B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
trying to boot my computer today, I seemed to get stuck on post code 71. I hit the reset button, it booted fine, went into Windows. I went "huh...weird".
Came to the forums, downloaded bios 1.12, rebooted, updated bios (from within bios), reloaded my profile from the USB, and then it wouldn't get past post code 71.
CPU was overclocked to 5.5 on 2 cores, 5.4 3-4 cores, 5.3 5-6 cores, 5.2 7 cores, 5.1 8 cores. Adaptive voltage, 1.275 target, +25 offset, -75% droop, mode 1, 750 PWM. Memory at XMP profile 1 (6000).
It had been running fine there on bios 1.10 for weeks, but clearly had a hitch when I first booted today before updating so I'm not sure it has anything to do with the update.
if I return the CPU to all OC settings "auto", it seems to boot fine. I ran the OC robot, just to see what it would do. It settled on 5.3GHz all-core. I applied, rebooted, went into the bios, it seemed to hold. Tried to reboot again to get to Windows and it stuck on code 71. I hit the safe boot button, and now I'm getting 6 of the angry beeps before it posts.
I can get into Windows if I put all the CPU settings at "auto".
6 angry beeps indicates "flash update failed". So I tried to re-flash the BIOS with 1.12 again and it seems like it took, but if I have to safe boot or clear CMOS I get the "CMOS Checksum error" message and have to go into setup again to re-apply all my settings. I also tried to clear CMOS, then load optimized defaults and do all my settings from scratch and it still seems to get stuck on post code 71 anytime my CPU isn't set at "auto" everything.
I was thinking that maybe my contact frame was causing an issue, so I took the cold block off (which is getting wonderful contact it seems from the paste pattern), loosened, and re-snugged the contact frame, re-pasted, booted back up, same situation.
I'm going to run memtest64 for a bit, but I wanted to get this post going while it was working just in case that kills it.
Ideas? Post code 71 is "South Bridge DXE SMM initialization has started" by the way.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/08 22:36:36
(permalink)
Also tried a different bios switch position. Same issue.
Ran memtest64 for about 10 minutes and found no errors. Ran cinebench on default cpu settings and it ran fine. Increased clocks in E-leet, ran fine, reset defaults in E-leet and the system froze. Tried to open xtu and it gives a warning about bit locker or something. Tried to scan vf curve and the system just reboots. It's 1:30am and I need to get some sleep, but I am at a loss here. I can get into Windows with default settings, but I'm not convinced it's stable there or that everything is working right at default clocks...just good enough to get into windows.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
Bee_Dee_3_Dee
SSC Member
- Total Posts : 598
- Reward points : 0
- Joined: 2004/11/20 02:40:52
- Status: offline
- Ribbons : 23
Re: stuck on post code 71
2022/07/08 22:37:05
(permalink)
ambient temperature fluctuations where i live have been drastic the last 24 hours. How about temps in ur local? ur post and temps here remind me: when i slowly built my system, started last few days of February all seemed fine for awhile, but by March there were higher than normal temps; and night time temps dropped below normal and the extra March Highs kept returning by noon. i had strange things occurring for more than 10 days. and didn't think twice about temperatures fluctuating so much. diagnosed everything. then one day i glanced at Windows Explorer on my 2nd of 4 monitors and barely witnessed out of the corner of my eye, that one of the seven SATA SSDs' Drives had its Drive Letter vanish in Explorer. Rebooted and it was back. Less than a minuet later it disappeared again. I said eureka! it's just a cable. Replaced the SATA SSD's Cable and all problems went away for good. i was so happy. i thought it was something big and it was so so simple. When the temps were low all was fine. when things got warm the faulty cable failed. GL  Edit: Could it be that BIOS v1.10 became corrupt -- and needed flashing -- and for some reason ur build doesn't like BIOS v1.12?
post edited by Bee_Dee_3_Dee - 2022/07/08 22:46:51
> PSU: Seasonic VERTEX GX-1200> Mainboard: EVGA Z690 CLASSIFIED (BIOS v1.03 12/3/2021)> CPU: Intel Core i9-12900KF> AIO: EVGA CLC 240mm> Physical Memory: G.SKILL Trident Z5 32GB (2 x 16GB) DDR5 5600> Physical Memory Model#: F5-5600U3636C16GX2-TZ5S> Monitor: Alienware AW2721D> Video Card: EVGA 3090 FTW3 ULTRA GAMING> Case: Cooler Master HAF X (942)> DAS: Sabrent DS-SC5B
|
Sajin
EVGA Forum Moderator
- Total Posts : 49164
- Reward points : 0
- Joined: 2010/06/07 21:11:51
- Location: Texas, USA.
- Status: online
- Ribbons : 199
Re: stuck on post code 71
2022/07/09 03:08:19
(permalink)
Sounds like the cpu is going south.
