Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
- Republic of Gamers Forum
- Motherboards
- Previous Generations
- Other Motherboards
- RIVE + BSOD 124 + solution ? + BIOS 4004 bugs
Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
RIVE + BSOD 124 + solution ? + BIOS 4004 bugs
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 01:48 AM
Hello,
I am in the process of upgrading my system:
RIVE (bought brand new), BIOS 4004
3930K (also new in my system - bought used one, previous owner claims that he used it on RIVE as well with no OCing and with water cooling)
Patriot Viper 3 4x4GB 1600 CL9 (ran fine on my P8Z68 + 2600K)
HD6950 (as above)
Chieftec 750W PSU
CM HYPER 212 EVO
From the very start I was having strange issue with BSOD 0x00000124. Memory dumps analysis points to Windows Hardware Error related to GenuineIntel (CPU I assume). It is fully repeatable/recreatable issue which happens in the following conditions:
- put a high load on CPU (Prime95 but any other torture test is good)
- wait till CPU heats up (50 C deg is fine)
- stop the torture test
then PC crashes immediately with BSOD 124 😞
Sometimes just a regular Windows boot up causes this because CPU is loaded when Windows boots and when booting is finshed, CPU clock drops an bang ! - BSOD 124.
I read multiple threads related to BSOD 124 on S2011 and they say about VTT, VCCSA and VCORE voltages. Actually I performed endless number of tests, BIOS setting changes, Optimal defaults, Safe defaults, etc. etc. and the only setting that cures the problem is VCORE. Manual/Auto or Offset set to any value (positive or negative) does not help. But setting Manual voltage to a fixed value (such as 1.15V) seems to be a cure for this issue but then there is a side effect - VCORE is fixed and does not follow clock changes.
Why is this happening ???:
- faulty RIVE
- faulty CPU
- memory or something else ?
Any ideas ?
I also noticed some obvious bugs in the 4004 BIOS:
1. Turbo Clock set to Auto does not mean stock clocks ! It sets all cores to the same max value, equal to Turbo frequency (3.8GHz). So you are OCing with Auto. Is it supposed to work like this ? On P8Z68 Auto meant stock clocks. If you want stock clocks on RIVE (3.4GHz max, more - up to 3.8GHz only if not all cores are loaded), Turbo Clock must be set to "by Core" and all settings (1-core, 2-core, etc.) must be set to Auto. Weird.
2. Setting AI Tuner to Auto sets incorrect voltage to DRAM. Auto should mean JEDEC settings for RAM (it worked like this on P8Z68) and timings are set correctly but voltage is set too high to 1.65V which is not correct because JEDEC setting in SPD is 1.50V ! AI Tuner set to XMP works fine - it sets timings to XMP (not very different to 800MHz JEDEC indeed) and DRAM voltage to XMP value, which is also 1.50V.
Best regards.
I am in the process of upgrading my system:
RIVE (bought brand new), BIOS 4004
3930K (also new in my system - bought used one, previous owner claims that he used it on RIVE as well with no OCing and with water cooling)
Patriot Viper 3 4x4GB 1600 CL9 (ran fine on my P8Z68 + 2600K)
HD6950 (as above)
Chieftec 750W PSU
CM HYPER 212 EVO
From the very start I was having strange issue with BSOD 0x00000124. Memory dumps analysis points to Windows Hardware Error related to GenuineIntel (CPU I assume). It is fully repeatable/recreatable issue which happens in the following conditions:
- put a high load on CPU (Prime95 but any other torture test is good)
- wait till CPU heats up (50 C deg is fine)
- stop the torture test
then PC crashes immediately with BSOD 124 😞
Sometimes just a regular Windows boot up causes this because CPU is loaded when Windows boots and when booting is finshed, CPU clock drops an bang ! - BSOD 124.
I read multiple threads related to BSOD 124 on S2011 and they say about VTT, VCCSA and VCORE voltages. Actually I performed endless number of tests, BIOS setting changes, Optimal defaults, Safe defaults, etc. etc. and the only setting that cures the problem is VCORE. Manual/Auto or Offset set to any value (positive or negative) does not help. But setting Manual voltage to a fixed value (such as 1.15V) seems to be a cure for this issue but then there is a side effect - VCORE is fixed and does not follow clock changes.
Why is this happening ???:
- faulty RIVE
- faulty CPU
- memory or something else ?
Any ideas ?
I also noticed some obvious bugs in the 4004 BIOS:
1. Turbo Clock set to Auto does not mean stock clocks ! It sets all cores to the same max value, equal to Turbo frequency (3.8GHz). So you are OCing with Auto. Is it supposed to work like this ? On P8Z68 Auto meant stock clocks. If you want stock clocks on RIVE (3.4GHz max, more - up to 3.8GHz only if not all cores are loaded), Turbo Clock must be set to "by Core" and all settings (1-core, 2-core, etc.) must be set to Auto. Weird.
