VOGONS


First post, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie

Hey everyone, I'm having an issue trying to get a Core 2 Quad Q9650 running on my ASUS P5GC-MX/GBL motherboard, and I'm hoping someone might have a solution.
The official CPU support list says the Q9650 is supported, and the board is running the latest BIOS version, 0407. Everything should work, but it doesn't.
When I install the Q9650, the computer has inconsistent booting. maybe 40% of the time, it show the initial gpu post text on the screen. However, when it does show up, it usually just reboots back to that same screen, cycles through that a few times, and then the screen goes completely black with no signal. It’s not totally dead though, because out of about 80 attempts, two times it managed to get to the asus logo bios screen, but it only lasted for about a second before blacking out.
the motherboard is fine. it boots with a core 2 Duo E4500. It also used to work with a E8400 (over clocked to 1333 fsb), but I don't have the E8400 anymore.
The RAM is 1x1GB DDR2-667, the GPU is WinFast PX8500 GT TDH low profile, and I'm running a 360w PSU. I hope to run a gtx 1050ti and 2x2GB ddr2-800 as I did with the e8400 successful. I have already cleared the CMOS, re-seated all the power connectors, and tried booting with minimal components, but the result is always the same.
Has anyone here successfully run a Q9650 or another core 2 quad on the P5GC-MX/GBL?
Any advice would be greatly appreciated.

Reply 1 of 15, by paradigital

User metadata
Rank Oldbie
Rank
Oldbie

Sounds power related to me. Either the motherboard capacitors or the PSU itself is struggling with the quad. Just because the board works with a dual-core doesn’t mean it isn’t failing.

Reply 2 of 15, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie
paradigital wrote on 2025-11-16, 08:29:

Sounds power related to me. Either the motherboard capacitors or the PSU itself is struggling with the quad. Just because the board works with a dual-core doesn’t mean it isn’t failing.

Hi, thanks for the reply. If this is the case, would vrm cooling or replacing the psu be the best next step to take?

Reply 3 of 15, by momaka

User metadata
Rank Oldbie
Rank
Oldbie
Moonsploit wrote on 2025-11-16, 10:41:

Hi, thanks for the reply. If this is the case, would vrm cooling or replacing the psu be the best next step to take?

I agree with paragidital, sounds like a power-related issue to me as well, quite possibly caused by bad electrolytic capacitors.

The Core 2 Duo chip are "67W" TDP chips that can peak up to around 100 Watts of power use. The Quads are usually "95W" TDP, but could peak as high as 130-150 Watts under load. So the fact that a higher load makes the system crash suggests there is unstable power from somewhere, which is usually a result of bad caps.

But before we try to guess any further, post a picture of the motherboard with the CPU heatsink removed (or alternatively, with good pictures of the caps around the CPU.) FWIW, I'm posting this from a P5GC-MX right now that I recapped some (10?) years ago. Both the UCC TMV and the UCC KZG caps just about everywhere on the board went bad. Yours, of course, might have different brand of caps, since ASUS used several different manufacturers. Sometimes they are good, sometimes not so much.
While at it, also post a picture of the power supply label... and if not under warranty, pop its top off (with power cord unplugged, of course) and take a picture. Might be some bad caps lurking in there too. 360 Watts sounds like an HP or Dell or some other OEM PSU. They are generally pretty good, but do get cap failures occasionally too.

By the way, if the mobo has onboard graphics, you could also try that with the Q9650 and see if it gets you any further in the post process. The GF 8500 GT isn't that power hungry of a card, but it's still somewhere in the range of 30-ish Watts TDP. So shaving that off could lower the ripple a little and allow the board to boot. If that does happen, it will further support the notion that there are bad caps somewhere.

Reply 4 of 15, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie
momaka wrote on 2025-11-16, 16:40:
I agree with paragidital, sounds like a power-related issue to me as well, quite possibly caused by bad electrolytic capacitors. […]
Show full quote
Moonsploit wrote on 2025-11-16, 10:41:

Hi, thanks for the reply. If this is the case, would vrm cooling or replacing the psu be the best next step to take?

I agree with paragidital, sounds like a power-related issue to me as well, quite possibly caused by bad electrolytic capacitors.