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/09 06:27:15
(permalink)
Also re-seated memory, checked power cables, and my temperatures are good. Idling around 32 last night and never hit higher than ~82 (while r23 benching). OC Robot only got to 76.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/09 07:34:54
(permalink)
Which contact frame and what torque specs? Anything with mem, pcie, or chipset in a no post situation where a frame is in the mix needs to have the frame ruled out before changing any bios or oc settings. Washer mod had the same debacle and all of the ilm mods come with “at your own risk” disclaimers. These aren’t cooling solutions. If the pin contact or pressure is off even a small bit in either direction you can lose oc headroom, drop channels etc
So it’s would start by ruling out the oc first. Reset cmos and see if it will post with the new bios. It is absolutely possible for a bios position to become corrupted. So the easiest way to rule out the switch position and bios flashed there is by trying a diff position. Maybe you tried it? If still no post at all, you may also have socket pin damage and will need to inspect the socket for this. LGA sockets are more vulnerable to damage than their cpus due to the pins being on the board and not the cpu. I’d be very surprised if the cpu is damaged unless you went real hard on voltage. Doubt that is the case.
Assume you do not have a second z690 board you can test with? As I told you earlier I’ll be home late tonight from the beach and will be happy to help with troubleshooting tomorrow and this week.
That should give you a few things to try though. Good luck and please report back with any updates or questions. About to board a ferry so will maybe have some service and can follow along.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/09 07:43:08
(permalink)
Hmm I just reread your op, 1.12 let you load a usb profile from 1.10? I was not able to do this between 1.07 and 1.09 so either they fixed this or this is another things to check. How to explain this, if some settings in 1.12 are removed or added from 1.10, it is maybe possible that this corrupts the new bios after loading the profile from 1.10. Reset cmos and try manually setting the oc again. I know it’s a pain but if you’re not posting, change bios switch to 2 or 3, flash to 1.12 and manually set things if/when it posts and see how that goes.
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/09 10:33:18
(permalink)
Yes, I had said that, but it was late so maybe it wasn't clear. I tried to go to the second bios switch position, load defaults, update bios, and then manually put all my settings in. It has the same behavior where if I leave the cpu on "auto" for everything, it will go into windows, but xtu is screwed up and causes the system to just reboot. If I change any cpu settings it just gives code 71 again and sits there. I also tried to let it sit for a few minutes on code 71 to see if it would ever figure itself out.
I had used up the last of my good thermal paste trying to re-seat the contact frame. I just ran to best buy and got some corsair paste that.. Well... It should be good enough to let me swap the original ilm back in to see if that let's me get back up and running to rule in/out the frame.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/09 10:42:12
(permalink)
Gotcha. I’m getting off the ferry now and will be driving the rest of the day. I’ll check in tonight. Good luck!
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/09 12:52:43
(permalink)
I put the stock ILM back in. I'm back up and running with a mild overclock. Seems to be getting through POST now. The temps aren't great, but that honestly could be due to a combination of higher ambient and this Corsair XTM50. It came out so watery at first I actually cleaned it off and applied again lol. Not a consistent tube and from reviews, it is nowhere near as good as the KPx, Prolimatech Pk-3, or even the kryonaut (gray) I've been using (was doing testing between the three of those on my 11900k...the TG was the worst one of the three, other two were near identical in performance for me). I had one crash, but I was trying to tweak things in ELEET. Not sure it always sets the voltage correctly when you don't try and manually replicate what was put into the BIOS. I'll have to run some more checks to see if this is good for now. Also, I ordered more KPx lol. Edit: I ran r23 for a while and it seemed ok...but hot...then I went to open CPUz and it crashed. So I definitely don't feel like I'm out of the woods yet.
post edited by B0baganoosh - 2022/07/09 13:03:50
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
Bee_Dee_3_Dee
SSC Member
- Total Posts : 598
- Reward points : 0
- Joined: 2004/11/20 02:40:52
- Status: offline
- Ribbons : 23
Re: stuck on post code 71
2022/07/11 01:59:06
(permalink)
B0baganoosh
I put the stock ILM back in. I'm back up and running with a mild overclock. Seems to be getting through POST now. The temps aren't great, but that honestly could be due to a combination of higher ambient and this Corsair XTM50. It came out so watery at first I actually cleaned it off and applied again lol. Not a consistent tube and from reviews, it is nowhere near as good as the KPx, Prolimatech Pk-3, or even the kryonaut (gray) I've been using (was doing testing between the three of those on my 11900k...the TG was the worst one of the three, other two were near identical in performance for me).
I had one crash, but I was trying to tweak things in ELEET. Not sure it always sets the voltage correctly when you don't try and manually replicate what was put into the BIOS.
I'll have to run some more checks to see if this is good for now. Also, I ordered more KPx lol.
