UPDATE: I did revisit this system, but I didn't feel like it warranted another video. Here's what happened: 1. I updated to the beta BIOS. Everything broke again to the point of no-posting and power cycling. 2. All attempts to invoke the backup BIOS failed. Tried shorting pins 1+6 on the BIOS chip while turning it on, no luck. 3. I manually programmed the beta BIOS and that fixed everything, including UEFI VBIOS. So we would've been 100% of the way there if I'd gone for the beta BIOS in this video, but also I still had no luck with the other suggested methods. Either way, it's all fixed now!
So when you say manually programmed the BIOS is that what you did in the video? only this time you used the beta version? but you used the same device? pretty new to computers. had to intel dh77eb motherboards. one the bios update worked the second went into black screen and never changed. might try to use the device you used in the video. also great video!!
This channel is great. Just straight troubleshooting issues. Most of the big PC tech channels do this rarely. Its great to have a channel largely dedicated to this.
Great videos coming lately, really enjoying watching 'em! :) I actually found your channel by searching guidance for bricked b450 motherboards as my mate had a bad bios flash and the board was basically a brick at that moment. After watching the b450 aorus video fully and even the in-depth video about the flasher tool, I ordered one for myself and after a lil bit of searching for some more guides I managed to save the board. :D
Having worked with SPI chips on TV and laptop board for a while now myself, I can confirm indeed that in-circuit reading doesn't work and I don't even bother - I just desolder the chip and leave it alone as it is on the table. I grab a new, blank chip and mess with that instead, especially if the device is working to some extent and I don't want to mess with the little life it's got left which may depend on the chip I just removed. This also confirms that the chip itself is not at fault, though it almost never is - what's actually on it gets messed up. In the case of TVs, it's caused by bad, noisy DC coming from the power supply once its caps start to fail. Happens on Samsungs all the time, esp. old LE series ones. Love your stuff - your style is similar to mine. Cheers !
Wow, that was a really interesting one. Thanks for showing how that EEPROM flashing stuff works. I am fairly good at PC troubleshooting but when it comes to taht kindof stuff I am still hesitant. I would have probably written this board off. And I still got a few boards sitting here, some for testing some for repair, that will probably not end up in the bin now.
Really enjoyed that! Learned how to take reasonable steps to diagnose and to be patient and willing to try different things. You are making great videos lately! Thanks very much for teaching me. Seeing the ref lash of the bios was helpful as well.
An interesting journey in this video. Had me worried for a while Graham, especially when you couldn't figure it out. I was saying to myself, "This is not good, Graham always figures it out!" Glad you're happy with the results. Thank you, always a pleasure to watch you work.💻🖥🔌⌨
That is sheer perseverance. I would have ditched the board after trying all the components and a fresh power supply. I have flashed an Eprom but without success and that put me right off.
I think it has dual bios just follow steps: 1. Unplug any hardware (recommended), only Monitor and Keyboard 2. Press Power Button, but HOLD. Computer will start, after a few second. Wait until computer back Off, and release Power Button. 3. Press Power Button again 4. Computer motherboard will update BIOS Corrupt automatic. Wait until finish and computer will turn off 5. Now Press Power On again This way I fixed many free given Gigabyte motherboards from people which thought its need to be in the trash can. Or almost for nothing sold desktop pcs, because of corrupted bioses. Other vendors also has bios backup but it also depends on motherboard models. Btw I forgot to mention you don't need physical switch, just use manual provided by vendor, btw I found your videos recently and its very entertaining and full of helpful information.
Troubleshooting PC's = eliminating possibilities. This video is the perfect example really. Also perseverance pays off? No better lesson than that for any technician.
with Most of giogabytes older boards (i think new ones too) if the Main Bios EEPROM is corrupt but to a point it cant hand over to the backup (which most models its main function is only to reflash the main) if you short pins 1 & 6 on the Main one, you'll force it to trigger this Recovery and reflash the main BIOS eeprom.
I have a Z68XP-UD3P that had a partially corrupt BIOS, and the only way to recover it in my case, was to desolder the main BIOS chip zero it out with a programmer, then flash F7 to it, and then do the same to the backup BIOS. In my situation the BIOS was so corrupted that simply flashing it wouldn't fix it, it required the chips to be zeroed and then flashed, and recovery never worked either, and I needed to manually perform each step when programming the chips. Basically exactly the same as what is shown in the video, I found that fix by accident because I was bored and wanted to finally get to the bottom of this issue. Once I did that it worked perfectly, and the boot loop issue that had plagued that board since I got it went away (or it had the last time I tried it about 2 weeks ago)
@@PaulTheFox1988 That's an extreme Case, other situations as far as i've experienced and seen, the auto recovery/jumping the pins to force it , works well enough.
@@CaelThunderwing Yeah I've had that work with a p55 motherboard, but in my case with that z68 board nothing would work, and the only fix was to nuke it from orbit :(
Very interesting video and you just never know what the cause might be so it pays to be patient in the troubleshooting process. Thanks for teaching us!
Just subscribed... Just working my way through your previous videos... Really enjoying your in-depth troubleshooting techniques and your thought process for this one ... Keep the videos coming.
I've only gotten one gigabyte board in my entire life which was the x570 elite, took me a week to boot into bios and after that the ram overclocking gave me audio problems. returned it immediately after that, never again
An easy trap to fall into is putting your build into the case when it appears OK BUT I've adopted the 'soak test' principle where I leave it out of the case for 3 days and regularly test the system ie that it posts, it runs OS fine et Thermals obviously out of the box should be a little higher depending on the temp in the room et but the main focus here is that it works fine and has over the three day tests. I do admire your tenacity re this old mboard as truth is generally most fail because of power outrages - clean replace the power and they can work fine, dust of course in old power supplies is murder and I am sure this guy has seen plenty of those where you clean em and its like having a mini dust storm on your hands. Keep up the good work buddy ☺‼
I think you're bang on the mark. Over the weekend as I published this video, I was watching and saw that in the final scene where it all works with the 960, the GPU wasn't booting EFI because the resolution had dropped. I must admit I don't know a lot about GPU EFI - I'm aware of its existence, but have never really investigated the differences and when you get a 'full EFI' and when you don't. I think the garbled video was a red herring the whole time - the problem was fixed as soon as the board started posting first try and wasn't cutting out or power cycling. I might make a follow up to this to explore GPU EFI BIOSes and a few other things people have said - there seems to be a bit more to learn from this that I missed at the time of recording!
