GIGABYTE Releases Statement on GP-P850GM & GP-P750GM PSUs

[watzupken]

I think that Intel, AMD and Nvidia also guilty. Because make design of products with much more power consumption and bad very big peaks and PSUs with little older design which is enough for older PC hardware is not enough for requirements of new models of PC hardware. I don't wish to think how older model PSUs will work with Intel Alder Lake or with AMD CPUs with ZEN4 architecture if rumors of how be it's consumption is true. There is problem with consumption of modern parts when overclock...Has big difference between consumption of OC of parts to 2014 when has gap to "play" OC without too big problems and parts after 2014 which when load is big work very close to limit. After this limit when OC consumption growing exponentially.

Its true that the power consumption of hardware have increased quite a fair bit in recent years, but I don't see that many issues with other PSUs coping with the power requirements without exploding. At least not something that is replicable with a high % of the PSU. So the issue is definitely with Gigabyte. Gigabyte don't manufacture these PSUs, but by slapping their brand on it, the basically failed QC. One can delegate responsibility, i.e, getting someone to make the product, but they cannot remove accountability for it.

In my opinion, Gigabyte should only do 2 things to clean up this mess,
1. Stop the sale of these PSUs,
2. Issue a replacement (after fully rectifying the issue) or refund to customers.

How can they continue to sell these crap PSUs that are potential fire hazard to consumer? Even if it doesn't burn down the house, there is a risk of it killing your PC.

 

[ArdWar]

I think that Intel, AMD and Nvidia also guilty. Because make design of products with much more power consumption and bad very big peaks...

I understand that the massive undocumented/unspecified peak in itself is a problem, but I can't see how increase of power requirement is their fault, or even a fault at all....

Anyway, downstream power shenanigans should not matter here. It's PSU's job to protect itself (or not). What is the point of advertising OPP, OVP, OVP, UVP, etc anyway if they don't actually give a protection. Or worse, cause the PSU to fail in non safe way taking everyone with them...

 

[watzupken]

I love it how they stress this part first:
for extended lengths of time repeatedly close to the 120% to 150% OPP trigger point

So it makes it seem like it's user/reviewer's fault that they went kaboom after surviving the OPP test and then they decide to lower OPP to 110-120%.

Whatever they say, the issue is specific to their now infamous exploding PSU. There are so many PSU makers and if it is a user problem, then why is the same issue not happening to other PSUs from reputable brands? So their suggestion here is for people buying their 850W PSU to use on Intel Celeron/ AMD Athlon processors, with Nvidia GT 1030 or AMD RX 550 GPUs so that the PSU will not overload.

They can say whatever they want, but will not change people's opinion on their PSUs, and potentially, their other products. When your QC is bad, it makes the quality of your other products also questionable.

 

[TumbleGeorge]

I understand that the massive undocumented/unspecified peak in itself is a problem, but I can't see how increase of power requirement is their fault, or even a fault at all....

Anyway, downstream power shenanigans should not matter here. It's PSU's job to protect itself (or not). What is the point of advertising OPP, OVP, OVP, UVP, etc anyway if they don't actually give a protection. Or worse, cause the PSU to fail in non safe way taking everyone with them...

Maybe issues is because more consumers buy this two PSUs for system with consumption and peaks close or above of limits of it's OCC because this models is cheaper than other trademarks "golden" PSUs with 750-850 watts? Is normal to use PSU that parameters has advantage than power consumption of system and work average on not more of 60-70% of it's upper limit. More purchases and more (bad)users=more causes.

 

[Tsukiyomi91]

it's more on "I wanna skimp on the PSU a little bit so I can save more to get my scalped RTX3080Ti" instead of "I wanna get the cheapest PSU and be happy with it" IMO.

 

[iancook221188]

fancy putting a poll up if people still trust psu or even the hole gigabyte as a brand now after this psu debacle

 

[Blue4130]

fancy putting a poll up if people still trust psu or even the hole gigabyte as a brand now after this psu debacle

For me, I don't trust or distrust any brand as a whole. Everybody has made shoddy parts. Need to review each part with blinders on. Choose based on the specifics of each part regardless of the brands past sins or wins.

 

[R0H1T]

because this models is cheaper than other trademarks "golden" PSUs with 750-850 watts?

How many users even on TPU need 80+ Gold/platinum/titanium PSU rated for 750-850W or more? Also it's not exactly cheap!

 

[TumbleGeorge]

fancy putting a poll up if people still trust psu or even the hole gigabyte as a brand now after this psu debacle

More people's in planet Earth has low education or education of kind which is not make them clever PC hardware parts buyers and users. They not choose good balanced configuration. The leading choice for them is to cram the most powerful hardware possible into their limited purchase budget and pay for the box and power supply with the rest of the pennies.

How many users even on TPU need 80+ rated Gold/platinum/titanium PSU rated for 750-850W or more? Also it's not exactly cheap!

"Gold" is far far behind platinum/titanium/diamond! Don't push it in same league!