Edit: I ran r23 for a while and it seemed ok...but hot...then I went to open CPUz and it crashed. So I definitely don't feel like I'm out of the woods yet.
i had trouble with CPU-Z like it locking-up PC before replacing that bad SATA cable bak in May. Currently, NP but it doesn't like to run a 2nd time in a session as far as the SPD Tab goes. The 1st time i launch it the SPD Tab properly displays all information. But the 2nd or any additional times in the same session it is launched, all is greyed out on SPD Tab.(I don't remember but i think Logging off and back on doesn't fix it -- i would have tried.) It is no prob. i just don't launch it a 2nd time in a session. (No reason to try newer BIOS just to fix it. All else is just fine.) But what matters is ur predicament, B0baganoosh. I was just thinking... B0baganoosh said yesterday that he could get a mild OC... but CPU-Z crashed it... so i thought it can't be Windows or Drivers for obvious reasons. And yesterday i was thinking Hardware and Software Hardware or Software... and Hardware constantly won out as culprit. Just now (well started this post because) i realized that i should re-read all this thread's posts because shouldn't u swap ur PSU? At 1st i'm thinking as i re-read how the running mild OC but Hot part isn't explained by a PSU. But then i'm thinking the 71, currently only being able to get a Mild OC, and it running a little Hot could be the PSU's fault. Try swapping it.
> PSU: Seasonic VERTEX GX-1200> Mainboard: EVGA Z690 CLASSIFIED (BIOS v1.03 12/3/2021)> CPU: Intel Core i9-12900KF> AIO: EVGA CLC 240mm> Physical Memory: G.SKILL Trident Z5 32GB (2 x 16GB) DDR5 5600> Physical Memory Model#: F5-5600U3636C16GX2-TZ5S> Monitor: Alienware AW2721D> Video Card: EVGA 3090 FTW3 ULTRA GAMING> Case: Cooler Master HAF X (942)> DAS: Sabrent DS-SC5B
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/11 07:10:00
(permalink)
I do think the contact frame might be part of the equation. One thing that I noticed, that most people who installed them probably noticed, but if you play with the stock ILM, unclamped, there's barely any pressure on the screws. When you lock down the clamp, it puts a great deal of pressure on the screws. Following the instructions from TG on the contact frame (even though I was using a different one) leads to a very similar amount of pressure as the stock ILM pre-clamping. I can't help but wonder if I just didn't have it clamped down nearly hard enough. That said, how would the water block being bolted on top not put even more pressure and make whatever pressure I had put on the contact frame irrelevant? I drew this up, and obviously it isn't to scale and doesn't include everything, but I feel like it shows what I mean:  Once you tighten down the cold-block mounting nuts on the posts, any pressure the frame is applying to the CPU should be irrelevant, right? The only thing I can see happening is if the frame-screws are too loose, when you tighten down the water-block, it would actually push the back-plate off the board, or at least completely release pressure, which would let the frame and back-plate slip down with gravity over time (with the motherboard in the up-right position). What I'm thinking is: I had the system up and running, set clocks, tested a whole bunch of benchmarks and stability tests, played hours of games, etc. and it all worked fine. Then the system sat for a week or so (unpowered), and started presenting issues after that time. I'm wondering if the frame I was using, in combination with being too loose after torqueing the cold-plate down, would have slipped with the back-plate and then caused some sort of poor contact on the CPU pins, or maybe even had a conduction path on some capacitors on the CPU or something. I don't think the back-plate would have touched the caps on the back of the motherboard, but I'm just not sure what could happen in that situation. I don't believe the CPU pins would have had bad pressure as the cold-block would have been pressing it into the socket quite hard. If the Thermal Grizzly contact frame ever returns to stock, I might purchase that one and give this another try, knowing they likely did a better job using a non-conductive anodize and that their machining tolerances might be a smidge better. I'll also try a bit more pressure I think, to try and replicate a clamped-ILM. I should have tried to shimmy the back-plate when I was seeing issues to see if it moved before I removed my block and the frame.
post edited by B0baganoosh - 2022/07/11 07:12:28
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/11 14:28:26
(permalink)
B0baganoosh
I do think the contact frame might be part of the equation. One thing that I noticed, that most people who installed them probably noticed, but if you play with the stock ILM, unclamped, there's barely any pressure on the screws. When you lock down the clamp, it puts a great deal of pressure on the screws. Following the instructions from TG on the contact frame (even though I was using a different one) leads to a very similar amount of pressure as the stock ILM pre-clamping. I can't help but wonder if I just didn't have it clamped down nearly hard enough.
That said, how would the water block being bolted on top not put even more pressure and make whatever pressure I had put on the contact frame irrelevant?
I drew this up, and obviously it isn't to scale and doesn't include everything, but I feel like it shows what I mean:
Once you tighten down the cold-block mounting nuts on the posts, any pressure the frame is applying to the CPU should be irrelevant, right? The only thing I can see happening is if the frame-screws are too loose, when you tighten down the water-block, it would actually push the back-plate off the board, or at least completely release pressure, which would let the frame and back-plate slip down with gravity over time (with the motherboard in the up-right position).
What I'm thinking is: I had the system up and running, set clocks, tested a whole bunch of benchmarks and stability tests, played hours of games, etc. and it all worked fine. Then the system sat for a week or so (unpowered), and started presenting issues after that time. I'm wondering if the frame I was using, in combination with being too loose after torqueing the cold-plate down, would have slipped with the back-plate and then caused some sort of poor contact on the CPU pins, or maybe even had a conduction path on some capacitors on the CPU or something. I don't think the back-plate would have touched the caps on the back of the motherboard, but I'm just not sure what could happen in that situation. I don't believe the CPU pins would have had bad pressure as the cold-block would have been pressing it into the socket quite hard.
