Z170 boot issues

samuelmorris

Supreme [H]ardness
Joined
Dec 20, 2010
Messages
5,506
I recently ordered two Z170 platforms to update two aging machines still using P55 and X48 respectively (Voyager and Intrepid in my signature). I ordered the following components:

Voyager: Core i5 6600 CPU, Gigabyte Z170X-UD5, 2x8GB Corsair Vengeance LPX 2400mhz, Corsair RM1000i
Intrepid: Core i5 6500 CPU, Gigabyte Z170-HD3P, 2x8GB Corsair Vengeance LPX 2400mhz

The PSU I realise is way over the top for the current 970 but both board and PSU were intended to future-proof the system for a high-end SLI setup potentially coming later.

I have failed an RMA for the UD5 board as it will not POST with memory in dual channel (cites a 55 error after several cycles through the early part of the POST routine), the only combinations I could get working were one stick only, then two sticks right next to each other (i.e. not in a pair of slots the same colour as they are supposed to be) - 4 slots didn't work, and neither did either set of the RAM I bought, in either of the proper channels (i.e. both black slots, or both grey slots). One particular stick didn't even seem to work on its own, so I've put that to one side for further testing when the RMA'ed board arrives.

That, however, isn't the issue that has me confused. One other issue the UD5 board has, which I originally disregarded due to this RMA, was that even with memory in one of the working configurations above, the system will only POST the first time the power button's pressed after mains power is applied. Every time the PC is to be powered up using the power button, the PSU needs to be switched off and then back on first, otherwise the POST stage will not complete (the dual 7-segment does not light up when this happens)

As a result of the RMA above, I set the 6500, second set of memory and the HD3P board up in the main PC, Voyager, so it could be properly tested before being installed in my file server, Intrepid, which would be a much more difficult machine to have stability issues with. This has none of the issues with memory the UD5 board has, and boots straight away with memory in the right place. However, although at the moment it seems to be OK with reboots, if this machine is re-powered after having been off for some time (say, more than 15 minutes) it too will not POST at first. I don't seem to need to re-power the PSU with this one, I just have to power the machine off two or three times, after which point it'll boot normally.

This really isn't something I want to be putting up with if the board is going into a server that's supposed to run 24/7. Is this a known issue with the Z170 chipset, perhaps Gigabyte's implementation of it, or do you think I have two defective boards?

Despite the boot issues, the system seems stable in windows, has been running for around 10 hours total and played an hour session of Rocket League without incident.

Bit puzzled by this one, and would appreciate any comments if anybody has seen something similar with these boards.
 
Are you using XMP profiles for the RAM? If so, try disabling that. Also try the latest BIOS on each board in case you have an issue they already fixed.
 
Thanks for the quick reply -
I've already upgraded the BIOS on the HD3P board to solve an issue with neither board being able to detect my SSD as bootable (That was my mistake, it really wasn't a bootable drive, I think my windows 8 bootloader was on the WD20EARS I also ditched at the same time I changed the board - whoops!) so that certainly doesn't seem to have solved the boot issue.

I didn't update the BIOS on the UD5 at the time, I could certainly try that, but my thinking is, would XMP being enabled really cause one board to POST, but another pretty similar board, using the same memory?

My other concern is that Gigabyte boards in the past (certainly my X38 and X48 I believe) had a propensity to revert back to the BIOS they shipped with if ever the system unexpectedly rebooted. A BIOS downgrading itself to a version that stops the system from POSTing and causing me to have to move the RAM sticks around to fix it would be extremely irritating.
 
Alright I have a similar problem now with a PC I finished putting together yesterday. I ran into this once before I wrote my post above and determined it was maybe a XMP issue, now it has happened a few times. My setup is similar for the first three components...

Gigabyte Z170X Gaming 3
i5 6500
Corsair LPX 2666 2x8GB
Intel SSD
Western Digital HDD
GTX950
XFX XTR 650

When I woke up this morning, apparently I had lost power overnight. PC would not boot past the "Gigabyte Ultra Durable" BIOS screen. I tried to power cycle a couple times and the keyboard was not responding on any attempt. I eventually "fixed" this by resetting the CMOS. Throughout the day I tried rebooting and cold booting the PC a few more times and had the issue at least two more times.