I'm glad I'm not the only person that has been stumped by GPU/EFI compatibility issues. I thought I was going mad. Quite a frustrating problem if you don't know about it.
I admire you so much for your persistent nature to get something working truly it is nice to see. that being said time vs £ probably would been better after 13 mins just paying £25-£40 (for £40 you can get a working GA-H61m with a working i3 processor on it and cpu cooler just fyi). I love pc repair I just think it needs to be said the money per hour imo is just not there especially when you in your own words didn't even think it would work. one can easily end up down a rabbit hole of 5-6 hours of work where £30 for a new part would guaranteed to solve the issue. don't get me wrong i respect your doing this for curiosity and fun just think it does need to be stated.
Ok, that was really interesting! Not too many do repairs that level so keep up the good work! : Oh and it's not too long. I'm used to 2 - 3 hours long :) Subbed
I really enjoyed this video, all the troubleshooting steps up to flashing the chips. I quite fancy playing with BIOS EPROM chips myself, that stuff was fasinating.
I'm suprised this worked out. The last Intel mobo I dealt with, a 965 one which was even older than yours, didn't use straight serial flash dumps for the update process. The bios update files provided by Intel instead did use some UEFI structure with parts that were written seperately into specific areas in the flash by an executable program included in the bios update file itself that gets executed by the bios on next reboot. Notably, the flash area containing device specific data like serial numbers and MAC addresses was not written to. It was pretty hard for me to deduce which parts from the file I had to manually write where to.
love the vids buddy!! im troubleshooting a dell t3500 mobo kinda doing same thing...was working...went to my girls for a da and came back and battery problems then next day no picture....tried different batteries...ram...power supplies cpu,,,still no pic...even used onboard usb to do the data to a external mining gpu pcie slot n didn't work...when I jump the 2 power pins it wont boot now but if I touch the rtcrst jumpers with a screwdriver itboots the cpu fan but fans don't spin on a rx580 n no pic...and on a 1gig ddr2 old card the fan does spin but no pic...not sure im feeling like a bad mobo...but love to watch your vids man!! definitely helps on ideas n troubleshooting..keep up the awesome work!!
Just a tip, those alligator clamp chip connectors wear out quickly from my experience, they have little plastic teeth at the tips that force the metal pins into contact with the chip leads and those teeth can get worn to the point that the pins will not make reliable contact. Another issue I've seen is that some motherboards may not have the proper circuit layout to allow in-place chip programming from an external programmer, and you have to desolder the chip anyway. In any case, I've been able to salvage many otherwise dead board by just reflashing the BIOS. It's crazy how much good equipment just gets tossed due to simple little problems like this.
I had ANOTHER attempt at using the clamp last week, and while I finally got it to talk to the chip, I think it may have upset the motherboard. Definitely done with it. Just going to desolder every time now. I wanted it to work because I understand that not everyone wants to take a hot air gun to their mobo, but yea... better to spend time learning to use a hot air station that learning how to use the damn clamp!
Congratulations on your stamina - I suppose whether this was worthwhile depends on how you value your time and how you think about e-waste. But a fascinating logical process anyway.
I've found that dual-bios boards can be easily resurrected by physically swapping the main and backup bios. After the swap, update the main bios (from the system itself, not an external flasher) and this updates the backup bios. The physical swap also helps incase the main bios chip is wearing out. Gigabytes in particular live up to the "ultra durable" title with this, they feel almost indestructible. Finally bad power supplies seem to cause the main bios to become corrupt.
Remember most of Gigabytes dual BIOS motherboards without a manual switch will switch to the backup BIOS after it cycles through like 3 boot errors. Cool vid. I have a ABit AG8 uGuru (915p Prescott) motherboard that does the same thing.
lol, my jaw dropped when the bios showed up perfect in the end.. the magic touch of a finger was in place.. second thought to add, i think that when you re-flashed the bios, you should have unplugged your power.. to refresh your system.. which you did when you plugged all back in the case together
That was a twisty one lol. I had to use the built in display to see what was wrong with a mates machine. It worked and as i touched the memory strips one of them upset the display, thin lines appeared. I would not trust memory as far as i could throw it, err well lol.
Hi adamant motherboards can be unpredictable they cease to amaze us humans just recently sub to your channel on PC hardware are really good and full of tips and tricks not just hardware software as well keep up the good work.
that's an interesting video to watch this kind of motherboard is what we learn to fix right now in our country we need more videos on this kind of motherboards fix (H61-H81-G41) and we need the diagnostic side
To read or write flash chips in circuit, it's a good idea to power up the board, and hold the CPU in reset, by jumpering the reset pins on the front panel button headers. You may have tried this already of course!
I have the same type of board; issues, power recycling, would not post, and no picture. Solution; graphics card bad, corrupted solid state drive, switched to onboard graphics, could not get into bios, pulled cmos battery to reset, one bios chip was corrupted, fortunately the backup bios kicked in, replaced solid state drive and was able to load windows 10. I have ordered the bios programmer, just in case I ever need it. (Great video)
HAY? Hoping all's well... First, I just cannot resist this.... you have RAM and CP-EWE ... and a HERD DRIVE... Back in the OLD DAYS, when I used to utilize EPROMs,the terms FLUSH and ERASE were interchangeable... BURN or PROGRAM were the opposite terms.. BEST TO YOU
I also always had some problems with the reading clip! Since the contacts of the bios chip are connected to the board and thus there are always some impedances between the contacts, the reading of the chip is barely possible! No matter the coin cell battery in place or pulled out, I never had luck to identify and read the chip correctly with the clip tool. A quick heat with the heat gun and removal of the chip from the board works for me all the time.