The Core 2 Duo chip are "67W" TDP chips that can peak up to around 100 Watts of power use. The Quads are usually "95W" TDP, but could peak as high as 130-150 Watts under load. So the fact that a higher load makes the system crash suggests there is unstable power from somewhere, which is usually a result of bad caps.

But before we try to guess any further, post a picture of the motherboard with the CPU heatsink removed (or alternatively, with good pictures of the caps around the CPU.) FWIW, I'm posting this from a P5GC-MX right now that I recapped some (10?) years ago. Both the UCC TMV and the UCC KZG caps just about everywhere on the board went bad. Yours, of course, might have different brand of caps, since ASUS used several different manufacturers. Sometimes they are good, sometimes not so much.
While at it, also post a picture of the power supply label... and if not under warranty, pop its top off (with power cord unplugged, of course) and take a picture. Might be some bad caps lurking in there too. 360 Watts sounds like an HP or Dell or some other OEM PSU. They are generally pretty good, but do get cap failures occasionally too.

By the way, if the mobo has onboard graphics, you could also try that with the Q9650 and see if it gets you any further in the post process. The GF 8500 GT isn't that power hungry of a card, but it's still somewhere in the range of 30-ish Watts TDP. So shaving that off could lower the ripple a little and allow the board to boot. If that does happen, it will further support the notion that there are bad caps somewhere.

Hello and thanks.
I've attached the images below. I read the PSU wrong since I only pulled it half out to check and I thought that the max load was the psus wattage, the PSU is stock from this antec fusion remote mediacenter. I also attached a photo of my p5gc-mx/1333 as it was the board that came with the computer and the capacitors are different and there are more. I swapped to the gbl due to a broken dimm slot and lack of quad support. Hope this helps

Reply 5 of 15, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++

Yeah, could definitely be power related.

I see at least one bulging capacitor in that picture of the motherboard.

Have you verified that the Q9650 is working as expected in another set up?

That ancient Antex PSU most likely needs a bunch of the capacitors replaced unless that has been done already.

I used Antex PSUs back when those were for sale and they all ended up needing capacitors replaced within about 5 years.

On another note, with only 1 stick of RAM, you are running the RAM in single channel mode which will hamper performance. Is this your standard setup or is this just that way for testing purposes?

Yamaha modified setupds and drivers
Yamaha XG repository
YMF7x4 Guide
Aopen AW744L II SB-LINK

Reply 6 of 15, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie
cyclone3d wrote on 2025-11-16, 17:50:
Yeah, could definitely be power related. […]
Show full quote

Yeah, could definitely be power related.

I see at least one bulging capacitor in that picture of the motherboard.

Have you verified that the Q9650 is working as expected in another set up?

That ancient Antex PSU most likely needs a bunch of the capacitors replaced unless that has been done already.

I used Antex PSUs back when those were for sale and they all ended up needing capacitors replaced within about 5 years.

On another note, with only 1 stick of RAM, you are running the RAM in single channel mode which will hamper performance. Is this your standard setup or is this just that way for testing purposes?

I don't have another setup to test. I don't see any problems with any of the capacitors, they all seem fine to me, could you point out the one you see the bulge on? Tomorrow I will try with a different PSU from a modern computer, if that still doesn't work would buying a different mobo fix it? I'm thinking of the P5G41T-M LX2 due to its more robust support for quads, and better memory support. And yes, the 1 stick of ram is just for testing, I plan to run it with 2x2gb, or if I get the other mobo, then 2x4gb ddr3

Reply 7 of 15, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++

Second picture from the bottom. The capacitor to the left of the socket near the edge of the picture.

I don't see any bulging capacitors in the power supply but I would still not really trust it from personal experience.
I try to only use power supplies in old systems that are at least 80+ bronze rated or that have gotten really good reviews on sites that actually do real testing such as the ones that JohnnyGuru or [H]ardOCP and a couple other sites used to do.

I think TheFPSReview does them now as previous [H]ardOCP employees founded that site.
NexusGamers I think also has a real tester now.

That being said, Seasonic have been my go-to for years now as they have always been really solid.

I will use other high-end brand power supplies as well.

As for a better motherboard, I would look at Intel P45:or X48 or Nvidia 780i or 790i based boards if you want really good power delivery and the ability to overclock to the moon.

Those tend to be pricier though.