I found a way to reproduce this, curious if you can do the same thing...
Turn off the PC
Turn off the power supply
Press the power button on the PC a few times (in theory this will discharge the stored electricity on the board)
Turn the power supply back on
Press the power button on the PC...

The PC is now stuck at the BIOS until I hit the power button a few times to power cycle it. It almost always comes back on the third power cycle. Doesn't matter if XMP is enabled or not on my PC, and it doesn't matter what the first boot device is either. I have nothing fancy set in the BIOS at all, it happens even if the lonely setting I change is the CPU OPT fan speed setting (for my water cooling pump).
 
Amazon posted my advance replacement out this evening so I should be able to collect it tomorrow - I'll have a look at the new board and see what happens. For what it's worth though, when the PC was stuck at the BIOS it did not produce an image on the screen - any time I was able to see the Gigabyte UltraDurable logo on screen, the system was good to go (apart from the aforementioned boot device issue which was most likely my fault). That said, I am using Displayport from the GTX970 rather than a simpler interface or the IGP. Does your board have the 7-segment display on it - if so, did it show a particular code when it got stuck? Mine would cycle through all sorts of codes but if it would fail to boot at all, it'd sit at 55.

However, with the common issue shared by both the bad board and the good board (not POSTing after the PC had been switched off for a while) the 7 segments did not illuminate at all - in the case of the UD5 board this also corresponds with the CPU fan running at 100% (loud!) and very slowly ramping down. In the case of the HD3P the CPU fan goes up to maybe 40-50% and sits there. Normal boot procedure it probably reaches 30% at most then ramps even further down immediately. Will update with my findings as soon as I can, but I'm having grave concerns that there are some major flaws either with Z170 itself, or Gigabyte's implementation of it.
 
Update: Trying the beta BIOS posted on Gigabyte's website now. It lists "Improve XMP Compatibility" as the change.
 
Yes that's the first thing I thought of when you mentioned XMP - thing is, if you get the same behaviour with XMP disabled (not tried it yet myself as I boxed all the kit back up for the UD5), would it necessarily make any difference?
 
Gigabyte has proven unable to write a BIOS without boot issues for literally years now - boot loops, memory recognition issues, you name it, ever since the Z68 chipset. Return the board and get a MSI or ASUS... No more boot problems.

I had similar problems with a Z68XP-UD4 and have read of others having those sorts of issues with Z77 boards from Gigabyte. It's not Z170, it's Gigabyte.
 
Last edited:
Gigabyte has proven unable to write a BIOS without boot issues for literally years now - boot loops, memory recognition issues, you name it, ever since the Z68 chipset. Return the board and get a MSI or ASUS... No more boot problems.

This goes back to at least the P35 days, I returned a GB P35 board for a MSI board. The Gigabyte board would completely forget all overclocking settings about 1/25 reboots. That MSI P35 still runs fine to this day and I trusted it enough to give it to my parents, still overclocked :p

My most recent MSI P67 board experience was no so great so I jumped ship to Gigabyte. I tend to avoid Asus since I ran into a board that misspelled Athlon wrong on the boot screen (LOL?)

I am still testing this beta BIOS now... I'll try MSI again if those goes poorly :(
 
Last edited:
Yes that's the first thing I thought of when you mentioned XMP - thing is, if you get the same behaviour with XMP disabled (not tried it yet myself as I boxed all the kit back up for the UD5), would it necessarily make any difference?

Could make a difference if the issue is simply initializing any RAM with such a profile.
 