Back in year 2000...when fixing no POST, the first thing i did was to remove all the RAM, HDD, just Mainboard and CPU and power it up...If you can hear a series of beeps, then the mainboard, BIOS and CPU is good. Then it just the matter of swapping RAM and Video card to see which one is fauty and stop the PC from booting. And 80% would be RAM problem....and most of the time, the RAM not actually bad, it just needs something to discharge (or refresh) the memory banks ( very different from just cleaning the contact pins) and what is did and work all the time, was using a piece of the protective thin metal coasted paper inside a cirgarette pack, with the metal side wrap around the pins of the RAM module, moving up and down the pins for 5 seconds and put the RAM back on the motherboard, and it WILL work...
As an electronics tech, i would be looking at the on-board voltages (maybe starting with Vcc on the bios chips) This is an old motherboard with "inconsistent" behavior... My first guess would be a marginal (or noisy) supply rail.... Very likely due to an old capcitor (or 2)
GA-H61M-S2PV boards (like the one you've got) are prone to main bios corruption for some reason. At the school I work at, we have about 250 odd Zoostorm branded PCs where 2 thirds are S2PV and the other third has GA-H61M-DS2 boards. I've had a good couple of dozen of the S2PV boards go corrupt where the DS2 boards have been fine. I suspect cheaper chips were probably used on the S2PV boards as to why this may happen.
your content is great. first time comming across it over the weekend... I have a HP Omen ryzen 7 1700 it started making beeping noises it power ups but nothing i don't want a £1200 paper weight
IIRC you can force gigabyte boards to boot from their backup bios in order to reflash the new one - I think you have to hold down the power button after clearing CMOS?
I use a WEP 862D+ at 400c. It's a cheap 2-in-1 station, and there are many similar stations with the same handle on them. It's great for beginners, but struggles with BGA work, so very much has its limitations.
That moment when I see on the thumbnail H61, and pronuncing the H as haitch /ˈheɪtʃ/, is when I know I've watched too many of your videos. :D Jokes aside, I'm glad that I've come across with your channel, keep up the good work.
SAME issue with my GA-H77M-D3H, in fact switching from IDE to AHCI was the final nail. "ULTRA-DURABLE" lol Initially appeared to be bad x16 slot: wouldn't fully post bare board unless vid card was removed (and using onboard, or using an ancient x2/x4 card). It finally got to the point where it corrupted the backup during recovery and all was lost. In process of re-flashing/replacing EEPROMs. I wonder if these boards are unusually susceptible to ESD damage.
one note to make, gigabyte has revisions of motherboards. like rev 1.0, then possibly a 1.1. if theres a revision version on the board it will be in the bottom left of the board and say rev x.x. and yes you need to get the bios for the revision of the board. it looks like it says rev 2.1
not to surprised about the bios on gigabyte motherboards doing this at lest on the old stuff. my old GA-5SMM had something like this happened where half of the bios went missing for no reason, lucky i was bale to re-flash a fresh copy on to it.
I remember a long while back I had a gigabyte P35 DS3R board that did that with certain types of ram, if I remember rightly back then the issue was to do with the ram and the fact that the ram voltage regulator was limited and was misreporting the ram voltage in the BIOS which caused leaving the setting to the SPD to end up with a bit of graphical corruption with descent ram, but swapping the corsair ram out to cheap units and setting the timings and voltage manually before shutting down and refitting the corsair units solved the issue so I left it at that
Good idea! I always take out the power cord, remove the battery and short both leads in the battery HOLDER (!!! HOLDER, not the battery itself) for 5 seconds, works always to drain any capacitor that still has charge in that part of the circuit.
You can stil buy those eeproms. If i had not ripped off my solder padsby unsoldering with a cheap iron, because the clamp didn´work, I would have tried this :D
If you just flash a bios like that you will definetly erase the serial number and sometimes even the MAC address. This means that the generated UUID will be different -> windows will see a different mainboard -> no longer activated
Naa... serial number maybe, but that alone won't be enough to disturb activation, the UUID needs to change significantly, a small alteration won't do it. MAC address will be in the PCH firmware, not the BIOS.
Nice video. Thank you! What's the recommended temperature for desoldering from motherboards? BTW, I remember when I used to play with EEPROM of routers, I had to desolder because the clamp would power not just the EEPROM but other chips when the EEPROM is still on the board so the data of those chips messed up the reading of the EEPROM.
Recommended temperature is "As much as you need"! I generally use 400c for motherboards, and I back off to 380c for simple PCBs (like peripherals). If you're spending 5mins trying to reach reflow temp, give it a bit more heat, if everything is melting with so much as a glance, back off a bit. With hot air, don't forget to pre-heat - blow the air around the general area you're working on first to heat everything up a bit, and then the component you're actually aiming for will flow a lot easier. But it's going to vary depending on the exact tools you use. My spare iron (WEP 862D+) is a lot less capable than my TS100, and generally needs about 420c to stay effective on heavy PCBs like mobos.
I got the feeling that the monitor (or video capture device) didn't like the resolution/frequency of the video display when it was in BIOS setup. I wonder if a different monitor would help? Maybe an old CRT monitor considering how old it is?
The problem with the garbled display might have been a grounding or EMI issue. If it had been a VGA connection I would have said it was a horizontal sync issue. I'm not sure if DVI/HDMI has a similar signalling design.
Nice video. The garbled graphics is similar to old arcade boards and sometimes reflashing the rom on those works. I wonder one of the bios pins is just missing a signal. Trying to figure out where that signal would be coming from is probably not worth the effort. If you remove the main bios chip will it just default to the backup chip? Lol I started writing this comment before watching the last minute... I’d take this as a win
There is this trick that you can use,its just that you have to short the 4 pins of the reset switch and the power switch and turn on the power switch,then you will see a message on the screen,saying recovering from corrupted bios,wait till it finishes and reboot,it will work properly.
Ah I could've tried that actually yea - I've been down that route on newer GB mobos and gotten nowhere, but those don't have a 'manual trigger' for the dual BIOS.
You say you wanna dabble in more repair stuff on this channel... maybe BIOS is bad on a MSI GTX 460 cyclone I have. Will post and put out video but it's all screwed up. if interested in trying to fix it lemme know and I'll send it to you, I just was gonna do a sli setup with it since I still had my original, bad one I got off eBay for like $40 lol. It was pretty banged up when I got it. Straightened out bent heatsink fins, fresh thermal paste and just clean up all I've done to it
My father had a AM2 lenovo that would be totally black with fan spinning if the cmos battery was low! Like you'd expect it to say 'cmos battery low', but that was not what it did. I pulled it out then tried a new battery later and it booted fine.
Clamp is not a reliable flashing method when the chip is in circuit. You might get corrupted read/write with errors etc. so it's always safer to flash the bios outside the circuit. (it does not relate to flashing on working machine by dedicated software)
Generally speaking with gigabyte, and a few other mother board manufacturers I've used, if the bios is shown as beta but there hasn't been a new bios for a year or more then there is a good chance it's a stable bios.
Whenever you re-flash the BIOS with a programmer, you should also pull the lithium CMOS battery and let the CMOS reset to factory. In fact, I would've done just that very early on, as CMOS does get corrupted, and can cause havoc on POST. I'm wondering if the video settings in the CMOS could have been misconfigured for that display. The scrambled display makes me think the speed of the video wasn't set right in CMOS, and therefore the video wasn't syncing with the display.
You have nice videos ,and You explain things well . Can You , for example , say how much heat/flow speed do You use on station at 23:40 because of plastic elements ? Thanks
I run my station at 400c max airflow for removal, and then I turn down airflow as much as needed to avoid blowing things away when replacing. I use angle to avoid melting things, you can see in this video that I'm aiming away from the plastic, and this avoids damage. It's worth noting that I have a cheap hot-air station - a more expensive one you'd probably want to turn the flow down.
Great video as always. I have a friend’s computer (Z87M Pro4, i5-4570) that has the same problem that you faced (boot loop cycles). I tried swapping ram, gpu and psu but still it does not post I don’t have another cpu or motherboard to try so I have bought another motherboard hoping that it will fix the problem. Maybe, It can be the CPU? Also you gave me the idea to use a eeprom programmer so I will try with that. If you have other advices let me know. Thanks for all your videos, they are very useful and interesting
have a similar issue on my Z97 board. but with me is boot loops sometimes, will not boot at all other times and when it boots bios says overvoltage on 5V.
At 8min, but I think you'll either short the main bios chip, so that the Backup bios kicks in, or you'll prob try flash the bios with a USB bios flash tool
so I'm guessing the bios was still using the onboard graphics despite having a gpu seated on the motherboard. some boards will still use the onboard graphics through the graphics card. had a similar issue with a ryzen apu build
when using the tweezer to read the chip in pcb lift vcc or reset pin from board this way you will always read the chip (basically i lift pin 8 vcc is easier i put a corner of a sheet of paper between the chip leg and pcb so that when i close the tweezer the leg does not touch again in the pcb)
Love your chanel, for scrumbled bios display i thing is the display cable or capture device you using try older displays or different cable hope will fix it
UPDATE: I did revisit this system, but I didn't feel like it warranted another video. Here's what happened: 1. I updated to the beta BIOS. Everything broke again to the point of no-posting and power cycling. 2. All attempts to invoke the backup BIOS failed. Tried shorting pins 1+6 on the BIOS chip while turning it on, no luck. 3. I manually programmed the beta BIOS and that fixed everything, including UEFI VBIOS.
So we would've been 100% of the way there if I'd gone for the beta BIOS in this video, but also I still had no luck with the other suggested methods. Either way, it's all fixed now!
]
ID LIKE TO SEE MORE STUFF LIKE THIS! especially gpus that don't post ect keep up the good work
So when you say manually programmed the BIOS is that what you did in the video? only this time you used the beta version? but you used the same device? pretty new to computers. had to intel dh77eb motherboards. one the bios update worked the second went into black screen and never changed. might try to use the device you used in the video. also great video!!
what are your thoughts on using debug/diagnostic post cards for motherboards w/o a debug led?
@@unlockeduk with 2nd 3rd gen, update the bios for 'current gpus' ive built 40+ systems, been there done that etc
This channel is great. Just straight troubleshooting issues. Most of the big PC tech channels do this rarely. Its great to have a channel largely dedicated to this.
Great videos coming lately, really enjoying watching 'em! :) I actually found your channel by searching guidance for bricked b450 motherboards as my mate had a bad bios flash and the board was basically a brick at that moment. After watching the b450 aorus video fully and even the in-depth video about the flasher tool, I ordered one for myself and after a lil bit of searching for some more guides I managed to save the board. :D
Nice bro
Having worked with SPI chips on TV and laptop board for a while now myself, I can confirm indeed that in-circuit reading doesn't work and I don't even bother - I just desolder the chip and leave it alone as it is on the table. I grab a new, blank chip and mess with that instead, especially if the device is working to some extent and I don't want to mess with the little life it's got left which may depend on the chip I just removed. This also confirms that the chip itself is not at fault, though it almost never is - what's actually on it gets messed up. In the case of TVs, it's caused by bad, noisy DC coming from the power supply once its caps start to fail. Happens on Samsungs all the time, esp. old LE series ones. Love your stuff - your style is similar to mine. Cheers !
Just finished watching the newest Star Wars before watching this, I enjoyed this twisty one more, lol
It's great to listen to you think your way through problems, even accepting partial fixes when it's not going to customer :)
Wow, that was a really interesting one. Thanks for showing how that EEPROM flashing stuff works. I am fairly good at PC troubleshooting but when it comes to taht kindof stuff I am still hesitant. I would have probably written this board off. And I still got a few boards sitting here, some for testing some for repair, that will probably not end up in the bin now.
Really enjoyed that! Learned how to take reasonable steps to diagnose and to be patient and willing to try different things. You are making great videos lately! Thanks very much for teaching me. Seeing the ref lash of the bios was helpful as well.
An interesting journey in this video. Had me worried for a while Graham, especially when you couldn't figure it out. I was saying to myself, "This is not good, Graham always figures it out!" Glad you're happy with the results. Thank you, always a pleasure to watch you work.💻🖥🔌⌨
That is sheer perseverance. I would have ditched the board after trying all the components and a fresh power supply. I have flashed an Eprom but without success and that put me right off.
As a newbie to computer repair I love the channel keep up the good work love watching I am a young 64 year old .
Good knowledge base, logic and analytical, recognized that, watching couple of your videos.
I think it has dual bios just follow steps: 1. Unplug any hardware (recommended), only Monitor and Keyboard
2. Press Power Button, but HOLD. Computer will start, after a few second. Wait until computer back Off, and release Power Button.
3. Press Power Button again
4. Computer motherboard will update BIOS Corrupt automatic. Wait until finish and computer will turn off
5. Now Press Power On again
This way I fixed many free given Gigabyte motherboards from people which thought its need to be in the trash can. Or almost for nothing sold desktop pcs, because of corrupted bioses. Other vendors also has bios backup but it also depends on motherboard models. Btw I forgot to mention you don't need physical switch, just use manual provided by vendor, btw I found your videos recently and its very entertaining and full of helpful information.
Troubleshooting PC's = eliminating possibilities. This video is the perfect example really. Also perseverance pays off? No better lesson than that for any technician.
with Most of giogabytes older boards (i think new ones too) if the Main Bios EEPROM is corrupt but to a point it cant hand over to the backup (which most models its main function is only to reflash the main) if you short pins 1 & 6 on the Main one, you'll force it to trigger this Recovery and reflash the main BIOS eeprom.
Nice think thanks for sharing
obviously a known issue on the h61 from gigabyte, happened to be several times.
I have a Z68XP-UD3P that had a partially corrupt BIOS, and the only way to recover it in my case, was to desolder the main BIOS chip zero it out with a programmer, then flash F7 to it, and then do the same to the backup BIOS. In my situation the BIOS was so corrupted that simply flashing it wouldn't fix it, it required the chips to be zeroed and then flashed, and recovery never worked either, and I needed to manually perform each step when programming the chips.
Basically exactly the same as what is shown in the video, I found that fix by accident because I was bored and wanted to finally get to the bottom of this issue.
Once I did that it worked perfectly, and the boot loop issue that had plagued that board since I got it went away (or it had the last time I tried it about 2 weeks ago)
@@PaulTheFox1988 That's an extreme Case, other situations as far as i've experienced and seen, the auto recovery/jumping the pins to force it , works well enough.
@@CaelThunderwing Yeah I've had that work with a p55 motherboard, but in my case with that z68 board nothing would work, and the only fix was to nuke it from orbit :(
"A good old GTX960" you makes me and my 550ti cry with that statement.
Hah same mate .
yeah ......same here I'm still running a gtx680
Very interesting video and you just never know what the cause might be so it pays to be patient in the troubleshooting process. Thanks for teaching us!
Just subscribed... Just working my way through your previous videos... Really enjoying your in-depth troubleshooting techniques and your thought process for this one ... Keep the videos coming.
After watching this video I think I ought to get a bios programmer and learn how to solder. Bloody amazing work. I'm thoroughly impressed.
These videos are helping me to stay away from Gigabyte motherboards.
bought b450 aorus elite, immediately regretted it, bow got an x570 gaming pro carbon from msi
I've used loads of gigabyte boards, only issue i had was a blown audio chip - which was my own fault lol.
I've only gotten one gigabyte board in my entire life which was the x570 elite, took me a week to boot into bios and after that the ram overclocking gave me audio problems. returned it immediately after that, never again
I've got z390 aorus pro, i9-9900k at 4.9Ghz, RAM at 3200, no problems whatsoever.
ASUUUUS
An easy trap to fall into is putting your build into the case when it appears OK BUT I've adopted the 'soak test' principle where I leave it out of the case for 3 days and regularly test the system ie that it posts, it runs OS fine et Thermals obviously out of the box should be a little higher depending on the temp in the room et but the main focus here is that it works fine and has over the three day tests. I do admire your tenacity re this old mboard as truth is generally most fail because of power outrages - clean replace the power and they can work fine, dust of course in old power supplies is murder and I am sure this guy has seen plenty of those where you clean em and its like having a mini dust storm on your hands. Keep up the good work buddy ☺‼
Best on-line resource ever. Very much appreciate your camera work and your ongoing trouble shooting narrative. Well done!! Cheers from Canada!
The problem with the output has to do with EFI and the GPU bios, had this problem once with a R9 290X updating it’s bios fixed it.
I think you're bang on the mark. Over the weekend as I published this video, I was watching and saw that in the final scene where it all works with the 960, the GPU wasn't booting EFI because the resolution had dropped. I must admit I don't know a lot about GPU EFI - I'm aware of its existence, but have never really investigated the differences and when you get a 'full EFI' and when you don't.
I think the garbled video was a red herring the whole time - the problem was fixed as soon as the board started posting first try and wasn't cutting out or power cycling.
I might make a follow up to this to explore GPU EFI BIOSes and a few other things people have said - there seems to be a bit more to learn from this that I missed at the time of recording!
