Lian Li SP850 First SFX PSU with ATX 3.0 + PCIe Gen 5, Including ATX12HPWR

[btarunr]

Lian Li unveiled the SP850, the first SFX form-factor power supply featuring ATX 3.0 specs readiness, and PCIe Gen 5 compliance, including a 12+4 pin ATX12HPWR connector for next-generation graphics cards. Available in black and white body-color variants, the it offers 850 W continuous output, including tolerance for +100% excursions for at least 100 µs intervals. and meets 80 Plus Gold efficiency. It features a single +12 V rail design, with most common electrical protections. The 92 mm fluid dynamic bearing fan keeping it cool offers fanless (zero RPM) cooling up to 340 W load (excluding excursions). The Lian Li SP850 offers full modular cabling. Besides the 12+4 pin ATXHPWR, you get two 8-pin EPS, and three 6+2 pin PCIe power connectors. Peripheral connectivity includes eight SATA power connectors. Lian Li is backing the SP850 with a 5-year warranty. Available now, it is priced at USD $150-170.



View at TechPowerUp Main Site

 

[wolf]

including tolerance for +100% excursions for at least 100 µs intervals

If I'm interpreting this correctly, that's mitigation for transient power spikes? and again if I'm interpreting correctly, that means ~1700w spikes for at least 100 µs are accounted for?

If so, I'd be interested to see that metric rated against popular existing models of PSU. I ran my 5900X/3080 on a Corsair SF600 Gold for well over a year, never saw the whole PC pull more than 500w from the wall, but I assume there were transient spikes over the max 600w rating that the PSU was able to deal with adequately.

 

[Valantar]

If I'm interpreting this correctly, that's mitigation for transient power spikes? and again if I'm interpreting correctly, that means ~1700w spikes for at least 100 µs are accounted for?

If so, I'd be interested to see that metric rated against popular existing models of PSU. I ran my 5900X/3080 on a Corsair SF600 Gold for well over a year, never saw the whole PC pull more than 500w from the wall, but I assume there were transient spikes over the max 600w rating that the PSU was able to deal with adequately.

This is a part of the ATX 3.0 spec. That obviously doesn't mean that no previous PSU will be able to also handle similar excursions, they just don't follow any type of standard or have any real rating to show for it. Variability in this has been on show in a lot of recent gaming builds, with some 750W and 850W PSUs shutting down intermittently when powering even an RTX 3080, while I've seen others run a 3090 on a 600W unit - and yes, the Corsair SF series is quite good in this regard. ATX 3.0 is trying to remove this insecurity.

 

[Blaylock]

What's odd to me is Lian Li shared images in its press release of a power supply without the 12+4 pin ATX12HPWR connector despite claiming that it is included.

 

[QinX]

What's odd to me is Lian Li shared images in its press release of a power supply without the 12+4 pin ATX12HPWR connector despite claiming that it is included.

It's for sale on the caseking website and it says:
2x PCIe 6+2-pin cable to 12+4-Pin 12 VHPWR cable (400 mm)
So no dedicated connector, just a conversion.

 

[WonkoTheSaneUK]

It's for sale on the caseking website and it says:
2x PCIe 6+2-pin cable to 12+4-Pin 12 VHPWR cable (400 mm)
So no dedicated connector, just a conversion.

So, it's like the one I bought from Corsair to power my 3080FE without using Nvidia's stupidly short splitter.

 

[wolf]

This is a part of the ATX 3.0 spec.

Cheers for the reply very helpful!

 

[Woomack]

It's for sale on the caseking website and it says:
2x PCIe 6+2-pin cable to 12+4-Pin 12 VHPWR cable (400 mm)
So no dedicated connector, just a conversion.

It's a dedicated cable from PSU = 2x8-pin connector to 12-pin PCIe. There is no 12+4-pin cable in the box. It's a mistake in an early version of the specs. The product page also has some mistakes that will be corrected soon. So again, in the package is a 2x8-pin to 12-pin cable rated at up to 400W.
I actually have a review sample (retail box) of this PSU (white version).

