HW News - It's Not AMD's Drivers, RTX 4060 Ti Rumors, MSI Afterburner Follow-Up
video description
Date: 2023-01-26
Related videos
Comments and reviews: 14
kenabi
while it may not be in this instance (though i'm iffy on the speculated reasons.. it's possible, but i'm not 100% on it.), the number of people i see commenting going 'it never could have been, because they can't bypass the bios because the settings are stored there!' or some variant, shows how little people understand that the drivers _can_ ignore those settings and dump everything wide open on the 'throttle', e.g. full open vregs, max clock settings, etc. if they couldn't.. we couldn't overclock/volt or underclock/volt them in software. so a random bug _can_ kill cards. it just doesn't seem to be the case here.
i'd actually think more along the lines of hard shock during shipping is more likely, causing microfractures, and then the heat/cooling cycle makes it worse up to failure. if it ran into the now known (for the most part minor) temp issue some cards are seeing with the latest drivers, that _would_ explain things.
12vhpwr;
nah. f this whole connector spec, don't care. we went from millions and millions of PCIE 6/8 pins and minimal occurrances, to a couple hundred thousand out in the wild with entirely too many. that points towards a failing in the mechanical design, how one side mates with the other, and how easy it is to not get it connected right, or possibly for the plug part to come out of the socket side.
even if i have to go in in the future due to no more PCIE 6/8 cards, i'll _build_ a board and convert things over to the PCIE 6/8, because i flat out refuse to use 12vhpwr. there was literally no reason beyond someone thinking 'hey, we need a new spec for reasons'.
no, we didn't.
it was a crap solution in need of a problem.
reply
while it may not be in this instance (though i'm iffy on the speculated reasons.. it's possible, but i'm not 100% on it.), the number of people i see commenting going 'it never could have been, because they can't bypass the bios because the settings are stored there!' or some variant, shows how little people understand that the drivers _can_ ignore those settings and dump everything wide open on the 'throttle', e.g. full open vregs, max clock settings, etc. if they couldn't.. we couldn't overclock/volt or underclock/volt them in software. so a random bug _can_ kill cards. it just doesn't seem to be the case here.
i'd actually think more along the lines of hard shock during shipping is more likely, causing microfractures, and then the heat/cooling cycle makes it worse up to failure. if it ran into the now known (for the most part minor) temp issue some cards are seeing with the latest drivers, that _would_ explain things.
12vhpwr;
nah. f this whole connector spec, don't care. we went from millions and millions of PCIE 6/8 pins and minimal occurrances, to a couple hundred thousand out in the wild with entirely too many. that points towards a failing in the mechanical design, how one side mates with the other, and how easy it is to not get it connected right, or possibly for the plug part to come out of the socket side.
even if i have to go in in the future due to no more PCIE 6/8 cards, i'll _build_ a board and convert things over to the PCIE 6/8, because i flat out refuse to use 12vhpwr. there was literally no reason beyond someone thinking 'hey, we need a new spec for reasons'.
no, we didn't.
it was a crap solution in need of a problem.
reply
semosesam
I think it's a bit unfair to say people not seating the new style connector properly are user error then go on to say that it's almost impossible to do the same thing with the old connector that we've been using for a decade+. This problem did not exist with the old connector, now it does with the new connector. Clearly, the problem is the new connector. I think the blames lies with Nvidia for choosing to use the new connector without proper testing, but ultimately with the standards organization that created it. There should be a more definitive click that fully seats the connector. It should not be possible for a user to partially seat the connector. This is a design error. Not a user error.
reply
I think it's a bit unfair to say people not seating the new style connector properly are user error then go on to say that it's almost impossible to do the same thing with the old connector that we've been using for a decade+. This problem did not exist with the old connector, now it does with the new connector. Clearly, the problem is the new connector. I think the blames lies with Nvidia for choosing to use the new connector without proper testing, but ultimately with the standards organization that created it. There should be a more definitive click that fully seats the connector. It should not be possible for a user to partially seat the connector. This is a design error. Not a user error.
reply
bloodem
I love my undervolted 4070Ti which hovers 50 - 55 degrees and 160 - 170W of power consumption. Won't have to worry about any melted cables.
I initially thought that I would regret purchasing it (it's THE MOST expensive GPU that I've ever bought), but it should serve me well for the next 4 - 5 years so... damn you, nGREEDIA, you fooled me... (still, a wonderful architecture, nonetheless)...
reply
I love my undervolted 4070Ti which hovers 50 - 55 degrees and 160 - 170W of power consumption. Won't have to worry about any melted cables.
I initially thought that I would regret purchasing it (it's THE MOST expensive GPU that I've ever bought), but it should serve me well for the next 4 - 5 years so... damn you, nGREEDIA, you fooled me... (still, a wonderful architecture, nonetheless)...
reply
Sokoloft
7:22 I prefer using MSI afterburner and RTSS over say precision x. I remember using precision x16 and rtss with my 560 se. When they removed it was a big issue for me. I still use it with my evga cards and will continue to. Hopefully if MSI cannot pay, the developer takes some time to re-release the software as open source so that others can contribute to the development of the software.
reply
7:22 I prefer using MSI afterburner and RTSS over say precision x. I remember using precision x16 and rtss with my 560 se. When they removed it was a big issue for me. I still use it with my evga cards and will continue to. Hopefully if MSI cannot pay, the developer takes some time to re-release the software as open source so that others can contribute to the development of the software.
reply
Zyx
The whole GPU industry is fSzked up, what is going on? It seems like the wild west now and everything is crazy, companies are trying to rob the customers and the midrange tier no longer even exists except at 700 dollars and stuff is blowing up and melting and other companies are exiting the market entirely, has it ever been this weird and unstable? This is not a smooth flight.
reply
The whole GPU industry is fSzked up, what is going on? It seems like the wild west now and everything is crazy, companies are trying to rob the customers and the midrange tier no longer even exists except at 700 dollars and stuff is blowing up and melting and other companies are exiting the market entirely, has it ever been this weird and unstable? This is not a smooth flight.
reply
Ardeof
Hey could you clarify one thing that's perplexing me a bit. I've heard that supposedly Intel B660/B760 boards only support boost/OC (i'm unsure which) for a limited amount of time. Say like 15 minutes, which is long enough for a benchmark but not long enough for a gaming session. Is this the key reason to go Z board over B boards if you're using an unlocked processor?
reply
Hey could you clarify one thing that's perplexing me a bit. I've heard that supposedly Intel B660/B760 boards only support boost/OC (i'm unsure which) for a limited amount of time. Say like 15 minutes, which is long enough for a benchmark but not long enough for a gaming session. Is this the key reason to go Z board over B boards if you're using an unlocked processor?
reply
Anonymous
To be fair, the Story of the 6000 Series AMD GPU's dying rode on the Back of the 7000 Series Vapor Chamber Issue. I don't blame anybody for believing it. A Company that doesn't Quality control their new Cards, might aswell be incompetent enough to blow up their old ones. Edit: My 7900 XTX MBA was DOA btw.
reply
To be fair, the Story of the 6000 Series AMD GPU's dying rode on the Back of the 7000 Series Vapor Chamber Issue. I don't blame anybody for believing it. A Company that doesn't Quality control their new Cards, might aswell be incompetent enough to blow up their old ones. Edit: My 7900 XTX MBA was DOA btw.
reply
Tinkle
So I was about to comment about the cards probably being mining cards right as you got to that part. When Der 8auer first went over it I had that suspicion to to how man there were at a single locale. Glad to know that that's ben confirmed, hopefully the people who bought them from the farm get replacements.
reply
So I was about to comment about the cards probably being mining cards right as you got to that part. When Der 8auer first went over it I had that suspicion to to how man there were at a single locale. Glad to know that that's ben confirmed, hopefully the people who bought them from the farm get replacements.
reply
Zera
So I was about to comment about the cards probably being mining cards right as you got to that part. When Der 8auer first went over it I had that suspicion to to how man there were at a single locale. Glad to know that that's ben confirmed, hopefully the people who bought them from the farm get replacements.
reply
So I was about to comment about the cards probably being mining cards right as you got to that part. When Der 8auer first went over it I had that suspicion to to how man there were at a single locale. Glad to know that that's ben confirmed, hopefully the people who bought them from the farm get replacements.
reply
Elijs
Gonna be honest, I saw that jayz2c video and rolled back my driver, just in case. Not particularly happy about how explicitly the messaging was drivers bad, roll back now (and no matter what Jay is trying to claim now, no, his initial video about it was very obviously suggesting to roll back drivers).
reply
Gonna be honest, I saw that jayz2c video and rolled back my driver, just in case. Not particularly happy about how explicitly the messaging was drivers bad, roll back now (and no matter what Jay is trying to claim now, no, his initial video about it was very obviously suggesting to roll back drivers).
reply
rex
That's old. I had the Diamond Viper V550 with the Riva TNT chip, and then the Diamond Viper V550 with the Riva TNT2 chip. I was happy to have it and play Forsaken as it was a game where I got hardware acceleration as it has OpenGL, and IIRC, it worked better on the TNT chip than the Voodoo cards.
reply
That's old. I had the Diamond Viper V550 with the Riva TNT chip, and then the Diamond Viper V550 with the Riva TNT2 chip. I was happy to have it and play Forsaken as it was a game where I got hardware acceleration as it has OpenGL, and IIRC, it worked better on the TNT chip than the Voodoo cards.
reply
Cam
I love how we are calling 12VHPR issues User error but we never had this problem with the 6/8 pin connector. if its so easy to screw up the 12VHPR connection then its Faulty you had to make it as simple and easy to do that even a monkey could do it
reply
I love how we are calling 12VHPR issues User error but we never had this problem with the 6/8 pin connector. if its so easy to screw up the 12VHPR connection then its Faulty you had to make it as simple and easy to do that even a monkey could do it
reply
KRISTIN
Man remember when the Ti cards were the last cards to be released, and it was only for the highest end cards. Now it seems to just be an excuse for Nvidia to release only the most expensive GPU's first and completley ditch non-Ti cards.
reply
Man remember when the Ti cards were the last cards to be released, and it was only for the highest end cards. Now it seems to just be an excuse for Nvidia to release only the most expensive GPU's first and completley ditch non-Ti cards.
reply
Jezza
Man remember when the Ti cards were the last cards to be released, and it was only for the highest end cards. Now it seems to just be an excuse for Nvidia to release only the most expensive GPU's first and completley ditch non-Ti cards.
reply
Man remember when the Ti cards were the last cards to be released, and it was only for the highest end cards. Now it seems to just be an excuse for Nvidia to release only the most expensive GPU's first and completley ditch non-Ti cards.
reply
Add a review, comment
Other channel videos
