Power Supply Quality

As part of our testing, we also check output parameters are within specifications, as well as voltage ripple and line noise.

Main Output
Load (Watts) 203.21 W 507.02 W 757.83 W 1009.06 W
Load (Percent) 20.32% 50.7% 75.78% 100.91%
  Amperes Volts Amperes Volts Amperes Volts Amperes Volts
3.3 V 2.27 3.45 5.66 3.45 8.5 3.4 11.33 3.37
5 V 2.27 5.18 5.66 5.16 8.5 5.09 11.33 5.06
12 V 15.1 12.16 37.75 12.14 56.62 12.11 75.5 12.1

 

Line Regulation
(20% to 100% load)
Voltage Ripple (mV)
20% Load 50% Load 75% Load 100% Load CL1
12V
CL2
3.3V + 5V
3.3V 2.4% 12 18 18 18 16 24
5V 2.2% 16 18 20 22 18 26
12V 0.5% 16 24 28 32 30 24

The electrical performance of the GIGABYTE UD1000GM PG5 is admittedly better than we initially surmised. There is strong filtering on the primary voltage output, with the GIGABYTE UD1000GM PG5 displaying a maximum ripple of 32 mV on the 12V line under maximum load, which is around 25% of the recommended design limit. The voltage regulation of the 12V line is equally impressive, at less than 0.5%. GIGABYTE’s engineers neglected the 3.3V and 5V lines a little, as the regulation is at about 2.2% and the voltage ripple goes up to 45% of the design limit. These power quality figures are very good for a product of this tier.

Conclusion

The announcement of Intel’s new ATX v3.0 specifications was bound to start a race, as every company is looking to establish itself among the early adopters as a means of growing with the larger market. GIGABYTE took a little different approach here – instead of developing a whole new platform capable of fulfilling all of ATX v3.0 requirements, they simply upgraded one of their existing units and added one 600 Watt 12VHPWR connector to it.

It is the first time we have ever seen a platform from Xiamen Metrotec, the OEM of this unit. The OEM obviously needs to be following the design and guidance of the final developer, which is GIGABYTE in this case. Nevertheless, it appears that they are doing a very good job with their core topology designs and assembly process, both of which we found to be very clean. However, the quality of the components used by GIGABYTE for the production of the UD1000GM PG5 is wanting. The PSU comes with a very long warranty but most of the components, passive and active alike, are of middling quality. Taking into account the high internal temperatures that the UD1000GM PG5 reaches under stress, we feel that GIGABYTE’s engineers are being a little too confident.

When it comes to performance, the GIGABYTE UD1000GM PG5 will not be entering the hall of fame. However, all things considered, it is fairly well-balanced. The power quality of the unit is very good, with excellent regulation and filtering on the primary voltage line and adequate figures on the secondary lines. It is adequately efficient so as to achieve an 80Plus Gold certification with an input voltage of 115 VAC, yet only barely, so there is room for improvement. The only real disadvantage of the UD1000GM PG5 is its small size, which forced the engineers to use a smaller 120 mm fan, making the unit run both hotter and louder than most equally powerful units. This only becomes a true issue while the unit is heavily loaded though, and should not be bothering typical users while they simply want to do their work or browse. The performance of the GIGABYTE UD1000GM PG5 is unimpressive by today’s standards but, taking into account the retail price of the unit, it is good overall.

The true advantage that GIGABYTE’s UD1000GM PG5 currently has is simple – no competition. It is virtually the only unit with a 12VHPWR connector currently available. If someone needs to purchase a powerful PSU right now and expects to be upgrading their graphics card in the near future and/or often, the UD1000GM PG5 is  essentially the only choice with respect to forward-compatibility. That being said, one has to remember that this unit does not fully fulfill the ATX v3.0 requirements - it is unlikely but not unfathomable that the UD1000GM PG5 may not be able to handle future graphics card releases that could produce excessive transient loads.

Regardless, the current retail price of $160 is reasonable for an 80Plus Gold certified unit with that kind of output, so buyers are not paying a severe premium for the new 12VHPWR connector. However, if you are not in a rush to purchase a PSU right now, we err on the side of caution, and recommend waiting for actual ATX v3.0 compliant units to pop into the market.

