VehiclesFashionRecipesBlogsHuntTravelsSportFunHandmadeITEducation
Mini-Games
x

x
zakruti.com » IT - Software » Gamers Nexus
HW News - ASUS' 600W Connector, NVIDIA Driver Issues, Tariff Impact on Computer Hardware

HW News - ASUS' 600W Connector, NVIDIA Driver Issues, Tariff Impact on Computer Hardware

FBTwitterReddit

video description

Rating: 4.0; Vote: 1
Sponsor: ID-Cooling Frozn A720 Black on Amazon https://geni.us/VDnou4Y In this news round-up, we recap the last two weeks of hardware and industry news. This episode includes ASUS' revision of its Q-Release system, its new 600W power connector re-emerging, rumors of Intel's high-end GPU being canceled (we'll see about that), impact from tariffs, and more. Check out the Just Josh laptop videos we were in! Includes one with Steve picking a laptop and another of overclocking! https://www.youtube.com/JustJoshTech
Date: 2025-04-07

Comments and reviews: 20


I watched the video on everyone using AI as something humanity couldn't possibly ever exist without, or well, it comes off that way. I have only used it once in Duck Duck Go to find an answer I was unsure how to express the question for, which you can have problems with in searching for something you don't know terminology or what something is called, and it was surprisingly good, but also too careful as to not be misleading and well it explained things I already understood in way to much detail until I told it to take it slow and only go into details when I ask it too.
What I liked about it was that it kept it real, instead of feeding me lines of BS it got from pseudoscience nut jobs of a very common types, since I hade questions in the realm of higher physics, and came for real answers, although I don't remember if I instructed it to also check for the accreditation as well as how many citations a paper has in other papers, and look for any rejections and state them too and most of all, the science journals it got information from, because there are a huge number of fake Science journals operated by pseudo scientists, and especially well funded christian religious science denying creationists AKA Intelligent design (exact same thing): see The SCOPES trials and how they tried to have it taught in schools as valid science under Intelligent design as they failed the first time using the well known word creationism and the second time by Intelligent design which they totally made up to try again and to fool our courts.
They try to fool gullible people who have no grasp of the laws of logic, nor do they themselves, because although they can make huge presentations where they misquote scientists, or quote them out of context even sometimes of field of study which is likely to be way above their heads, as if refuting all of science is equal to Proving gods existence despite doing no such thing, as that requires the claim stand on it's own merit:
Refutation is only proof of their attempt to refute, even if they did (never do or ever have in all recorded history), not evidence for the counter argument, and with that in mind you can easily not even get into a long debate, but just demand the body of evidence, since claims are not evidence. They even try to back up claims with even more claims throughout, and never bring any actual evidence, nothing but piles of logical fallacies, and knowing that will help you not fall for their or other pseudo sciences utter absurd BS and waste valuable time explaining the actual science, as that will get you nowhere, and they failed pretty much by the mere attempt:
That which can be claimed without evidence, can be dismissed without evidence, and the burden of proof is on the ones making the claim, not you to show evidence against it! Besides when someone pulls that shit they are way too invested to care about the truth anymore if they ever did. They don't stick to the rules of scientific methods, nor sound philosophical principles, and with it are violating science ethics. They aren't even trying to convince you, but anyone listening who is a fool and falls for it, so they can use them as a tool, and income source, and it works way too good for them too, because a huge part of indoctrination is thwarting children's natural tendency to learn HOW to think by drilling into their heads WHAT to think and with gullible people who don't have any education in the matter.

reply

AMD's 9000 series launch has been really good, so good in fact that it's making me consider switching back to Nvidia. I use Linux, and 7900 GRE felt like a great card for use with Linux. Then I bought the 9070 XT since they had been ready for months before the launch so they must've had the time to get drivers working, and they had just been boasting about the ROCm support and performance in general being great at AMD, and I also would benefit from the increased performance for those things.
Got the GPU, desktop unstable, ROCm doesn't work at all. Waited for updated drivers, tried again. Did that 4 times now, and it keeps getting worse. There's no official word from AMD about ROCm support ever coming for RDNA4, the support for RDNA3 took eons so maybe it won't be coming anytime soon. The desktop has degraded to a level where if I log into X11 the system locks up a couple of seconds after launching any application. On Wayland the displays are severely corrupted and the system is completely unusable.
Decided to test how poorly my chosen Linux distro would work on my gaming PC with a 4090 and .. not a single issue, even old games have less weird stutters in Linux. Drivers work perfectly fine. Things that can use CUDA/ROCm of course run like 2.5x faster too. Now I'm thinking maybe I should stop wasting my time and energy supporting Radeon GPUs and just get another 4090 or something. And everyone keeps claiming that AMD is massively better on Linux and generally has great Linux support. Laughable.

reply

well that's telling amd gpus are so ass that they're only desireable when the price is right... that's a low bar, a VERY LOW bar... I hope that AMD LEARNS something from this, as personally I'd rather have an Intell B780 available than the 9070 as it would cream the AMD GPUs in compute! But at least AMD provided 16GB VRAM which is what I consider to be minimal VRAM for cards today or at leas those that aspire top greater than 1080p perf... and well maybe not even then IF they can provide compute perf... sorry AMD that EXCLUDES you! (and it USED to be nVidia artificially gimping compute on consumer cards back in the day! Meanwhile AMD GPUs just sucked balls for compute!)
Someday one will hope, AMD addresses compute for consumer cards as thats an IMPORTANT part of the card nowadays for a VARIETY of applications INCLUDING GAMING FFS! But hey if AMD wants to slip into the deep dark quietly, more power to them!

reply

Asus .. that's so cute and quaint. Being back to the Asus MediaBus shenanigans. Can I have another SCSI && Audio Card I still have my T2P4 somewhere.
Special slot sockets was a sht idea in the 90's and still is a sht idea.
Also this BTF is a pile of poop, to put those on the PCB has be soldered 4 times, top SMD (in a reflow oven), bottom SMD, THT from the bottom (Back IO panel connectors, PCIe connectors, etc.), and then AGAIN top THT, just for really heat intensive stuff, that soak head because they are directly attached to the ground and voltage planes, let alone those thick pins, via a so called selective for the latter two. Hope they checked their components, because those have a limited amount of solder cycles they are allowed to have.
How about putting money in better parts and for lower class boards instead of this BS

reply

The 600W GPU HiPWR bus on the MoBo actually does sound quite nice in theory... not only would it completely eliminate both the always awkward GPU Power cable snaking around the GPU, but more so any issues with power connector seating and melting.
On top of that though, one potential major bonus could also be increased stability for beefy GPUs when mounted in the traditional way; that extra slot at the rear-most end of the GPU could help quite a bit with the sagging issues, if it is sufficiently beefy.
That said, one major downside would be that all GPUs, no matter their performance tier/actual PCB size would need to fit this form factor, thus if this became an industry-wide standard, it would mean the end of shorter cards when the PCB itself also is shorter/doesn't require massive cooling design.

reply

How else would they sell 5000-7000 laptop without the higher number no matter how misleading it is (in terms of expectations of performance by the buyer). And then you'll have someone telling you how oH BuT BuILdINg A dEsKtoP wiTH A 5090 WilL Be MoRe ExpEnsIVe sO iT maKES sensE ThE Way iT Is. 7000 for a 5070 maybe 5070Ti performance - nice. You can go all in - Asus Crosshair mb, Asus Ryujin (because of how dumbly expensive it is at least in EU), Corsair QX fans, Dominator ram, 1200w Asus PSU (again, ROG tax), 9100 Samsung m.2, Asus most expensive 5070Ti and a 480hz OLED monitor, 9950X3D and still will save money compared to buying 5090 laptop and likely get superior performance.
reply

so the 600W, instead of going from PS directly into a video card via a suitable connector (not the ones catching fire recently), they introduced another couple of connectors on the motherboard and then from the motherboard to the video card, so one extra sliding connector... and 50 amps going through the motherboard...
so - new standard, much more expensive motherboards that require videocards with larger pcbs to fix a problem with a connector which supposedly solely was used to reduce the size of the video card pcb
the question is - wth are they all smoking there, stop smoking that and smoke something else, because this current product is clearly causing brain damage

reply

Man I really, REALLY hope that Intel does not cancel their high-end plans. I gave up on Nvidia with the 40-series; I've let too much $ht from their part slide. I purchased an RX7900XTX and was looking forward to a 9900XTX until AMD announced they weren't doing it (Ik, no need to upgrade, but I have the capital and was willing to hold on to it for a long time). Since that didn't happen, I thought I might give Intel a hand and try them out. Clearly not this generation, but hopefully they don't cancel the next generation as well. Some of us are rooting for you guys!
reply

GamersNexus, I recently had an unrelated failure on a XFX 7900XTX that also had cosmetic damage by the Q-Release and while they did mention the damage, they still warrantied it will a replacement. I kept some detailed records, if you need any 'receipts' of the whole process. XFX has a crazy good warranty return process and support service. I was extremely worried that the cosmetic damage would void the warranty but they assured me as long as there was no actual damage to the pin contact it usually is not the cause of any failures.
reply

Observation regarding nVidia drivers. After the latest upgrade a few days back I suddenly saw several YT videos having sound sync issues. I have seen that before, but disabling either the nVidia or the Gamebar overlay solved that. I did check and they were both still disabled so I thought it might have been the videos were simply faulty. Now the latest video where I saw this was this one, but after downgrading to 566.36 everything works as it should. No sound sync problems anymore in this video!
reply

For some reason, i kept hearing AI AI O to the tune of Old McDonanald had a farm.
edit:
Old Nvidia had a farm
AI AI O
And in that farm he had some chips
AI AI O
With a melt melt here and a Melt melt there, Here a melt, there a melt, everwere a Melt melt
Old Nvidia had a farm
AI AI O
And in that farm he had some Rops
AI AI O
With a Rop fail here and a Rop fail there, Here a fail, there a fail, everywere a fail fail
AI AI O
ok, i;m sure someone can do better. :)

reply

I have a 4090 TUF and these driver issues have legitimately made me insane to the point of contemplating trading it for a Switch 2 because of the amount of headaches it's given me
566.14 (November) was the only one that worked for more than a couple hours, DDU'd in safe mode each time I tried one (ignoring the W11 sign-in bug that got patched) and even still I had a complete freeze from DP signal loss just earlier.
Considering the 9070xt if a higher variant isn't released by the fall tbh

reply

GamersNexus Not only InZOI and The First Berserker: Khazan, but also Path of Exile 2 (Grinding Gear Games) has reported spending months of their engine development team attempting to fix an nVIDIA/DX12 crash. So much so that they implemented a workaround that pauses the game & caches everything when the driver dies, and then loads it all up again after it recovers, as to not completely interrupt gameplay. They also recommend using older drivers to combat the problem. It's pretty wild.
reply

I'm not surprised that Radeon is killing it.
We're back in that 2020 era market where the only thing they need in order to compete is to A, have a product... B, that exists... and C, that people can actually buy. (That last one is debatable though.)
I just hope that this actually encourages consumers to take Radeon seriously instead of just saying We want Radeon to be good because it pressures nVidia to lower their prices on the cards we were already going to buy.

reply

the problem with building factories in the US is that no one knows what new tariffs or old tariffs will be changed without warning. No one would invest in a market that changes this drastically on a whim. There was chip manufacturing being planned to the US, but DOGE defunded the CHIPS act, and now that is in limbo as well. It is very likely that some of those deals will fall through, and we will have less manufacturing going to the US as a result.
reply

I remember seeing headlines about Arc being cancelled as soon as the A-series cards launched. Then the B-series cards came along and, as soon as they launched, similar headlines. If for no other reason than to remind nVidia that competition is a thing, I hope they carry on for years to come. Being realistic about these things, the Arc cards are probably what's keeping Intel afloat, right now, though. Because their CPUs mostly suck
reply

Now, lets talk about Asrock with their idiot decisions about their PCIe release mechanism. On their AM5 ITX boards. To be exact: Their great idea of putting the BIOS reset jumper pins right beneath the GPU release lever. Damaging Yes. My system got a fancy BIOS reset when I accidently mushed my GPU against them when trying to remove it from the case. Those pcie slot stripping issues of Asus pale against that one.
cu, w0lf.

reply

You've cited the possibility of yet another way for PC components to weld themselves to each other with the BTF motherboard format. I have an idea. What if they included a component that was designed to fail before everything else can and is easily replaceable As it's there to prevent the parts fusing together we could call it a stay-apart or something Not sure if it would ever catch on though...
reply

Was/is the stock for the 9070XT in US that bad In the EU (by my experience Belgium,Netherlands,France) these are available since launch but the prices were so jacked up that there were always several models on the shelves. If you wanted one you could buy one. Even now you can get any brand/model you want just as long as you are willing to pay 800-950 (which by today's standards isn't THAT bad).
reply

Seeing the results about nVidia gpus being more and more reduced from the flagship card with each generation has made me think for a few years now that nVidia will eventually just release flagship cards. I can just see Jen-Sun saying, You want a second tier card Buy last generation... It's still $1,400... And we tender all second hand sales... Would you like an nVidia credit card to buy it
reply
Add a review, comment






Other channel videos