Build Advice New build questions ?

[Aeacus]

what about where the screen isnt full height like with laptops? do you give the height, or what the height would be if full height?

You can select different reso presets or when making custom preset, you can select from multiple different resos.

Also, i tend to run my GPU benches at full screen. But if you have another aspect ratio screen than listed, you can run in Windowed mode to still test the GPU. E.g if your monitor would be 2560x1080 (ultrawide 1080p), then running 1920x1080 in Windowed mode still works.

what is needed is UPB! (universal parallel bus).

UPB isn't cost effective.

If you want N × the speed you are getting, it is more economical and reliable to make your electronics N × faster, or use N × more sophisticated compression or prediction, or refactor the system that uses the link, or all of the above, than to make a physical cable N × thicker, with N² more insulation and crosstalk and racing.

Parallel busses are more expensive and, as the speed rises, parallel busses are actually worse than serial ones. Keeping the data lined up when the length of a single bit is less than the length of the cable is really hard.

Serial busses are a solved problem. The silicon for a sophisticated serial bus is, by current standards, small. If you want more bandwidth, the modern solution is several serial busses, not a parallel bus.

Source: https://www.quora.com/Why-dont-we-have-universal-parallel-bus

that does answer one question I was going to ask, which is if SCSI is parallel, the answer is yes! which maybe explains why SCSI has fat cables!

Modern day SCSI is serial.

Recent physical versions of SCSI‍—‌Serial Attached SCSI (SAS), SCSI-over-Fibre Channel Protocol (FCP), and USB Attached SCSI (UAS)‍—‌break from the traditional parallel SCSI bus and perform data transfer via serial communications using point-to-point links.
Although much of the SCSI documentation talks about the parallel interface, all modern development efforts use serial interfaces. Serial interfaces have a number of advantages over parallel SCSI, including higher data rates, simplified cabling, longer reach, improved fault isolation and full-duplex capability. The primary reason for the shift to serial interfaces is the clock skew issue of high speed parallel interfaces, which makes the faster variants of parallel SCSI susceptible to problems caused by cabling and termination.

Further reading: https://en.wikipedia.org/wiki/SCSI

 

[Richard1234]

progress report, or should I say frustration report!

first, I did the graphics benchmarks with unigine and the results are embarrassing:

the mobo USB C to DP outdid the RTX 4060 DP to DP on EVERY test, namely 1080 and 4K, I havent done 8K yet as I thought 4K was the top speed, but now looking at the box it does say it does 8K @ 60Hz.

when I changed cable to USB C from the mobo back panel, there was no audio, I had to change the audio to an identical setting then I got audio, so I did the mobo test with both no sound and sound just in case lack of sounds sped things up.

now the only explanation I can give is that the mobo routes the graphics to the USB C socket,

but even then it is embarrassing, that this is FASTER than the RTX 4060's inbuilt DP ports!

here are the screenshots if you want to decipher them, as its currently over my head:

RTX 4060 1080 test:

http://www.directemails.info/tom/graphics_card/benchmark_RTX4060_1080.jpg

18826

then mobo CPU without audio 1080 test:

http://www.directemails.info/tom/graphics_card/benchmark_cpu_1080.jpg

19523, which is 3.7% faster

then mobo CPU with audio 1080 test:

http://www.directemails.info/tom/graphics_card/benchmark_mobo_1080_with_sound.jpg

19521, approx the same,

where I did all the RTX 4060 tests first, then all the mobo ones and have left it connected to the mobo now as this is faster,

here are the 4K tests:

RTX 4060 4K test:

http://www.directemails.info/tom/graphics_card/benchmark_RTX4060_4K.jpg

7807,

then the mobo 4K without audio:

http://www.directemails.info/tom/graphics_card/benchmark_mobo_4K.jpg

7878, 0.9% faster,

then the mobo 4K with audio:

http://www.directemails.info/tom/graphics_card/benchmark_mobo_4K_with_sound.jpg

7884, which is in fact slightly faster.

with the tests, I noticed they have a photo of Tesla on the far left, and then a photo of Einstein, I dont know who the other people are.

then the really frustrating fiasco: I decided to flash the BIOS, I had to power on many times before the mobo even booted, then to the UEFI controls, selected the M-flash as advised by the pdf from MSI, the machine reboots, and says something about entering M flash, then nothing happens. after 30 minutes I quit.

switched off the PC with its own switch, then at the mains, then powered up, and it says exitting flash.

then nothing happens for ages, so I powered down by force. then on again, it does boot, but says Windows didnt load properly, I reboot it, and it now boots properly.

it is ridiculous this, that the damned flashing doesnt even function properly!

now unless I didnt wait long enough like my advice in another topic long ago?

You can select different reso presets or when making custom preset, you can select from multiple different resos.

Also, i tend to run my GPU benches at full screen. But if you have another aspect ratio screen than listed, you can run in Windowed mode to still test the GPU. E.g if your monitor would be 2560x1080 (ultrawide 1080p), then running 1920x1080 in Windowed mode still works.


UPB isn't cost effective.


Source: https://www.quora.com/Why-dont-we-have-universal-parallel-bus

Spoiler: expand for off topic comments on serial versus parallel

the only thing is you shouldnt blindly believe what you read on the internet or anywhere else,

you need to do your own verification, there is a ton of disinformation,

if you havent tested something for yourself, then it might not be true.

all that such info means, is the public protocols are what they say, but the stuff done behind closed curtains might do the stuff they say you shouldnt do.

during our degree project, we were sometimes told: the way to do things is X, you shouldnt do Y, because of problem Z.

but after the degree, programming at home, I tried some of the things they said you shouldnt do and are no good, and in fact they are good, yes they have the problem mentioned, but they have advantages the advised MO doesnt have.

and eg in the case being discussed, N parallel wires has a BLATANT advantage over 1 wire that it is Nx as fast! it was just like that with the things we were advised against, they all had a screaming advantage!

with my degree project itself, I wanted to program part of the project a certain way, and my supervisor strongly advised against this. and the guy was an expert in the field, having been part of history, so I followed his advice. but after the project, I eventually reworked the project from scratch and tried out what he strongly advised against, as I am into testing what is advised against, because this is where the interesting ideas lie, and not only did it work, but it eclipsed the way he advised!

with the racing mentioned, you can easily mitigate against, because what you do is have each wire send say several bytes eg say 1 kilobyte, which goes to a buffer, it then flags that its data has arrived,

when say all 9 wires have flagged their data has arrived, then an acknowledge signal is sent to the sender,
who now sends the next 9 x 1 kilobyte. you would also have timeouts in case the data is taking too long.

so that is that counterargument countered. the guy is thinking in terms of bit level synchronisation which would be time expensive, but I am thinking en masse synchronisation.

and as regards cross talk, well, that's an easy one, use fibre optics, then no crosstalk as the light is trapped in the fibres. but there are probably ways to mitigate electronic cross talk, this kind of thing requires lab experimenting as it is about the specifics, the level of cross talk of a specific arrangement and whether you can filter out the cross talk. eg say putting the wires in a ribbon rather than a cord, means you just have to mitigate neighbouring ones, as the other ones are further apart. I am not an electronics person, but first thing I'd check is whether dummy wires between the active wires will absorb some of the cross talk,

merely putting the wires further apart in the ribbon will reduce cross talk, and sending a weaker signal will probably also, where you use relay boosters for longer distances. with this kind of thing you have to do lab testing to see what distances are safe, and what is the quietest signal which is loud enough.

if you use lower resistance wires then you can send a weaker signal.

there are probably research papers written on how to mitigate cross talk, and SCSI accomplished this.

also just because SCSI did things a certain way with a certain argument, doesnt mean that is the only option, most experts overlook a ton of stuff, they work with the known, they dont know so much about the unknown! someone like Tesla was inventing a torrent of stuff because he was an experimentalist.


you'd have to look also at how electronics is shielded.

cost effectiveness isnt a proper argument, because a lot of cables are short, eg for USB and SATA these arent very long. SATA itself seems to use a ribbon cable anyway.

what you said about bit length being longer than wire length, that sounds interesting, but I'd have to see this played out with actual equipment, because in all honesty its over my head. for me, a bit doesnt have a length, but has a time! eg if I send high voltage over 1cm of wire for 10 minutes, I dont see why that is a problem, in terms of distance travelled that is probably further away than the sun!


the fact is that the buses the CPU uses is parallel, at least I think these are, eg the lanes of PCIe?

and as you know, the more lanes, the faster the bus.

you could have a bendy bus in a cable, but what you have to do is use buffers per lane, and wait till each buffer has its data before merging this. you mustnt just merge each bit on arrival, as the bits on the one wire could arrive faster.

Modern day SCSI is serial.

Further reading: https://en.wikipedia.org/wiki/SCSI

the public side may be thus, but the military dont follow the normal rules, and they will do things that we are told doesnt work!

they limit and restrain the technology that gets in the public domain, the military and the police are always several steps ahead of the public.

they actually actively shut down technology which is too advanced, buying it up and then removing it from the public. our own Unix based department got shut down a few years after I completed my degree and got replaced by a Microsoft based department!

 

[Aeacus]

but even then it is embarrassing, that this is FASTER than the RTX 4060's inbuilt DP ports!

here are the screenshots if you want to decipher them, as its currently over my head:

Based on the screenshots, it looks like when you have RTX 4060 installed inside the PC and you use MoBo display port, the benchmark actually uses both GPUs; iGPU and dGPU at once, to complete the task. That's why you see performance uplift when display is connected to the MoBo.

Laptops usually do it, utilizing both the iGPU (inside the CPU) and dGPU (dedicated GPU). But desktop PCs are also capable of doing that.

To validate it once and for all:
* Remove RTX 4060 from the PC, completely
* Have monitor hooked to MoBo and run the presets again

This way, there is no way, that PC can use RTX 4060 alongside Radeon graphics when you benchmark your iGPU.