Last edited:
I've had some DOA stuff from Gigabyte in the past but not much. By contrast, every single product I've owned from Asus has failed within 18 months, usually within a year - and I've owned in excess of 10. More frustrating, since they always fail several months into their use and not immediately, you're blissfully unaware you own a defective product. I usually try and give manufacturers 2 or 3 chances to get it together but Asus have had more chances than any other and have repeatedly demonstrated to me they're bottom of the pile when it comes to hardware reliability, and the comprehensive stats recorded by hardware.fr agree. The only alternative brand I'd really be exploring at this stage is MSI, but I'll see what happens with the RMA board first. If that all works fine that'd give me confidence enough to keep using it. If that still doesn't work I will try a BIOS update with trepidation just in case, then if all else fails, MSI it will be. Generally though I've used Gigabyte exclusively for years because I've had the least issues - my previous build (not in sig) used a 4690S and a Z97N-WiFi, no trouble at all, assembled it, pressed the power button and off it went. The previous builds these Z170s are supposed to replace are either approaching or over 6 years old so they've lasted very well. But, times do change, certain companies (like Asus, for example!) that were once good can go bad, so I'm willing to entertain an alternative product, but the 6 1/2 year old X48-DS4 replaced a failed 6-month old Maximus II Formula which was in turn replacing the DOA X48-DS5, so history suggests a replacement board might work out fine. We shall see - but good luck, I'm very interested to hear your findings, it's rare to find the obscure issues I encounter with hardware are also happening to someone else...
 
I found a way to reproduce this, curious if you can do the same thing...
Turn off the PC
Turn off the power supply
Press the power button on the PC a few times (in theory this will discharge the stored electricity on the board)
Turn the power supply back on
Press the power button on the PC...

I tried this with gaming 7 with G.skill ram, no problem at all. It seems people with problem all have Corsair Vengeance LPX ram, those with G.skill ram are mostly problem free. I think those who want to go with Gigabyte 170 boards should pick G.skill ram for painless experience.
 
I tried this with gaming 7 with G.skill ram, no problem at all. It seems people with problem all have Corsair Vengeance LPX ram, those with G.skill ram are mostly problem free. I think those who want to go with Gigabyte 170 boards should pick G.skill ram for painless experience.

Had I expected this may happen I would have gone with G.Skill and saved a lot of frustration! Thanks for trying that out though.

The beta BIOS seems to have fixed my issue for now, so possibly Gigabyte is aware of their stuff not working. I was ready to send this board back tomorrow if I ran into a single issue tonight with this beta BIOS. I've been going through a series of reboots and power cycles while cooking, cleaning, and watching Netflix. Everything has been working:

[XMP enabled for all tests]
3x shut off the power supply, press power button to discharge, wait 10 minutes, turn on power supply, power on (issue I listed out above)
3x normal reboot from Windows
3x reboot from the BIOS setup using F10
3x power off from Windows, wait 10 minutes, power on
3x power off using power button, wait 10 minutes, power on
3x shut off the power supply, press power button to discharge, wait 10 minutes, turn on power supply, power on
3x turn on from hibernate
3x power off from Windows, wait 10 minutes, power on

So that cover 24 power on / reboot cycles for everything I could think of with no issues. I am using "F5b" BIOS from here:
http://www.gigabyte.com/products/product-page.aspx?pid=5497#bios

Looks like damn near all Z170 Gigabyte boards also have beta BIOSes for RAM compatibility and they aren't in sync at all. Some board got beta release two months ago and it doesn't look like they added the support to non beta builds yet. Boards like the UD5 belonging to the OP have had multiple beta releases for RAM issues.

I'll have an eagle eye out for any issues between now and the end of my 30 day return window. Unfortunately this sucked a lot of the joy out of my first new build for myself in years.

In other fun news, I went out got got a UPS for my PC earlier in the day as I thought maybe a nasty brownout or surge caused the initial problem. So at least I finally have a UPS for my PC again...

Build shot before I cleaned up the cables a bit more and added another hard drive, I just had to get Corsair RAM to match the CPU cooler and new LED fans:
mb663Ar.jpg
 
Last edited:
The only beta BIOS for the HD3P is F3d. I installed F4, which I'm assuming would include anything contained within F3d? Either way, the system's behaviour is absolutely predictable - first boot the fan operates correctly but doesn't POST, second boot the fan operates at a fixed speed and no POST, third boot the system will operate normally.

I must admit I hadn't considered a different RAM brand, I might try and pick up a set of G-Skill instead and see if that changes anything.
 
