ty_ger07
Insert Custom Title Here
- Total Posts : 21171
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 270
Re: Dead 3090 / Power Balance Issues
2022/07/26 14:35:37
(permalink)
Michapolys
yaymz
The capacitors on a maxed out PSU aren't big enough to handle those load spikes coming from the GPU. Not the PSU's fault if the power demand goes way over the wattage limit of the PSU, even for a split millisecond.
That is the problem. The power demand is within the recommended limit as defined by the ATX spec. It is just that most PSUs currently out there rely on a loophole of the spec to reduce their transient load capabilities.
Which is still irrelevant to cards that burn up.
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
My EVGA Score: 1546 • Zero Associates Points • I don't shill
|
NobleHonor7
New Member
- Total Posts : 13
- Reward points : 0
- Joined: 2022/03/21 10:20:41
- Status: offline
- Ribbons : 0
Re: Dead 3090 / Power Balance Issues
2022/07/27 06:30:21
(permalink)
I’m hoping I receive a Rev 1.0 from this RMA (I requested said Rev 1.0 specifically), and hopefully the updated power balancing will keep the card from blowing up. The main card type I’ve been seeing recently is the old EVGA posts that were everywhere, and gigabyte products, so we shall see. I’ll be stress testing the card fully upon delivery, and update this post with results.
Planning on testing with Super Position, Furmark, Port Royal and several games (mix of old and new games that I feel will stress the card in different ways). Taking recommendations as well to test it and see if it will burn up. I’m hoping it survives, but I seriously have my doubts.
|
ty_ger07
Insert Custom Title Here
- Total Posts : 21171
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 270
Re: Dead 3090 / Power Balance Issues
2022/07/27 06:57:37
(permalink)
I don't think that you should expect the new card to burn up right away. It's not an event that appears to be too common. Let's say (figuritively) that it happens every 1 in a million transients. But transients happen with these cards all the time; it is guaranteed by its design. So, in my opinion, it is more like playing russian roulette. As far as purposely trying to prove that the card survives, World of Warships has been reported by some to be super effective at showing a card's vulnerability to transient loads. There's also of course New World and Halo MCC which have brought many cards to their VRM knees.
post edited by ty_ger07 - 2022/07/27 06:58:43
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
My EVGA Score: 1546 • Zero Associates Points • I don't shill
|
UndesiredSanity
New Member
- Total Posts : 74
- Reward points : 0
- Joined: 2020/11/19 18:41:33
- Status: offline
- Ribbons : 1
Re: Dead 3090 / Power Balance Issues
2022/07/27 08:38:47
(permalink)
This thread reminded me that I still have an original rev 0.1 "Red Lip" 3090 FTW3 Ultra Gaming originally purchased Nov 2020 from Microcenter and serial number registered. I read somewhere about a Load Balancing issue and a specific XOC exchange program/RMA. This is my wife's card now and has been the past year while purchased another 3090, rev.10 (black lip w/ XOC 500W) bios. I ran a few 3Dmark benchmark tests and can confirm it's drawing 88+W on the PCIe power bus.
Do you guys think I can still submit this GPU for a RMA on received a newer rev. 1 version so that we won't have any issues or concerns in the past? I now have a 3090ti KPE and thinking of a SLI setup with the two 3090's but they are different rev versions. What is the proper name of this exchange/RMA program specific to these 3090 load balancing older cards?
|
yaymz
SSC Member
- Total Posts : 736
- Reward points : 0
- Joined: 2006/09/08 07:14:31
- Status: offline
- Ribbons : 4
Re: Dead 3090 / Power Balance Issues
2022/07/27 09:55:55
(permalink)
ty_ger07
yaymz
The capacitors on a maxed out PSU aren't big enough to handle those load spikes coming from the GPU. Not the PSU's fault if the power demand goes way over the wattage limit of the PSU, even for a split millisecond.
Yeah, sure. But then the PSUs will just trip like some of those Seasonic and EVGA PSUs were doing. Or the GPU will black screen due to undervoltage protection kicking in. But the cards which are burning up, they were successfully able to draw enough power to kill themselves; so it is illogical to blame the PSU for the GPU's death or say that the under-sized PSU somehow could possible cause that condition to happen.