 

[illusion archives]

So it's the first PSU that is ATX3.0-compatible but not with a native 12VHPWR cable?
And it seems that it passes the test for those with native 12VHPWR cables?

For those ATX3.0-compatible PSUs without a native 12VHPWR cable but with a connector, the excursion request is formly 150% or 200%?

---

This is a part of the ATX 3.0 spec. That obviously doesn't mean that no previous PSU will be able to also handle similar excursions, they just don't follow any type of standard or have any real rating to show for it. Variability in this has been on show in a lot of recent gaming builds, with some 750W and 850W PSUs shutting down intermittently when powering even an RTX 3080, while I've seen others run a 3090 on a 600W unit - and yes, the Corsair SF series is quite good in this regard. ATX 3.0 is trying to remove this insecurity.

Maybe in ATX 2.X there is a request for about +30% excursion.
Tt uses this in their PR for GF3: https://www.thermaltake.com/toughpower-gf3-1650w-gold-tt-premium-edition.html

But I cann't get the request for the interval.

 

[Valantar]

So it's the first PSU that is ATX3.0-compatible but not with a native 12VHPWR cable?
And it seems that it passes the test for those with native 12VHPWR cables?

For those ATX3.0-compatible PSUs without a native 12VHPWR cable but with a connector, the excursion request is formly 150% or 200%?

---

Maybe in ATX 2.X there is a request for about +30% excursion.
Tt uses this in their PR for GF3: https://www.thermaltake.com/toughpower-gf3-1650w-gold-tt-premium-edition.html

View attachment 261183

But I cann't get the request for the interval.

1.3x sounds like it would be pretty close to fitting within OCP/OPP territory for many PSUs though, so that isn't a particularly big ask.

 

[Flydommo]

Hopefully, my Corsair SF750 will be sufficient to provide power for an RTX 4080 plus Ryzen 5700X and x570 motherboard...

 

[Valantar]

Hopefully, my Corsair SF750 will be sufficient to provide power for an RTX 4080 plus Ryzen 5700X and x570 motherboard...

That depends entirely on the power target, but hopefully, yes. At the very least it shouldn't require much of an UV/UC to get there. The SF series seems to have very good transient handling (I've seen several people run RTX 3090s on SF600s), and the rest of your system isn't exactly power hungry. Though of course predicting this is extremely difficult as transients not only depend on the GPU power target, but the specific GPU AIC design and BIOS (spikes differ a lot between vendors for seemingly identically specced GPUs), and also for some obscure reason they vary a lot depending on the motherboard of the system they're installed in. Still, given how good of an effect a slight undervolt and underclock tends to have on transient power spikes, I wouldn't worry too much.

 

[TheGigaCat]

My only, ONLY issue with Lian Li's SFX PSUs is that they're very audible when the fan kicks in. Hope they dealt with that.

 

[jonnyGURU]

Supports 1700W power excursions? I'll believe it when Intel or Aris confirms it.

 

[delshay]

Supports 1700W power excursions? I'll believe it when Intel or Aris confirms it.

Are you the person that used to review PSU? As I have a Question.

 

[jonnyGURU]

Are you the person that used to review PSU? As I have a Question.

We all have questions. What am I doing with my life? Where was God when I needed him most? Why did TOGO's discontinue their cuban sandwich.....

 

[Valantar]

Supports 1700W power excursions? I'll believe it when Intel or Aris confirms it.

Cue teardown shots showing half the internal volume of the PSU being output capacitors?

 

[Nanochip]

So given that this power supply doesn't have a physical ATX12HPWR slot (and seems to be using converter cables), I have two questions.

• 1) Are there any potential overcurrent issues with using the adapter cable (I read yesterday that the PCIe-SIG is warning of potential for issues with some PSUs/cables that convert 8-pin PCIe power cables to a ATX12HPWR cable)?
• 2) Does this PSU comply with the same power excursion requirements of ATX3.0 PSUs that come with the ATX12HPWR port? If not, then how does this PSU's power excursion capability compare to the more stringent requirements of ATX3.0 PSUs that come with the ATX12HPWR port?

 

[jonnyGURU]

So given that this power supply doesn't have a physical ATX12HPWR slot (and seems to be using converter cables), I have two questions.