At this point I think it would be worth trying F3d on the HD3P, I don't think it's clear the non beta BIOS releases include the beta code. Plus it would be good for science :)
 
Quick update - I have yet to flash the BIOS but it appears the boot issue is heat related. With the room a good few degrees cooler it took 9 power cycles before the system would POST. Bit of a long drawn out procedure...

Another quick update - G-Skill is hard to get hold of in the UK at short notice since most retailers exclusively sell Corsair RAM (at least for DDR4). I've gone with Kingston Hyper-X as it was in stock somewhere that would give me free delivery soon-ish, so I'll see what happens with that. Will try the BIOS update on the HD3P shortly.
 
Last edited:
Another quick update - BIOS update to F3d has bricked the board. Will no longer POST at all.
 
Another quick update - BIOS update to F3d has bricked the board. Will no longer POST at all.

Tried puling the BIOS battery, all the RAM, one boot attempt without RAM, clear CMOS again, then one stick of RAM? Sometimes my board would boot after that. Like... once.

Well, hopefully you can RMA/exchange? Sorry about all the science.
 
I'll RMA for an MSI I think - there's no way that updating the BIOS through the official tool should have that effect, so I'm pretty confident both boards are defective. I'll keep hold of it for an extra day - the replacement UD5 arrived this evening way too late to collect from amazon locker so I'll pick that up tomorrow and try that. The HyperX is due to arrive tomorrow too so hopefully I'll be able to test the second UD5 with the Corsair first and if that fails, try it with the Kingston instead. If that works, I'll get a second set of Kingston for the HD3P and see if that too behaves correctly. If it does, both sets of Corsair RAM will go back along with the first UD5. If, however, even Kingston RAM does not fix one or both boards, all three of them will be returned (though I'll keep the three sets of RAM initially) and then replace both boards with something other than Gigabyte.

Working in IT and having access to spare machines has enabled me to do all this diagnosis with a cool head but really, this would be beyond frustrating to a regular system builder. Either Gigabyte or Corsair have seriously screwed up this generation, and to be honest, given the first UD5's issues, I'm rather unconvinced it's Corsair at fault at this stage.
 
RIP HD3P motherboard, it's final attempt to use that RAM is appreciated.

I think this is Gigabyte's fault more than Corsair's. LPX DDR4 has been on the market for a while and generally well reviewed.
 
Tried the HD3P board again last night to no avail - popped the Kingston RAM in today without doing anything else, it booted into the BIOS, rebooted itself once then went straight into windows... Intriguing stuff - will see how it fares over the next couple of days - if it works out OK I might try it in the UD5, the replacement for which I also picked up today but haven't got the time to try out just yet.
 
Tried the HD3P board again last night to no avail - popped the Kingston RAM in today without doing anything else, it booted into the BIOS, rebooted itself once then went straight into windows... Intriguing stuff - will see how it fares over the next couple of days - if it works out OK I might try it in the UD5, the replacement for which I also picked up today but haven't got the time to try out just yet.

Glad that is working! Which RAM did you buy?
 
Kingston HyperX 2400mhz CAS15. However, that was a false alarm, this morning the system is back to how it was before. I suspect that it booting yesterday was triggered by me changing the RAM rather than anything to do with which RAM was used. I have a feeling if I were to swap the Corsair back again I'd see the same result. That's enough info to justify RMA'ing the HD3P I feel. Tonight I'll try fitting the replacement UD5 and see if that's any different.
 
Kingston HyperX 2400mhz CAS15. However, that was a false alarm, this morning the system is back to how it was before. I suspect that it booting yesterday was triggered by me changing the RAM rather than anything to do with which RAM was used. I have a feeling if I were to swap the Corsair back again I'd see the same result. That's enough info to justify RMA'ing the HD3P I feel. Tonight I'll try fitting the replacement UD5 and see if that's any different.

Yep -- pulling RAM and reseating / only using one or two slots when 4 had been used before sometimes brought my previous gigabyte board back for a boot cycle or two. I don't know what they've been doing lately, but it's a mess. Best of luck with the replacements. I may go to Skylake in the next few months so I'm watching this process closely -- hopefully we can figure out a good combo for you and then I'll just follow that exactly. Tired of the song and dance after too many late nights swearing at my Z68.
 