The GPUs which fried their VRM or burnt their fuses were able to successfully draw too much power. That is solely the fault of the GPU.
Correct. We are talking about 2 separate issues here. But both were being unfairly blamed on the PSU.
cpu: Intel 12900k-EK Quantum Magnitude waterblockmobo: Asus z690 Apexram: G.skill DDR5 @ 6000+. gpu: MSI 4090 Suprim Liq x on EK-Quantum Vector² Trio ABP waterblockssd: Samsung m.2 980 Pro 2TB (x2)psu: beQuiet Dark Power Pro 1500wcase: Lian-li o11d xl monitor: Asus ROG Swift 27" 1440p @240hz (PG279QM).[/
|
Michapolys
Superclocked Member
- Total Posts : 199
- Reward points : 0
- Joined: 2021/12/25 09:26:18
- Status: offline
- Ribbons : 2
Re: Dead 3090 / Power Balance Issues
2022/07/27 16:00:46
(permalink)
They are essentially one issue... on the PSU...
|
ty_ger07
Insert Custom Title Here
- Total Posts : 21171
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 270
Re: Dead 3090 / Power Balance Issues
2022/07/27 16:04:31
(permalink)
Michapolys
They are essentially one issue... on the PSU...
Your refusal to use logic and reasoning is impressive. In our 4 discussions you have yet to provide a logical explanation for the holes in your theory. Lets make it really simple: The card has undervoltage protection. The card does not have functioning overcurrent protection. How does the VRM fry or the VRM fuses blow? It's a simple question. Don't answer it for me. Answer it for yourself. Look at the facts, and then decide what could or could not cause the failure. We are talking about a power transient. Power. Watts. Not a voltage transient. A power transient is guaranteed to happen when the load increases, due to NVIDIA's design. A voltage transient is protected against, a power transient is not protected against. Electrical theory states that a source of power cannot force power into a load. The load dictates the power draw. The power transient absolutely, without a doubt, originates from the GPU. It is not caused by the PSU. A GPU which fries during a power transient was, without a doubt, successfully able to receive all the power it demanded. IT demanded. There is no question. So, how can a PSU which successfully provided all the power demanded of it, be blamed when the GPU fries due to its demand of too much power? It is illogical. Whose fault is it that the GPU demands more power than its VRM is capable of handling? Whose fault is it that the VRM is under uneven load and thus more likely to overload one input before the other(s)? I think those questions have two different answers, and neither one has anything to do with a power supply.
post edited by ty_ger07 - 2022/07/27 16:31:39
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
My EVGA Score: 1546 • Zero Associates Points • I don't shill
|
Michapolys
Superclocked Member
- Total Posts : 199
- Reward points : 0
- Joined: 2021/12/25 09:26:18
- Status: offline
- Ribbons : 2
Re: Dead 3090 / Power Balance Issues
2022/07/27 16:53:08
(permalink)
It fries by having repeated transient loads with a duration lower than 5 us where the voltage takes a dive and the current increases exponentially. Edit: This actually refers to just the VRM output voltage as the PWM only has under-voltage protection just for that. A note here. A few decades ago or so, PWMs did not have any protections. You were left with whatever voltage the PSU would provide. If you fed the wrong voltage to the hardware, it would just fry. Always. Having such protections is an extra. And still the hardware is blamed. :P Thing is, I use logic and reasoning and still avoid statements based solely on those. It is evidence that clear things up. Something like this here for example: https://i.imgur.com/y0LwY...1920&fidelity=highThat is the behavior of an above average, kilowatt level modern PSU (a 1200 watts unit). One using an LLC resonant topology on its primary, with a PWM supporting efficient low load switching (burst mode). It is barely within spec with a low frequency load and a 40% step size from a low load transition... and that is with a capacitive load 1/3 lower than it is supposed to. Most modern LLC resonant based PSUs currently out there perform even worse than that, a few perform slightly better though. This does not include the OP's supernova G5 as it uses a different primary topology with even worse transient response capabilities.
post edited by Michapolys - 2022/07/27 18:26:42
|
badboy64
SSC Member
- Total Posts : 921
- Reward points : 0
- Joined: 2006/06/05 15:11:40
- Location: Fall River USA
- Status: offline
- Ribbons : 0
Re: Dead 3090 / Power Balance Issues
2022/07/27 16:57:51
(permalink)
Michapolys
They are essentially one issue... on the PSU...