There is a way to disable iGPU from UEFI when you have dGPU installed but it is difficult to do. Better to remove RTX 4060 completely from the system, easier this way (you doesn't have to dabble in UEFI and there is always a possibility that disabling from UEFI doesn't actually work).

I also see that you do not have latest GPU drivers. Benchmark shows your GPU driver version to be 536.67 while the latest is 552.22
Link: https://www.nvidia.com/download/driverResults.aspx/224154/en-us/

then nothing happens for ages, so I powered down by force. then on again, it does boot, but says Windows didnt load properly, I reboot it, and it now boots properly.

it is ridiculous this, that the damned flashing doesnt even function properly!

now unless I didnt wait long enough like my advice in another topic long ago?

As long as you can boot to OS, you're good.

UEFI flash shouldn't take longer than ~5 mins. As of why it hangs or doesn't display anything when it completed - i can't tell.

Though, since there are high risks involved with UEFI flash (update), i reserve it as a last resort option and only if you know, for a fact, that newer UEFI version fixes the specific issue you're having.

Oh, when you boot to UEFI, does the version name now say it to be the latest? The one you flashed (updated) it to?

 

[Richard1234]

I have trouble now!

each attempt to flash the BIOS failed, ie no option to select a BIOS file.

so I decided to use the "flash BIOS" button, where I have to rename the BIOS file to MSI.ROM,
where I reformatted a flash drive to FAT format, and to have just the 2 files from the BIOS archive, both at the top level, renaming the BIOS one to MSI.ROM

see p 71 of the mobo manual, "updating BIOS with Flash BIOS Button",

I powered down the machine, with the mains on. then with the flash drive attached to the "flash BIOS" usb socket on the back panel of the mobo, where the manual doesnt even show where this "flash BIOS" socket is, and the font is too small to read, I had to use a huge reading lens to read it.

I pressed the Flash BIOS button, and the led began flashing, but it is still flashing now 1 hr 27 minutes later, so the later comment saying flashing the BIOS should only take about 5 minutes, means this has failed!

I am writing this from the 2010 PC.

I will leave the led flashing for the moment, I think I will have to use the multi-BIOS switch to get out of this,
see p 55,

what is the way to do this?

should I switch the machine off at the mains, then push the switch to the other BIOS, then power up and boot the machine?

how will I fix the current BIOS when I switch to the other one, or will this one be up in smoke?

I am very annoyed by this, MSI told me to flash the BIOS, and I have tried the M-flash option, and tried that also with the BIOS file renamed to MSI.ROM, and have tried the flash bios button, and everything has failed, I think I was sold a dud!

worst case I have to junk this mobo, if that happens I will get a Gigabyte mobo, as I cant take any more of this stress. it has been problematic from the start, often not booting.

their manual is hopeless, not even saying which socket is the flash BIOS port, this is only written on the back panel, and its a bit ambiguous a bit like Lidl pricing, where you dont know if it refers to the socket above or the socket below the text. eventually I figured it must be the socket above the text.

Based on the screenshots, it looks like when you have RTX 4060 installed inside the PC and you use MoBo display port, the benchmark actually uses both GPUs; iGPU and dGPU at once, to complete the task. That's why you see performance uplift when display is connected to the MoBo.

Laptops usually do it, utilizing both the iGPU (inside the CPU) and dGPU (dedicated GPU). But desktop PCs are also capable of doing that.

To validate it once and for all:
* Remove RTX 4060 from the PC, completely

I dont want to remove it, as I cannot access the latch without removing the CPU cooler, and then I think I would have to apply the thermal paste again, its not worth it.

* Have monitor hooked to MoBo and run the presets again

This way, there is no way, that PC can use RTX 4060 alongside Radeon graphics when you benchmark your iGPU.

I should have benchmarked it before I installed the 4060, its too much hassle to benchmark it now.

There is a way to disable iGPU from UEFI when you have dGPU installed but it is difficult to do. Better to remove RTX 4060 completely from the system, easier this way (you doesn't have to dabble in UEFI and there is always a possibility that disabling from UEFI doesn't actually work).

I also see that you do not have latest GPU drivers. Benchmark shows your GPU driver version to be 536.67 while the latest is 552.22
Link: https://www.nvidia.com/download/driverResults.aspx/224154/en-us/


As long as you can boot to OS, you're good.

at the moment the machine is marooned!

I will have to switch BIOSes, and I dont know if the current BIOS is unsalvageable.

can you flash the "other" BIOS? or only the current BIOS?

UEFI flash shouldn't take longer than ~5 mins.

the low level flashing via the Flash Bios Button, has had the led flashing for more than 90 minutes now, so this BIOS is kaput.

I will probably need to action the warranty, as I have done everything by the book, where it is obviously defective goods.

As of why it hangs or doesn't display anything when it completed - i can't tell.

it must have been kaput right from the start, probably a returned item which they offloaded onto me

Though, since there are high risks involved with UEFI flash (update), i reserve it as a last resort option and only if you know, for a fact, that newer UEFI version fixes the specific issue you're having.

it was MSI who told me to reflash the UEFI, I wouldnt decide to do that without advice,

and it basically cannot be reflashed,

Oh, when you boot to UEFI, does the version name now say it to be the latest? The one you flashed (updated) it to?

no chance!

 

[Aeacus]

I will leave the led flashing for the moment, I think I will have to use the multi-BIOS switch to get out of this,
see p 55,

what is the way to do this?

From MoBo manual, page 71:
Updating BIOS with Flash BIOS Button
1. Please download the latest BIOS file that matches your motherboard model from
the MSI® website.
2. Rename the BIOS file to MSI.ROM, and save it to the root of the USB storage device.
3. Connect the power supply to CPU_PWR1 and ATX_PWR1. (No need to install CPU
and memory.)
4. Plug the USB storage device that contains the MSI.ROM file into the Flash BIOS
Port on the rear I/O panel.
5. Press the Flash BIOS Button to update BIOS, and the LED starts flashing.
6. The LED will be off when the process is completed.

where the manual doesnt even show where this "flash BIOS" socket is

Sure it does.

Page 11, item #11 is the flash BIOS USB port, type-A.
It is located just above USB type-C port.

should I switch the machine off at the mains, then push the switch to the other BIOS, then power up and boot the machine?

how will I fix the current BIOS when I switch to the other one, or will this one be up in smoke?

Flipping the PSU switch should suffice in terminating power to the PC.

So, flip the PSU switch (you can disconnect from the mains as well), wait for ~5 mins, plug the power cord back in (or flip the PSU switch) and power on. It is possible that MoBo could continue flashing the BIOS.

If no image is seen and the Flash BIOS button doesn't blink, things get hairy.
Power down, flip PSU switch, remove USB thumb drive with MSI.ROM.
Flip PSU switch to ON, power back up. Look if you get an image.

If 0 life and no image, power off, flip PSU switch. Open up PC case and move the dual-BIOS switch to position B (it should be in position A).
MoBo manual, page 55 for further reading.
After that, flip the PSU switch to ON and power on. Dual-BIOS LED above the switch, should change from white LED to red LED. Look if you now get the image.

 

[Aeacus]

can you flash the "other" BIOS? or only the current BIOS?

You should be able, but this means you have to download same or earlier version of the BIOS, compared to what the MoBo came with.

Then, power off, flip PSU switch, move dual-BIOS switch to position A, flip the PSU switch to ON.
Prepare the USB thumb drive with MSI.ROM in root directory (as described in MoBo manual). Format the USB thumb drive in FAT32 format. Don't format the file system into FAT (aka FAT16). Also, better to have just single file on the root directory, rather than two or more (like you had).

DO NOT POWER ON THE PC, plug the USB thumb drive into item #11 USB port, press Flash BIOS button on the back of the MoBo. LED should start blinking. Once LED stops blinking (goes off), BIOS flash is complete.

the low level flashing via the Flash Bios Button, has had the led flashing for more than 90 minutes now, so this BIOS is kaput.

You have 2nd BIOS to save you.

I will probably need to action the warranty, as I have done everything by the book, where it is obviously defective goods.

Warranty doesn't cover user made mistakes. It only covers manufacturing defect.
You knowingly going through the process of updating BIOS, while very well knowing the risks, is not something MSI is responsible of.

I should have benchmarked it before I installed the 4060, its too much hassle to benchmark it now.

Yeah, you should have.

it must have been kaput right from the start

No, it was not. BIOS worked. You got your build running. OSes installed. It worked fine.

it was MSI who told me to reflash the UEFI, I wouldnt decide to do that without advice,

Well, you can contact MSI again and tell them what happened. Unsure if MSI rep would accept the RMA or not. If they do, they provide you the shipping parcel and you have to pay for the shipping to them. Easier would be using 2nd BIOS and not touch the BIOS update anymore, at all.

 

[35below0]

then the really frustrating fiasco: I decided to flash the BIOS

Richard, you were warned this was risky.

Why did you decide to flash BIOS? It's rhetorical.
Immediately before your latest post Aeacus warned you not to do this because risk is high, and said to only do it if you know for a fact the new version fixes your problem.

MSi support replied with a generic reply. They told you to use the latest BIOS just to get you to go away. There was no fix in this version and only a prayer it might somehow, some way solve a problem.

You should have known better. This has been a very convoluted and overly complicated process to go through. I hope it wasn't all just to lead you to a bricked motherboard.

You can return the motherboard since it was MSi support that urged you to flash it. Sort of.


You have shown you are not one to be held back by rules, warnings, advice or common sense. Sometimes that is fine, like the left handed guitar conversation we had.
But if something isn't safe, like firearms or medicine then having an indipendent mind about it is as wise as jumping off a cliff.

But that's besides the point.
Helping you complete a build and put it together should have been finished ages ago. You were interested in experimenting with lots of different things, and that's fine. Sometimes it was not what i would advise trying but there was no risk.
Guiding you through unneccessary BIOS flashes or possible bricked motherboard issues is not something i hoped to see.
You have every right to treat your possesions as your own playground but ...i wish you were more willing to keep it simple and just get from A to B, and B to C and so on, and enjoy a new PC instead of looking for new ways to get mired in OS installs and reinstalls, display tests and who knows what else

Then again, it's all part of a new PC. I guess. Going against the grain as often as you do, you might end up breaking something. Esp. taking chances with BIOS flashes.
Please don't.


What's this about the motherboard never booting properly?

 

[Aeacus]

Why did you decide to flash BIOS? It's rhetorical.
Immediately before your latest post Aeacus warned you not to do this because risk is high, and said to only do it if you know for a fact the new version fixes your problem.

This wasn't actually the 1st time when i warned about BIOS updates.

In this very topic, 1st page, in my 7th reply (reply #16), i wrote this:

Some high-end MoBos can have dual BIOS, as a redundancy, while most MoBos only have one BIOS. And if BIOS update fails, without having backup BIOS, MoBo will be bricked, where only fix is MoBo replacement.

Since most MoBos have only one BIOS, updating BIOS isn't something you can do on a whim. BIOS update, as such, is only viable when you know for a fact that newer BIOS fixes the specific issue you have with your PC. (E.g if i want to use Kaby Lake CPU with my Z170 chipset MoBo, i need to update my MoBo BIOS.) If there are no issues, there is no reason, what-so-ever, to update BIOS. Just because you "can" update BIOS doesn't mean that you "have to". Or in other words: "If it ain't broke - don't fix it."

Most MoBos never get their BIOS updated and work fine until they are obsolete. Also, do note that when BIOS update would be interrupted for whatever reason (e.g power loss), your MoBo will be bricked since PC won't run when MoBo has corrupt BIOS. And only fix is MoBo replacement. If lucky, you can roll back the BIOS update, if your MoBo has that feature. Or use 2nd BIOS which some high-end MoBos have.

Anyone who are updating their BIOS on a whim and brick the MoBo, will be paying stupid tax afterwards.

 

[35below0]

It probably got buried. There was a lot of information and advice. Also it was something for the future, once the assembly is complete.
I wish Richard didn't resort to it.

I'm interested in this motherboard not booting properly problem though. Could be RAM training or could be that chipset issue? It is a high end z790 MSi board.
Am not sure.

It's a specific problem that maybe requires it's own thread.

 

[Aeacus]

It is a high end z790 MSi board.

Not an Intel, but instead AMD. MSI MEG X670E ACE,
specs: https://www.msi.com/Motherboard/MEG-X670E-ACE

Unless... R9 7950X3D now works with Z790 chipset MoBo as well.

 

[Richard1234]

progress update:

I eventually forced shut the PC at the mains. waited many minutes to be sure any memory including firmware memory was cleared, then mains on, and tried the flash BIOS button again, and the led was flashing ongoingly.

more than 8 hours later after breakfast, the led was still flashing, the word flashing is ambiguous here as could mean the led flashing, or the flashing of the BIOS. anyway, the led still flashing, forced the machine off directly at the mains. waited many minutes, then mains on, waited several minutes, then tried booting. and this time booted really quickly to Windows, so maybe the flashing has worked. will have to check later as I have a shopping expedition later.

if it has worked, the mobo isnt functioning the way the manual says, which does show incompetence of MSI in not properly documenting, or the mobo is defective. because p71 the 2nd item 6. says the led will be off when the process is completed. FALSE.

From MoBo manual, page 71:
Updating BIOS with Flash BIOS Button
1. Please download the latest BIOS file that matches your motherboard model from
the MSI® website.
2. Rename the BIOS file to MSI.ROM, and save it to the root of the USB storage device.
3. Connect the power supply to CPU_PWR1 and ATX_PWR1. (No need to install CPU
and memory.)
4. Plug the USB storage device that contains the MSI.ROM file into the Flash BIOS
Port on the rear I/O panel.
5. Press the Flash BIOS Button to update BIOS, and the LED starts flashing.

done all that, and its as described

6. The LED will be off when the process is completed.

false! even after 8 hours the led was still flashing. but maybe it has reflashed the rom. this shows either a defective mobo, which then means poor quality assurance, or incompetence of incorrect documentation.

Sure it does.

Page 11, item #11 is the flash BIOS USB port, type-A.

its not on p11!

p11 is "connecting the power connectors"

but I did find it mentioned on p 24, but it is well hidden, it ought to be labelled on the diagram.

It is located just above USB type-C port.

there are 3 USB C ports, and "above" is highly relative! relative to the diagram? relative to the tower case?

Flipping the PSU switch should suffice in terminating power to the PC.

So, flip the PSU switch (you can disconnect from the mains as well), wait for ~5 mins, plug the power cord back in (or flip the PSU switch) and power on. It is possible that MoBo could continue flashing the BIOS.

If no image is seen and the Flash BIOS button doesn't blink, things get hairy.
Power down, flip PSU switch, remove USB thumb drive with MSI.ROM.
Flip PSU switch to ON, power back up. Look if you get an image.

If 0 life and no image, power off, flip PSU switch. Open up PC case and move the dual-BIOS switch to position B (it should be in position A).
MoBo manual, page 55 for further reading.
After that, flip the PSU switch to ON and power on. Dual-BIOS LED above the switch, should change from white LED to red LED. Look if you now get the image.

looks like forcing the machine off at the mains, then mains on, and flash BIOS button again may have worked, it has either worked or it has done nothing. it booted much faster than it ever has, so I think it may have worked. if so, that is good, but shows the manual is unreliable.

Then, power off, flip PSU switch, move dual-BIOS switch to position A, flip the PSU switch to ON.
Prepare the USB thumb drive with MSI.ROM in root directory (as described in MoBo manual). Format the USB thumb drive in FAT32 format. Don't format the file system into FAT (aka FAT16).

somewhere I read it can be FAT or FAT32, but I cannot locate the quote. as FAT is more ancient, I would expect that to be the safest. I think FAT has been designed to be a bootstrap conformant filesystem, ie you can use a bit of the format easily directly without filesystem eg when the UEFI saves screenshots. a general filesystem might be unfeasible without an OS to hand.

Also, better to have just single file on the root directory, rather than two or more (like you had).

DO NOT POWER ON THE PC, plug the USB thumb drive into item #11 USB port, press Flash BIOS button on the back of the MoBo. LED should start blinking. Once LED stops blinking (goes off), BIOS flash is complete.

I cannot rely on the blinking stopping! will just have to leave it on for several hours to be absolutely sure.
will find out later if the BIOS got reflashed.

Richard, you were warned this was risky.

Why did you decide to flash BIOS? It's rhetorical.
Immediately before your latest post Aeacus warned you not to do this because risk is high, and said to only do it if you know for a fact the new version fixes your problem.

I didnt want to, but the machine often doesnt boot. generally rebooting doesnt work, the mobo led counter just either keeps going round and round, or it halts at some number, and never boots, not even after 30 minutes.

only with mains off, then mains on, it usually boots, but not always, sometimes it freezes up.

I have been using my 2010 Gigabyte mobo for 14 years, and never had this problem. nor the other PCs I have used, all boot first time. now the OS boot might freeze up. but not the mobo boot, ie the process up till the OS starts booting.

MSi support replied with a generic reply. They told you to use the latest BIOS just to get you to go away. There was no fix in this version and only a prayer it might somehow, some way solve a problem.

You should have known better.

MSI didnt deny there was a problem, but wouldnt help unless I flashed the BIOS.

lack of denial is generally interpreted as confession.

denial here would be "that is normal, the machine wont always boot up".


thus the mobo was in unacceptable condition, and so this either has to be fixed or the mobo junked, as I cant spend the next 10 years having to hope the machine boots each time I use it.

so the only way forwards was to flash the BIOS.

This has been a very convoluted and overly complicated process to go through. I hope it wasn't all just to lead you to a bricked motherboard.

luckily it isnt bricked.

You can return the motherboard since it was MSi support that urged you to flash it. Sort of.


You have shown you are not one to be held back by rules, warnings, advice or common sense.

Sometimes that is fine, like the left handed guitar conversation we had.
But if something isn't safe, like firearms or medicine then having an indipendent mind about it is as wise as jumping off a cliff.

following advice also is highly risky!

I was told in 2006 by a financial advisor that over any 20 year period, the stock market outdoes all investments. but in fact the FT100 didnt outdo where it was at start of 2000 until 2015, and was outdone from 2000 to 2020 by gold.

eg Aeacus said the USB C mobo port will just be CPU graphics, but in fact is faster than the 4060's DP port.

either that, or the benchmarking software is benchmarking the wrong thing. but the benchmarking software is also his advice.

the mobo manual says the led flashing will halt once the flashing is complete, well either the flashing doesnt work or the led doesnt halt flashing! so if I followed the manual's instructions I might wait 10 years!

But that's besides the point.
Helping you complete a build and put it together should have been finished ages ago. You were interested in experimenting with lots of different things, and that's fine. Sometimes it was not what i would advise trying but there was no risk.
Guiding you through unneccessary BIOS flashes or possible bricked motherboard issues is not something i hoped to see.
You have every right to treat your possesions as your own playground but ...i wish you were more willing to keep it simple and just get from A to B, and B to C and so on, and enjoy a new PC instead of looking for new ways to get mired in OS installs and reinstalls, display tests and who knows what else

yes, but from these experiments, I now know how to have Windows 10, Windows 11, Linux Mint 21.1 all booting directly from the same disk.

these experiments arent mobo tourism, its to understand a lower level of stuff which isnt documented.

Then again, it's all part of a new PC. I guess. Going against the grain as often as you do, you might end up breaking something. Esp. taking chances with BIOS flashes.
Please don't.

I did take into account that there was a spare BIOS, if there wasnt, I would have been more hesitant.

What's this about the motherboard never booting properly?

when I switch on the mains, then press the power button on the tower,

it somtimes never reaches the UEFI screen, where there is the dragon and it says press delete for the UEFI, and press ???? to reflash the rom, and ????? for boot options.

instead, the fans come on, and all the continual led colour changes, the led counter on the mobo might cycle round and round, or it might halt at some number. and this just continues forever, eg wait 30 minutes or an hour and still no boot.

this happens virtually every time with reboots, ie when you select restart on Windows. the machine powers down, then powers up, and it never reaches the UEFI options.

so its not "never" but "sometimes" from mains on, and "usually" if you reboot.

now with the possible reflashing, it booted much faster.

This wasn't actually the 1st time when i warned about BIOS updates.

In this very topic, 1st page, in my 7th reply (reply #16), i wrote this:

you warned me plenty, this is why I was hesitant. but MSI's lack of denial told me the mobo was malfunctioning, and they wouldnt help me unless I reflashed. I asked you various times and you were unable to help me with the problem that the mobo often doesnt boot to the OS.

the fact the mobo has a spare BIOS meant I could take the risk. but if that failed, I wouldnt risk flashing the 2nd BIOS.

it must have been kaput right from the start, probably a returned item which they offloaded onto me

No, it was not. BIOS worked. You got your build running. OSes installed. It worked fine.

incorrect!

it DIDNT work fine, it often doesnt boot. I would have to hope it booted each time I used the machine.

sometimes I had to power off at the mains maybe 4 times before it would boot properly.

this is like the Fiat car my parents bought in 1973, where they would have to keep trying the ignition for 5 minutes before the car would start!

as kids we used to say:

toyoyoyoyo ... for trying the ignition,
then
brrrm brrm for when the engine started!


I have used several PCs over the years, the 2004 HP Pavilion tower, the 2006 Gigabyte PC, the 2007 Fujitsu-Siemens PC, the 2010 Gigabyte, the 2023 HP Laptop, a few PCs of other people, and EVERY PC BOOTS EVERY TIME TO THE OS BOOT. the OS boot can fail, where it might freeze up.

with the Gigabyte, the moment you power on, you instantly get hardware info, with the message that press Delete to go to the early startup.

there is no waiting and no doubt.

so I know what to expect, and the mobo didnt meet my expectations. same way the mobo graphics via HDMI was WORSE than my Intel Celeron powered HP Pavilion of 2004 with ATi Radeon!

I am not going to pretend this or any hardware is better than it is, nor pretend it is worse than it is.

I will describe it as it is, good or bad or mediocre. because part of the value of the internet is to warn people of bad stuff, and also to point out good stuff. but if you do this dishonestly, either overrating or underrating, it deteriorates the internet.

I once was talked into an HP printer, then delays arriving, so I read amazon reviews, and these were lukewarm, with one guy saying "its alright, but not a patch on my Epson". I then cancelled the purchase as it wasnt delivered when promised, and no clear date when it would, and bought an Epson Ecotank, best decision I made with printers!

if it has been fixed by reflashing, where I have only tried this once, then the BIOS must have been corrupted, or bugged.

and the MSI person didnt deny the description.

its not good enough if some hardware or software sometimes works, sometimes doesnt.

what you might call temperamental.

it is defective if this happens.

now in the old days, eg with the Amiga 500, there was sometimes indispensible software, where there were no alternatives, and which did have bugs. but from continual use I knew where the bugs were and would step around those.

eg I found a bug even with the GNU C compiler, gcc, where it processed a certain kind of expression incorrectly. but in the modern era, gcc is the only option, so I worked around the problem.

Warranty doesn't cover user made mistakes. It only covers manufacturing defect.
You knowingly going through the process of updating BIOS, while very well knowing the risks, is not something MSI is responsible of.

it wasnt a user made mistake, I havent made any mistake. right off the bat it wasnt always booting.

the very first boot was alright. but right from day 1, it often wouldnt boot. which was a new experience for me as I have been using PCs for 20 years, and I used real Amigas from 1988 till 2003, which is 15 years, and never had such a problem. 2004 onwards I started using emulated Amigas, as that enabled me to use a much higher res screen, AND have say 1 gigabyte of memory. my real Amiga 1200 had 6MB, 2MB factory installed, and 4MB expansion.

Its a fully valid warranty claim, legally they either have to replace it with the same or better, OR issue a full refund. they can refuse to refund, but only if they replace it.

to deny a refund, they would have to prove I made a mistake, the law in Britain is always on the side of the consumer, the consumer doesnt have to prove they didnt make a mistake, the firm has to prove they did make a mistake.

so without scarring of the machine, its not viable for them to do this. I am ultra careful in the handling of all equipment, so there are no scars on this machine.

I only went to update the BIOS because MSI by not denying the description was a problem, tacitly confessed to there being a problem, and they werent going to help further unless I reflashed the BIOS.

eg if I said: I bought one of your apples, and it was bitter.

and you reply: you need to boil the apple for one hour.

you have tacitly confessed that the bitterness is a problem, as you havent denied this.

you are essentially saying an apple being bitter is a problem, and some further processing is necessary to remedy the bitterness.

whereas if I said: I bought one of your apples, and it was a delicate sour taste,

and you reply: that is how better quality apples are meant to be.

then you are denying there is a problem.

you are essentially saying an apple being sour isnt a problem.

 

[Richard1234]

I shut down via Windows, then later to write this, I tried powering on at the PC and it wouldnt boot, with the led at the top of the tower flickering away like crazy,

so then mains off, then powered on, and it booted nicely.

Well, you can contact MSI again and tell them what happened. Unsure if MSI rep would accept the RMA or not. If they do, they provide you the shipping parcel and you have to pay for the shipping to them. Easier would be using 2nd BIOS and not touch the BIOS update anymore, at all.

I am experienced at getting refunds, I know how the system works.

as its not my fault, if the problems persist I WILL either get a replacement or a refund, first from the vendor,

if the vendor doesnt cooperate I will escalate the dispute to Barclays VISA.

if VISA doesnt resolve it, I will take the vendor to the small claims court, in most cases companies then issue a refund, because to dispute further, they have to trek to the Bristol small claims court, and its not viable for them.

most firms will only attend court via a lawyer, for technical reasons, and lawyers for court will cost them more than the amount disputed, so they have no option but to issue a refund.

now many firms will try and bluff their way out that no refund is due, and most customers will either get bluffed or not have the time to dispute further.

now if I damaged the item, then I wouldnt try to get a refund, as all firms will do a visual on returned goods.

but eg I got a speedy no quibble refund from Alibaba for the fake 64T drives.

firms dont lose money on returned goods, they just return them to the manufacturer as defective, if the manufacturer disputes this, the firm then forwards the dispute back to the customer.

manufacturers who get a lot of refunds, dont have the time to fully dispute everything, so many things will just get a quick visual, and maybe some quick tests. they dont have the time to do a 30 minute assessment of every return.

although that mobo maybe cost me £700, it probably only cost MSI £100. its just some sand for the silicon and some minerals.

 

[35below0]

Not an Intel, but instead AMD. MSI MEG X670E ACE,
specs: https://www.msi.com/Motherboard/MEG-X670E-ACE

Unless... R9 7950X3D now works with Z790 chipset MoBo as well.

Sorry. I got lost in the shuffle. Well that eliminates the z790 problem.

THERE'S TOO MANY 7s & 9s & 0s!

 

[35below0]

I am experienced at getting refunds, I know how the system works.

as its not my fault, if the problems persist I WILL either get a replacement or a refund, first from the vendor

I think this is headed in that direction. Possibly the PSU is screwy but it's more likely that the motherboard is faulty. Maybe not, but...

Sometimes motherboards are dead in the box. Try to get ir repaired or replaced or refunded. The refund route is the simplest since you're fed up with MSi. If you want to, you can give them another chance, but i would still ask for a refund and buy a new one.

Or get whichever Gigabyte you're interested in.

 

[Aeacus]

if it has worked, the mobo isnt functioning the way the manual says, which does show incompetence of MSI in not properly documenting, or the mobo is defective.

Or your USB thumb drive is defective.

Whatever the case, it is always high risk when updating BIOS. And BIOS updates have failed in quite high degree, else-ways, i/we won't warn users about it and the risks it has, if the success rate would be 99.999% and there would be little, if any concern.

its not on p11!

p11 is "connecting the power connectors"

but I did find it mentioned on p 24, but it is well hidden, it ought to be labelled on the diagram.

Yeah, my bad. It's on page 24.

there are 3 USB C ports, and "above" is highly relative! relative to the diagram?

Once you see where the item #11 is and look the diagram provided with the manual, it is located above USB type-C port.

Which way the MoBo is rotated inside the PC case and/or which way the PC case is put on the table, is relative.

it has either worked or it has done nothing.

Validate it by checking the UEFI version from UEFI.

somewhere I read it can be FAT or FAT32, but I cannot locate the quote.

From MSI support page:

1. Connect the USB flash drive to the computer, right-click on the USB flash drive, and select “Format.”

Note: Formatting will clear all data contained within the USB drive, and this step requires converting the USB format to FAT32. Using a USB flash drive with other system formats such as NTFS or exFAT will not allow you to update the BIOS.

If your flash drive is already in FAT32, you don’t need to format it and directly continue from Step 3.

Source: https://us.msi.com/support/technical_details/DT_BIOS_Update

I didnt want to, but the machine often doesnt boot. generally rebooting doesnt work, the mobo led counter just either keeps going round and round, or it halts at some number, and never boots, not even after 30 minutes.

How come you say this now, after (failed) BIOS update? I don't recall you saying it to be the issue earlier in this topic.

following advice also is highly risky!

I was told in 2006 by a financial advisor that over any 20 year period, the stock market outdoes all investments. but in fact the FT100 didnt outdo where it was at start of 2000 until 2015, and was outdone from 2000 to 2020 by gold.

eg Aeacus said the USB C mobo port will just be CPU graphics, but in fact is faster than the 4060's DP port.

either that, or the benchmarking software is benchmarking the wrong thing. but the benchmarking software is also his advice.

the mobo manual says the led flashing will halt once the flashing is complete, well either the flashing doesnt work or the led doesnt halt flashing! so if I followed the manual's instructions I might wait 10 years!

Your example of financial advisor is poor one, since no-one can tell what future brings. So, any advice regarding what may happen in the future, is uncertain.

By default, iGPU should be disabled when dGPU is installed. This is so with most desktop PCs and for Intel. Now, i don't know if you enabled iGPU from UEFI once you installed your dGPU or not. Or perhaps it is automatically enabled when you plug monitor to MoBo, despite having dGPU. But PC can use both GPUs at the same time. Many laptops do it already.

Regarding Unigine Superposition; you asked a program to measure GPU performance. Unigine Superposition is the best GPU benchmark program there is. But there are others as well, like 3DMark, PassMark, Cinebench.

With Flash BIOS LED blinking, i'd contact MSI and ask them what the issue is.

instead, the fans come on, and all the continual led colour changes, the led counter on the mobo might cycle round and round, or it might halt at some number.

This much i also did say. In my very 1st reply in this topic. Namely:

If you don't see any image on monitor, despite MoBo being powered on, look what code the DEBUG LED is showing. To locate the DEBUG LED, look MoBo manual page 61.
Pages 61 to 65 has a chart of all the DEBUG LED values and what they mean.

Now, it may get seemingly stuck on either 15, 16, 17, 18 or 2E, but wait since this is memory training and takes a bit of time. Shouldn't take more than 5 mins though. MoBo can do the memory training twice, but once POST is complete, PC should boot directly to UEFI (BIOS).

I take that you didn't memorize in general, what info MoBo manual contains. E.g it listing DEBUG LED codes.

I asked you various times and you were unable to help me with the problem that the mobo often doesnt boot to the OS.

Would you kindly point out where you asked me this?

if it has been fixed by reflashing, where I have only tried this once, then the BIOS must have been corrupted, or bugged.

IF given a chance that BIOS A is bugged, why not try with BIOS B and look if it fixes your issue? Far better option than trying to update the BIOS.
This is, actually, something you can also try. Won't hurt to try, since there is no risks involved (other than your BIOS settings going away since BIOS B doesn't have user defined settings saved in it).

as its not my fault, if the problems persist I WILL either get a replacement or a refund

Before MoBo RMA, i'd call PC repair tech on site (or haul PC to PC repair shop) for diagnostics. Since what we can do here, over the forums, is little, compared to actually being on site and having access to the PC.

THERE'S TOO MANY 7s & 9s & 0s!

Yes, there are. Even i, sometimes, get the chipset naming wrong. Especially B-series. E.g B660 vs B650. Which is Intel, which is AMD? Tough to tell. So, i usually look it up to be sure. But Z-series is Intel, H-series is Intel, A-series is AMD and X-is also AMD, unless you count older Intel X99, X299 server chipsets.

Or get whichever Gigabyte you're interested in.

I don't think it would be any better with Gigabyte. Knowing Gigabyte history, it actually could be worse.

Also, not many options to choose from either, only 5x X670 chipset MoBos. Or 3x, if looking at X670E chipset.
Lineup: https://www.gigabyte.com/Motherboard/All-Series?fid=2735

 

[35below0]

By default, iGPU should be disabled when dGPU is installed. This is so with most desktop PCs and for Intel. Now, i don't know if you enabled iGPU from UEFI once you installed your dGPU or not. Or perhaps it is automatically enabled when you plug monitor to MoBo, despite having dGPU. But PC can use both GPUs at the same time. Many laptops do it already.

This may not be user choice. The setting is in BIOS/UEFI but what is default is anybody's guess.
In my case both the iGPU and dGPU are enabled. I could and probably should disable the iGPU because the dGPU does all the work all the time anyway, and the iGPU does nothing except hog up to 32Gb of RAM.

But nothing is broken and i do not relish a trip to BIOS that may or may not screw something up, so i'll leave it be for now.

Anywa, unless the iGPU is specifically disabled in BIOS, it will be available and it's up to the OS to do something with it. Like give it RAM it doesn't need. *sigh*
Whether or not a user will make any changes in BIOS, and what is or isn't default depends on the motherboard and the person building/using the PC.

Even i, sometimes, get the chipset naming wrong. Especially B-series. E.g B660 vs B650. Which is Intel, which is AMD? Tough to tell. So, i usually look it up to be sure. But Z-series is Intel, H-series is Intel, A-series is AMD and X-is also AMD, unless you count older Intel X99, X299 server chipsets.

I presume you know but just to make sure, i wasn't being serious. I just screwed up. ^^

Not to be too anti-AMD, but their naming conventions suck. Except for the GPUs, i can handle those. Mostly. They suck less, let's put it that way.

 

[Richard1234]

ok, I just had a brief amount of time in the afternoon after shopping, and did some troubleshooting, not enough time to post here about this.

but it looks like the BIOS wasnt updated, comparing a new screenshot with the old ones, its identical. but this may be the comment later that the flash drive was FAT and not FAT32. so I will try FAT32 last thing tonight.

tomorrow we have some elections in Britain, but not a major election, eg local police commissioner, and probably local council, where I will go and vote, but hopefully the rest of the day working on the computer.

the thing is Tuesday the entire day on the computer and didnt accomplish much, eg I got quagmired with the Roxio technical support where today's version of their bluray writing software would take many hours to download, but only download + install would work, they have now fixed that where if already downloaded, I can install without downloading again, requires the licence key linked to my name and email address. I have a different bluray writer from the quad layer writer drive, where with both I opted into upsell, but this takes a lot of time to download. I also got quagmired with customer support for some non computer electronics, where they sold me a replacement part.

sometimes it will boot the OS really fast eg some 20 seconds, but the problem again that it often wont boot.
eg I easily had to try and fail to boot 7 times to boot this time to post this message.


I think their advice is unwise, where they say:

1. download latest BIOS file 7D69v1E
2. reflash
3. clear CMOS

when in fact as also suggested by you later, you should go for the easiest things first, either

a. try the other BIOS, which is what you suggested. because if this is also problematic, it means the problem maybe isnt the BIOS. whereas if it works fine, then the other BIOS is the probable problem.

b. clear CMOS on the existing BIOS. because if this fixes it, then no need to reflash the BIOS.

instead their advice is to go in with maximum force.

same way if a machine doesnt function, first check the power socket, by plugging something else in. then check the plug eg a fuse. then check the PSU. only then to consider the machine itself.

if the power socket doesnt work, check the neighbouring ones, then check the fusebox.

various things on my 1985 2nd hand hifi didnt work, eg the cassette deck didnt work. so I took it to a repair shop, and the guy tested it and said everything fine. but it didnt work, so I took in the amplifier. the guy tested it and said the DIN socket to the cassette deck was fine. but it didnt work. eventually I decided to try replacing the DIN cable, and now it worked perfectly!

right now also tricky to get the UEFI to load, so the fans are at full blast. I tried the clear CMOS button at the back, and first reboot, I got a screen of text which I havent seen before, with some options but not enough time to read. then unable to boot. repeatedly unable to boot, even with mains off then on. eventually have rebooted without an option to go to the UEFI.

anyway, this session I will reformat the flash drive to FAT32, and try again last thing tonight, just in case to leave it running overnight before quitting.

recently before the failed reflashing, I couldnt access desktop folders. I eventually did a reset of Windows, and it looks like some MSI software is now not installed, so have been reinstalling those. also had to reinstall Firefox.

also with that, I specifically opted in to restore points, and created one. but after the reset, the restore points all gone, where I couldnt undo the reset.

Or your USB thumb drive is defective.

Whatever the case, it is always high risk when updating BIOS. And BIOS updates have failed in quite high degree, else-ways, i/we won't warn users about it and the risks it has, if the success rate would be 99.999% and there would be little, if any concern.

I did follow this advice against reflashing, in that I only did it as a last resort, and very cautiously. with my 2010 PC, Gigabyte once said to reflash, and I couldnt figure out how to do that, as I wasnt aware there was a manual. and so I didnt reflash. but I wasnt exercising any caution as I didnt know it was high risk.

I think a lot of people dont know about the mobo manual, and when advised to reflash will give up!

Yeah, my bad. It's on page 24.


Once you see where the item #11 is and look the diagram provided with the manual, it is located above USB type-C port.

Which way the MoBo is rotated inside the PC case and/or which way the PC case is put on the table, is relative.


Validate it by checking the UEFI version from UEFI.

yes, it is unchanged.

somewhere I read it can be FAT or FAT32, but I cannot locate the quote. as FAT is more ancient, I would expect that to be the safest. I think FAT has been designed to be a bootstrap conformant filesystem, ie you can use a bit of the format easily directly without filesystem eg when the UEFI saves screenshots. a general filesystem might be unfeasible without an OS to hand.

1. Connect the USB flash drive to the computer, right-click on the USB flash drive, and select “Format.”

Note: Formatting will clear all data contained within the USB drive, and this step requires converting the USB format to FAT32. Using a USB flash drive with other system formats such as NTFS or exFAT will not allow you to update the BIOS.

If your flash drive is already in FAT32, you don’t need to format it and directly continue from Step 3.

From MSI support page:


Source: https://us.msi.com/support/technical_details/DT_BIOS_Update

ok, I will try this, setting up the drive to FAT32 this session, and trying another reflashing last thing tonight.

the BIOS file is exactly 32 Megabytes, I dont know if such a large size will take a lot longer to reflash?

I didnt want to, but the machine often doesnt boot. generally rebooting doesnt work, the mobo led counter just either keeps going round and round, or it halts at some number, and never boots, not even after 30 minutes.

How come you say this now, after (failed) BIOS update? I don't recall you saying it to be the issue earlier in this topic.

I'd have to trawl through to find where I mentioned this, I did find these mentions:

(A)

one other thing, I emailed MSI technical support, that the mobo often doesnt boot properly, and they said to download the latest 7D69v1E BIOS from the MSI website, and use the M-Flash function to update it.
to then clear the CMOs, to enter BIOS and press F6 to load optimized defaults, then press F10 to save and reset to test again.

(B)

also when booting to win 10 via the win 11 boot, the machine sometimes wont boot, to fix this I have to power off at the mains then repower.

(C)

at the moment I have wasted a lot of time trying to boot this machine, where it wont boot, I think I may be getting this under control, where the correct peripherals have to be attached and on and you may have to press delete repeatedly immediately on powering on.

Your example of financial advisor is poor one, since no-one can tell what future brings. So, any advice regarding what may happen in the future, is uncertain.

its just an example of a financial expert, the advisor at Yorkshire Building Society, telling me with all certainty that the stock exchange outdoes all other investments over any 20 year timeframe. a few years before the stock market crashed!

I have gotten into major jeopardy at certain points in time by blindly following advice of experts or authority figures.

in Germany, a lot of people have their feet and legs amputated, because of the medical advice.

in fact that isnt necessary, but is because they arent fixing the causes of problems, where a person's foot or leg deteriorates because of something the person is taking, either food or medicines, where if they changed what they take, the problem will resolve.

eg there was a guy in a hospital, who looked perfectly healthy, but then I noticed his foot was missing. he said he was there for big dialysis sessions each week. this kind of thing relates to the german obsession with very sour foods, which causes kidney problems, they then remove the kidney, and eventually amputate. when all they have to do is stop eating sour foods!

By default, iGPU should be disabled when dGPU is installed.

you need to clarify what iGPU an dGPU are, and also maybe some screenshots of say an MSI UEFI setup where the iGPU would be disabled, or other screenshots if not from the UEFI.

you probably explained about iGPU but it would be work to locate that, it might be in the earlier topic!

This is so with most desktop PCs and for Intel. Now, i don't know if you enabled iGPU from UEFI once you installed your dGPU or not. Or perhaps it is automatically enabled when you plug monitor to MoBo, despite having dGPU. But PC can use both GPUs at the same time. Many laptops do it already.

Regarding Unigine Superposition; you asked a program to measure GPU performance. Unigine Superposition is the best GPU benchmark program there is. But there are others as well, like 3DMark, PassMark, Cinebench.

With Flash BIOS LED blinking, i'd contact MSI and ask them what the issue is.

it does look like it wasnt flashed, so maybe this is why it kept flashing, the flash BIOS button is very low level, where it doesnt need a CPU, so the interface is very rudimentary.

instead, the fans come on, and all the continual led colour changes, the led counter on the mobo might cycle round and round, or it might halt at some number. and this just continues forever, eg wait 30 minutes or an hour and still no boot.

this happens virtually every time with reboots, ie when you select restart on Windows. the machine powers down, then powers up, and it never reaches the UEFI options.

so its not "never" but "sometimes" from mains on, and "usually" if you reboot.

This much i also did say. In my very 1st reply in this topic. Namely:


I take that you didn't memorize in general, what info MoBo manual contains. E.g it listing DEBUG LED codes.

it seems often to halt at A6, which is "SCSI detect".

I asked you various times and you were unable to help me with the problem that the mobo often doesnt boot to the OS.

Would you kindly point out where you asked me this?

it would be work to locate where I might have put this as a question, but I have found 3 mentions before I did the reflashing, namely (A), (B), and (C) above in this message.

IF given a chance that BIOS A is bugged, why not try with BIOS B and look if it fixes your issue? Far better option than trying to update the BIOS.

yes, I agree 100%, they gave me bad advice. this would be the least energy thing to check for first. and it would also determine the problem is the default BIOS if the other BIOS doesnt have the problems.

the 2nd lowest energy thing to try is to clear the CMOS or to load default settings, not sure if these 2 are identical?

it reminds me of with my 2004 HP Pavilion intel celeron machine, that one day the DVD drive vanished from the My Computer list of disk drives. I contacted HP customer support, and they said I have to reinstall Windows, which I did. but in fact much better just to buy a new optical drive, and install its driver. no need to reinstall windows!

their advice was terrible.

the thing is I had spent the entire day on the computer and had become fatigued and exasperated by the problems, and when fatigued we make bad decisions. its also what happens with compulsive gamblers, that they eventually get fatigued, and continually gamble more money.

This is, actually, something you can also try. Won't hurt to try, since there is no risks involved (other than your BIOS settings going away since BIOS B doesn't have user defined settings saved in it).

I will try that soon, will try reflashing with FAT32 also.

Before MoBo RMA, i'd call PC repair tech on site (or haul PC to PC repair shop) for diagnostics. Since what we can do here, over the forums, is little, compared to actually being on site and having access to the PC.

there is an Apple focussed shop in town, which fixed my Fujitsu Siemens laptop when it stopped booting. they said it was merely that the cable to the inbuilt monitor had become detached. whereas when I took it to PCWorld they told me the mobo was fried, and I had to buy a new computer!

the guy told me they have diagnostic equipment. but they have a fixed £60 diagnostic fee, even if they cant fix the problem. PCWorld had a £20 diagnostic fee! they are an example of experts giving incorrect advice, where they said it cannot be repaired. but as the laptop had cost me 470, I didnt want to abandon it so fast, and decided to get a 2nd opinion before abandoning it.

Yes, there are. Even i, sometimes, get the chipset naming wrong. Especially B-series. E.g B660 vs B650. Which is Intel, which is AMD? Tough to tell. So, i usually look it up to be sure. But Z-series is Intel, H-series is Intel, A-series is AMD and X-is also AMD, unless you count older Intel X99, X299 server chipsets.


I don't think it would be any better with Gigabyte. Knowing Gigabyte history, it actually could be worse.

Also, not many options to choose from either, only 5x X670 chipset MoBos. Or 3x, if looking at X670E chipset.
Lineup: https://www.gigabyte.com/Motherboard/All-Series?fid=2735

naming is a skill, and although Intel create very advanced hardware, their naming skills are rubbish.

in both maths and computer programming, naming is all important, as both subjects involve a LOT of naming, eg with programming, each subroutine needs a name, as do many other things.

if you dont get the naming right, the system becomes unworkable, as people wont remember most names.


bad naming leads to confusion, and your comments above are exactly this confusion.

good naming leads to names that you will NEVER forget. eg with AmigaOS, the programming command to open a window on the Workbench (AmigaOS desktop) is OpenWindow(), you will NEVER forget that!

you get this with cars, eg my car is a "Peugeot 308 SW SR HDI 92 FAP (D1)"!

I will never remember that, which means the car will never get recommended.

whereas if I say a Renault Megane, that is easy to remember.

you get this also with book titles, eg I have a great book on demographics called "Zero Hour", easy to remember title.

but I have a lot of books, which I cannot remember the title of.

 

[Richard1234]

SeaSonic M12II-620Bronze (replaced after 12! years)

 

[Richard1234]

From MSI support page:

1. Connect the USB flash drive to the computer, right-click on the USB flash drive, and select “Format.”

Note: Formatting will clear all data contained within the USB drive, and this step requires converting the USB format to FAT32. Using a USB flash drive with other system formats such as NTFS or exFAT will not allow you to update the BIOS.

If your flash drive is already in FAT32, you don’t need to format it and directly continue from Step 3.

Source: https://us.msi.com/support/technical_details/DT_BIOS_Update

ok, this has worked!

I reformatted to FAT32, then put just the BIOS file renamed as MSI.ROM,

powered down, mains off, waited, mains on, waited, then pressed flash BIOS button with flash drive in the flash BIOS socket,

the led flashes, then the mobo lights and fans come on, which they didnt with the earlier reflashing attempt.

after about 6 minutes, the flashing stops.

I think the machine powered down, then powered up, but didnt boot anywhere.

I eventually powered it down.

powered it up, but it didnt boot, getting stuck at A6.

repeated mains off, then mains on, power on,

and eventually it booted to Windows, no UEFI option.

as unable to get to UEFI, unable to snapshot or examine what BIOS it has.

Now I havent tried clearing the CMOS yet. also MSI still havent emailed a reply. I have to put a Trust Pilot review of MSI at some point, to warn people of their inadequate manual, and dubious customer support.
maybe put an amazon review.

eg the manual doesnt mention that it has to be FAT32. Really they should supply the manual with the mobo.

the mobo cost some extortionate amount, I expect a bit of service for that kind of price!

 

[Aeacus]

the BIOS file is exactly 32 Megabytes, I dont know if such a large size will take a lot longer to reflash?

That may take a while.

Also, i'm unsure of the firmware update process. Namely, does it format the storage capacity and then writes the new one into empty space, OR does it overwrite all the data in BIOS chip. But this much is known from MoBo manual, that BIOS chip capacity is 256 Megabits, which = 32 MB.

I'd have to trawl through to find where I mentioned this, I did find these mentions:

Right, inside the progress reports, which i didn't read fully. You talked there about Windows installs and issues with OS, so, i mostly skipped over those. I'm specialized on hardware and it's issues, so, i try to help out hardware wise. Software wise, i better leave that to someone else who knows better.

you need to clarify what iGPU an dGPU are

I have clarified it several times. This time, write it down;

iGPU - internal GPU - the one inside the CPU - your Radeon Graphics. It is also known as "IGD" - integrated graphics device (IGD naming is used within your UEFI)
dGPU - dedicated GPU - your RTX 4060
eGPU - external GPU - usually used with laptops to give them more "oomph" (more graphical processing power)

and also maybe some screenshots of say an MSI UEFI setup where the iGPU would be disabled

Located on this screenshot of yours:

http://www.directemails.info/tom/mobo/uefi/MSI_SnapShot_06.bmp

Looks like iGPU is enabled and is also default boot option.

SeaSonic M12II-620Bronze (replaced after 12! years)

Brain fart?

Seasonic M12II-620 - if not the EVO version, then that's one ancient PSU. Predating even Seasonic S12II series. Talk about a trip down memory lane.

Though, M12II series, back in the day (around 2004 or so), was actually a good PSU lineup. It's successors (in 2009) were S12II and M12II EVO series. Both of which i've used for many years. And both of them, back in the day, were also good PSUs. Heck S12II-520 was the best group-regulated PSU ever made. Even today, many modern day DC-DC regulated PSUs (namely the cheaper/lower quality ones) can't surpass what S12II-520 achieved back in 2009.

Interesting fact: While M12II-520 EVO and M12II-620 EVO were group-regulated, then M12II-750 EVO and M12II-850 EVO were DC-DC regulated. A strange thing to see within same PSU lineup.

powered it up, but it didnt boot, getting stuck at A6.

repeated mains off, then mains on, power on,

and eventually it booted to Windows, no UEFI option.

as unable to get to UEFI, unable to snapshot or examine what BIOS it has.

Probably need to clear CMOS. Better to do it from jumper pins. Latter is written in page 54.

Found a topic with very same issue,
topic: https://forums.tomshardware.com/thr...tect-prevents-me-from-accessing-bios.3750132/

 

[35below0]

Seasonic M12II-620 - if not the EVO version, then that's one ancient PSU. Predating even Seasonic S12II series. Talk about a trip down memory lane.

That is my old PC's PSU. Bought in 2012. Replaced it last month with a G12 GM-550 that is not nearly as good. For an old Ivy bridge PC that isn't important, it'll do.

Though, M12II series, back in the day (around 2004 or so), was actually a good PSU lineup. It's successors (in 2009) were S12II and M12II EVO series. Both of which i've used for many years. And both of them, back in the day, were also good PSUs.

I remember wanting a good PSU back then. This one was not cheap at all and it's proved itself over time.
That PC never worked too hard but i would leave it on for months at a time. Though, i'm not sure if switching it on/off may be worse?

The exact model was SS-620GM Active PFC F3. And i have no idea what any of that means.

Heck S12II-520 was the best group-regulated PSU ever made. Even today, many modern day DC-DC regulated PSUs (namely the cheaper/lower quality ones) can't surpass what S12II-520 achieved back in 2009.

I didn't know too much about that. People whose knowledge i respected recommended SeaSonic and i went with their advice.
I don't remember M12II, or even S12IIs. It was 2012 so the details have faded. I can't remember what else was being considered only that i took a long time to pick components.
I might dig up the old receipt just to see what it cost. I remember it was the 2nd or 3rd most expensive component in that PC. And i was happy for it really, because i wanted a safe one.

Interesting fact: While M12II-520 EVO and M12II-620 EVO were group-regulated, then M12II-750 EVO and M12II-850 EVO were DC-DC regulated. A strange thing to see within same PSU lineup.

Pretty unusual, yeah. Wonder why they didn't split the lineup? Maybe because 750w and 850w were a LOT of w back in the day. A higher end lineup consisting of DC-DC regulated units, that isn't going to sell so well due to a high price, maybe didn't seem like a great idea at the time.

But it is interesting.

Edit
The G12 GM550 is fine for what it is. A bottom of the barrel PSU. Is it safe? Yes. Is it anything else? No, but what do you want for .... er. $90??
Ok, it's not even cheap.

Well, it's supposed to be a cheap and safe PSU. I find it far too noisy but beggars can't be choosers. That said, if it's going for this ridiculous price, i'd much rather have a beQuiet! Pure Power 11 400w or SeaSonic FOCUS GX-750.
No really, lol @ this G12 GM550. How does it cost so much? I paid 70 euro for it, which is normal. Still too high but, eh. Europe.

 

[Richard1234]

first a progress report,

I woke up with a new idea, try the "elimination diet" for PCs,

Spoiler: expand for idea of elimination diet

elimination diets are where you determine which food if any is causing a health problem, by firstly going for a really simple safe diet, eg say carrots and rice. take that for some days, where if the problem goes, it was likely from some food you eat. you then reintroduce foods one at a time, eg one new food each day, till the problem re-emerges, then you know that food is the problem food. Doing that I found that tomatos were causing me a problem. Later I found if I removed the skins by soaking in boiling water for some minutes first, then no problem. and even later found if I wash the skins thoroughly to remove the powder which forms on tomatoes, then also no problem.

by removing all peripheral hardware and seeing if no problem, then reintroduce one peripheral, try booting and rebooting, then reintroduce another, until the problem re-emerges. if removing all and still a problem, then its not the peripheral hardware, similarly with health problems, if the really simple diet doesnt fix the problem, its probably not a food problem. you'll be surprised how many health problems are just nutritional, but some are also caused by not eating certain foods.

I then thought to maybe try a reverse elimination diet for the PC, to remove one peripheral at a time, until maybe the problem is fixed, then that last item is the cause.

the advantage of the elimination diet versus the reverse elimination diet is it determines in one step that the problem is a food or a peripheral. where the ensuing steps are to determine which food or peripheral. whereas with reverse elimination, you could spend weeks removing one item at a time, to find it wasnt any!

anyway, looking at the peripherals, I noticed the optical drive is connected with an AliBaba extender, and an AliBaba male to male adapter to the MSI SATA cable with 90 degrees.

so I disconnected the optical drive. and now it booted in something like 36 seconds to Windows. restarted, and also reasonably quick, rebooted in I think less than 1 minute. tried this a few times, each time rebooting correctly. BUT there is still a problem, that if I keep pressing delete, it doesnt boot. so I cannot get to the UEFI, but I may try the jumper tip mentioned. thus at the moment very noisy as all fans at max power.

Now is the problem the 2 AliBaba connectors? or is it the optical drive? needs a further elimination diet. so I connected the mobo SATA cable directly to the optical drive, and it boots in maybe 36 seconds.

so the problem is the AliBaba cable and connector, now I may eliminate further because the Alibaba cable is given as Sata 3, but the connector is unspecified. I think the problem could be the connector.

the label was unreadable. the only problem is that AliBaba cable is a very short one, so it wont reach the mobo. so for the moment I wont test that.

you did warn me about AliBaba cables, but in fairness I ought to test the one item which was unspecified.

with those 2 Alibaba items, the bluray drive does work, and I have burnt blurays, with the full verification option, where they are verified byte by byte to be correct.

as mentioned, there is still the problem of getting to the UEFI options.

That may take a while.

Also, i'm unsure of the firmware update process. Namely, does it format the storage capacity and then writes the new one into empty space, OR does it overwrite all the data in BIOS chip. But this much is known from MoBo manual, that BIOS chip capacity is 256 Megabits, which = 32 MB.

it took almost exactly 6 minutes.

it could be that with the BIOS chip, specific functions are at fixed places within a size limit, if that is the case, then it would be safer to overwrite sequentially, and not format first. because say the overwriting is currently:

[new1][new2].....[new100] [ being_overwritten101] [old102][old103]........

then the not yet overwritten old102, old103, ... will probably work fine, with just being_overwritten101 being incoherent.

the other MO would be to have 2 copies of the BIOS, where they put say a checksum at the end of each,

then when booting, the firmware can check if the first checksum is correct, and use that copy, but if incorrect, to check the 2nd checksum, which ought to be alright, to then use that if alright, which would be the old BIOS. but if that also corrupt, then the boot fails.

this kind of thing is a "decision", and it depends on what decision was made, whether they prioritise safety over memory efficiency. their putting of 2 BIOSes is a similar decision.

Right, inside the progress reports, which i didn't read fully. You talked there about Windows installs and issues with OS, so, i mostly skipped over those. I'm specialized on hardware and it's issues, so, i try to help out hardware wise. Software wise, i better leave that to someone else who knows better.

it is a hardware problem, but I was also talking of the Windows boot, so its a problem of information overload.

I once said to someone that such and such problem is a hardware problem not a software problem, and he replied: hardware is software, because the hardware is based on firmware which is software!

I have clarified it several times. This time, write it down;

iGPU - internal GPU - the one inside the CPU - your Radeon Graphics. It is also known as "IGD" - integrated graphics device (IGD naming is used within your UEFI)
dGPU - dedicated GPU - your RTX 4060
eGPU - external GPU - usually used with laptops to give them more "oomph" (more graphical processing power)

ok, I have noted it with the HDMI formats eg 4K!

I wont forget now because I can see i is for i.nternal, d is for d.edicated, and e is for e.xternal.

I was always mystified why in chemistry, C is for c.arbon, O is for o.xygen, but Na is for sodium and K is for potassium. until decades later in Germany I bought some fertiliser, and found that in german, sodium is Natrium, and potassium is Kalium, ie Na = Na.trium, K = K.alium! which must mean the discoverer of these elements was german!

Located on this screenshot of yours:

http://www.directemails.info/tom/mobo/uefi/MSI_SnapShot_06.bmp

ok, very obscure control. what should I set those other options to?

should I upload screenshots of the options? at the moment I cannot reach the UEFI!

Looks like iGPU is enabled and is also default boot option.


Brain fart?

Seasonic M12II-620 - if not the EVO version, then that's one ancient PSU. Predating even Seasonic S12II series. Talk about a trip down memory lane.