2. Setting AI Tuner to Auto sets incorrect voltage to DRAM. Auto should mean JEDEC settings for RAM (it worked like this on P8Z68) and timings are set correctly but voltage is set too high to 1.65V which is not correct because JEDEC setting in SPD is 1.50V ! AI Tuner set to XMP works fine - it sets timings to XMP (not very different to 800MHz JEDEC indeed) and DRAM voltage to XMP value, which is also 1.50V.
Best regards.
Labels:
- Labels:
-
Other Motherboards
8,004 Views
20 REPLIES 20
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 02:14 AM
First and foremost stop tying to compare the way X68/Z77 BIOS works with X79, there is no profit in that they are not the same. There might be some cosmetic similarities but that is coincidental. However your 'bugs' are noted and will be replicated by me to confirm. 🙂 So thanks for that.
Second - when you system gives you these BSOD are you at defaults, and that includes memory? And out of interest is your OS a fresh install or the old Sandy OS that you have just put into the new system?
Here is what I would do. Download the bios again. Use a fat formatted USB device and the flashback method I would re-flash the bios. Let your system do the full firmware upgrade if required.
Then I would clear cmos once the entire process has completed. That done into BIOS and set your SATA mode if running RAID. Now boot to OS and see if your system is stable, play your favourite game for a couple of hours, not overly worried about stress tests as they prove nothing to me.
Here is what I would do. Start with the memory. Use the XMP profile as your starting base, you can either set the profile values manually or let the system do it for you. I like the manual method but both work just fine. Ideally I would suggest your follow the guide to memtest86+ in my sig to the letter that will take out a whole heap of the variations that might be the cause for your issue.
With your cooler you are not going to get too much OC done with that CPU. But here is a good instance where a stress test has value. See how your Cooler copes after 30 minutes of Prime or similar. If still manageable then all good.
0124 is 99% of the time vCore.
Will check back later.
To help us help you - please provide as much information about your system and the problem as possible.
Second - when you system gives you these BSOD are you at defaults, and that includes memory? And out of interest is your OS a fresh install or the old Sandy OS that you have just put into the new system?
Here is what I would do. Download the bios again. Use a fat formatted USB device and the flashback method I would re-flash the bios. Let your system do the full firmware upgrade if required.
Then I would clear cmos once the entire process has completed. That done into BIOS and set your SATA mode if running RAID. Now boot to OS and see if your system is stable, play your favourite game for a couple of hours, not overly worried about stress tests as they prove nothing to me.
Here is what I would do. Start with the memory. Use the XMP profile as your starting base, you can either set the profile values manually or let the system do it for you. I like the manual method but both work just fine. Ideally I would suggest your follow the guide to memtest86+ in my sig to the letter that will take out a whole heap of the variations that might be the cause for your issue.
With your cooler you are not going to get too much OC done with that CPU. But here is a good instance where a stress test has value. See how your Cooler copes after 30 minutes of Prime or similar. If still manageable then all good.
0124 is 99% of the time vCore.
Will check back later.
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 02:45 AM
I compared RIVE to P8Z68 BIOS just to indicate that for me Auto is supposed to be default/stock setting and Z68 behave exactly like I expected 🙂 Two mentioned settings on RIVE does seem to work rather as Auto=OC.
I am getting BSODs even if I am resetting all settings to Safe or Optimized, wihout touching anything.
Windows is not a fresh install - over years I mastered a procedure of moving entire installation to even completely different hardware. This same Windows 7 copy worked fine without reinstall, since the date Windows 7 was oficially released and I bought it. I think it already ran on 4-5 mobos, both AMD and Intel. The procerdure includes uninstalling everything (tools, drivers) related to old hardware, replacing hardware, installing new tools/drivers. To remedy different boot controller driver issue (BOOT DEVICE NOT FOUND BSOD) I use addon SATA controller card. This procedure never failed me so far. To find the reason of the BSOD I already thought about installing a fresh Windows 7 copy on separate disk, just for test but it does not seem a good target solution for me as I got a lot of apps and settings on my current install...
Basically I left any OCing for later, when my system is stable on stock settings. I am aware that 212 Evo was good for 2600K OCing but on 3630K it is good only for stock. Anyway I was doing OCing unaware of the results of Turbo Clock set to Auto. At full load it caused all cores to run at 3.8GHz (did not notice that immediately !), VCORE was automatically set by CPU to around 1.25V at such clocks and temperature was rising as high as to 70+ C deg. With Turbo Clock set to By-Core/all Auto, at full load all cores run as the should at 3.4GHz and I am getting much lower temp (eg. 60) but the BSOD issue persists. If I set VCORE to fixed 1.15, then at full load all cores run at 3.4, temp is about 52-55, no BSOD issue and system seems to be stable even though VCore is less than CPU would need.
I am going to replace the fan with H80i anyway and borrow some other mem sticks for test.
I was trying to use some older verion of BIOS for comparison, I could not flash it anyway because BIOS EZ Flash told me that image is outdated. So I switched to BIOS 2 which contained some old version (I upgraded BIOS 1 to 4004 instantly after installation) but the BSOS issue is the same with that old BIOS.
I am getting BSODs even if I am resetting all settings to Safe or Optimized, wihout touching anything.
Windows is not a fresh install - over years I mastered a procedure of moving entire installation to even completely different hardware. This same Windows 7 copy worked fine without reinstall, since the date Windows 7 was oficially released and I bought it. I think it already ran on 4-5 mobos, both AMD and Intel. The procerdure includes uninstalling everything (tools, drivers) related to old hardware, replacing hardware, installing new tools/drivers. To remedy different boot controller driver issue (BOOT DEVICE NOT FOUND BSOD) I use addon SATA controller card. This procedure never failed me so far. To find the reason of the BSOD I already thought about installing a fresh Windows 7 copy on separate disk, just for test but it does not seem a good target solution for me as I got a lot of apps and settings on my current install...
Basically I left any OCing for later, when my system is stable on stock settings. I am aware that 212 Evo was good for 2600K OCing but on 3630K it is good only for stock. Anyway I was doing OCing unaware of the results of Turbo Clock set to Auto. At full load it caused all cores to run at 3.8GHz (did not notice that immediately !), VCORE was automatically set by CPU to around 1.25V at such clocks and temperature was rising as high as to 70+ C deg. With Turbo Clock set to By-Core/all Auto, at full load all cores run as the should at 3.4GHz and I am getting much lower temp (eg. 60) but the BSOD issue persists. If I set VCORE to fixed 1.15, then at full load all cores run at 3.4, temp is about 52-55, no BSOD issue and system seems to be stable even though VCore is less than CPU would need.
I am going to replace the fan with H80i anyway and borrow some other mem sticks for test.
I was trying to use some older verion of BIOS for comparison, I could not flash it anyway because BIOS EZ Flash told me that image is outdated. So I switched to BIOS 2 which contained some old version (I upgraded BIOS 1 to 4004 instantly after installation) but the BSOS issue is the same with that old BIOS.
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 02:53 AM
THe key to keeping single Windows copy alive and kicking over years is intense use of disk imaging tools and frequent backups, prior to any app or dirver install. I used Norton Ghost for years but its development obviously stalled so I switched to Acronis True Image. Whenever something goes wrong, I just restore the system from image. Saved countless hours of work for me... Today I could switch back to P8Z68 with just an reimage.
124 BSODs on RIVE corrupted my system drive (C: ) filesystem a few times (!), I am not relying on filesystem autocheck/repair. If that happens I rely on my ATIH images 🙂
Anyway the 124 BSOD issue does not seem to be software-related to me but installing fresh Win 7 would resolve this dispute 😉
124 BSODs on RIVE corrupted my system drive (C: ) filesystem a few times (!), I am not relying on filesystem autocheck/repair. If that happens I rely on my ATIH images 🙂
Anyway the 124 BSOD issue does not seem to be software-related to me but installing fresh Win 7 would resolve this dispute 😉
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 02:55 AM
OK that is key bit of information that your system is on defaults.
I strongly request that before you do anything else you run the memtest86+ as per my guide.
To help us help you - please provide as much information about your system and the problem as possible.
I strongly request that before you do anything else you run the memtest86+ as per my guide.
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 04:52 AM
If you have moved memory from your P8Z68 to the RIVE you will be running two dual channel kits right? And the RIVE is very picky about RAM and running two different kits together can give these kind of headaches. Best to run a set of 4 sticks that have been binned together...i.e. a quad channel kit. But at 1600 You will probably be able to get them whipped into shape tweaking VCCSA/VTT and/or vcore.
If you separate the kits and just run one of them at a time, I wonder would your problem stop?
But the Memtest86+ should tell you if they are stable....as single sticks and as dual channel kits and all together....
If you separate the kits and just run one of them at a time, I wonder would your problem stop?
But the Memtest86+ should tell you if they are stable....as single sticks and as dual channel kits and all together....
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 08:08 AM
As per RAM - I used quad 4x4GB=16GB Patriot Viper 3 1600 CL on P8Z68/2600K. I moved this quad DIMM set to RIVE (indeed I bought quad kit with having move to S2011 in mind 😉 ).
Patriots are NOT on the ASUS QVL. Why ? Because earlier I used 2x4GB Corsair XMS 3 1600 CL9 kit on P8Z68. Corsairs were on QVL for P8Z68 (as well as they are on QVL for RIVE) but I had a bunch of issues with them - random BSODs, once a day or two. Every time different code. Certainly a DRAM issue... Nothing helped for this... Eventually I sold them to a close friend and he runs them rock-stable on some AMD board.
Now I borrowed this Corsair kit back from him an will let you results later.
Meanwhile I am sure IT IS a HW issue, not related to my old Win 7 install which survived multiple HW changes. I temporarily installed FRESH Win 7 copy on a SATA classic disk (normally I use SSD), with minimal driver set to clear out any exclamation marks in Device Manager (eg. Intel chipset divers Intel LAN, Intel MEI, ATI HD, unneeded H/W disabled in BIOS so no need to install BT or ASMEDIA drivers). Unfortunately the same 0x00000124 BSOD when stopping any torture test.
Patriots are NOT on the ASUS QVL. Why ? Because earlier I used 2x4GB Corsair XMS 3 1600 CL9 kit on P8Z68. Corsairs were on QVL for P8Z68 (as well as they are on QVL for RIVE) but I had a bunch of issues with them - random BSODs, once a day or two. Every time different code. Certainly a DRAM issue... Nothing helped for this... Eventually I sold them to a close friend and he runs them rock-stable on some AMD board.
Now I borrowed this Corsair kit back from him an will let you results later.
Meanwhile I am sure IT IS a HW issue, not related to my old Win 7 install which survived multiple HW changes. I temporarily installed FRESH Win 7 copy on a SATA classic disk (normally I use SSD), with minimal driver set to clear out any exclamation marks in Device Manager (eg. Intel chipset divers Intel LAN, Intel MEI, ATI HD, unneeded H/W disabled in BIOS so no need to install BT or ASMEDIA drivers). Unfortunately the same 0x00000124 BSOD when stopping any torture test.
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 08:20 AM
klosz007 wrote:
indeed I bought quad kit with having move to S2011 in mind 😉
Ooops :rolleyes: that'll teach me to jump to conclusions......still, 124 is classic memory problem so follow HiVizMan's suggestion and thoroughly test various combinations.....but it's funny that you crash on dropping voltage at end of test.
have you tried tweaking some of the DIGI+ RAM power options...
Like current capability 120% Power phases optimized....LLC on high
By the way, what VCCSA and VTT is the board setting on AUTO....I have heard of some values being set at AUTO that are too low and outside of the normal delta with Vcore (that normally being VTT and VCCSA within 0.3v of Vcore)
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
06-01-2013 09:13 AM
As per my notes VTT on Auto is set to 1.15V, VTT 2nd 1.05, VCCSA 1.15. VCORE is 1.15V. I set those values to manual but it does not improve the issue.
It is better with Corsairs - I get BSOD only on every 5-th tests, so it has improved but it is not ideal. So it would mean memory is the issue or at least it has important role.
Good news is that BIOS sets correct voltages on memeory on Auto (1.5) and XMP (1.65) for these Corsairs.
I looked at the warranty card for the CPU that I received form the seller. He bought entire system in one take (so everything is on the warranty card) - 3930K + RIVE + PSU, GPU, SSD etc etc. He bought G.SKILL RipjawsZ DDR3 2133MHz 16GB quad channel set (F3-17000CL9Q-16GBZH). This set is on ASUS QVL and is available for purchase in Poland (yet not cheap). Has anybody been using this set and can confirm they are OK/stable for RIVE ?
It is better with Corsairs - I get BSOD only on every 5-th tests, so it has improved but it is not ideal. So it would mean memory is the issue or at least it has important role.
Good news is that BIOS sets correct voltages on memeory on Auto (1.5) and XMP (1.65) for these Corsairs.
I looked at the warranty card for the CPU that I received form the seller. He bought entire system in one take (so everything is on the warranty card) - 3930K + RIVE + PSU, GPU, SSD etc etc. He bought G.SKILL RipjawsZ DDR3 2133MHz 16GB quad channel set (F3-17000CL9Q-16GBZH). This set is on ASUS QVL and is available for purchase in Poland (yet not cheap). Has anybody been using this set and can confirm they are OK/stable for RIVE ?
Related Content
- Z370 Strix F - Aura sync nor showing Cooler and IO shroud itself in Z370 / Z390
- Rampage V Windows lockup fault codes fd and 61 solved! in X99
- Maximus V Formula - very slow boot, hangs 1 min on "starting windows" in Other Motherboards
- Rampage IV formula unable to power on in Other Motherboards
- [Maximus V Gene] - CSM disabled + Windows UEFI mode = Wake On Lan does not work in Other Motherboards