No it is not the PSU that is the problem and as stated below this post of yours. You see the correct answer. Earlier on when the 3090 came out their where games that actually killed the card. Like the New World game from Amazon which was a big issue with these cards.
14th Intel® Core™ i9 14900KF CPU 3.2GHz@6.0ghz, Memory 2x24GB GSkill Trident Z Trident Z5 7200 mhz DDR5 Ram,4,000 GB MSI M480 PRO 4TB , Motherboard eVga 690 Dark , Operating System Windows 11 Pro 64-Bit, Msi Suprim X24G 4090, Monitor Acer CG437K, Logitech G910, Razer Lancehead Tournament Edition, Thermaltake View 91 RGB plus, eVga 1600w P2 PSU, Custom watercooling.   Speed Way Score 11,055 points. https://www.3dmark.com/sw/1112818
|
ty_ger07
Insert Custom Title Here
- Total Posts : 21171
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 270
Re: Dead 3090 / Power Balance Issues
2022/07/27 17:20:38
(permalink)
It's an interesting theory, but what about evidence?
The card has built-in undervoltage protection. Why is that not tripping? Why is the PSU's undervoltage protection or overcurrent protection not triggering? Refer to my earlier posts.
Power is proportional to voltage, not inversely proportional. How can it consume more power when the voltage is dropping? Where is the source of that power? Repeat: how does the GPU's buck converter circuitry continue to function in such a situation? How much current could actually be drawn from the PSU, if the PSU's output is momentarily exhausted?
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
My EVGA Score: 1546 • Zero Associates Points • I don't shill
|
UndesiredSanity
New Member
- Total Posts : 74
- Reward points : 0
- Joined: 2020/11/19 18:41:33
- Status: offline
- Ribbons : 1
Re: Dead 3090 / Power Balance Issues
2022/07/27 18:03:31
(permalink)
ty_ger07
It's an interesting theory, but what about evidence?
The card has built-in undervoltage protection. Why is that not tripping? Why is the PSU's undervoltage protection or overcurrent protection not triggering? Refer to my earlier posts.
Power is proportional to voltage, not inversely proportional. How can it consume more power when the voltage is dropping? Where is the source of that power? Repeat: how does the GPU's buck converter circuitry continue to function in such a situation? How much current could actually be drawn from the PSU, if the PSU's output is momentarily exhausted?
You are half right. Power is only proportional to Voltage if Current or Resistance is constant. The PSU supply current and the GPU load is not constant or fixed.
|
Michapolys
Superclocked Member
- Total Posts : 199
- Reward points : 0
- Joined: 2021/12/25 09:26:18
- Status: offline
- Ribbons : 2
Re: Dead 3090 / Power Balance Issues
2022/07/27 18:25:21
(permalink)
OK, guess I'll entertain you. ty_ger07
It's an interesting theory, but what about evidence?
I posted a pic. Take ot or leave it... it seems that you left it though. I would take it. :/ ty_ger07
The card has built-in undervoltage protection. Why is that not tripping?
Because it is just for the VRM output voltage. Which is fine... as the input voltage goes haywire... Which means that the post I made previously, with the 5 us statement, actually refers to just the VRM output voltage. My bad. I (falsely) assumed that you knew what you were talking about when stating that the card had under-voltage protection on the VRM input. ty_ger07
Why is the PSU's undervoltage protection or overcurrent protection not triggering?
Cause the trigger is not instant. It takes some time for the supervisor IC to detect what is going on. ty_ger07
Power is proportional to voltage, not inversely proportional. How can it consume more power when the voltage is dropping?
By the electrical current increasing adequately. ty_ger07
Where is the source of that power?