 

[R0H1T]

"Gold" is far far behind platinum/titanium/diamond! Don't push it in same league!

Of course but you said they are cheap as compared to other "Golden" PSUs ~ do you want me to show how they're not cheap, at least price wise?

Also there's no diamond in there, unless you meant it as a joke?

 

[TumbleGeorge]

Of course but you said they are cheap as compared to other "Golden" PSUs ~ do you want me to show how they're not cheap, at least price wise?

When years ago "gold" standard was on the top, has enough top quality capacitors and other parts to made all "gold" PSU really "gold". Next years when on market was offered most rated standards, this limited top parts is used for make it and for production of "gold" PSUs began to use lower quality parts, which are produced in larger quantities.

 

[freeagent]

How many users even on TPU need 80+ Gold/platinum/titanium PSU rated for 750-850W or more? Also it's not exactly cheap!

Even with my old GTX 980 I can get my rig to pull 500w from the wall with something like F@H. But when I bought mine 750 was pretty much all I could find, unless I wanted a thermaltake 850.. no thanks I already have a dead one of those.

 

[R0H1T]

When years ago "gold" standard was on the top, has enough top quality capacitors and other parts to made all "gold" PSU really "gold". Next years when on market was offered most rated standards, this limited top parts is used for make it and for production of "gold" PSUs began to use lower quality parts, which are produced in larger quantities.

So you basically meant something else? The reason I'm highlighting this is two fold ~
1) Crap PSUs are generally almost always the cheapest in a bunch i.e. it's relatively easy to spot garbage products with their prices. These two aren't exactly cheap so not a dead giveaway!
2) Your premise that people cheaped out, on quality parts, & bought this also doesn't hold weight because it's generally 5~15% cheaper than better quality parts. Which is every much acceptable from different (quality) brands.

So most people who bought these didn't buy because it was "that cheap" & more because it was a quality brand & supposedly gold rated PSU.

So Gigabyte had really sh!t QC or they intentionally tried to deceive the buyers! Either way Gigabyte should pay for this.

 

[maxfly]

Gigabytes clueless pr dept couldnt have possibly made things any worse if they tried.
I can see it now...
Heres our strategy people, we'll just blame the ones that reported the problem in the first place! Brilliant sir! No way anyone's going to see through that smoke screen!
Cough cough gasp cough
Yep, those are the guys im buying my next gpu, mb and psu from. You betcha! They know what their doing over there

 

[Sithaer]

For me, I don't trust or distrust any brand as a whole. Everybody has made shoddy parts. Need to review each part with blinders on. Choose based on the specifics of each part regardless of the brands past sins or wins.

Same here, I don't judge a whole brand because of a bad product they happened to relase or some questionable PR related crap. 'show me a brand thats free of that or never done anything in the past, best case scenario they are yet to be caught'

I do my research before buying something and if its okay in general and fits my needs then its all good, regardless of the brand.
I'm on my third Gigabyte GPU in a row now and so far they got the job done just fine even if I had to tweak my RX 570's power settings manually.

Wouldn't have any issue buying another GPU from them if it comes to that. 'say it has the best pricing where I live,etc'

 

[Dr. Dro]

Tone deaf, lame attempt at damage control... truth is, it doesn't look like Gigabyte cares.

Which may well be true, after all, these power supplies have been notorious for their problems for quite some time, and they probably had a generous stock of unsold units, which is probably why it went into the Newegg bundles to begin with. These forced combo sales only happen because of poor consumer protection legislation anyway. In my country, this is illegal (article 39, I of our federal consumer protection law) and will result in legal trouble plus serious fines for any establishment that attempts to pull something like that.

The sad part is once all this backlash is over and unsold stock is returned, it'll just be rebranded and offloaded on third world "emergent" markets, and they'll be the talk of the town in our hardware forums for the next few years.

 

[TumbleGeorge]

So you basically meant something else? The reason I'm highlighting this is two fold ~
1) Crap PSUs are generally almost always the cheapest in a bunch i.e. it's relatively easy to spot garbage products with their prices. These two aren't exactly cheap so not a dead giveaway!
2) Your premise that people cheaped out, on quality parts, & bought this also doesn't hold weight because it's generally 5~15% cheaper than better quality parts. Which is every much acceptable from different (quality) brands.

So most people who bought these didn't buy because it was "that cheap" & more because it was a quality brand & supposedly gold rated PSU.

So Gigabyte had really sh!t QC or they intentionally tried to deceive the buyers! Either way Gigabyte should pay for this.

I have obviously said the situation is not same and change over the years.

 

[Tomorrow]

It's always funny to see haters jumping on board when a company produces (or in this case rebadges) a bad product. Yeah so because two on their PSU's are bad the rest they produce must be utter crap too?

Their AM4 boards are some of the best out there. Hardware Unboxed just said their M28U monitor is pretty good value for a 4K 144Hz model etc.
You cant take one product and say all the companys products are s*it (well maybe Biostar) Yet that is exactly how some people behave. Obviously they rarely follow their own advice. I mean should we also avoid all ASUS products because they made a bad laptop or bad cooling system on some AMD cards? Should we avoid EVGA because some of their 3090 designs can blow up in one game?