 
Hot Test Results (~45°C Ambient)
Comments Locked

49 Comments

View All Comments

  • Khanan - Friday, June 24, 2022 - link

    This makes a lot of sense and I agree with this. Same shit they did with recent PSUs, didn’t accept it’s trash and wanted to sell it anyway via Newegg bundles. Hot trash, just sell it fast instead of recalling. Everyone knows what happened after, a PR disaster, there are multiple videos about in on gamersnexus YouTube. Gigabyte nowadays seems to be like a hit or miss, you really should know what’s good and what’s not with them or you risk buying trash or a mediocre product. With this PSU it’s pretty fine, I just don’t get why they put such a cheap fan in it.
  • Leeea - Thursday, June 23, 2022 - link

    Anandtech, you are familiar with the rather recent Gigabyte power supply debacle? The recall for "known to occasionally catch fire"?

    The one where it appeared they realized they had a massive issue with PSU quality
    then worked with newegg to dump this defective stock on unsuspecting customers
    were outed by Gamers Nexus and others for this blatant anti-consumer behavior

    and lastly, recalled the defective models because they were "known to occasionally catch fire"

    That was September 2021, not long ago.
    https://www.google.com/search?q=gigabyte+power+sup...
    https://www.tomshardware.com/news/gigabyte-full-re...

    I would think the most important detail in any Gigabyte power supply review would be to address the question:
    Will this light on fire?

    with pretty much everything else being a distant second concern.
  • Threska - Thursday, June 23, 2022 - link

    Clearly running Crysis would be a bad idea.
  • Khanan - Friday, June 24, 2022 - link

    He did a high current test and the PSU passed so yes it’s fine. The issue was with cheaper models, this is a bit more expensive.
  • Leeea - Friday, June 24, 2022 - link

    He did many very good tests, and for all the questions not related to "Will this light on fire?" the review was excellent.

    However, if I remember right, to spontaneously ignite a Gigabyte PSU they were triggering the OCP. It would work once or twice, but the majority of tested units failed with considerable drama.

    Thing is, when a GPU pushes its transients up and triggers the OCP, most users do not realize why their computer just shut down. This results most users who trigger OCP triggering it several times in a row.

    Which on Gigabytes previous models, was rolling the dice on an unfortunate and dramatic failure. Or as Tom's Hardware put it: "known to occasionally catch fire"

    While E. Fylladitakis's testing was excellent and exhaustive, it does not appear he tested the known failure mode for Gigabyte power supplies.

    It is also unlikely anyone could answer the question: "Will this light on fire?", because considering Gigabytes recent history, who knows?
  • DanNeely - Friday, June 24, 2022 - link

    Anandtech editors work from home. Deliberately trying to trigger a fire starting failure mode almost certainly goes against E. Fylladitakis's insurance and/or rental agreements. It's the sort of thing that needs to be done in commercial/industrial spaces with fire suppression and enhanced insurance coverage.

    After the recent debacle, I hope GN does attempt to trigger similar failures in other PSUs but disagree that it should be part of a standard for all reviewers going forward.
  • Leeea - Friday, June 24, 2022 - link

    You make good points.

    Gamers Nexus did try a whole bunch of other brands, including no name Amazon Brands. They discovered a bunch of Amazon/Ebay brands were lying about their 80+ efficiency, but they were not able to get any of the other brands to self combust.

    I suppose for a home reviewer getting a RMA on GPU with scorch marks not going to be doable.

    I guess it is one of those catch 22 moments, where if he mentions it people are going to ask why he did not test it, and if he does not mention it people like me are going to pitchfork him.

    That said, I do feel consumers should know Gigabyte has a recent history of selling very questionable PSUs while appearing to ignore reports of disaster with said PSUs.
  • E.Fyll - Friday, June 24, 2022 - link

    If that is enough to put your mind at ease, I always test all of the primary safety mechanisms of every PSU. This includes OPP and it worked as expected. If it did not, I would be yet another reviewer with a dead sample on my bench anyway.

    I am familiar with the issues GB had to deal with. The new revisions, although far from perfect, at least will not blow up while playing minecraft.
  • Leeea - Friday, June 24, 2022 - link

    Thank you!
  • Kaggy - Thursday, June 23, 2022 - link

    Wonder if ATX PSUs will eventually graduate from ATX and become smaller through GAN technology.

Log in

Don't have an account? Sign up now