To be fair my Z68 Gigabyte system is good as gold, the only issue I have with that is that unless I switch the PSU ofc at the wall overnight it has a habit of sometimes powering back up all by itself. Never had any boot or stability issues with it though, that said the Z170 board isn't unstable either, once you get to windows it's absolutely fine whichever way round, it's just getting to that stage that's hard work. I almost wonder if they're sending boards out with dodgy BIOS chips...
 
I think I'm the only one lamenting Intel leaving the consumer motherboard market now. Most people hated on their Z68/Z77 boards, but I've loved this thing from minute one. It's really well designed, has a great layout, and after a BIOS update (which was, to be honest, way harder than it should've been) has been completely problem free. This is my first Intel board, and unfortunately it seems it will be my last. I just wish they'd brought the little LED-lit skull over from the Z68 variant. ;)

Edit: WAIT. I just saw a Z87 Intel board review on Youtube... are they still making desktop boards? Did they make a Z170 board? Let me google that for me!

Edit 2: No. No they don't. Back to being sad.
 
Last edited:
I never used them because I recall them having a reputation for being useless for overclockers, but if they still made them now I'd strongly be considering them - Intel may make some questionable design decisions sometimes but overall I'm far more impressed with the standard of their hardware than most companies in the industry.
 
The Intel Z77 70K OCs very, very well based on the reviews I've read. I've just been too lazy to push my chip back up to where it was or higher since the board swap. Maybe tomorrow? I'll let you know how it goes, as if you care. ;)
 
Isn't Z77 a bit old hat these days? ;)
I must admit the Z68 system was originally supposed to be Z77 but that board was all I could find 'in store' at the time, no such luxuries like doddle.it and amazon locker back in 2012 so was lumbered with what the local stores had available. After a (wrongly diagnosed) 'faulty' board, the only alternative was a Z68MX-UD2H, a bit spartan but it's worked out fine in practice. That system wasn't plain sailing when it was first built as a SATA data cable was actually somehow shorting data to earth, the PSU used to squeal every time any I/O occurred to the disk (and shut down a lot!) - changed the SATA cable and all was good :D
 
Isn't Z77 a bit old hat these days? ;)
I must admit the Z68 system was originally supposed to be Z77 but that board was all I could find 'in store' at the time, no such luxuries like doddle.it and amazon locker back in 2012 so was lumbered with what the local stores had available. After a (wrongly diagnosed) 'faulty' board, the only alternative was a Z68MX-UD2H, a bit spartan but it's worked out fine in practice. That system wasn't plain sailing when it was first built as a SATA data cable was actually somehow shorting data to earth, the PSU used to squeal every time any I/O occurred to the disk (and shut down a lot!) - changed the SATA cable and all was good :D

Wasn't willing (or able) to buy new CPU/RAM/MB when my Gigabyte Z68 died, so I picked up the cheapest Z77 I could find that wasn't terrible -- The Intel Z77 70K was the only one I could find with dual 16X PCIe (8x/8x split) under $100 new. Other manufacturers, people were asking $200+ for a 2+ year old board -- sometimes used. Idiots. Yeah, no.

Having now worked with an Intel "Extreme" series board, I wish I could use nothing else from here on out. Sadly that won't be the case.

I'll probably go to Skylake in the next few months... and I'll be looking at MSI when I do, more than likely.
 
New F5f beta BIOS has just be released for UD5 to improve memory compatibility.
 
Going to be trying the replacement UD5 either tonight or tomorrow - not had the time to do any tinkering this week so I've stuck with the HD3P board, which I can use as long as I don't reboot the PC during the daytime. If I switch the PSU off when I turn the PC off, the few hours that pass between then and when I next need it are sufficiently long enough that when the PSU is switched on again, the PC will boot.

Analog onboard audio on this board is also getting rather tiresome, not having S/PDIF (my receiver doesn't have proper analog inputs set up, I use optical so that the surround speakers work) on a £95 board is pretty lame, considering the Z97N-WiFi in the LAN machine has it and that board was only £7 more expensive. The amount of current the Z170-HD3P's onboard audio can source for headphones is rather 'lacklustre' shall we say...
 
Test results:

Background environment: HAF 932 case, Corsair RM1000i PSU, Gainward GTX970 GPU - all other peripherals unplugged for testing
Variations specified in tests below if made

Original (for reference):

Z170X-UD5 board #1, i5 6600, dual-channel Corsair set 1 (1/2): No POST (Error 55 - memory)
Z170X-UD5 board #1, i5 6600, dual-channel Corsair set 1 (3/4): No POST (Error 55 - memory)
Z170X-UD5 board #1, i5 6600, dual-channel Corsair set 2 (1/2): No POST (Error 55 - memory)
Z170X-UD5 board #1, i5 6600, dual-channel Corsair set 2 (3/4): No POST (Error 55 - memory)
Z170X-UD5 board #1, i5 6600, single-channel Corsair set 1, stick 1 (1): No POST (Error 19)
Z170X-UD5 board #1, i5 6600, single-channel Corsair set 1, stick 1 (3): No POST (Error number not recorded)
Z170X-UD5 board #1, i5 6600, single-channel Corsair set 1, stick 2 (3): POST on first power-up only


Z170-HD3P board, i5 6500, dual-channel Corsair set 1 (1/2): POST on first power-up only
Z170-HD3P board, i5 6500, dual-channel Corsair set 2 (1/2): POST on first power-up only
Z170-HD3P board, i5 6500, dual-channel Corsair set 2 (3/4): POST on first power-up only
Z170-HD3P board flashed to F4 BIOS, i5 6500, dual-channel Corsair set 2 (3/4): POST on first power-up only
Z170-HD3P board flashed to F3d BIOS, i5 6500, dual-channel Corsair set 2 (3/4): No POST
Z170-HD3P board flashed to F3d BIOS, i5 6500, dual-channel Corsair set 2 (1/2): No POST
Z170-HD3P board flashed to F3d BIOS, i5 6500, dual-channel Kingston set (1/2): POST on first power-up only

Z170-UD5 board #2, i5 6600, dual-channel Corsair set 2 (1/2): POST on first power-up only
Z170-UD5 board #2, i5 6600, dual-channel Corsair set 2 (3/4): POST on first power-up only
Z170-UD5 board #2, i5 6600, dual-channel Kingston set (1/2): POST on first power-up only
Z170-UD5 board #2, i5 6600, dual-channel Kingston set (1/2), Zalman ZM850-HP PSU used: POST on first power-up only
Z170-UD5 board #2, i5 6600, dual-channel Kingston set (1/2), Zalman ZM850-HP PSU used, GTX970 moved to second PCIe slot: POST on first power-up only
Z170-UD5 board #2, i5 6600, dual-channel Kingston set (1/2), back to RM1000i, graphics card removed and using IGP: POST OK first 6 tests, 7th test POST failed with error A6, 8th-10th tests OK
Z170-UD5 board #2, i5 6600, dual-channel Kingston set (1/2), RM1000i, XFX HD6970 graphics card used: POST OK first 5 tests, tests 6-10 all POST failed with error A6

What are my conclusions from these tests?

1. The first Z170X-UD5 board is legitimately DOA with a memory fault not seen on the other two boards
2. The Z170-HD3P and both Z170X-UD5 boards are examples of a design defect with Gigabyte Z170 boards, which if I had to guess, is something to do with the PCI express bus. The fact that it doesn't make any difference which PCIe slot is used, but does distinctly matter what graphics card (if any) is plugged in makes me think there is a PCIe problem with these boards.
The fact that different CPUs were used as part of the tests also tells me it is very unlikely that the issue is related to the PCI express controller in the CPU. With PSU, GPU, RAM and CPU all swapped out and providing identical results each time (with known working components in the case of PSU and GPU), board is the only logical answer I can come up with.
3. The second Z170X-UD5 might also be faulty due to the A6 error as the manual tells me that's SCSI? On the other hand it could perhaps be another manifestation of the issue above.