If the Thermal Grizzly contact frame ever returns to stock, I might purchase that one and give this another try, knowing they likely did a better job using a non-conductive anodize and that their machining tolerances might be a smidge better. I'll also try a bit more pressure I think, to try and replicate a clamped-ILM. I should have tried to shimmy the back-plate when I was seeing issues to see if it moved before I removed my block and the frame.
i'm not 100% on how the frame works, but here's my understanding: the stock ILM is putting too much pressure on the middle of the ihs to the point where it is bending the pcb like a half pipe. this dip in the middle where the stock ILM presses it down causes paste spread to be uneven with more in the middle and less (better contact/spread) on top and bottom. to counter this, some of the oc community was using spare boards to cut the socket out to lap the ihs flat IN the socket with stock ilm pressure to get it flat despite the aforementioned half pipe scenario. there's an article about this written by Splave on toms hardware https://www.tomshardware.com/news/intel-12900k-world-record-with-hacksaw so after this, and the buildzoid washer mod thread in the community attempting to address the bowing pcb from stock ILM, the contact frame ideas were born. how i think the frames work? they apply ENOUGH pressure (less than the stock ILM as you said) to flatten the pcb without creating a concave situation, but to still maintain pin pressure and contact across the entirety of the land grid array. so the torque specs of the frame matter in this respect. too much pressure is maybe the better case, but here i've seen people claim they have lost oc headroom or channels so i can't speak to what causes this other than maybe it creates a unique concave situation like the stock ILM with additional pin contact/pressure issues? too little pressure of course will lead to poor contact on the pins closest to the center where the stock ILM tabs would be. any update on performance? things smoother now? titan rig is where i snagged my frame, but with the kingpins arriving tomorrow and wednesday i'll be working on that before i get in to the frame mounting.
post edited by zippytek - 2022/07/12 06:34:55
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/12 06:59:52
(permalink)
I haven't poked at it since I got it up and running the other day. My KPx order should be here Friday so I'll do some more testing. I'm going to be just using the stock ILM until a TG frame comes in stock. MemoryC apparently had 5 today, but they sold out in the 10 minutes from when I got my email notice to when I got on the site. I understand what goes wrong with the stock ILM and I know what the aftermarket frames are trying to do, but my question is really about how much less pressure the aftermarket frames put on than the stock ILM. Maybe I'm bolting my cold block down too tight, but the only thing I can think of for why it worked great one weekend and not a few days later would be that the time and gravity caused the frame and backplate to slide due to it not being under pressure anymore after the cold block was bolted down. My Optimus water block just has 4 little nuts on the back of the motherboard, so it isn't a solution that also holds the back-plate in place. I feel like none of this would matter if I had a back plate that did that. My old board, but same cold block and mounting nuts:   You can see how the frame's back-plate wouldn't have anything holding it in place if suddenly the screws were loose. The tolerance of the screw threads and the holes would be the only thing keeping it in place and I know there's definitely a bit of wiggle room there on the z690. You can see here what the classy looks like:  (it opens bigger if you open it in a new tab) I don't think the back plate will touch anything if it slides down. it is heavy though, so if the screws are suddenly loose, I think it could slide down with gravity (it is installed in this orientation too). So then it would be a question about the front of the CPU and where the frame sits:  I don't think anything is shorted out because the frame I have is anodized (so it shouldn't be very conductive) and it isn't like the system won't boot at all. That said, I'm wondering what would happen if you put an unwelcomed impedance path along the top of the CPU somewhere (even if it was a relatively high impedance), say, should the contact frame slide down and make contact with the CPU's PCB edge somewhere. It would also generate a capacitance there if it was close enough. So I'm wondering if it is possible that it is just close enough to cause interference when the voltages or frequencies exceed a certain level. In any case, the "fix" would have to be more pressure on the screws and/or a frame that has different dimensions underneath or is made in a way that prevents it from sliding. I'm curious to know how snug the IHS tabs fit in the TG contact frame. There was a little bit of slop in the one I bought. Not a ton, but I'm wondering now if it was too much. Of course, this is all just theory for now. I know Raptor lake is already too far along, but I'm really hoping that future gens come with a different mounting method. For now I'll update this thread when I know more about my current situation.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/13 08:55:57
(permalink)
Just managed to snag a Thermal Grizzly contact frame off MemoryC.com. They had 2 in stock and I had just so happen to refresh the page before I went to lunch when they showed up.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/13 10:50:17
(permalink)
nice! hoping the lapping frames appear at some point too. hope that one gives better results!
