[SOLVED] Asus tuf 3080 MLCC

SaarVFX

Prominent
Sep 24, 2020
37
0
530
Hey :)

First I would like to apologize for my English grammar mistakes 😊

I saw that Asus said they didn’t use the Poscap capacitors at all on their Nvidia 3080, and only used the MLCC capacitors, but then I saw that the Asus is crashing kind of the same like the other manufactures, so my question is,
Why the Asus RTX 3080 is crashing too, if the common opinion is that it’s for sure a hardware issue causing the crashes (especially the Poscap/MLCC thing) and is it really make Asus 3080 better than the other manufacturers?

Second question is:
I just built a new PC for me, in a high budget In the goal to make it an editing monster, and I didn’t bought a Graphic Card yet because I waited for the RTX 3080 to get out ..
Is it still worth to buying it? I am now have all the components of my new PC but can’t start using it because I still didn’t buy the GPU..

The main purpose of the computer is going to be for heavy 3D modeling and rendering, heavy After Effect use, and heavy compositing on a software like Nuke..

and I will also want it to run the popular games on the best or almost the best settings :)

Thanks for reading and I would really appreciate any help,
Thank you!

My other new PC parts that I bought are:


CORSAIR Vengeance RGB Pro 64GB (2 x 32GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel XMP 2.0 Desktop Memory Model CMW64GX4M2E3200C16

Western Digital WD BLACK SN750 NVMe M.2 2280 1TB PCI-Express 3.0 x4 64-layer 3D NAND Internal Solid State Drive


Intel Core i9-9900K Coffee Lake 8-Core, 16-Thread, 3.6 GHz (5.0 GHz Turbo) LGA 1151 (300 Series) 95W BX806849900K Desktop Processor Intel UHD Graphics 630

Seagate BarraCuda ST4000DM004 4TB 5400 RPM 256MB Cache SATA 6.0Gb/s 3.5" Hard Drives Bare Drive - OEM

GIGABYTE Z390 AORUS PRO WIFI LGA 1151 (300 Series) Intel Z390 SATA 6Gb/s ATX Intel

Gigabyte PSU AORUS 850W PFC Gold 80+ Modular P850W

Corsair iCUE H115i RGB PRO XT Liquid CPU Cooler

Phanteks Case Eclipse P400 TG Black/White

Samsung LCD 28" UHD Gaming Monitor U28R554UQU
 
Solution
The more this has gone on the more it looks to me like this is not a hardware issue but a lack of testing of peak boost speeds by AIB’s. I’ve seen a review where they changed 2 POSCAPS for MCLL and it made a 30mhz difference. I’ve seen other reviews saying the maximum impact would be 50mhz. The annoying thing is the boost in question only runs for a couple of seconds and is not used 99% of the time, it only seems to exist so 1 card can say it boosts higher than the other in reviews but has no practical improvements in performance. This seems to be the result of AIB’s racing to get the highest boost on paper with a gpu that is very consistent for maxing out at 2000-2100mhz making it very difficult to differentiate cards of different tier...
3D modeling is when the 3090 makes sense, if you can afford, but yes, RTX 3080 is a perfectly good investment, see, this POSCAPS x MLCC debacle is still nothing short of speculations, with the release of the latest drivers things seem to have stabilized, still, they probably have relation with the problem, so much so that indeed asus with 6 MLCCs arrays also crash, but things seems to have hit hard less quality cards like zotacs with their 6 POSCAPS, but no one is or was really absutelly safe. Regardless, if the problem was a combination of driver, poscaps, mlccs, manufacturers are changing their design in attempts to solve the issue, MSI for instance was using an array of 5 POSCAPS and 1 MLCC, while NVIDIA uses 4 + 2.... go figure, this will be fixed no doubt, so frankly if you were to go with any ASUS/MSI/EVGA you should be fine, I wouldnt go with gigabyte or palit or zotac simply because I dont trust their overall quality.
 
The more this has gone on the more it looks to me like this is not a hardware issue but a lack of testing of peak boost speeds by AIB’s. I’ve seen a review where they changed 2 POSCAPS for MCLL and it made a 30mhz difference. I’ve seen other reviews saying the maximum impact would be 50mhz. The annoying thing is the boost in question only runs for a couple of seconds and is not used 99% of the time, it only seems to exist so 1 card can say it boosts higher than the other in reviews but has no practical improvements in performance. This seems to be the result of AIB’s racing to get the highest boost on paper with a gpu that is very consistent for maxing out at 2000-2100mhz making it very difficult to differentiate cards of different tier levels. I’ve also seen 1 claim but I haven’t seen it mentioned anywhere else or been able to verify it but it said POSCAPS have a lower failure rate which would an interesting point if true.
 
Solution