Quoted

Initially, only the version shown will be available, either with the latch facing the PCB or away from it

Tell me please

It is planned to purchase the asus rog astral 5090 liquid edition

Standard cables from the seasonic power supply unit with direct connectors will be used for connection.

Which version of your product is suitable for me?

Thanks for the answer!

Will you be shipping this outside of Germany ? I am pretty scared of using my 5090 lately with the amount of melted cables i have seen on social media lately, so I really hope I can get my hands on this asap so I can go back to enjoying gaming

As I understand it, some sort of extension cable would be required for this to work with a 5090 FE? If so, is there any specific type of cable you would recommend?

On the point of using an extension cable, it was said that: "...you have to be aware that
additional contact transitions can have a further negative impact on the
current flow." What does this mean / what are the implications of this exactly for a lay person? Thanks!

Just wanted to drop in and show my support for this product! very excited to finally have a proper solution compared to the other solutions before from other brands.
Also wanted to share my interest in the white model!

Do you know if this will be compatible with the MSI 5090 Suprim (the air cooled one), and which orientation I would need?

Quoted from "vyoh"

Do you know if this will be compatible with the MSI 5090 Suprim (the air cooled one), and which orientation I would need?

have the same question.

As November is in a couple of days, and this was indicated to ship mid-Nov, when can we expect to have it available for pre-order?

Quoted from "jwingy"

As I understand it, some sort of extension cable would be required for this to work with a 5090 FE? If so, is there any specific type of cable you would recommend?

On the point of using an extension cable, it was said that: "...you have to be aware that
additional contact transitions can have a further negative impact on the
current flow." What does this mean / what are the implications of this exactly for a lay person? Thanks!

For every interconnect, there's some loss/added resistance to the whole chain. The extension is required due to how the socket exists on the FE. Any appropriate cable should work. it's literally just an extension cable. Though I'd not use one.


YMMV.

I'm very interested in that device.
But most of the time, that kind of thing isn't designed for compatibility with riser in mind (which is too bad as riser avoid to put too much weight on motherboard port).
For exemple, I use the Cooler Master Vertical Graphics Card Holder Kit V3. Like most of riser holder, there is a vertical plate behind the graphic card.
On this model, between the verge of the 12VHPWR plug of the graphic card and the edge of the vertical plate of the holder, there is a thickness of nearly 2,5 cm.
it means there should be 2,5cm between the 2 connectors of the AMPINEL to make the device compatible (for exemple, it's impossible to use the WireView Pro 2).

Do you think it's possible with the AMPINEL?

Also i'm very interested in the white version to match the color of my rig, or even a silver one.

Your competition just unveiled their new product and in the video they say, that they decided against active load balancing because they dont know what is going to happen if you push higher current through pin that has higher resistance. What do you think about that?

Quoted from "Kazdi"

Your competition just unveiled their new product and in the video they say, that they decided against active load balancing because they dont know what is going to happen if you push higher current through pin that has higher resistance. What do you think about that?

What is going to happen if you push more current through the conductor that has a higher resistance is not a mystery. More current through a higher resistance means more voltage drop across that resistance, which results in more power being dissipated as heat. I don’t think the Ampinel works that way.

If one circuit path has a higher resistance, less current will flow through it. The Ampinel does not try to balance the current flowing through the 6 conductors by forcing more current through the high resistance path. Doing that would cause more heat to be generated, which further increases the resistance of the contact. A feedback loop is formed that results in a melted connector. My understanding is that the Ampinel adjusts the voltage drop across the total resistance in each circuit path. If 1 conductor has a higher resistance due to poor contact alignment, oxidation, or whatever reason, less current will flow through that conductor, not more, because V = I2R, so if the resistance is higher, the current must be reduced to keep the voltage drop the same. If the resistance is so high that the Ampinel cannot compensate for it, the Protection mechanisms are activated. That is how I think it works after reading Igor's Lab articles about the Ampinel. If I am wrong, perhaps someone from Aquacomputer will jump in here and clarify.

I am curious about this other new product. Can you provide a link to the video you mentioned?

Quoted from "Speedy-VI"

Quoted from "Kazdi"

Your competition just unveiled their new product and in the video they say, that they decided against active load balancing because they dont know what is going to happen if you push higher current through pin that has higher resistance. What do you think about that?

What is going to happen if you push more current through the conductor that has a higher resistance is not a mystery. More current through a higher resistance means more voltage drop across that resistance, which results in more power being dissipated as heat. I don’t think the Ampinel works that way.

If one circuit path has a higher resistance, less current will flow through it. The Ampinel does not try to balance the current flowing through the 6 conductors by forcing more current through the high resistance path. Doing that would cause more heat to be generated, which further increases the resistance of the contact. A feedback loop is formed that results in a melted connector. My understanding is that the Ampinel adjusts the voltage drop across the total resistance in each circuit path. If 1 conductor has a higher resistance due to poor contact alignment, oxidation, or whatever reason, less current will flow through that conductor, not more, because V = I2R, so if the resistance is higher, the current must be reduced to keep the voltage drop the same. If the resistance is so high that the Ampinel cannot compensate for it, the Protection mechanisms are activated. That is how I think it works after reading Igor's Lab articles about the Ampinel. If I am wrong, perhaps someone from Aquacomputer will jump in here and clarify.

I am curious about this other new product. Can you provide a link to the video you mentioned?

https://youtu.be/IY5Ak33rPg8?t=1096

Quoted from "Kazdi"

https://youtu.be/IY5Ak33rPg8?t=1096

Oh OK you were referring to a product, that I am already aware of. I thought there was another player entering this market. He expresses concern that the Ampinel will push more current through a conductor/contact that already has a problematic resistance. According to Igor’s Lab, that is not what the Ampinel does. Igor says the Ampinel does not attempt to push the same amount of current through all 6 conductors. Instead, it attempts to make the voltage drop across the effective resistance of all 6 circuits be the same. He says it does this by slightly increasing the resistance of the other 5 current paths. This is the part I don't understand. If the Ampinel increases the resistance of the other 5 current paths, and the GPU power demand remains the same, then the amount of current that has to flow to deliver that power stays the same. If the resistance of the other 5 current paths is increased, then more current has to flow through the bad path. I am hoping someone from Aquacomputer, or another forum member can clarify this point for me.

Here is a quote from his article about this. I also provided the link so you can read it yourself if you like.

“Update: the load balancer does not arbitrarily change the current on a line but seeks the electrical operating point at which the voltage drops across all lines are as equal as possible. If a pin carries less current due to higher contact resistance, the system is initially unstable because the other five paths take on more load. If nothing were done, the difference would grow ever larger, and that is exactly what is observed in unbalanced connectors. In this situation the balancer does not try to actively load the weak pin more but reduces the dominance of the overloaded paths by slightly raising the effective resistance there. The total current remains almost constant, the source, that is the power supply or the GPU load request, determines it, but the voltage distribution at the connector becomes equalized. This electrically reintegrates the “bad” contact without necessarily requiring it to carry its full one sixth load. The system does not aim for equal current shares but for minimal voltage deviations, in other words a stable state of equal distribution.”

Igor’s Lab reviewed the Ampinel a few weeks ago. That review is HERE.

There was a lot of confusion about how the device works, so he did a follow-up article that goes through the whole 12VHWPR/12V2X6 connector saga, what Nvidia did (or more accurately did not do), what Asus did, what der8auer did, and what the Ampinel does. He compares the strategies of just monitoring the current versus actively controlling it. In part 4 of the follow-up article, he addresses the concern that is expressed in that video. Maybe the author of that video still does not understand how the Ampinel works, or maybe he does...

Here is a LINK to part 4 of Igor's Lab follow-up article.

Quoted from "Speedy-VI"

Oh OK you were referring to a product, that I am already aware of. I thought there was another player entering this market. He expresses concern that the Ampinel will push more current through a conductor/contact that already has a problematic resistance. According to Igor’s Lab, that is not what the Ampinel does. Igor says the Ampinel does not attempt to push the same amount of current through all 6 conductors. Instead, it attempts to make the voltage drop across the effective resistance of all 6 circuits be the same. He says it does this by slightly increasing the resistance of the other 5 current paths. This is the part I don't understand. If the Ampinel increases the resistance of the other 5 current paths, and the GPU power demand remains the same, then the amount of current that has to flow to deliver that power stays the same. If the resistance of the other 5 current paths is increased, then more current has to flow through the bad path. I am hoping someone from Aquacomputer, or another forum member can clarify this point for me.

Here is a quote from his article about this. I also provided the link so you can read it yourself if you like.

“Update: the load balancer does not arbitrarily change the current on a line but seeks the electrical operating point at which the voltage drops across all lines are as equal as possible. If a pin carries less current due to higher contact resistance, the system is initially unstable because the other five paths take on more load. If nothing were done, the difference would grow ever larger, and that is exactly what is observed in unbalanced connectors. In this situation the balancer does not try to actively load the weak pin more but reduces the dominance of the overloaded paths by slightly raising the effective resistance there. The total current remains almost constant, the source, that is the power supply or the GPU load request, determines it, but the voltage distribution at the connector becomes equalized. This electrically reintegrates the “bad” contact without necessarily requiring it to carry its full one sixth load. The system does not aim for equal current shares but for minimal voltage deviations, in other words a stable state of equal distribution.”

Igor’s Lab reviewed the Ampinel a few weeks ago. That review is HERE.

There was a lot of confusion about how the device works, so he did a follow-up article that goes through the whole 12VHWPR/12V2X6 connector saga, what Nvidia did (or more accurately did not do), what Asus did, what der8auer did, and what the Ampinel does. He compares the strategies of just monitoring the current versus actively controlling it. In part 4 of the follow-up article, he addresses the concern that is expressed in that video. Maybe the author of that video still does not understand how the Ampinel works, or maybe he does...

Here is a LINK to part 4 of Igor's Lab follow-up article.

Thx for detailed response. I dont understand everything, but i read all what you said and the igors article and i believe all is good. Active load balancing seems like the way to go. Iam already on the list to get notified when ampinel is up for preorder. I just dont want my 5090 to burn )