Can I run an EPS 2x 8pin m/b on an EPS 1x 8pin PSU?

Jedibeeftrix

Weaksauce
Joined
Dec 1, 2016
Messages
126
I bought a 650W Seasonic Prime Ultra Platinum for £100 (a bargain).

Intending to use it on an MSI x570 Tomahawk m/b (when released).

Realised afterwards the 650W model only has one EPS 8pin CPU power connector, and the x570 Tomahawk has two EPS 8pin CPU power connectors...!

Is this a problem (given I intend to get a 3800x and have no intention of O/C'ing)?
 

Armenius

Fully [H]
Joined
Jan 28, 2014
Messages
21,339
The cable may run hotter than normal, but it should be fine if you're not overclocking. A single 8-pin is designed to handle up to 20A at +12V, or about 235W. In stock condition the 3800X uses up to 200W in multithreaded workloads, and that power is already going to be distributed by the ATX mainline in addition to the EPS. "Unofficially," a single 8-pin can handle up to 28A or 336W, I just wouldn't hit that level of power continuously.

Whatever you do, do not try supplementing with a PCI-E cable because the pinout is different.
 

Wooshoo

Weaksauce
Joined
May 27, 2019
Messages
92
The cable may run hotter than normal, but it should be fine if you're not overclocking. A single 8-pin is designed to handle up to 20A at +12V, or about 235W. In stock condition the 3800X uses up to 200W in multithreaded workloads, and that power is already going to be distributed by the ATX mainline in addition to the EPS. "Unofficially," a single 8-pin can handle up to 28A or 336W, I just wouldn't hit that level of power continuously.

Whatever you do, do not try supplementing with a PCI-E cable because the pinout is different.
I believe you can buy adapters as well to utilize other cables as needed but this may only be possible for gpu cables. Or for 8 to 4 pin cpu situations
 

Jedibeeftrix

Weaksauce
Joined
Dec 1, 2016
Messages
126
Separate questions:

1. If I were to get the 2x 8pin tomahawk - would i still benefit from the better power circuitry when only using it with 1x 8pin eps?
2. Back to the Elite - would i be able to use the 2 x USB 3.2 Gen 1 headers on a case which only has USB 3.0 5GB front sockets?
 

Jamie Marsala

Limp Gawd
Joined
Mar 9, 2016
Messages
227
As far as I can tell you do not need to actually plug both in. I saw somewhere that, I think it was MSI boards, that the two connectors were in parallel anyway. So unless you are drawing way more power than one plug can supply then there should be no issues.
 

sirmonkey1985

[H]ard|DCer of the Month - July 2010
Joined
Sep 13, 2008
Messages
21,835
I bought a 650W Seasonic Prime Ultra Platinum for £100 (a bargain).

Intending to use it on an MSI x570 Tomahawk m/b (when released).

Realised afterwards the 650W model only has one EPS 8pin CPU power connector, and the x570 Tomahawk has two EPS 8pin CPU power connectors...!

Is this a problem (given I intend to get a 3800x and have no intention of O/C'ing)?
second 8 pin EPS only matters if you're doing extreme overclocking. the single 8pin EPS can handle 384w which your cpu will never draw. nothing to worry about. there are some asus boards that have external pcie power draw pulled from the extra eps connector but only really matters if you're trying to run quad sli/cfx.
 

Jedibeeftrix

Weaksauce
Joined
Dec 1, 2016
Messages
126
thank you all,

delighted that i might still benefit from the better power circuitry on the x570 tomahawk even with only one 8pin eps.

now just this question: :)

2. Back to the Elite - would i be able to use the 2 x USB 3.2 Gen 1 headers on a case which only has USB 3.0 5GB front sockets?
 

sirmonkey1985

[H]ard|DCer of the Month - July 2010
Joined
Sep 13, 2008
Messages
21,835
now just this question: :)

2. Back to the Elite - would i be able to use the 2 x USB 3.2 Gen 1 headers on a case which only has USB 3.0 5GB front sockets?
check the manual since USB 3.2 Gen 1 is now technically the old USB 3.0 5Gbps so it depends on what the header is actually certified for.
 
Joined
Dec 1, 2004
Messages
1,008
would i still benefit from the better power circuitry when only using it with 1x 8pin eps?
Sigh. "better power circuitry"

*warning* opinion alert, opinion alert *warning*

Look, if you're not doing something extreme with a motherboard - and the premise of this thread is that you're not - then power delivery is very much a "you must be this tall to ride this ride" kind of thing. Either the system is going to deliver stable power to your CPU, or it isn't. There is no "extra stable" power, or "better" power - it's either going to be held within tolerances or it isn't. Just like being an extra 10 inches taller doesn't make a roller coaster any more fun than being just over the line to be able to hop on.

The entire lead-up to the Ryzen 3000 series launch and the absolutely stupid amount of attention that was focused on VRMs and power delivery for the 3000 series CPUs was a borderline scam to sell more motherboards. Once the 3000 series came out, it became pretty obvious pretty fast that core-for-core the 3000 series had *less* intensive power requirements than the 2000 series that came before it, and that as long as you weren't overclocking or chasing down the very last 3% of performance even an average 300 or 400 series motherboard could run the entire set of 3000 series CPUs with no problem. For fuck's sake you can run a 3950X on a whole suite of A320 chipset motherboards, let alone their beefier B and X series big brothers.

I'm not saying bigger boards with better power delivery don't have their niches; I'm just saying that running at stock is *not* that niche.
 