I can't definitively prove at this stage that most Gigabyte Z170 boards are bad, but it currently seems a more likely outcome than two CPUs, entirely stable from within windows, that have the same defect (CPUs in my experience are relatively rarely defective).
The final stage of proof is to introduce a different brand of motherboard into the equation and see if I get the same result there. If I don't, the above is pretty much proven solid. If I do, then I really would be stumped.
 
A coworker of mine is going through some similar issues with a Z170 Asus board, and the board is his second Z170 board (the first board wasn't Asus, may have been a Gigabyte). Really frustrating that some people doing nothing out of the ordinary are having a lot of issues with Z170 boards and power on. I wonder if the chipset wasn't ready for prime time or board makers just dropped the ball. When P67 "B3" boards came out my friends and I built three PCs with different hardware and every one of the systems had no issues booting from any previous power states. Two of us had XMP issues to the point where I just gave up and used my board for more than four years at default DDR3 speeds.

I am happy my PC works fine now, however also a bit worried that my bad boot behavior may return in the future with another BIOS update. I would return my board for a refund if there was a bulletproof option on the market, however people are having issues with everything. My setup only works with a beta BIOS and I'd rather be using normal releases and have confidence that I can update as needed over the next couple years. And I don't feel the issues I was encountering with the previous BIOS releases were acceptable given my very simple and common hardware load out.

This goes back to the Intel mainboard discussion earlier. When Intel released boards they were often down on features, had few expansion slots, and we're pretty ugly. Damn near every one I have encountered got the basics right though. I originally looked for an Intel Z170 board and unfortunately Intel is out of the market now. I am past the point in my life where I care about gaming features or "extreme" unicorn fur capacitors marketing hoopla. I just wanted something that works normally and could run RAM at XMP speeds to utilize my Skylake CPU. I won't feel good about this Gogabyte board I have until I get a couple more BIOS releases without encountering boot issues.

My previous MSI P67 board was fine aside from never working with my XMP RAM and the fact that the screwed up the BIOS and CPU support over time. Best of luck to you with your new mainboard.
 
Last edited:
Very disconcerting - I will give another manufacturer a shot to see if that's going to work - Amazon have now processed my refund request for the two dud UD5 boards - my request to Scan for the HD3P seemed to vanish without trace, so I'll speak to them Monday about resurrecting that. I'm going to try the equivalent featured MSI board for the UD5 and if that doesn't work, I am sorely tempted to either go Z97 or X99 and write off Z170 as a bad job. I can't escape the thought though that "Surely Intel couldn't have screwed up this badly, could they?"

I'll check to see if I can spot the refund in - if I can I'll order the MSI this weekend for delivery either Monday or Tuesday and have a go at that, pff, whenever I get time, possibly not until Wednesday night given changing motherboards is a rather time consuming process. Will of course report back with my findings. As reassuring as it is to see others with the same issue, it's baffling that these systems are getting even more problematic as time passes.
 
Well this seems a little... scary. Have you looked around and noticed this on many other Z170 boards?
 
Only seen my refund come in tonight so will be ordering an MSI board tomorrow for delivery Thursday. Will see if that makes any difference.

Edit: MSI Z170A Gaming Pro ordered for delivery Thursday. Will see if that's any good. Going to send the UD5s back today. Scan have now sent me a huge long email asking me to fully diagnose the HD3P before it goes back, so I imagine I'll be subjected to more testing there, hardly seems worth the effort for the sake of £95 so I'm tempted to just ebay it instead, but I figure despite the aggravation more testing may help firm up my hypothesis that there is a genuine design defect with the Gigabyte boards. Hopefully the MSI one comes up with the goods or I really am stumped, and will probably return all the kit including the CPUs and RAM and go with a Z97 build instead.
 
Last edited:
For what it's worth, I got an ASUS Z170-K and have been trouble-free for a couple of months now. It's $25 cheaper than an ASUS Z170-A, only downside being no SLI support. It will do Crossfire. I always go nVidia for GPU but I never SLI, so I'm fine there.

I hadn't had a modern-style BIOS until now, I love it.
 
Back
Top