Yamaha modified setupds and drivers
Yamaha XG repository
YMF7x4 Guide
Aopen AW744L II SB-LINK

Reply 8 of 15, by Towncivilian

User metadata
Rank Newbie
Rank
Newbie

I opened up the latest BIOS 0407 with MC Extractor, and for the Q9650's CPUID 1067A, the microcode is not the most current version available (BIOS has A07 vs latest A0E). Do you want me to update the microcode in the BIOS to see if that makes a difference?

abit BX-133 RAID, P3-S 1.4Ghz, 768MB PC133, Radeon 9000 PRO, AWE64, 2x40GB RAID1, Sony SDT-9000 & Connor CTD-8000 SCSI DDS2 DAT drives, Realtek 8169 1GB NIC, Adaptec DuoConnect AUA-3121, Win2k SP4
Depeche Mode Live Wiki

Reply 9 of 15, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie
cyclone3d wrote on 2025-11-16, 19:15:
Second picture from the bottom. The capacitor to the left of the socket near the edge of the picture. […]
Show full quote

Second picture from the bottom. The capacitor to the left of the socket near the edge of the picture.

I don't see any bulging capacitors in the power supply but I would still not really trust it from personal experience.
I try to only use power supplies in old systems that are at least 80+ bronze rated or that have gotten really good reviews on sites that actually do real testing such as the ones that JohnnyGuru or [H]ardOCP and a couple other sites used to do.

I think TheFPSReview does them now as previous [H]ardOCP employees founded that site.
NexusGamers I think also has a real tester now.

That being said, Seasonic have been my go-to for years now as they have always been really solid.

I will use other high-end brand power supplies as well.

As for a better motherboard, I would look at Intel P45:or X48 or Nvidia 780i or 790i based boards if you want really good power delivery and the ability to overclock to the moon.

Those tend to be pricier though.

That's the /1333 that I switched for the /GBL, the GBL is the one I'm trying the q9650 on because the 1333 doesn't support any quads. Will see about the power supply after trying a different one, that's for the recommendations. And my case only supports micro atx motherboards, which means almost no p45s with ddr3, I could only find a single one on eBay for $450

Reply 10 of 15, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie
Towncivilian wrote on 2025-11-16, 20:37:

I opened up the latest BIOS 0407 with MC Extractor, and for the Q9650's CPUID 1067A, the microcode is not the most current version available (BIOS has A07 vs latest A0E). Do you want me to update the microcode in the BIOS to see if that makes a difference?

Yes please. could be useful, thanks

Reply 11 of 15, by momaka

User metadata
Rank Oldbie
Rank
Oldbie
Moonsploit wrote on 2025-11-16, 18:18:

I don't see any problems with any of the capacitors, they all seem fine to me, could you point out the one you see the bulge on?

The p5GC-MX/1333 definitely has at least one bulging capacitor (Chemicon KZG - probably rated 6.3V 820 uF). And the caps around the CPU are Chemicon TMV, just like on my P5GC-MX non-1333 version, all of which are bound to go bad too.
Better recap that motherboard if you plan to use it more. The "broken" memory slot could well be due to bad caps on the memory power rail.

As for the P5GC-MX/GBL, that one appears to use solid polymer caps for the CPU V_core and Panasonic FJ or FJS for the other caps on the board (the caps with the curved "T" grooves stamped on top), all of which are pretty good quality and shouldn't be the problem.
In this case, if trying out another PSU does not solve the problem, then Towncivilian might be onto something that perhaps the BIOS microcode for the CPU is possibly the cause of the issue.

Moonsploit wrote on 2025-11-16, 18:18:

Tomorrow I will try with a different PSU from a modern computer, if that still doesn't work would buying a different mobo fix it?

Yes, try the other PSU and see if that changes anything. If it does, then that Antec PSU is probably due for a recap. If not, then perhaps it is a BIOS CPU microcode issue.

BTW, I have the same PSU - actually, both the Earthwatts 430 and Earthwatts 500. These are very good PSUs, but do suffer from bad caps. Fairly easy to repair if you've down to do some soldering and hunting for new caps online. Here are two threads that document the repairs pretty well:
https://www.badcaps.net/forum/troubleshooting … sy-bad-caps-fix
https://www.badcaps.net/forum/general-topics/ … s12ii-based-psu

cyclone3d wrote on 2025-11-16, 19:15:
I don't see any bulging capacitors in the power supply but I would still not really trust it from personal experience. I try to […]
Show full quote

I don't see any bulging capacitors in the power supply but I would still not really trust it from personal experience.
I try to only use power supplies in old systems that are at least 80+ bronze rated or that have gotten really good reviews on sites that actually do real testing such as the ones that JohnnyGuru or [H]ardOCP and a couple other sites used to do.