Last edited:

vick1000

[H]ard|Gawd
Joined
Sep 15, 2007
Messages
2,004
Sigh. "better power circuitry"

*warning* opinion alert, opinion alert *warning*

Look, if you're not doing something extreme with a motherboard - and the premise of this thread is that you're not - then power delivery is very much a "you must be this tall to ride this ride" kind of thing. Either the system is going to deliver stable power to your CPU, or it isn't. There is no "extra stable" power, or "better" power - it's either going to be held within tolerances or it isn't. Just like being an extra 10 inches taller doesn't make a roller coaster any more fun than being just over the line to be able to hop on.

The entire lead-up to the Ryzen 3000 series launch and the absolutely stupid amount of attention that was focused on VRMs and power delivery for the 3000 series CPUs was a borderline scam to sell more motherboards. Once the 3000 series came out, it became pretty obvious pretty fast that core-for-core the 3000 series had *less* intensive power requirements than the 2000 series that came before it, and that as long as you weren't overclocking or chasing down the very last 3% of performance even an average 300 or 400 series motherboard could run the entire set of 3000 series CPUs with no problem. For fuck's sake you can run a 3950X on a whole suite of A320 chipset motherboards, let alone their beefier B and X series big brothers.

I'm not saying bigger boards with better power delivery don't have their niches; I'm just saying that running at stock is *not* that niche.
VRM temps are a thing though, not just for system temps, but overall longevity of the mobo. Efficient power design is a great indicator of quality.
 

Jamie Marsala

Limp Gawd
Joined
Mar 9, 2016
Messages
227
thank you all,

delighted that i might still benefit from the better power circuitry on the x570 tomahawk even with only one 8pin eps.

now just this question: :)

2. Back to the Elite - would i be able to use the 2 x USB 3.2 Gen 1 headers on a case which only has USB 3.0 5GB front sockets?
Yes. I have mine connected just fine on my X570 Ace MB and Corsair case. Same with my previous two cases USB 3 Front Panel connectors and the MB's USB 3.2 front headers.
 
Joined
Dec 1, 2004
Messages
1,008
VRM temps are a thing though, not just for system temps, but overall longevity of the mobo
I'm not saying they aren't. What I'm saying is that existing motherboards from the 300 and 400 series were designed to run processors from the 1000 and 2000 series of Ryzen chips with a given power usage, and that unlike what seemed to be widely expected prior to release when the 3000 series chips came out they consumed roughly similar amounts of power (or less). That would mean the 3000 series puts approximately the same load on the power circuitry as their 1000 and 2000 series predecessors. This is all when talking about running at stock values, where processors from AMD restrict their actual power usage to be relatively close to the rated TDP of the CPU, even under load. That means a 300 or 400 series board would be equally reliable / have an equivalent lifespan under a 3000 series CPU as the 1000/2000 it was originally 'designed' for. My point is the near hysteria the AMD community reached regarding power circuitry and VRMs was overblown; not that it's completely unimportant, just that its importance was vastly overstated for a huge market segment (non overclockers).
 

pitingres

Weaksauce
Joined
Jul 25, 2018
Messages
101
VRM temps are a thing though, not just for system temps, but overall longevity of the mobo. Efficient power design is a great indicator of quality.
I'll give you the first point. I don't think the second point follows at all. "Quality" meaning what exactly? Efficient power delivery design doesn't ensure proper chipset heat-sinking or proper manufacture or anything else, except the efficiency of the VRM subsystem.
 
Joined
Dec 1, 2004
Messages
1,008
"Quality" meaning what exactly?
I can follow along his meaning, and even generally agree with it. If a motherboard "gets the details right", then you can have better confidence that overall things are done in a high quality fashion. I find it easier to think in the reverse order; if you notice glaring spelling errors all over a document, you start to suspect the quality of the information within.
 

pitingres

Weaksauce
Joined
Jul 25, 2018
Messages
101
I can follow along his meaning, and even generally agree with it. If a motherboard "gets the details right", then you can have better confidence that overall things are done in a high quality fashion. I find it easier to think in the reverse order; if you notice glaring spelling errors all over a document, you start to suspect the quality of the information within.
I guess my issue is calling it a "great indicator", I think that is too strong. Good design of one subsystem of a board gives one hope that equal care was applied across the board. Whether that's true or not depends on how many different engineers or teams were involved in the overall design. Your document example makes the same sort of implicit assumption that the same people responsible for spelling were responsible for the content, and that's not necessarily true. It's an indicator, but not necessarily a good one.
 

Jedibeeftrix

Weaksauce
Joined
Dec 1, 2016
Messages
126
Yes. I have mine connected just fine on my X570 Ace MB and Corsair case. Same with my previous two cases USB 3 Front Panel connectors and the MB's USB 3.2 front headers.
smashing, thank you.
also bought a second FT05 case for an absolute song.
my new pc is coming together in dribs and drabs whenever a bargain presents itself...
 
Last edited:

vick1000

[H]ard|Gawd
Joined
Sep 15, 2007
Messages
2,004
I'll give you the first point. I don't think the second point follows at all. "Quality" meaning what exactly? Efficient power delivery design doesn't ensure proper chipset heat-sinking or proper manufacture or anything else, except the efficiency of the VRM subsystem.
In my experience, power design and quality components used in that design, ARE an indicator of overall build quality. Maybe not firmware quality or other software factors, but absolutely for build quality of the hardware. This applies to other electronics as well.
 
Top