For the load, assuming you use a switching mode power supply to power it, it is the decoupling capacitors connected to its circuit, between that and the SMPS. The inductors in-between also hold some charge so those too I guess. If the charge on the aforementioned stuff is gone then you are left with the fets/diodes on the secondary providing directly to the load. If the load needs really loose regulation, it can actually work. Modern electronics don't. They fry unless a protection kicks in in time. ty_ger07
how does the GPU's buck converter circuitry continue to function in such a situation?
Refer to the previous statement. ty_ger07
How much current could actually be drawn from the PSU, if the PSU's output is momentarily exhausted?
As much as the card needed until either something gets fried or the card shuts down. Depends on whether the PSU's input is also exhausted, amongst other factors.
|
ty_ger07
Insert Custom Title Here
- Total Posts : 21171
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 270
Re: Dead 3090 / Power Balance Issues
2022/07/27 19:06:46
(permalink)
The image you provided says nothing. There is no context. Give some context. Tell us your test setup. Show us some pictures. By proof, I meant, show us a GPU drawing excessive current while the PSU's voltage droops. Everyone knows that a PSU has output ripple, and we agree that PSUs don't respond instantaneous to load, but where is the proof about it killing these cards? Why is it just certain video cards dying? Why aren't CPUs or motherboards dying from causing transients or during transients? Why isn't anything else dying from causing a transient or during a transient? Why is the rest of the system completely unaffected? PSUs can only have voltage droop when GPUs are the load, and only GPUs cause and die from these issues? It's an interesting theory, but there is no evidence to say that any of these GPUs died because a PSU's voltage dipped the perfect amount. To the contrary, we know that the power transient is caused by the GPU, we know why it is caused by the GPU, and we know that specific GPUs suffer deaths from it while others don't, and we know that those specific GPUs die regardless of all the different PSUs people were using while they died.
By the electrical current increasing adequately.
Come on man! You know that there is no current source! According to you, the PSU's output voltage dropped due to slow response, which means its LC reserves have been consumed. If the output voltage is dropping, there is no source of excess current to supplement the GPU's demands. It can't be both ways at the same time. If there is a high current source, the PSU is not drooping as drastic as you claim. All you can blame is the GPU for not having adequate power regulation and drawing more power than it should.
post edited by ty_ger07 - 2022/07/27 19:21:50
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
My EVGA Score: 1546 • Zero Associates Points • I don't shill
|
Michapolys
Superclocked Member
- Total Posts : 199
- Reward points : 0
- Joined: 2021/12/25 09:26:18
- Status: offline
- Ribbons : 2
Re: Dead 3090 / Power Balance Issues
2022/07/27 19:20:01
(permalink)
Look. You have been telling me that the GPU in question has under-voltage protection on the VRM input. The associated datasheets say otherwise. So I am not keeping this up with you anymore.
I am wrong, you are right. Whatever. I do not want to have another conversation with you in the future. For anything. Like, ever.
|
NobleHonor7
New Member
- Total Posts : 13
- Reward points : 0
- Joined: 2022/03/21 10:20:41
- Status: offline
- Ribbons : 0
Re: Dead 3090 / Power Balance Issues
2022/07/27 19:36:18
(permalink)
Well, what psu would you say is good enough? I’ll purchase it, and test it out with the Rev 1.0 psu. Anything I can try to get it to not blow a fuse or **** up a power stage.
|
Michapolys
Superclocked Member
- Total Posts : 199
- Reward points : 0
- Joined: 2021/12/25 09:26:18
- Status: offline
- Ribbons : 2
Re: Dead 3090 / Power Balance Issues
2022/07/27 20:28:11
(permalink)
NobleHonor7
Well, what psu would you say is good enough? I’ll purchase it, and test it out with the Rev 1.0 psu. Anything I can try to get it to not blow a fuse or **** up a power stage.
Well, the good news are that the PWM on Rev 1.0 cards seem to have Over Current Limiting (OCL) protection, which is not something you see every day. It looks like Onsemi went the extra mile and put that there to protect against catastrophic scenarios, like the one affecting these cards. This seems to be what saves these cards when Rev 0.1 cards would fry, as the UPi PWM on those lacked such a protection, as most PWMs do. Personally, I would not risk it and use a different PSU. Especially if using a G5 like you do. The topology on the primary of that PSU is not meant for such working conditions. I do not know why EVGA went with that FSP platform, apart from cutting manufacturing costs. As for a replacement with the same wattage, if I were you I would ask EVGA for an RMA replacement with a different PSU line, like the newer G6 platform that seems to be able to handle such loads fine, despite being from Seasonic (most of Seasonic's other platforms also have issues, since the Radeon Vega days). The newer Corsair RMX line seems to be fine, so are the G2/P2 and P6 lines from EVGA, the PTS line and Ti rev 2 lines from Silverstone too, among others. From what I saw, those still have relatively low capacitance on the output capacitors so I do not know how they will keep up with newer 12VHPWR GPUs, and apart from the Silverstone units that use mostly aluminum polymer capacitors on the output, the rest have most of the output capacitance on wet aluminum capacitors that will die sooner under the circumstances. There are about three modern LLC resonant based PSUs I have seen with relatively large output capacitance with the same wattage as your current PSU, but at least two of those do not use japanese capacitors. The MEG Ai1000P from MSI which might use japanese caps or not, but I have not seen its internals, the Gigabyte UD1000GM PG5 and the Pure Power 11 FM 1000w. Personally I would not buy a new PSU at all and ask EVGA for a PSU replacement under warranty, with one compatible PSU like the G6, P6, G2 and P2. Preferably the newer G6/P6. This is not your problem. Both the PSU and the GPU you use are by them, so EVGA should be the ones who solve the issue with their components.
|
ty_ger07
Insert Custom Title Here
- Total Posts : 21171
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 270
Re: Dead 3090 / Power Balance Issues
2022/07/27 20:42:50
(permalink)
Michapolys
Well, the good news are that the PWM on Rev 1.0 cards seem to have Over Current Limiting (OCL) protection, which is not something you see every day. It looks like Onsemi went the extra mile and put that there to protect against catastrophic scenarios, like the one affecting these cards. This seems to be what saves these cards when Rev 0.1 cards would fry, as the UPi PWM on those lacked such a protection, as most PWMs do.
Except It is set so high that it is effectively disabled, no? The reason the cards aren't black-screening during power transients due to OCP triggering is because they purposely set the OCP way too high. They knew that the transients are ridiculous and didn't want all the calls for constant black-screens.
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
My EVGA Score: 1546 • Zero Associates Points • I don't shill
|
ty_ger07
Insert Custom Title Here
- Total Posts : 21171
- Reward points : 0
- Joined: 2008/04/10 23:48:15
- Location: traveler
- Status: offline
- Ribbons : 270
Re: Dead 3090 / Power Balance Issues
2022/07/27 21:34:29
(permalink)
Michapolys
There are about three modern LLC resonant based PSUs I have seen with relatively large output capacitance with the same wattage as your current PSU, but at least two of those do not use japanese capacitors.
I don't think that bulk output capacitance is a magic bullet. Too much capacitance is undesirable for transient response and efficiency. If too little capacitance is bad for transient response, too much is too. Why not just throw a super capacitor on that output? Because it isn't beneficial. Otherwise they would, and market it and charge for it.
ASRock Z77 • Intel Core i7 3770K • EVGA GTX 1080 • Samsung 850 Pro • Seasonic PRIME 600W Titanium
My EVGA Score: 1546 • Zero Associates Points • I don't shill
|
squall-leonhart
CLASSIFIED Member
- Total Posts : 2904
- Reward points : 0
- Joined: 2009/07/27 19:57:03
- Location: Australia
- Status: offline
- Ribbons : 24
Re: Dead 3090 / Power Balance Issues
2022/08/16 03:14:11
(permalink)
NobleHonor7
I saw that the original issue with a weak solder joint on a mosfet was taken care of, but I’ve also seen issues with 3080ti having this issue with halo mcc, and I’ve seen 3090s drawing 135% power limit, or the cards pulling 75 - 85% GPU power while the limit is set to 50%. It’s just a terrifying thought. I had seen where other brands gpus stayed much closer to the power limit.
I hope the issues have been fixed because I would prefer to stick with EVGA. I have a 2080, 3080 ti and 3090 from EVGA. I just don’t want to deal with continuous RMA issues, and the constant fear/possibility of a failure dealing permanent damage to other components.
There was never a solder issue, EVGA lied about this publically, all Rev 0.1 parts will fail eventually because the up9511+AOZ combination ends up super heating under rapid load changes. NobleHonor7
So your telling me Halo MCC won’t blow up a Rev 1.0 card?
1.0 parts use Onsemi+Onsemi and react to load variations like a champ.
CPU:Intel Xeon x5690 @ 4.2Ghz, Mainboard:Asus Rampage III Extreme, Memory:48GB Corsair Vengeance LP 1600
Video:EVGA Geforce GTX 1080 Founders Edition, NVidia Geforce GTX 1060 Founders Edition
Monitor:BenQ G2400WD, BenQ BL2211, Sound:Creative XFI Titanium Fatal1ty Pro
SDD:Crucial MX300 275, Crucial MX300 525, Crucial MX300 1000
HDD:500GB Spinpoint F3, 1TB WD Black, 2TB WD Red, 1TB WD Black
Case:NZXT Phantom 820, PSU:Seasonic X-850, OS:Windows 7 SP1
Cooler: ThermalRight Silver Arrow IB-E Extreme
|
UndesiredSanity
New Member
- Total Posts : 74
- Reward points : 0
- Joined: 2020/11/19 18:41:33
- Status: offline
- Ribbons : 1
Re: Dead 3090 / Power Balance Issues
2022/08/17 04:49:35
(permalink)
2nd try... I have one of these OG Red Lip v0.1 3090's. It does exhibit the load balance issue and drawing 85W+ on the PCIe slot. I have 437 days left on warranty, bought from Micro Center Nov 2020. I don't know what I should do with this card. Should I RMA it and try to get a v1.0 but that would leave me with the 1 year warranty. Is that XOC exchange program still valid and ongoing and should I try to exchange it for a v1.0 or refurbished v0.1? Should I push it in gaming and try to blow it up? Kind of sucks knowing I have one of these time bomb short fused OG 3090's and feel it would be my luck it would crap out after the warranty is done.
post edited by UndesiredSanity - 2022/08/17 04:52:10
|
rjohnson11
EVGA Forum Moderator
- Total Posts : 102262
- Reward points : 0
- Joined: 2004/10/05 12:44:35
- Location: Netherlands
- Status: offline
- Ribbons : 84
Re: Dead 3090 / Power Balance Issues
2022/08/17 06:01:27
(permalink)
Michapolys
yaymz
The capacitors on a maxed out PSU aren't big enough to handle those load spikes coming from the GPU. Not the PSU's fault if the power demand goes way over the wattage limit of the PSU, even for a split millisecond.
That is the problem. The power demand is within the recommended limit as defined by the ATX spec. It is just that most PSUs currently out there rely on a loophole of the spec to reduce their transient load capabilities.
...which should corrected by ATX 3.0 complaint PSUs.
|
Michapolys
Superclocked Member
- Total Posts : 199
- Reward points : 0
- Joined: 2021/12/25 09:26:18
- Status: offline
- Ribbons : 2
Re: Dead 3090 / Power Balance Issues
2022/08/17 06:33:00
(permalink)
rjohnson11
Michapolys
yaymz
The capacitors on a maxed out PSU aren't big enough to handle those load spikes coming from the GPU. Not the PSU's fault if the power demand goes way over the wattage limit of the PSU, even for a split millisecond.
That is the problem. The power demand is within the recommended limit as defined by the ATX spec. It is just that most PSUs currently out there rely on a loophole of the spec to reduce their transient load capabilities.
...which should corrected by ATX 3.0 complaint PSUs.
Yeah, ATX 3.0 got rid of the loophole. They were really direct on the definition with no room for misinterpretation.
|