• 1) Are there any potential overcurrent issues with using the adapter cable (I read yesterday that the PCIe-SIG is warning of potential for issues with some PSUs/cables that convert 8-pin PCIe power cables to a ATX12HPWR cable)?
• 2) Does this PSU comply with the same power excursion requirements of ATX3.0 PSUs that come with the ATX12HPWR port? If not, then how does this PSU's power excursion capability compare to the more stringent requirements of ATX3.0 PSUs that come with the ATX12HPWR port?

1. No. The wccftech piece was complete crap. PCI-SIG never said not to use adapters. Read the actual letter in the article. wccftech extrapolated from the letter that they shouldn't be used, but whoever leaked the letter to them did not leak the attachment that explained the "situation" further.

2. We won't know until it's tested, but I have my doubts. ATX 3.0 currently "asks" for a PSU to do 2x it's rated capability in a power excursion event; which is very unrealistic. Especially when you get into higher wattage and/or smaller footprint PSUs.

 

[Bomby569]

This is a part of the ATX 3.0 spec. That obviously doesn't mean that no previous PSU will be able to also handle similar excursions, they just don't follow any type of standard or have any real rating to show for it.

am i wrong if i say ATX 3.0 is a standard but not a rating, so it might do what they say or not, you still have no idea besides their word?

 

[jonnyGURU]

am i wrong if i say ATX 3.0 is a standard but not a rating, so it might do what they say or not, you still have no idea besides their word?

Yeah. You can "claim" ATX 3.0 and not get in trouble for saying so incorrectly because it's not a rating. It's a standard.

Intel does have a testing service. AND IT'S FREE!!!!! But hardly anyone sends them their PSUs for testing... EVEN THOUGH IT'S FREE.... and despite this, they're so behind that their "PSU selector" doesn't even include any ATX 3.0 or ATX12VO 2.0 units!!!! (and I've sent them ATX 3.0 and ATX12VO 2.0 units, so I know they have some to test!!)

"Power Supply Selector" on this page are the units they've "validated": https://www.intel.com/content/www/us/en/collections/topics/power-supply-selector.html?s=Newest

 

[oxidized]

I've been out of this world for a while now, i'm just trying to catch up on some things. Can someone quickly explain what's all this "power excursion" about exaclty? How are current and future video cards going to draw 1,8KW randomly once in a while? Maybe i got it all wrong

 

[R-T-B]

I've been out of this world for a while now, i'm just trying to catch up on some things. Can someone quickly explain what's all this "power excursion" about exaclty? How are current and future video cards going to draw 1,8KW randomly once in a while? Maybe i got it all wrong

It's more for future video cards than current (though Ampere is a bit of an example of what to expect), and I picture it has quite the buffer built in, like most standards.

 

[Valantar]

I've been out of this world for a while now, i'm just trying to catch up on some things. Can someone quickly explain what's all this "power excursion" about exaclty? How are current and future video cards going to draw 1,8KW randomly once in a while? Maybe i got it all wrong

It's mainly about power spikes, also often called transient power draw, which are spikes of very high power draw (some times up to 2x the rated power of the GPU) that last for very short periods of time, typically in the microsecond range. It's a consequence of the massive size (in transistors/cores) of current GPUs, their high clocks, and their aggressive boost algorithms which try to clock as high as possible at all times. When that is the case and the workload changes from something light to something heavy (which happens often in games), there's a risk of a massive power spike before the boost regulation system has the time to lower clocks in response to the heavier workload.

This isn't really a massive concern, and only really applies to high end GPUs (as they are very large and have huge power budgets), but it can cause system crashes/shutdowns if the PSU isn't able to handle the spikes, either through OCP/OPP triggering or through voltage dropping too low for the system to stay on. And it varies a lot between PSUs - I've seen 850W units shut down when powering a 3080, but I've also seen 600W units powering 3090s with no issues. Hence why the ATX 3.0 spec is trying to standardize this somewhat.

 

[jonnyGURU]

2x is normal. 2.5x is Ampere. They say the new Lovelace is 3x.