP bus being locked, my understanding is that AGP/PCI is locked at 66/33 as long as you lock PCI-E at 100 (Async overclocking).
Well, after some more testing, it seems like my Rev. 3 is the only stick that shows any errors when it is being tested, alone, on the mobo under memtest86. I can't imagine the board would have issues with this one piece of memory and not the Rev. 2 stick, so maybe I'll try getting a new stick of memory from OCZ to replace it. I can't understand why I never got any errors on the NF7-Sv2 though, unless it just suddenly decided to go bad.
Originally posted by: SynthDude2001
The "conductive pen" Vcore mod should work with any BIOS version as far as I'm aware...I've briefly run OCWbeta2 with the mod and it worked fine....Which variation did you do? I did the VID4 (I think) one where you have to select 1.15V in the BIOS to get 1.55V actual. I had to do it two or three times to get it to "take"; I probably didn't actually connect the traces/solder pads on the first try (the area is really tiny and my vision isn't the greatest
As for your question about the AGP bus being locked, my understanding is that AGP/PCI is locked at 66/33 as long as you lock PCI-E at 100 (Async overclocking).
Hey thanks Synth for replying. I actually used the conductive grease supplied in the "HighSpeed PC Athlon XP Unlocking Kit". Remember those? It had the brown conductive grease, white gap filler, and a cleaner. So I tried the VID4 mod, the contacts at the top of those contacts. Anyways I used both solutions to no avail.
I will try again but its a headache because I have to work around a Big Typhoon! So I masked off the areas and tried again (because it did spread apart blobbing in other places) again to no avail for both grease and gap filler.
I'm at a point now where I think I'm going to solder it. Soldering is good because the solution doesn't spread at all and stays at a stationary place.
BTW thats what I thought, should work with all/any BIOS revisions.
Thanks, I'll let you know how it turns out.
Any help with this? I just need to know if there's a confirmation prompt at the end of the ULi drivers installation.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter