herb

Hello, i have one of those Chinese "x89" opteron motherboards. recently i had one of these boards running for an extended period of time and one of the mosfets blew up and the board was killed. i had multiple of these and now im starting to notice that VRM temps/load is a really the main problem on these boards. My question is is it possible to swap out the mosfets for something a bit beefier to try and either reduce the strain and/or lower the temps? The Vrms are made up of a mdu1511 and an mdu1514. are there better chips that i could just replace these with to get a better overall power delivery without making massive modifications to the board? i do not have much technical knowledge when it come to power delivery so please try to explain in as comprehensive of a way as possible.

im pretty sure i have the latest chipset drivers. i dont think its a windows issue because my friends pc didnt have any issues with the card just mine, they both are on the latest version of windows 11. im going to install the chipset driver again anyway but i dont think thats it. im to the point where i think there might be a motherboard incompatibility or something. i know that not a usual issue but ive done everything at this point and the internet has not been able to provide any answers. if the card is fine in one computer and not fine in mine then the platform has to be to blame since every other factor has not fixed the issue. am i missing something?

okay so at this point i have done A LOT of tests to figure out what is wrong. ill save you the whole story because i just wasted like 4 hours of my life but pretty much my rx 6700 xt is not running at its rated clock speeds. when i play a game like star wars battlefront II the gpu only runs at 1000mhz. its supposed to run at 2600-2700. when i manually input the clockspeeds in the amd control panel the performance does get better but its still not perfect. im still noticing drops into the low hundreds of mhz when the clockspeeds are set manually. im also still getting some frametime issues when loading in more open scenes but my 1080 ti never had this issue. my friend who i live with has a 6600 xt and i decided to test my card in his pc and nothing is wrong with it, it is just my computer. im glad im closer to figuring out what is wrong but please help me.

i can try that later. the gpu and the motherboard are pcie 4.0 capable and i have it in the correct spot. shouldnt it negotiate the speed automatically? i know thats an issue with risers but i dont use one.

i updated my bios today but i didnt have free time to test it yet. I have resizable bar enabled already, thats what SAM is. i have the latest drivers and running ddu again probably wont do anything.

So yesterday i upgraded my gpu from a gtx 1080 ti to an rx 6700 xt. i ran ddu for amd and nvidia before installing the card because i wanted to make sure the drivers would be clean. after i got the amd drivers installed and setup for my preferences i started noticing that the frame time in general is very inconsistent and seemingly not correlated to anything. does anyone know what the usual culprit is for frametime issues? the particular games i tested were Battlefield V, Battlefront II, Asphalt 9, and Geometry Dash. all of these games showed frametime issues. battlefront II was the worst and geometry dash was the best, but i think the fact that a super lightweight game like geometry dash is having issues is evidence that something else is wrong. i think its important to note that i do have SAM enabled i have not done much testing with it disabled but based on my quick check in BF5 its not the problem. i am running an R7 3700x and 32gb of ddr4 3600 cl 16 ram on an msi x570 mpg gaming edge wifi motherboard. another thing i considered but haven't tested is the fact that my copy of windows is heavily modified and the gui has been modified as well. i didnt have any issues with the 1080 ti but amd might be different. i plan to reinstall windows later once i have backed up my data because i want to anyway. let me know what you guys think is the issue.

i need help with creating a custom bios for a barco MXRT-7500. i bought this card because it is the same as a w7000 but i got it for $25. the issue is that with the barco bios certain functionality is cut off and causes the AMD w7000 drivers to error out upon install. i tried flashing a stock w7000 bios onto the card but it gave me an error saying that the bios could not be written because the vendor id did not match. I also tried the techpowerup RBE to see if i could edit the bios with the amd vendor id but because its technically a "barco" card, RBE wouldn't let me read and edit the bios. so, with that said how can i grab an image of the bios, copy the vendor id and then create a new w7000 bios for this card so it can be used with regular drivers? FYI the card works with the barco drivers its just lacking certain functionality that is important for a desktop gpu.

thanks for putting this in the thread. i figured this out back in 2021 when i posted this but i never added it to the thread.

That is very interesting. thank you for this information. would you have any insight as to what the hacks are that make this work and why it only works on lynnfield?

i had a 4790k back in 2018 and i ended up getting it stable at 4.7 ghz. that cpu was still a beast and will definitely be a good bump over the 4690k

hi so recently i bought an Asus p7f7-e ws supercomputer motherboard because i like to collect weird hardware of the past. the thing that drew me to this board in particular is that it is a first gen motherboard that natively supports both core and xeon cpus and ECC RAM on a consumer platform. But the #1 thing that made me buy this is the REALLY strange ram setup. This motherboard has 6 ram slots, that might not sound strange but this isn't an x58 board. These are lga 1156 cpus which means they only have ram channels for 4 dimms. My question is how does this work? if the cpus were never designed to handle this many dimms how did asus make it work? id love to know for the sake of preserving the history of this board. i am reasonably well versed in computer hardware so i do have a theory of my own. my theory is that the blue slots on the motherboard function like you would expect, 2 dimms per channel for a nice dual channel setup. my theory for the black slots is that they work in pairs. So the pairs of slots would almost function like a single stick that is a dual rank dimm. this might not be the case of course since dual rank dimms would have existed back then but that my best guess as to how they could allocate more to less. Please let me know your thoughts as this is a super cool piece of history!!

ive been using this mic everyday since i posted this and pretty much the solution is to get a boom arm with a pop filter, turn the volume all the way up, and do the rest of your sound adjustment in OBS (i record so thats what i use but default volume has generally been fine for things like discord or video chat). its a really good mic for the price just needs a little extra love to make it the best i can be.

so recently i was looking at some old core 2 quads because i was looking for an old system to overclock, but after looking at chips like the q9400 vs the q9450 i noticed the only real difference is the cache size. i know cache can make a big difference when it comes to performance overall so that got me thinking: Assuming the q9400 is the same silicon as a q9450 with half the cache disabled, would it be possible to re-enable it in some way? if the q9400 isnt the same silicon then obviously my question is irrelevant but i just have to entertain the idea that this might be possible.

im very familiar with building pc's. Ive already done every trouble shooting step that i know of. it just doesn't work

so this is a really weird issue that i have never seen before. im putting together a budget gaming PC to sell. i found a really good deal on an old hp rp5800 at a garage sale and decided to fix it up. however while upgrading it i ran into some weird issues which initially seemed hardware related but later i would find out that that was not the case. sometimes the PC would randomly not boot properly and it would get stuck trying to post but it would just have a black screen. other times it would ramp up the fans and then shut itself off. and then a different time it would randomly come back to life after 4 or 5 reboots. sounds like a bios issue right? nope. i flashed the latest bios twice. i tried every combination of hardware i had and nothing seemed to fix the problem. then i noticed once i had finally gotten into windows a few times the PC would always die after i had restarted instead of just shutting down. if i shut down normally the PC works without a hitch but as soon as i restart i have to spend another 20 mins seemingly doing random things to get it to come back. so now Ive been able to successfully reproduce the problem by "restarting". my only question now is wth do i do? i cant sell a pc in this condition even if it is easily avoidable. most drivers require you to restart to install them properly and i cant figure out how to fix this. any ideas? P.S. i know that the rp5800 is a POS system and its not meant to do what im making it do but it is essentially a elite 8200 with more features so it should still be fine.