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/17 10:34:58
(permalink)
Update today. I received the Thermal Grizzly Frame. Here are some picture comparisons:    Some notable differences: The opening in the TG frame is larger. This means that once installed, it actually doesn't press on the top and bottom edge of the CPU. You can see this in the bottom picture. In the second picture, you can see that there's actually a slightly deeper section where the side-tabs of the CPU sit. The Amify frame does not have that. The TG frame didn't come with any sort of rubber sticky-pads that go where the frame could touch the motherboard. The Amify and Thermalright frames both come with those. What seems to be quite different with the TG frame is that I don't believe it actually touches the motherboard. Once pressured in place, it seems to just sit on the CPU and not squish the rubber pads at all. It is possible the Amify pads didn't touch either, but I was paying more attention to this one considering it's metal hovering right over the board. Results: 1st boot - POST code 71. I did the snug, then add 90°. So I continued to have my hunch that it wasn't enough pressure. 2nd Boot - Didn't re-paste, just took the block off, added 45° turn on frame screws, put the block back on and it actually booted up great. It was a bit hot, so I figured my quick no-paste run was the culprit there, so I took the block back off, cleaned everything, re-pasted, and then: 3rd boot - POST code 71 again. I took the block back off, gave the screws another 45°, cleaned it up nice this time, re-pasted... 4th boot - Good! Still running a few tests, but I'm back to 5.5 on 1-2 cores, 5.4GHz on 3-4 cores, 5.3GHz on 5-6 cores, 5.2GHz 7 cores, 5.1GHz 8 cores. I'm sure I could do some benchmarks with a higher all-core, but Cinebench would run hotter than I'm comfortable with. Any of the game testing I've done has shown that getting 1-4 cores faster has the most impact, so I really wanted that 5.5 1-2 cores, 5.4 3-4 cores to happen. So far so good. We'll see what happens as the next week or two go by, because that's when it failed on me last time. It's possible I didn't have the Amify one screwed down tight enough. The screws felt like they were maxed out on that one though and it didn't feel like I could safely add more pressure, which I'm thinking might have meant that the non-conductive feet they gave me for it were in fact on the motherboard and preventing it from tightening anymore. I'm also curious about my cold-block pressure. I have the Optimus Foundation block on it and I snug that thing up pretty good. I'm curious if that might be more pressure than an AIO with a mounting frame that bottoms out or something and that's part of why this is so sensitive to how much pressure I put on the screws. Any tips (sanity checks lol) there would be appreciated.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
Bee_Dee_3_Dee
SSC Member
- Total Posts : 598
- Reward points : 0
- Joined: 2004/11/20 02:40:52
- Status: offline
- Ribbons : 23
Re: stuck on post code 71
2022/07/17 11:42:55
(permalink)
congratz!!! Play Far Cry 6!  try story mode and stick to the Operations (Main Story). Amigos Guapo and Boom Boom were cool. i'm no longer afraid of crocodiles (Guapo).lol Boom Boom (dog) is perfect for flanking enemies.
> PSU: Seasonic VERTEX GX-1200> Mainboard: EVGA Z690 CLASSIFIED (BIOS v1.03 12/3/2021)> CPU: Intel Core i9-12900KF> AIO: EVGA CLC 240mm> Physical Memory: G.SKILL Trident Z5 32GB (2 x 16GB) DDR5 5600> Physical Memory Model#: F5-5600U3636C16GX2-TZ5S> Monitor: Alienware AW2721D> Video Card: EVGA 3090 FTW3 ULTRA GAMING> Case: Cooler Master HAF X (942)> DAS: Sabrent DS-SC5B
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/18 06:37:23
(permalink)
I suppose to continue my post above:
Boot ~6: POST code 71 again...just hit the reset button on the case...
Boot 7: fine. I ran a few tests, then played 7 days to die for about 4 hours and it was fine.
Confidence is...not high. I'm not really sure how to combat this. It's definitely related to the frame and having adequate pressure, but something seems to move over time and that causes this glitch.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/18 06:41:56
(permalink)
B0baganoosh
Update today. I received the Thermal Grizzly Frame. Here are some picture comparisons:
Some notable differences: The opening in the TG frame is larger. This means that once installed, it actually doesn't press on the top and bottom edge of the CPU. You can see this in the bottom picture. In the second picture, you can see that there's actually a slightly deeper section where the side-tabs of the CPU sit. The Amify frame does not have that. The TG frame didn't come with any sort of rubber sticky-pads that go where the frame could touch the motherboard. The Amify and Thermalright frames both come with those. What seems to be quite different with the TG frame is that I don't believe it actually touches the motherboard. Once pressured in place, it seems to just sit on the CPU and not squish the rubber pads at all. It is possible the Amify pads didn't touch either, but I was paying more attention to this one considering it's metal hovering right over the board.
Results:
1st boot - POST code 71. I did the snug, then add 90°. So I continued to have my hunch that it wasn't enough pressure.
2nd Boot - Didn't re-paste, just took the block off, added 45° turn on frame screws, put the block back on and it actually booted up great. It was a bit hot, so I figured my quick no-paste run was the culprit there, so I took the block back off, cleaned everything, re-pasted, and then:
3rd boot - POST code 71 again. I took the block back off, gave the screws another 45°, cleaned it up nice this time, re-pasted...
4th boot - Good!
Still running a few tests, but I'm back to 5.5 on 1-2 cores, 5.4GHz on 3-4 cores, 5.3GHz on 5-6 cores, 5.2GHz 7 cores, 5.1GHz 8 cores. I'm sure I could do some benchmarks with a higher all-core, but Cinebench would run hotter than I'm comfortable with. Any of the game testing I've done has shown that getting 1-4 cores faster has the most impact, so I really wanted that 5.5 1-2 cores, 5.4 3-4 cores to happen. So far so good. We'll see what happens as the next week or two go by, because that's when it failed on me last time.
It's possible I didn't have the Amify one screwed down tight enough. The screws felt like they were maxed out on that one though and it didn't feel like I could safely add more pressure, which I'm thinking might have meant that the non-conductive feet they gave me for it were in fact on the motherboard and preventing it from tightening anymore.
I'm also curious about my cold-block pressure. I have the Optimus Foundation block on it and I snug that thing up pretty good. I'm curious if that might be more pressure than an AIO with a mounting frame that bottoms out or something and that's part of why this is so sensitive to how much pressure I put on the screws. Any tips (sanity checks lol) there would be appreciated.
great detail and a happy ending! just what we love to see. ime, for tightening water blocks the most important part is applying even pressure via cross tightening. they need to be snug, but you don't have to crank them down to where the springs are 100% compressed. let the springs apply the pressure, and if you were careful to cross tighten equally, you should be a nice even sponge-like pattern on the ihs for paste spread. if it shows a corner or edge with thicket application than the rest of the spread, that's an indication you tightened too much (or too little) in an area. and man, i thought i asked this before but what is Amify? i've never heard of that and would def not recommend using that frame. i figured you were using the Thermalright frame before you managed to snag the TG. in this case i think the devil is 100% in the details and when it comes to mounting pressure and pin contact, the tight clearances in the cad model for a frame like this really requires a design from someone who went to the trouble to test things after it was finished. i would hope thermalright did this, but i know for a fact that Roman did and also sent the frame samples to a bunch of xoc guys in the community to get real world feedback. Enjoy!
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/18 06:44:50
(permalink)
B0baganoosh
I suppose to continue my post above:
Boot ~6: POST code 71 again...just hit the reset button on the case...
Boot 7: fine. I ran a few tests, then played 7 days to die for about 4 hours and it was fine.
Confidence is...not high. I'm not really sure how to combat this. It's definitely related to the frame and having adequate pressure, but something seems to move over time and that causes this glitch.
oof, looks like i spoke too soon. what paste are you using? def cross tightening the block? can you share any pics of the paste spread to see if there are clues about pressure? i know the optimus blocks are HEAVY. i don't think your backplate screws are a problem, but does optimus not make a backplate for their blocks? i swear my foundation block has a backplate but it did not come with the block. i ordered it separately
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/18 07:12:16
(permalink)
No backplate for their block. It just comes with this hardware. I haven't seen one anywhere on their site either. Yes, definitely cross-tightening. Using Kingpin KPx. I don't have a picture, but generally it looks pretty good. It does look like there's a bit of a high spot in the middle, but the lower spots (where any paste remains after removing the block) are mostly on the very outermost edges (left and right mainly). It's a bit hard to tell if they're actually that much lower or if the suction when I pull the block off sort of sucks it back onto the CPU right at the edge when I pull the block off. Last time I removed it with the ILM on there, I could see it look a lot like some of the images on GN's video. upper left here:  I had attributed that to I must have been rushing and accidentally overtightened the left-side more than the right side. There are no springs with Optimus. You just have to push down on the center of the block to hopefully center the thing, tighten the thumb screws just until they all kiss the bracket, then you cross-tighten to do it evenly. about 9/10 times I get a very centered squish-out, but I have seen a couple like the upper left picture above and assume that it wasn't my best.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/18 07:55:52
(permalink)
so i'm clear, these photos are NOT yours?
first thought on those images: that is way too much paste :)
if i have to choose which is the best spread? i'm prob going for top right. still a bit thick for my liking. are you spreading the paste before you put the block on? i know there's a big debate about this but i always take 2 minutes and spread the paste evenly so it covers the entire ihs. tightening the block down after this should produce an even spread since without needing to force it all to the edges using pressure alone, it will remove the excess and leave the best layer of transfer without too much excess.
if it mostly looks even when you remove the block, you're probably fine. generally paste spread isn't much of a concern but with 12th gen it can be a good indicator of how the ILM and block mounting pressure may influence temps and performance since it varies so wildly between stock ILM and frames.
isn't this fun? argh
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/18 08:40:25
(permalink)
zippytek
so i'm clear, these photos are NOT yours?
first thought on those images: that is way too much paste :)
if i have to choose which is the best spread? i'm prob going for top right. still a bit thick for my liking. are you spreading the paste before you put the block on? i know there's a big debate about this but i always take 2 minutes and spread the paste evenly so it covers the entire ihs. tightening the block down after this should produce an even spread since without needing to force it all to the edges using pressure alone, it will remove the excess and leave the best layer of transfer without too much excess.
if it mostly looks even when you remove the block, you're probably fine. generally paste spread isn't much of a concern but with 12th gen it can be a good indicator of how the ILM and block mounting pressure may influence temps and performance since it varies so wildly between stock ILM and frames.
isn't this fun? argh
Correct, not my pictures, these are from a Tom's article of the lga1700 issue or something (pulled from google search). I agree, too much paste! When opened up, mine tend to look like the upper left image, but squished out on about 90% of the CPU, only thicker spots where you're not just looking at the IHS are the outer edges and maybe corners. Like I said, I'm not sure if that's where it's pulled back in when removing the block. Here's one of mine from 11900k just to show how much paste I tend to use and what I was talking about with uneven pressure from overtightening one side by accident. As I've said, I've learned from that and am much more consistent these days about centering the pressure, I just don't have a recent picture:   So with this picture, While you can obviously tell there was good pressure in the center and right side, the left side had crappy contact and when I pulled the block off, it definitely suctioned there into a much higher spot than it would have been once installed. This is what I was talking about on the new CPU. If you look at the upper left image from the set above pulled off google images, then imagine that the good contact area covers the whole middle of the CPU, but all edges look like the left edge in that picture...that's what I get now generally. I don't know why I always forget to take a picture of that lol. Also, I've tried all sorts of paste methods, usually I get pretty bad results from spreading the paste manually before putting the block on. I always end up with too much paste. With 11900k I had the best results with a sort of oblong blob in the middle. It always feels like too much paste to me, but it performed the best in my experiments. 12900k so far I've had the best luck with a single grain-of rice thick line down the middle and a tiny dot in each corner. Also, as far as I've tested with 11900k (and now 12900k), the only paste I'm interested in using is Prolimatech pk-3 nano or KPx. Kryonaut (gray) isn't as good, but not bad, and everything else I've tested is much worse than those. I haven't tested the new TG Kryonaut extreme (pink) stuff. Most of the TG stuff seems way over priced compared to similar stuff from other companies. The aftermarket backplate sounds like a reasonable idea, but Optimus is pretty adamant about not using one and I'd have to go find some hardware that I could replace these posts with. I've heard that I could use some M4 bolts and standoffs. I just wonder at what point am I creating more problems than solving.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/18 10:25:26
(permalink)
if they don't recommend one, i wouldn't bother. i swear my amd block has one tho...maybe they just don't do them for intel? i've been using HK blocks and never bothered to test my optimus blocks. one day i will!
spread in those pics looks pretty good, that's what you'd want across the whole ihs if possible. admittedly, i've gotten in the habit of using the spatula to spread bc when you run cold you need a thicker layer to avoid cracks, and if you don't have coverage across the whole ihs it can mean tearing down and repasting after cooldown which is not ideal. for ambient setups, kryo extreme is overkill imo. it's so expensive that it really only makes sense to use it for LN2 on expensive GPUs. for how tacky it is, i really hate putting that stuff on cpus. way worse to clean off around socket and ilm than kpx.
lastly, if you see mostly even spread but one side thicket like in the photo, that can sometimes just be explained by how you removed the block. suction can cause it to detach from one side entirely while that quick moment where it is angled a bit can move some of the paste towards what appears thicker. so i wouldn't imagine your paste application is at all to blame for the post codes.
did we rule out any possible socket pin damage? i think so but don't remember. if it mostly works i wouldn't expect that but if it were me i might even do another 1/8-1/4 turn on the frame to see if it helps (assuming the intermittent post codes continue)
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/18 10:46:15
(permalink)
I confirmed there's no socket damage. Searching online, there are some folks on overclock.net who are quite unpleased with Optimus when it comes to LGA1700. Optimus had said "it's great, our testing showed better than backplate" (paraphrasing), but a few users have said they've ran into OC instability and that Alder-lake would make that even worse and Optimus has apparently ghosted them on that forum even though they used to post regularly. I'm not inclined to think a backplate will do a whole lot, considering the Classified board (and pretty much all EVGA boards) is much thicker than many on the market and stiffer too. So it already doesn't flex as much as some. In any case, I saw a few people recommend either the heavy Watercool backplate (that also replaces the socket backplate...which made me a bit nervous and is $40+shipping) and also the EK Velocity LGA1700 backplate. The EK one is cheap enough (found a US store, only paid $7 for it with shipping) to give it a shot. One of the biggest concerns I've had with this whole frame thing is that it'd be a giant PITA to lay my case down and try to do this install horizontally. Added to that, the Optimus hardware involves no extra backplate, so I have to basically put some tape on my backplate to hold it in place and then veeeery gently tighten the contact frame screws. There's a chance that ends up with a crooked frame because it could pull one corner of the backplate in off-center. Every video or instruction set I've seen for the contact frame starts by saying "well hopefully you have a cooler back-plate on the board that holds the CPU backplate in place." Soo...I'm just fairly convinced that all the opposing pressures of this current setup are contributing to this intermittent issue. I ordered the EK backplate and I'm going to see if the Optimus studs just so happen to thread into it. If not, I can get some screws easily enough to make it work for a test. My main concern looking at it is there aren't a lot of threads in that metal plate. I might need to put a screw through from the rear and then tighten it to the motherboard with a fiber washer and nut so that the head of the screw from the rear is what actually does the holding. I'm just currently a little concerned that I go to tighten the block on and it pulls the screw right out of the EK backplate lol. So my plan for that: 1. figure out how to install that backplate, get it fairly snug on the back of the motherboard. 2. loosen up the contact frame, basically start that process over, but before tightening anything: 3. with the Intel backplate loose, but centered, tighten up the EK backplate so that it firmly holds the Intel backplate in place. 4. snug up the TG contact frame on the CPU. 5. paste. install block. cross fingers.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
Bee_Dee_3_Dee
SSC Member
- Total Posts : 598
- Reward points : 0
- Joined: 2004/11/20 02:40:52
- Status: offline
- Ribbons : 23
Re: stuck on post code 71
2022/07/19 06:17:44
(permalink)
> PSU: Seasonic VERTEX GX-1200> Mainboard: EVGA Z690 CLASSIFIED (BIOS v1.03 12/3/2021)> CPU: Intel Core i9-12900KF> AIO: EVGA CLC 240mm> Physical Memory: G.SKILL Trident Z5 32GB (2 x 16GB) DDR5 5600> Physical Memory Model#: F5-5600U3636C16GX2-TZ5S> Monitor: Alienware AW2721D> Video Card: EVGA 3090 FTW3 ULTRA GAMING> Case: Cooler Master HAF X (942)> DAS: Sabrent DS-SC5B
|
B0baganoosh
CLASSIFIED Member
- Total Posts : 2363
- Reward points : 0
- Joined: 2009/08/04 04:27:18
- Status: offline
- Ribbons : 39
Re: stuck on post code 71
2022/07/19 06:23:34
(permalink)
Bee_Dee_3_Dee
i saw a JayzTwoCents YT yesterday... wasn't sure if pertinent to ur prob, but remembered two places in it to mention:
(See 6:28 and 10:32.)
Fixing Intel's Mistakes...
Jul 16, 2022 I had to tear down my new personal 12th Gen Intel rig again!
i didn't watch much else but a few minuets at 6:28 and 10:32. so other stuff in it might apply to stuck on Code 71 prob and other ones.
Yeah, I've watched those. The thing I've noticed about Jay's video, as well as GN and Der8our, is that they all use a cold-block backplate behind the z690 socket backplate. So I've ordered one of those as I think it may aid in the proper leveling of the contact frame, as mentioned above.
6Q6CPFHPBPCU691 is a discount code anyone can use. i9 13900k - EVGA Z690 Classy - Nvidia RTX 4090 FE - G.Skill 32GB DDR5-6000 - WD SN850 2TB NVMe Gen4 - Be Quiet! Straight Power 12 1200W - Be Quiet! Dark Base 900 Pro. MO-RA3 420 Pro. Dark Palimpsest MODS RIGS post for build notes.
|
Bee_Dee_3_Dee
SSC Member
- Total Posts : 598
- Reward points : 0
- Joined: 2004/11/20 02:40:52
- Status: offline
- Ribbons : 23
Re: stuck on post code 71
2022/07/19 06:29:44
(permalink)
B0baganoosh
Bee_Dee_3_Dee
i saw a JayzTwoCents YT yesterday... wasn't sure if pertinent to ur prob, but remembered two places in it to mention:
(See 6:28 and 10:32.)
Fixing Intel's Mistakes...
Jul 16, 2022 I had to tear down my new personal 12th Gen Intel rig again!
i didn't watch much else but a few minuets at 6:28 and 10:32. so other stuff in it might apply to stuck on Code 71 prob and other ones.
Yeah, I've watched those. The thing I've noticed about Jay's video, as well as GN and Der8our, is that they all use a cold-block backplate behind the z690 socket backplate. So I've ordered one of those as I think it may aid in the proper leveling of the contact frame, as mentioned above.
cool. i'm finishing watching the vid @3x speed (normal pitch) with Enhancer for Youtube. And it doesn't seem much else matters in the vid. interesting how he spreads paste and the brand he uses. i think i'll switch to using it if i ever need paste. i remember when it was 1st on the market and sounds cool.
> PSU: Seasonic VERTEX GX-1200> Mainboard: EVGA Z690 CLASSIFIED (BIOS v1.03 12/3/2021)> CPU: Intel Core i9-12900KF> AIO: EVGA CLC 240mm> Physical Memory: G.SKILL Trident Z5 32GB (2 x 16GB) DDR5 5600> Physical Memory Model#: F5-5600U3636C16GX2-TZ5S> Monitor: Alienware AW2721D> Video Card: EVGA 3090 FTW3 ULTRA GAMING> Case: Cooler Master HAF X (942)> DAS: Sabrent DS-SC5B
|
zippytek
FTW Member
- Total Posts : 1185
- Reward points : 0
- Joined: 2020/10/06 17:25:35
- Status: offline
- Ribbons : 9
Re: stuck on post code 71
2022/07/19 06:42:01
(permalink)
|