Beat me to it, spot on.
@@Adamant_IT You should have flashed that beta bios, it probably fixes the GPU problem.
I'm glad I'm not the only person that has been stumped by GPU/EFI compatibility issues. I thought I was going mad. Quite a frustrating problem if you don't know about it.
I'm not convinced it was an EFI/GPU incompatibility, as the onboard video had the same issue.
I admire you so much for your persistent nature to get something working truly it is nice to see.
that being said time vs £ probably would been better after 13 mins just paying £25-£40 (for £40 you can get a working GA-H61m with a working i3 processor on it and cpu cooler just fyi). I love pc repair I just think it needs to be said the money per hour imo is just not there especially when you in your own words didn't even think it would work. one can easily end up down a rabbit hole of 5-6 hours of work where £30 for a new part would guaranteed to solve the issue. don't get me wrong i respect your doing this for curiosity and fun just think it does need to be stated.
Ok, that was really interesting! Not too many do repairs that level so keep up the good work! : Oh and it's not too long. I'm used to 2 - 3 hours long :) Subbed
I really enjoyed this video, all the troubleshooting steps up to flashing the chips. I quite fancy playing with BIOS EPROM chips myself, that stuff was fasinating.
I'm suprised this worked out. The last Intel mobo I dealt with, a 965 one which was even older than yours, didn't use straight serial flash dumps for the update process. The bios update files provided by Intel instead did use some UEFI structure with parts that were written seperately into specific areas in the flash by an executable program included in the bios update file itself that gets executed by the bios on next reboot. Notably, the flash area containing device specific data like serial numbers and MAC addresses was not written to. It was pretty hard for me to deduce which parts from the file I had to manually write where to.
love the vids buddy!! im troubleshooting a dell t3500 mobo kinda doing same thing...was working...went to my girls for a da and came back and battery problems then next day no picture....tried different batteries...ram...power supplies cpu,,,still no pic...even used onboard usb to do the data to a external mining gpu pcie slot n didn't work...when I jump the 2 power pins it wont boot now but if I touch the rtcrst jumpers with a screwdriver itboots the cpu fan but fans don't spin on a rx580 n no pic...and on a 1gig ddr2 old card the fan does spin but no pic...not sure im feeling like a bad mobo...but love to watch your vids man!! definitely helps on ideas n troubleshooting..keep up the awesome work!!
IT at its finest work through it and when it works just take it and roll with the good times.
Just a tip, those alligator clamp chip connectors wear out quickly from my experience, they have little plastic teeth at the tips that force the metal pins into contact with the chip leads and those teeth can get worn to the point that the pins will not make reliable contact.
Another issue I've seen is that some motherboards may not have the proper circuit layout to allow in-place chip programming from an external programmer, and you have to desolder the chip anyway.
In any case, I've been able to salvage many otherwise dead board by just reflashing the BIOS. It's crazy how much good equipment just gets tossed due to simple little problems like this.
I had ANOTHER attempt at using the clamp last week, and while I finally got it to talk to the chip, I think it may have upset the motherboard. Definitely done with it. Just going to desolder every time now.
I wanted it to work because I understand that not everyone wants to take a hot air gun to their mobo, but yea... better to spend time learning to use a hot air station that learning how to use the damn clamp!
Thanks for sharing all this knowledge with noobs like me.
Congratulations on your stamina - I suppose whether this was worthwhile depends on how you value your time and how you think about e-waste. But a fascinating logical process anyway.
I've found that dual-bios boards can be easily resurrected by physically swapping the main and backup bios. After the swap, update the main bios (from the system itself, not an external flasher) and this updates the backup bios. The physical swap also helps incase the main bios chip is wearing out. Gigabytes in particular live up to the "ultra durable" title with this, they feel almost indestructible. Finally bad power supplies seem to cause the main bios to become corrupt.
Remember most of Gigabytes dual BIOS motherboards without a manual switch will switch to the backup BIOS after it cycles through like 3 boot errors.
Cool vid. I have a ABit AG8 uGuru (915p Prescott) motherboard that does the same thing.
lol, my jaw dropped when the bios showed up perfect in the end.. the magic touch of a finger was in place.. second thought to add, i think that when you re-flashed the bios, you should have unplugged your power.. to refresh your system.. which you did when you plugged all back in the case together
That was a twisty one lol.
I had to use the built in display to see what was wrong with a mates machine.
It worked and as i touched the memory strips one of them upset the display, thin lines appeared.
I would not trust memory as far as i could throw it, err well lol.
Hi adamant motherboards can be unpredictable they cease to amaze us humans just recently sub to your channel on PC hardware are really good and full of tips and tricks not just hardware software as well keep up the good work.
that's an interesting video to watch
this kind of motherboard is what we learn to fix right now in our country
we need more videos on this kind of motherboards fix (H61-H81-G41)
and we need the diagnostic side
I don't have a need right now, but I have to get one of those programmers... Cool video, great troubleshooting! Just subscribed!
Before I bought this Gigabyte Aorus Z390 board I had been using the Intel LGA1155 motherboards. Nothing fancy but rock solid performers.
7:13 "Are we power cycling?"
-- car sound outside !
" yes we probably are"
LOL that sound !
To read or write flash chips in circuit, it's a good idea to power up the board, and hold the CPU in reset, by jumpering the reset pins on the front panel button headers. You may have tried this already of course!
I have the same type of board; issues, power recycling, would not post, and no picture. Solution; graphics card bad, corrupted solid state drive, switched to onboard graphics, could not get into bios, pulled cmos battery to reset, one bios chip was corrupted, fortunately the backup bios kicked in, replaced solid state drive and was able to load windows 10. I have ordered the bios programmer, just in case I ever need it. (Great video)
You really did your best to fix the motherboard, it was a computer ghost.
what a crazy outcome. I definitely did not expect this!
HAY? Hoping all's well...
First, I just cannot resist this.... you have RAM and CP-EWE ... and a HERD DRIVE...
Back in the OLD DAYS, when I used to utilize EPROMs,the terms FLUSH and ERASE were interchangeable... BURN or PROGRAM were the opposite terms..
BEST TO YOU
Good work, champ.. I'm really getting more knowledge every time I watch +1 video of yours, thanx a lot man ❤️
Another great video, kept us guessing as to what was the fault !
I also always had some problems with the reading clip! Since the contacts of the bios chip are connected to the board and thus there are always some impedances between the contacts, the reading of the chip is barely possible! No matter the coin cell battery in place or pulled out, I never had luck to identify and read the chip correctly with the clip tool. A quick heat with the heat gun and removal of the chip from the board works for me all the time.
Back in year 2000...when fixing no POST, the first thing i did was to remove all the RAM, HDD, just Mainboard and CPU and power it up...If you can hear a series of beeps, then the mainboard, BIOS and CPU is good. Then it just the matter of swapping RAM and Video card to see which one is fauty and stop the PC from booting. And 80% would be RAM problem....and most of the time, the RAM not actually bad, it just needs something to discharge (or refresh) the memory banks ( very different from just cleaning the contact pins) and what is did and work all the time, was using a piece of the protective thin metal coasted paper inside a cirgarette pack, with the metal side wrap around the pins of the RAM module, moving up and down the pins for 5 seconds and put the RAM back on the motherboard, and it WILL work...
I really would like to see that little gadget work
As an electronics tech, i would be looking at the on-board voltages (maybe starting with Vcc on the bios chips) This is an old motherboard with "inconsistent" behavior... My first guess would be a marginal (or noisy) supply rail.... Very likely due to an old capcitor (or 2)
GA-H61M-S2PV boards (like the one you've got) are prone to main bios corruption for some reason. At the school I work at, we have about 250 odd Zoostorm branded PCs where 2 thirds are S2PV and the other third has GA-H61M-DS2 boards. I've had a good couple of dozen of the S2PV boards go corrupt where the DS2 boards have been fine. I suspect cheaper chips were probably used on the S2PV boards as to why this may happen.
your content is great. first time comming across it over the weekend... I have a HP Omen ryzen 7 1700 it started making beeping noises it power ups but nothing i don't want a £1200 paper weight
I just realised you have the same watch as me. Thats cool
IIRC you can force gigabyte boards to boot from their backup bios in order to reflash the new one - I think you have to hold down the power button after clearing CMOS?
That really works. I've tried this myself. I hold the power button for a couple of power loops and the bios backup kicked in.
which hot air work station do you use and at what temperature. thank you for the videos great work
I use a WEP 862D+ at 400c. It's a cheap 2-in-1 station, and there are many similar stations with the same handle on them. It's great for beginners, but struggles with BGA work, so very much has its limitations.
Im even didnt know how to turn on the motherboard without a switch.. lol
Great Channel. I'm a subscriber now! Love it! Dutch regards, Nico.
That moment when I see on the thumbnail H61, and pronuncing the H as haitch /ˈheɪtʃ/, is when I know I've watched too many of your videos. :D
Jokes aside, I'm glad that I've come across with your channel, keep up the good work.
SAME issue with my GA-H77M-D3H, in fact switching from IDE to AHCI was the final nail. "ULTRA-DURABLE" lol Initially appeared to be bad x16 slot: wouldn't fully post bare board unless vid card was removed (and using onboard, or using an ancient x2/x4 card). It finally got to the point where it corrupted the backup during recovery and all was lost. In process of re-flashing/replacing EEPROMs. I wonder if these boards are unusually susceptible to ESD damage.
one note to make, gigabyte has revisions of motherboards. like rev 1.0, then possibly a 1.1. if theres a revision version on the board it will be in the bottom left of the board and say rev x.x. and yes you need to get the bios for the revision of the board. it looks like it says rev 2.1
not to surprised about the bios on gigabyte motherboards doing this at lest on the old stuff.
my old GA-5SMM had something like this happened where half of the bios went missing for no reason, lucky i was bale to re-flash a fresh copy on to it.
I remember a long while back I had a gigabyte P35 DS3R board that did that with certain types of ram, if I remember rightly back then the issue was to do with the ram and the fact that the ram voltage regulator was limited and was misreporting the ram voltage in the BIOS which caused leaving the setting to the SPD to end up with a bit of graphical corruption with descent ram, but swapping the corsair ram out to cheap units and setting the timings and voltage manually before shutting down and refitting the corsair units solved the issue so I left it at that
When I take out the bios battery I hold the power button on for a minute two or three times to drain all power from the board
Good idea! I always take out the power cord, remove the battery and short both leads in the battery HOLDER (!!! HOLDER, not the battery itself) for 5 seconds, works always to drain any capacitor that still has charge in that part of the circuit.
I live in Brazil and i admire your channel, is excelent
i love your work ! ive learned so much! thank you so much adamant. worth it watching your videos!
Great videos, nice background music
You can stil buy those eeproms. If i had not ripped off my solder padsby unsoldering with a cheap iron, because the clamp didn´work, I would have tried this :D
If you just flash a bios like that you will definetly erase the serial number and sometimes even the MAC address. This means that the generated UUID will be different -> windows will see a different mainboard -> no longer activated
Naa... serial number maybe, but that alone won't be enough to disturb activation, the UUID needs to change significantly, a small alteration won't do it. MAC address will be in the PCH firmware, not the BIOS.
Nice video.
Thank you!
What's the recommended temperature for desoldering from motherboards?
BTW, I remember when I used to play with EEPROM of routers, I had to desolder because the clamp would power not just the EEPROM but other chips when the EEPROM is still on the board so the data of those chips messed up the reading of the EEPROM.
Recommended temperature is "As much as you need"!
I generally use 400c for motherboards, and I back off to 380c for simple PCBs (like peripherals). If you're spending 5mins trying to reach reflow temp, give it a bit more heat, if everything is melting with so much as a glance, back off a bit. With hot air, don't forget to pre-heat - blow the air around the general area you're working on first to heat everything up a bit, and then the component you're actually aiming for will flow a lot easier.
But it's going to vary depending on the exact tools you use. My spare iron (WEP 862D+) is a lot less capable than my TS100, and generally needs about 420c to stay effective on heavy PCBs like mobos.
@@Adamant_IT Thanks!
I got the feeling that the monitor (or video capture device) didn't like the resolution/frequency of the video display when it was in BIOS setup. I wonder if a different monitor would help? Maybe an old CRT monitor considering how old it is?
The problem with the garbled display might have been a grounding or EMI issue.
If it had been a VGA connection I would have said it was a horizontal sync issue. I'm not sure if DVI/HDMI has a similar signalling design.
ah those old pop can heatsinks were fun
Nice video. The garbled graphics is similar to old arcade boards and sometimes reflashing the rom on those works. I wonder one of the bios pins is just missing a signal. Trying to figure out where that signal would be coming from is probably not worth the effort. If you remove the main bios chip will it just default to the backup chip? Lol I started writing this comment before watching the last minute... I’d take this as a win
😂😆Remember what you once said, Graham, "It's never the CPU."🤔
I enjoy watching your videos.
“Valid backup ROM detected”
That’s why I like servers
There is this trick that you can use,its just that you have to short the 4 pins of the reset switch and the power switch and turn on the power switch,then you will see a message on the screen,saying recovering from corrupted bios,wait till it finishes and reboot,it will work properly.
Ah I could've tried that actually yea - I've been down that route on newer GB mobos and gotten nowhere, but those don't have a 'manual trigger' for the dual BIOS.
You say you wanna dabble in more repair stuff on this channel... maybe BIOS is bad on a MSI GTX 460 cyclone I have. Will post and put out video but it's all screwed up.
if interested in trying to fix it lemme know and I'll send it to you, I just was gonna do a sli setup with it since I still had my original, bad one I got off eBay for like $40 lol. It was pretty banged up when I got it. Straightened out bent heatsink fins, fresh thermal paste and just clean up all I've done to it
My father had a AM2 lenovo that would be totally black with fan spinning if the cmos battery was low! Like you'd expect it to say 'cmos battery low', but that was not what it did. I pulled it out then tried a new battery later and it booted fine.
Clamp is not a reliable flashing method when the chip is in circuit. You might get corrupted read/write with errors etc. so it's always safer to flash the bios outside the circuit. (it does not relate to flashing on working machine by dedicated software)
Moral of the story: don't be scared of Beta Bioses.
Generally speaking with gigabyte, and a few other mother board manufacturers I've used, if the bios is shown as beta but there hasn't been a new bios for a year or more then there is a good chance it's a stable bios.
Just be scared of Gigabyte
Whenever you re-flash the BIOS with a programmer, you should also pull the lithium CMOS battery and let the CMOS reset to factory. In fact, I would've done just that very early on, as CMOS does get corrupted, and can cause havoc on POST. I'm wondering if the video settings in the CMOS could have been misconfigured for that display. The scrambled display makes me think the speed of the video wasn't set right in CMOS, and therefore the video wasn't syncing with the display.
👍👍❤ I love these puzzling cases!!!!
You have nice videos ,and You explain things well . Can You , for example , say how much heat/flow speed do You use on station at 23:40 because of plastic elements ? Thanks
I run my station at 400c max airflow for removal, and then I turn down airflow as much as needed to avoid blowing things away when replacing. I use angle to avoid melting things, you can see in this video that I'm aiming away from the plastic, and this avoids damage. It's worth noting that I have a cheap hot-air station - a more expensive one you'd probably want to turn the flow down.
Thank you for sharing that was an inteligent move...
really enjoy your description good work
As soon as I saw Gigabyte and H61 I knew what was wrong with it. Mine ended up in trash can , replaced with H81 MSI , NEVER looked back.
awesome bios fix
Great video as always.
I have a friend’s computer (Z87M Pro4, i5-4570) that has the same problem that you faced (boot loop cycles).
I tried swapping ram, gpu and psu but still it does not post
I don’t have another cpu or motherboard to try so I have bought another motherboard hoping that it will fix the problem.
Maybe, It can be the CPU?
Also you gave me the idea to use a eeprom programmer so I will try with that.
If you have other advices let me know.
Thanks for all your videos, they are very useful and interesting
have a similar issue on my Z97 board. but with me is boot loops sometimes, will not boot at all other times and when it boots bios says overvoltage on 5V.
"It's never the CPU - except when it is."
A wired keyboard is a good thing to have laying around. Sometimes the fault is the usb hub driver. Happened to me.
Did you clean the motherboard with any fluid, any days prior to the video? 70% alcohol will cause problems for days until totally dry under chips.
At 8min, but I think you'll either short the main bios chip, so that the Backup bios kicks in, or you'll prob try flash the bios with a USB bios flash tool
so I'm guessing the bios was still using the onboard graphics despite having a gpu seated on the motherboard. some boards will still use the onboard graphics through the graphics card. had a similar issue with a ryzen apu build
Hi bro i have the same problem and my motherboard GA-H61M-DS2 the bios not apear at all it's still on and off do thinsk shorting pins is good?
when using the tweezer to read the chip in pcb lift vcc or reset pin from board this way you will always read the chip (basically i lift pin 8 vcc is easier i put a corner of a sheet of paper between the chip leg and pcb so that when i close the tweezer the leg does not touch again in the pcb)
As soon as I saw all these problems I KNEW it was a Gigabyte. LOL that is the same POS I had , but REV. 2.2
Love your chanel, for scrumbled bios display i thing is the display cable or capture device you using try older displays or different cable hope will fix it