• 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.

FINALLY A DEAL ON GPU'S - IT HAS BEEN TOO LONG (NVIDIA 3060 & 3060Ti) !!

Squink, than for the clarification.
When I saw post saying folks had grabbed FE’s I did assume they had purchased via OCUK.
Thanks for putting me right.
 
Received my zotac 3060ti oc today. Luckily it wasn't stolen as dpd driver left it on my doorstep at 8 o'clock this morning, nobody was home as we were staying at the inlaws. Card is working fine, bit noisy and high temp out the box. I have just undervolt and overclocked it slightly and now running cooler and quieter. Still tinkering to fine tune.

Would you care to post your before and after for temps and voltage/clock settings, would give me an idea of what to expect when mine arrives.
 
Would you care to post your before and after for temps and voltage/clock settings, would give me an idea of what to expect when mine arrives.

This zotac card does run abit hot and noisy out the box but mine is installed in a lian li tu150 itx case. I managed to get it stable at .875v at 1875mhz with fan curve that keeps the card at 76c at 60% fan speed on average, mem is at +500 but I haven't pushed it higher yet. Every card will behave differently so u might get different results. I will post some pics if you want.
I have my best undervolt overclock but I will probably make another profile with abit more juice and higher frequency for games that are more demanding.
 
To be honest, you shouldn’t need to worry on non GDDR6X cards. It’s only the 3080’s and above that have memory temp. issues.

The zotac 3060ti from what I've read online and have experienced with my own card has a bad out the box fan curve and higher than needed voltage. I fully recommend msi afterburner and ctrl+f to set a better voltage/frequency and adjust fan curve. My card is very quiet at 60% fan speed but I can hear it at 70% and any higher becomes to loud for my preference.
 
This zotac card does run abit hot and noisy out the box but mine is installed in a lian li tu150 itx case. I managed to get it stable at .875v at 1875mhz with fan curve that keeps the card at 76c at 60% fan speed on average, mem is at +500 but I haven't pushed it higher yet. Every card will behave differently so u might get different results. I will post some pics if you want.
I have my best undervolt overclock but I will probably make another profile with abit more juice and higher frequency for games that are more demanding.

Thanks for replying, mine will be in a Fractal Design Define 7 Compact with 2 140mm fans at the front (240mm AIO on the top for exhaust), so indeed mine may be dfifferent, Ill report back my findings as mine its due Monday. Initally I will bump up the Power Limit in Afterburner to see how high the boost will go on its own before I start to look at Overclocking. I am upgrading from a 1070 so even stock will be a hefty boost I am sure, its served me and likely will continue to do so back in my old 3770K rig.
 
The zotac 3060ti from what I've read online and have experienced with my own card has a bad out the box fan curve and higher than needed voltage. I fully recommend msi afterburner and ctrl+f to set a better voltage/frequency and adjust fan curve. My card is very quiet at 60% fan speed but I can hear it at 70% and any higher becomes to loud for my preference.
Interesting to hear about that on the Zotac 3060.

If you ever get your hands on a 3080 or above, that will give you some worrying VRAM temps. without the thermal pad mod. My previous MSI 3070 ran fine, however, the 3080 FE... that's a different story haha.
 
Why should users have to MOD their own, brand new cards to get acceptable temps? Nvidia are really treating customers like fools

If there's an actual issue (as opposed to temps that are high but within tolerances when run at stock) with cooling the GDDR6X then why is nvidia and only nvidia to blame?

The claim is that there's enough heat energy to raise the temps high enough to melt the thermal pads. Completely liquify them. There's no way such a serious flaw could exist without being detected by adequate product testing. Or anything vaguely approaching adequate product testing.

The obvious first port of call for blame is Micron, since they're the only manufacturers of the memory. It's proprietary and non-standard, so it's all on Micron. Micron openly and publically state an operating temp of up to 105C for GDDR6X. I don't know how much detail they would supply to companies buying GDDR6X or companies making graphics cards using GDDR6X, but I don't see a reasonable way to place all the blame on nvidia alone. If Micron gives those companies inaccurate information on Micron's products, it's Micron's fault. If companies using it don't ask for the necessary details or don't take that information into account when making their cards, it's their fault.

The only scenario in which I can see nvidia being mainly to blame is if Micron gave nvidia accurate information on GDDR6X and then nvidia lied to graphics card manufacturers and none of those graphics cards manufacturers checked for themselves with Micron. But even in that rather implausible scenario it still wouldn't be fair to blame nvidia alone because even in that scenario any vaguely adequate product testing by the graphics card manufacturers would uncover such a monumental flaw. Releasing a product that melts in its intended use can't happen unless it's not tested in its intended use.

As for the earlier statement that led to the post that I think you replied to (you didn't quote anything, so it's not certain what you were replying to):

The zotac 3060ti from what I've read online and have experienced with my own card has a bad out the box fan curve and higher than needed voltage. [..]

that's not unique to nvidia's own cards or to cards from other manufacturers using nvidia's chips. For example, I had it in spades on the Radeon 7950 I had a while back. The fan curve was OK in itself as it was based on temps, but the startlingly large out of the box over-volting jacked the temps up and that had a knock-on effect on the application of the fan curve. It also jammed the card hard up against the power draw throttling, even at stock. I got something like a 30% performance increase just from reducing the voltages to the right amounts while keeping the clocks at stock, just because it reduced power limit throttling. I also got far lower temps (~15C IIRC) and far lower fan speeds and thus noise.
 
Nvidia, AMD and all the manufacturers it just doesn't matter because every card turns out different and all of these companies have up and downs with their products or marketing. Some cards/chips are designed better. Read reviews and look on reddit etc so you know what to expect from the card your interested in. Nothing is perfect and everything can be improved. I knew that the zotac 3060ti had a few short comings before I bought it but it was easily remedied with a bit of fine tuning on the voltage and fan curve. I do this on every card I buy anyway as there is always an improvement to be made.
 
Nvidia, AMD and all the manufacturers it just doesn't matter because every card turns out different and all of these companies have up and downs with their products or marketing. Some cards/chips are designed better. Read reviews and look on reddit etc so you know what to expect from the card your interested in. Nothing is perfect and everything can be improved. I knew that the zotac 3060ti had a few short comings before I bought it but it was easily remedied with a bit of fine tuning on the voltage and fan curve. I do this on every card I buy anyway as there is always an improvement to be made.

I agree. I didn't know about it until I encountered it with that 7950 I had some years ago, but that was my ignorance. It's now something I look into.

Having said that, I didn't find any room for improvement with the card I'm using now. MSI 1070 Ti Gaming 8G. As far as I can tell, it's perfectly tuned straight out the box. I've just been running it at stock everything all the time after a couple of hours of overclocking for sport when I first got the card and I didn't touch the voltages then because the version of Afterburner available at the time couldn't reduce the voltage for a 1070Ti and I had no interest in increasing the voltage. The card overclocked very well on stock volts, past some overclocked 1080s let alone other 1070Ti cards while keeping a max GPU temp of 70C without going over 60% fan speed, but I dropped it back to stock after a couple of hours of sport overclocking as stock was enough performance. I was overclocking for sport, tweaking and testing for the fun of it and benchmarking just to get a measurement, a score. Donkey King overclocking - how high can you get? I recently tried a little undervolting (newer version of Afterburner) and didn't find any scope for any significant improvement. Maybe I could shave a couple of millivolts off the voltages and maybe get an improvement of a fraction of a degree or a fraction of a decibel, but the card automatically varies voltages in what seems to be pretty much spot on efficiency. GPU voltage on that card is only 1.045V under full load. ~1.06V under full load is more common on 1070Ti cards and the max spec for the 1070 Ti GPU is 1.093V, so I think it's fair to say that MSI have already done the fine tuning as efficiently as is practical on that card.
 
well having gotten my hands on the Zotac 3060Ti OC version and having had a chance to play, my first observation is that it seems very quiet. Admittedly, coming from an R9 290X, I've been tempted to put a room fan on because the room is now eerily quiet.
I agree with @tommills that it's a bit hot and noisy using the default settings, but after trying first Zotac Firestorm OC scanner, which locks up every time I try to use it and I ended up having to force close it every time, even after a reboot (I tried six times), I gave up and switched to MSI Afterburner.
Afterburner took about 30 minutes to come up with a GPU frequency curve. Not touched the memory timing yet. I also made my own fan curve.
I don't want the fan kicking in at all till the GPU hits 35'C and by 75'C I want the fan at 100%.
It's definitely a vast improvement over the out of the box settings and it absolutely it is worth the small amoutn of hassle. Having never used Afterburner before (Radeon before) I don't know how to tell what settings its locked in, but after a wee spin up of (modded) Skyrim SE (because it's my most played game, even if it is old) I noticed that OHM showed it had hiked up to 2025MHz.
I'll maybe finally fire up Cyberpunk 2077 later.
 
My card arrived today also!
https://www.dropbox.com/s/dvjprgnj5vx9uh8/20210809_132254.jpg?dl=0
https://www.dropbox.com/s/h6fepdtllr7w7d6/20210809_144137.jpg?dl=0

(for some reason the images wouldn't embed)

Its not as loud as I thought it would, but I think my old 7950 Twin Frozr III set the benchmark of a how loud a GPU could go. Not had much chance to play around with it yet because of work, but looking to try and work out what a good Overclock/Undervolt will get me. Did a quick FireStrike run with the Power Limit at 110% and the highest GPU temp I saw was 71c so not bad so far. Will fire up Doom Eternal and Shadow of the Tomb Raider with some Ray Tracing enabled and see what happens there.
 
Why should users have to MOD their own, brand new cards to get acceptable temps? Nvidia are really treating customers like fools
The money spent on these custom cooled GPUs, they seem okay, however, some new forum member managed to put an Arctic Accelero IV to his Zotac 3060 Ti, even though its not on the compatability list, he shaved off like 10°C off max temps.:eek:
For anyone interested...

Tonight I've fitted my trusty Arctic Accelero IV to my Zotac 3060 Ti. The huge backplate wouldn't fit, so I just reused the stock Zotac one and added a few thick thermal pads to the rear of the PCB to help cooling.

Testing with Heaven benchmark. Ambient temp 21C.
With the stock heatsink/cooler, GPU temp maxed out at 74C with fans at 83%, was a little noisy.
With the Accelero IV, temps maxed at 64C with fans at 53%, almost silent.

The cooler temps mean you get a nice little performance boost too, even without overclocking.
Link from Latest News, Deals & Special Offers section: https://www.overclockers.co.uk/forums/posts/35019827/
 
Last edited:
Back
Top Bottom