I think TheFPSReview does them now as previous [H]ardOCP employees founded that site.
NexusGamers I think also has a real tester now.

Unfortunately, many of these review sites really only ever focused on the performance of the PSUs and rarely talked too much in depth about the caps or the overall parts quality. And the truth is, it's the caps that will almost always be the ones to fail first after some years of use. So having good quality Japanese caps (and enough of them too) is more important than having a few % higher efficiency or better load regulation. After all, the latter would be irrelevant if the PSU isn't working.

Also, JohnnyGuru, when it existed (it's now closed down), really only focused on higher-end and/or higher-power PSUs in the last few years of its operation. So finding a good review for a more budget PSU model wasn't really what they did... which is a shame, because there are A LOT of people out there who would have benefited from this kind of info.

cyclone3d wrote on 2025-11-16, 19:15:

That being said, Seasonic have been my go-to for years now as they have always been really solid.

I never go picking anything by the brand anymore.
Seasonic, like everyone else, can make some pretty unreliable PSUs if they are cheaper/lower-end models. Take a look at this Seasonic B12 BC-550, which wasn't even 2 years old and already developed bad caps and was running completely unstable.

cyclone3d wrote on 2025-11-16, 19:15:

As for a better motherboard, I would look at Intel P45:or X48 or Nvidia 780i or 790i based boards if you want really good power delivery and the ability to overclock to the moon.

Intel P45 and x48: YAY!
nVidia 780i and 790i: NAY, run away! These chipsets are all problematic due to being from the bumpgate era. They *will* all fail (and most already are.)
In general, anything past nForce 3 should be avoided.

Reply 12 of 15, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie

The "broken" memory slot could well be due to bad caps on the memory power rail.

I doubt it as one of the pins in the slot is brown and has melted plastic around it.

Yes, try the other PSU and see if that changes anything. If it does, then that Antec PSU is probably due for a recap. If not, then perhaps it is a BIOS CPU microcode issue.

I tried out the good modern psu and still the exact same issue, the duo E4500 works perfectly fine, but the quad Q9650 still only boots to the graphics card post and then cuts out as before. So I think it's either an issue with the weak motherboard itself or the problematic microcode.
Will wait for Towncivilian to finish up with that and if it still doesn't work then I'll get that P5G41T-M LX as it's good enough for me, certainly better than the p5gc-mx

Reply 13 of 15, by Towncivilian

User metadata
Rank Newbie
Rank
Newbie

Here is BIOS 0407 for the ASUS P5GC-MX/GBL with the microcode for CPUID 1067A updated from A07 to A0E (cpu1067A_platB1_ver00000A0E_2015-07-29_PRD_59BF808E.bin). MMTOOL 3.26 was used. No other changes were made.

The BIOS is untested since I do not have this board to test with. Flash at your own risk.

abit BX-133 RAID, P3-S 1.4Ghz, 768MB PC133, Radeon 9000 PRO, AWE64, 2x40GB RAID1, Sony SDT-9000 & Connor CTD-8000 SCSI DDS2 DAT drives, Realtek 8169 1GB NIC, Adaptec DuoConnect AUA-3121, Win2k SP4
Depeche Mode Live Wiki

Reply 14 of 15, by Moonsploit

User metadata
Rank Newbie
Rank
Newbie
Towncivilian wrote on 2025-11-17, 23:39:

Here is BIOS 0407 for the ASUS P5GC-MX/GBL with the microcode for CPUID 1067A updated from A07 to A0E (cpu1067A_platB1_ver00000A0E_2015-07-29_PRD_59BF808E.bin). MMTOOL 3.26 was used. No other changes were made.

The BIOS is untested since I do not have this board to test with. Flash at your own risk.

I updated the bios, but the issue still remained the same. Thanks for trying but guess I'm getting a new motherboard

Reply 15 of 15, by Horun

User metadata
Rank l33t++
Rank
l33t++

Agree it sounds like a weak cpu VRM section, had similar issues with a 848 based board not liking the max supported cpus...

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun