• 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 RTX 50 SERIES - Technical/General Discussion

Murphy said:
IDK who Jon Gerow is, i think it's the person PCMag spoke to, however he seems knowledgable (not that I'd know) and he has some more words to say on the subject for anyone who's interested...
Click to expand...

I’m not sure if intended or coincidence but this seems to be the very same guy I posted making comments on Reddit (I’m confused with the laughing smiley you made on that post after this comment though?).
 
You're forgetting or misrepresenting me @Murphy. To review, I said:
philo-sofa said:
....[It] should have been part of the spec for the PSU side to load balance, or at least for the 4000 and 5000 series cards to have 3+ shunt resistors to load balance there.
Click to expand...
You then replied, ignoring that I said it should be part of the PSU spec, saying:
Murphy said:
You can't expect the PSU to do the load balancing because it has no knowledge of what's been connected to it, things that use electricity draw power.
Click to expand...
Huh. Do you understand PSUs are aware of the power that they're provisioning, or that they can be set to load balance to specific lines and wires? Because what you're saying here makes little sense. Hence:
philo-sofa said:
Not true. The PSU could be designed to be able to sense the current in each of the six live wires, and to load balance them. It doesn't need to know what its connected to, just the total load request and power going down each live wire.
Click to expand...
Murphy said:
What does true or false have to do with what expectations?

If you're suggesting that the entire ATX specification should be changed simply because Nvidia can't design a connector then maybe you're being a bit unreasonable. Can it be done, sure if we put our minds to it most things can. Should it be done, heck no because expecting an entire industry to bend over backwards just to accommodate a single organisation is ridiculous.
Click to expand...
To conclude:
1) The ATX power supply specification was revised in 3.0 and 3.1 chap, for the PCIe 5 12V-HPWR and 12V-2x6, respectively. So to bring things full circle, yes I'm saying when that end of the connector was designed it should have been designed to allow a PSU load balance its wires, rather than allow 600w to potentially go through one wire. Nvidia was the most active part of PCI-SIG when designing the revisions so a lot of the blame rests with them.
2) I noted at the very start "or at least for the 4000 and 5000 series cards to have 3+ shunt resistors to load balance there" so I covered that side of things. Not sure why you're trying to pick and argument with a strawman.
 
Last edited:
philo-sofa said:
You're forgetting or misrepresenting me.
Click to expand...
Pretty sure I'm not...
philo-sofa said:
To conclude:
1) The ATX power supply specification was revised in 3.0 and 3.1 chap, for 12V-HPWR and 12V-2x6, respectively. So to bring things full circle, yes I'm saying when that end of the connector was designed it should have been designed to allow a PSU load balance its wires, rather than allow 600w to go through one wire. Nvidia was the most active part of PCI-SIG when designing the revisions so a lot of the blame rests with them.
2) I noted at the very start "or at least for the 4000 and 5000 series cards to have 3+ shunt resistors to load balance there" so I covered that side of things. Not sure why you're trying to pick and argument with a strawman.
Click to expand...
Yea, nope. Defiantly not forgetting or misrepresenting you, you're saying that power supplies should be redesigned to balance power draw.
 
Murphy said:
Pretty sure I'm not...

Yea, nope. Defiantly not forgetting or misrepresenting you, you're saying that power supplies should be redesigned to balance power draw.
Click to expand...
No mate, I'm actually saying:

1) PSUs can and do balance power draw
2) The PSU connector and the GPU side has been poorly designed, in not allowing a single 600w port to balance power draw
3) Given I was pretty clear and you're trying to muddy the waters, I'm now also saying you're being outright dishonest
 
Last edited:
Nitefly said:
My current hypothesis - purely because I think it’s the most rational explanation - is that the H++ cable spec hasn’t changed, yet Moddiy, Seasonic, Coolermaster and Cablemod are causing confusion by suggesting they have made some changes to cables because of a purported change in spec… when actually what has happened is they have made minor changes to make cables more robust which are not actually required by the H++ spec. But anyone’s guess is as good as mine.
Click to expand...

Seems like you're right. Jonny Guru has said in some of his recent messages that the spec hasn't changed. It very much does feel like it's a bunch of companies with cables to sell telling everyone that they really should buy their cables.
 
Cosmic Schwung said:
Seems like you're right. Jonny Guru has said in some of his recent messages that the spec hasn't changed. It very much does feel like it's a bunch of companies with cables to sell telling everyone that they really should buy their cables.
Click to expand...

If some manufacturers (Moddiy) admit that they have been talking rubbish, and then a universally clear set of messaging from everyone is given on what’s safe, that would be deffo be the best outcome for consumers - so let’s hope that’s it… and not some other weird, worse outcome :o

The missing puzzle piece would then be getting to the bottom of what mechanical / electrical issues are still causing the higher temps, as shown by Der8auer (imperfect connections, crimping, materials? etc.). We already know why the cable subsequently ends up catastrophically failing = unnecessarily limited redundancy in the overall design.
 
Last edited:
Its like what happened with the 4090 though. Someone kicks off with melt proof. People slam multiple cards on rigs testing and come back with hmm no melts. All we can tell though is it was never fixed but PSU manufacturers and nvidia need to get this finished properly.
 
MrH said:
I've been fiddling with the OC for a few days and I'm happy to stop here. Absolutely rock solid in every game and benchmark. Overall I'm very impressed with the OC ability of the Palit Gamerock. I found Ninja Giden 2 Black is extremely sensitive to instability, it would crash immediately with a slightly higher core clock, whereas every other game and benchmark was fine with it, but after dialling it back a tiny bit it's also completely stable.


Nice score. What your core and memory at
MrH said:
3,127 MHz core, 16,601 MHz memory.


Go all the way there is more left. If only I knew how to get past afterburner +2000 limit. Astral 5080 owners can overclock to 18000mhz if I could get that far I would begivin' it large to the astral owners :p./QUOTE]
Click to expand...
Click to expand...
 
Last edited:
Locky said:
Any driver after 561.09 will crash for me on my 4090, thats running completely stock.. and its a wide spread issue too.
Click to expand...

Weird. What CPU? Stupidly, I’ve always updated drivers as soon as a new one becomes available, that’s currently with a 4090FE and 5950X, and I’ve not had a crash for maybe six months or so.
 
You must log in or register to reply here.
Back
Top Bottom