EVERY company has had a bad product at one point or another including the big RGB (AMD, Nvidia, Intel).
Its how a company respons and proceeds is important. With big companies it is not unusual for some products to be better than others. I would not buy an AsRock Z590 board but i'll gladly buy one of their miniPC designs like the X300. It's also not unusual for a big company to have great product one gen and a bad product the next gen.

 

[Chrispy_]

This statement from Gigabyte is inflammatory, pun not intended.

The reason they catch fire is because the OPP is set far too high (150%) for the quality of the components used, they can barely handle 100% of the rated load for periods that should be considered "normal operation"

Repeating that their OPP is set far too high in an official statement doesn't address anything at all and just demonstrates Gigabyte's tone-deaf approach to feedback and lack of understanding about how their reputation is at stake here.

 

[eidairaman1]

Which is pretty typical Gigabyte.

Yeah they havent had great boards since about 2012/13/14.

It's always funny to see haters jumping on board when a company produces (or in this case rebadges) a bad product. Yeah so because two on their PSU's are bad the rest they produce must be utter crap too?

Their AM4 boards are some of the best out there. Hardware Unboxed just said their M28U monitor is pretty good value for a 4K 144Hz model etc.
You cant take one product and say all the companys products are s*it (well maybe Biostar) Yet that is exactly how some people behave. Obviously they rarely follow their own advice. I mean should we also avoid all ASUS products because they made a bad laptop or bad cooling system on some AMD cards? Should we avoid EVGA because some of their 3090 designs can blow up in one game?

EVERY company has had a bad product at one point or another including the big RGB (AMD, Nvidia, Intel).
Its how a company respons and proceeds is important. With big companies it is not unusual for some products to be better than others. I would not buy an AsRock Z590 board but i'll gladly buy one of their miniPC designs like the X300. It's also not unusual for a big company to have great product one gen and a bad product the next gen.

As i wrote above, also why do their boards go through so many revisions? Why not do it right the first time?

 

[Tomorrow]

Yeah they havent had great boards since about 2012/13/14.
As i wrote above, also why do their boards go through so many revisions? Why not do it right the first time?

Like i said. Most of their AM4 boards are great. Especially X570. Why is having revisions a bad thing? Most motherboard makers have them.
It's not always possible to get things right the first time. I mean look at X570 vs X570S. That's not Gigabyte's fault.

In case of my board X570 Aorus Master it has been rock solid. The only complaint i can make is that there a BIOS bug that with CSM disabled the BIOS UI becomes laggy for some reason. They have tried to work around that. Gigabyte is also one the first to release Beta versions of AGESA patches and the revision 1.1 and 1.2 (i have 1.0) have added only TPM header and improved the memory topology layout plus some cap changes. Nothing earth shattering because i have firmware fTPM and 3733 C14 running daily (1:1).
I dont feel like i've been screwed with 1.0 in any way.

 

[xtreemchaos]

a good company would do a recall of said products, thay will in the end if thay are smart im not holding my breath.

 

[Redwoodz]

I've never really been a Gigabyte fan. I bought an ATI/Gigabyte gpu which turned out to be a horrible overclocker, I think I could only get another 17 mhz.And it had an ugly blue pcb. Not really a good reason to blacklist a company,but I have shied away from them ever since.

That being said, I actually applaud this press release. They directly addressed the problem. Every single product out there is a compromise, sometimes product targets are off a bit. To be honest I am much more likely to choose one of their PSU's now then before, because I know they had to research the problem.

p.s. Has their been any actual users reporting explosions? Or is this all based on review failures?

 

[TheinsanegamerN]

Like i said. Most of their AM4 boards are great. Especially X570. Why is having revisions a bad thing? Most motherboard makers have them.
It's not always possible to get things right the first time. I mean look at X570 vs X570S. That's not Gigabyte's fault.

In case of my board X570 Aorus Master it has been rock solid. The only complaint i can make is that there a BIOS bug that with CSM disabled the BIOS UI becomes laggy for some reason. They have tried to work around that. Gigabyte is also one the first to release Beta versions of AGESA patches and the revision 1.1 and 1.2 (i have 1.0) have added only TPM header and improved the memory topology layout plus some cap changes. Nothing earth shattering because i have firmware fTPM and 3733 C14 running daily (1:1).
I dont feel like i've been screwed with 1.0 in any way.

x570 vs x570s is two different chipsets, not revisions. What's he's talking about is the same board from gigabyte having sometimes upwards of 6 revisions within a year. THAT is not normal. That is poor QC beta quality boards being sold as final products. One or two revisions is one thing, but 5-6 was ridiculous even 20 years ago when QC was non existent compared to today.

 

[TumbleGeorge]

x570 vs x570s is two different chipsets,

Off topic/Why not rename it like something x575 if different chipset?/