• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

NVIDIA 4000 Series

srekal34 said:
If you are saying that the way it is soldered in this particular adapter is better than crimping, you are the one with no clue.
And in molex connectors, soldering is typically not wanted in general - its screems amateur to say the least.
Click to expand...
No it doesn't it's much more robust than crimping and much better for higher loads. It provides a low resistance path.
Crimping adds another potential local resistance and a local hot spot. The guy has no idea.
 
Shaz12 said:
Nope. I am taking the card out until I get the CableMod cable next week. What’s the point of getting a new adapter when that will also melt anyway.
Click to expand...
Lol you sir are a bit odd, how is the manufacturer going to know the cables are faulty if they don't get returned, what makes you think the cablemod cable is any better.
Just raise an rma and see what they say.

I'm not sure you even have a card.
 
srekal34 said:
If you are saying that the way it is soldered in this particular adapter is better than crimping, you are the one with no clue.
And in molex connectors, soldering is typically not wanted in general - its screems amateur to say the least.
Click to expand...
Just to clarify moles connectors are cheap as chips and run with extremely low currents. No where near the power required for the 4090. Probably a thousand times less.
 
  • Like
Reactions: Rup
Slam62 said:
Lol you sir are a bit odd, how is the manufacturer going to know the cables are faulty if they don't get returned, what makes you think the cablemod cable is any better.
Just raise an rma and see what they say.

I'm not sure you even have a card.
Click to expand...
I do have the card. See how I have been using it so far.



Did you see what Igor and Buildzoid had to say about this adapter? It has questionable build quality which uses low quality solders making it prone to breaking. Contacting the AIB would achieve nothing because nvidia has advised them to keep silent for fear of legal liability.

There have been zero cases of melting reported with Cablemod and Corsair cables as they use crimped terminals instead of solder. The only solution to the 4090 burning is a right angle adapter and a native cable from Cablemod or Moddiy.
 
srekal34 said:
Molex Microfit is rated to 10.5A, same as Amphenol used in 4xxx series.
Click to expand...
I'd like to see you put 10.5a I to a hard drive.
A hard drive is what a couple of Watts, these connectors have to take at least half a kilowatt completely different beast.
Like comparing an electric bar fire to an led light bulb.

It's not surprising they melt if seated incorrectly or faulty.
 
Purgatory said:
But the best part of the joke is yet to come...4080 16GB the real 3080 replacement is starting at £1450-1500 AIB for probably 3090-3090ti performance.. Unless you can get the FE for £1270......

The 3080 went from £650 to now 4080 £1270...

Then they try sell a 4060ti-4070 for £1000.... :cry:
Click to expand...

Oh how the masterplan failed! :cry:
 
Slam62 said:
I'd like to see you put 10.5a I to a hard drive.
A hard drive is what a couple of Watts, these connectors have to take at least half a kilowatt completely different beast.
Like comparing an electric bar fire to an led light bulb.

It's not surprising they melt if seated incorrectly or faulty.
Click to expand...
You really have no clue if you think Molex Microfit is for harddrives :D
Molex Minifit JR for example is the one used for CPU EPS - 4x12v wires rated for 366W in total - and even there is grossly underspecced.
Please show me any industrial example where Molex/Amphenol connector is soldered instead of crimped. Certain connectors from Molex are used in cars for up to 50A current and even then its not soldered.
 
Last edited:
srekal34 said:
You really have no clue if you think Molex Microfit is for harddrives :D
Molex Minifit JR for example is the one used for CPU EPS - 4x12v wires rated for 366W in total - and even there is grossly underspecced.
Click to expand...
Very good, but solder is better than crimping, fewer losses and more expensive.
The connectors you refer to are not rated as high as you say for normal uses.
'8 pin EPS12V CPU power receptacle can supply 235 continuous watts'
 
Slam62 said:
Very good, but solder is better than crimping, fewer losses and more expensive.
The connectors you refer to are not rated as high as you say for normal uses.
'8 pin EPS12V CPU power receptacle can supply 235 continuous watts'
Click to expand...
I have a feeling your work in the industry where ROHS doesnt apply.
ROHS effectively killed the reliability of wire-to-wire or wire-to-pcb connections using solder.
Please see example of crimp only connection with 26A per wire(more than twice than 16pin pcie):

Modern solder where ROHS apply is less reliable than crimps in applications where pressure buildup cannot be avoided, its as simple as that.
 
I do find it interesting that nobody has been able to replicate it yet. I mean clearly there is a problem but if it is as bad and as dangerous as everyone is saying you would think someone could create the problem if they were trying.
 
srekal34 said:
I have a feeling your work in the industry where ROHS doesnt apply.
ROHS effectively killed the reliability of wire-to-wire or wire-to-pcb connections using solder.
Please see example of crimp only connection with 26A per wire(more than twice than 16pin pcie):

Modern solder where ROHS apply is less reliable than crimps in applications where pressure buildup cannot be avoided, its as simple as that.
Click to expand...
Fair enough I am a mechanical engineer. I just don't trust anyone with 3.7m subscribers on YouTube.
 
ExcitableBoyUK said:
I do find it interesting that nobody has been able to replicate it yet. I mean clearly there is a problem but if it is as bad and as dangerous as everyone is saying you would think someone could create the problem if they were trying.
Click to expand...
Because no one is trying to replicate a real world scenario for testing. They are all using open benches. The adapter cable needs to be inside a typical PC case with the side panel closed so that not only is the cable bent but there is also pressure applied on it by the side panel consistently for days on end along with internal heat. That is what is causing the terminals to shift over a couple of days of gaming.

Just bending the cable on an open bench isn’t reflective of real world conditions.
 
Last edited:
ExcitableBoyUK said:
I do find it interesting that nobody has been able to replicate it yet. I mean clearly there is a problem but if it is as bad and as dangerous as everyone is saying you would think someone could create the problem if they were trying.
Click to expand...

I’d imagine Nvidia are currently taking their cable supplier to task as you’d think they would have requested better cables than they’ve been supplied with.

Just find it hard to believe that they would have specced a cable that wasn’t suitable for their needs, IMO.

There’s penny pinching and there’s flagrant disregard for standards, I’d like to think that it was penny pinching rather than the latter.

However, in the unlikely event that I were to receive a GFE invite to buy a FE, I still would right now, “cablegate” or not. I’m not one to keep installing and removing cards so it would be a one time fit for me until the card got replaced. Fortunately I have a big enough case that cable placement isn’t an issue.
 
You must log in or register to reply here.
Back
Top Bottom