Though, M12II series, back in the day (around 2004 or so), was actually a good PSU lineup. It's successors (in 2009) were S12II and M12II EVO series. Both of which i've used for many years. And both of them, back in the day, were also good PSUs. Heck S12II-520 was the best group-regulated PSU ever made. Even today, many modern day DC-DC regulated PSUs (namely the cheaper/lower quality ones) can't surpass what S12II-520 achieved back in 2009.

Interesting fact: While M12II-520 EVO and M12II-620 EVO were group-regulated, then M12II-750 EVO and M12II-850 EVO were DC-DC regulated. A strange thing to see within same PSU lineup.

I commented on that, because its unusual to give a configuration item which isnt there!

but it looks like he did that because the replaced item was a legend!

Probably need to clear CMOS. Better to do it from jumper pins. Latter is written in page 54.

Found a topic with very same issue,
topic: https://forums.tomshardware.com/thr...tect-prevents-me-from-accessing-bios.3750132/

ok, will have a study of doing this.

I dont know if the MSI center is any use for controlling the UEFI, because p71 of the mobo manual says you can update the BIOS with the MSI Center (in british english we would write "centre", "center" and "meter" (versus "metre") are more logical spellings as the r is pronounced after the e).

I have to get ready now to go and vote, and I may do a small shopping excursion after that to free up time on Friday, havent decided.

.... Gigabyte z790 UD .... Gigabyte Eagle OC 4060 .... Gigabyte z77X-UD3H ....

you have several Gigabyte products! presumably the 4060 is the Gigabyte version of the RTX 4060?

Seasonic M12II-620 - if not the EVO version, then that's one ancient PSU. Predating even Seasonic S12II series. Talk about a trip down memory lane.

the thing is the moment you buy something, it is heading towards memory lane!

life is a conveyor belt, which carries brand new stuff to obsolescence.

this is what I like about pure maths, that the ideas are eternal, eg the angles of a triangle will always add up to 180 degrees, 1000 years ago, 1000 years in the future.


with some things, you can slow down the deterioration side, but generally not with in use items.

 

[35below0]

I commented on that, because its unusual to give a configuration item which isnt there!

but it looks like he did that because the replaced item was a legend!

Ok. It took me a while but i understand what Richard is talking about.

He pulled the quote

SeaSonic M12II-620Bronze (replaced after 12! years)

from my signature. Not from an actual post.

I knew this. I didn't know why he'd comment on it, or what the comment was supposed to be. I left it be.

Then Aeacus picks it up, and so we have a chat about the old PSU.

So what it was was this: Richard wanted to comment on a line in my sig about a component that i replaced. While not a legend, it worked without fault for 12 years and probably still works. I have it in a box at home and really have no intention of using it anymore. When my fever subsides i will take a walk to a nearby recycling center.

So yeah, hat tip to SeaSonic for that quality and durability.

Also, i think i'm going to edit my signature because there's too much in there. But right now i don't feel like it.
Specs don't normaly include components that have been taken out but that is an old PC, and i listed it's original components. And besides, i replaced the PSU last month so it's a pretty new change.

It's still a good enough PC. Good enough for me. If it could run 11 reasonably well or if Win 7 had support, i wouldn't need a new one for another 5-6 years.
But anyway, i'm quite pleased with the new one, and we can get back to the topic.

 

[35below0]

.... Gigabyte z790 UD .... Gigabyte Eagle OC 4060 .... Gigabyte z77X-UD3H ....

you have several Gigabyte products! presumably the 4060 is the Gigabyte version of the RTX 4060?

It is.

z77X-UD3H is the super sexily named 3rd gen intel era motherboard. I don't remember why i bought it. It did well in reviews i guess and it had ultra durable printed on it. I had a string of cheap and crappy PCs, acouple of which exploded and this time i wanted something nice.
Real quality. And durable.

I had money at the time so i built a PC that was not totally crap and totally a waste of money. The only thing that was cheaper was the monitor. And it had a dead pixel on day one, and another one two weeks later. In fact it's hard to even find a mention of that Phillips monitor online.
So yeah. I panicked because i thought i was spending too much money and cheaped out on the one thing i thought it would be ok to cheap out on.

I also never bought a GPU because i didn't need one at the time and i thought i'd pick something up later on. Never happened because the price was always just too high for me to justify the purchase, and the cheap GPUs were hardly an improvement on the Intel HD 4000.
The HD 4000 could run X3 Reunion beautifully btw, and only choked on Anno 2070 and Cities Skyline. Most of my other games were Papers, Please and FTL. So no real GPU needed.

Anyway, this Gigabyte product did well for me, so when i was shopping for a new PC late last year, i considered their lineup. I briefly considered MSi, Asus and AsRock but didn't really find what i was looking for.
What i wanted was a basic, durable and reliable motherboard that has no features that i don't need and don't want to pay for. The z790 UD is the rock bottom of the z790 lineup for Gigabyte. It wasn't even all that cheap. All the more expensive models had a few things or improvements i didn't need. This is also true for other motherboard manufacturers. I could have gone with almost any of them except the cheap AsRock and Asus ones. So i went with what i was satisfied with last time.

I'm pretty happy with my choice. The nicest surprise is the Gigabyte Control Center bundled software. I was 99% sure i was going to never touch it, but i just gave it a chance. It's purpose is threefold, but mainly it collects drivers for all components such as audio, video, etc. etc. It made no mistakes and it's made keeping drivers up to date a breeze.
Other things it does is control system/CPU fan speeds and curves, and controll RGB.

The old version of GCC was so terrible, it was flagged as malware. lol Also, it never could find anything.

And the 4060 GPU, that's Gigabyte largely because they all had the same price (more or less) so i just rolled with it. Also, i thought there would be less chance of a driver conflict if it was in a Gigabyte motherboard but that is not true. A Gigabyte GPU in a Gigabyte motherboard is just as likely to suffer from buggy drivers of other glitches as any other GPU.

The Eagle OC was the best value in Gigabyte's 4060 lineup. The Windforce is the same GPU without a factory overclock (and for a 4060 that is amazingly crucial, har har). Windforce is just their name for the counterspinning fans design anyway. The Gaming OC 4060 has a higher clock. Slightly higher clock.
There is such little difference between them, and yet the Eagle had a clearly superior price/performance.

I didn't need a GPU. The only reason i bought one at all was because when i had a PC with no GPU, i regretted it a couple of times. One game i wanted to play, UFO Aftershock, didn't want to run at all with no GPU.
The 4060 ticked enough boxes for me and was monstrously powerfull for my needs. Also doesn't draw too much power for it's performance (4070 is nearly double).


While i'm rabbiting, i may as well comment on the PSU. I was sure i was going to buy a SeaSonic. Same again please. Why not, i was satisfied.
But i happened to notice that fractal design had a line of PSUs too. Since i was buying another case from them (yes, satisfied customer. Also they make such nice cases to work with) why not PSU. Because it's not reputable, surely?
So i read some reviews and the Ion +2 Platinum did well. And it was made by a good OEM. One of the Seasonic units i was looking at was a little bit noisy. I don't know how trustworthy that part of the review process is, but the Fractal was silent.
And then the store i was buying it from cut the price 30% for black friday.

So i bought a Fractal Design Ion +2 Platinum 760w.
It's quiet. During the day, when i walk into my room, i cannot tell if the PC is on or not.
Also, the bit about the cables being velvety and soft is true!

Would i recommend it? Eh... Well, it's very good. But.
For a gamer, the lack of ATX 3.0 or 3.1 is a deal breaker.
For a non gamer, the high price is worth paying but it has to be said an equally good PSU can be had for less money, and also that for the same money a better PSU can be had. "better" in this case is a little bit untrue. You cannot really do "better", only more watts.
If you care about efficiency, and silence, and your store drops the price 30% then buy it ^^


tl;dr

It is.

 

[Richard1234]

progress report: I connected the clear CMOS jumper, I borrowed a jumper from my 2004 system IDE drive to do that! but I still cant get the UEFI settings.

maybe its the resetting of windows which has corrupted things, I am going to do the next Windows 10 or 11 install to see if that fixes things. if not, its the multi bios switch.

at the moment saving the bookmarks, passwords and history of browsers on each system, and this time saving Firefox history via the MS browsers, to save the MS browser history + passwords + favourites, then delete these, then import these from Firefox, then save these from MS Edge. will test what happens if I then load these to MS Edge on the next Windows install, then import over back to Firefox, to see if the history is re-established. will also test whether MS merges different histories. eg if I import both my XP history, and also say my 32 bit Windows 10 history (of Firefox).

Spoiler: expand for potentially off topic comments

He pulled the quote

from my signature. Not from an actual post.

I knew this. I didn't know why he'd comment on it,

I commented because it seemed a bit wild to give a configuration item which wasnt there!

the comments you give in the later message are interesting context, and give some perspective to the choices, but if things go off topic probably best to put the message in a spoiler to prevent the thread proliferating. then you can talk at leisure without annoying the moderators. my earlier topic I think got halted because it veered off topic majorly! I have put my comments here in a spoiler.

That PC never worked too hard but i would leave it on for months at a time.

that could be why it lasted so long! I think these machines need to be used regularly otherwise they rust up.

with aeroplanes, they need to be in almost continual use right from the start otherwise they have to be junked. eg if a plane was unused for some months it probably has to be junked. talking of jetliners and not the small executive jets.

I bought some stuff on ebay from a buy with a classic open topped VW beetle, and he drives it a few times each year in good weather to keep it in working condition. its in perfect condition.

I think the continual use keeps them "primed", its probably like that with PC hardware. eg my 2004 and 2006 PC have only been used a few times since I built the 2010 one, and that mostly 10 years ago, and now neither function. whereas the 2010 one has been in continual use until I got Win 10 installed on this new machine. so maybe I need to use it once a month or something to prevent it rusting up.

continually inserting and detaching items will wear out sockets, I have had to discard so many USB2 hubs over the years. I now use extenders wherever possible. but with some items, the extender attenuates the power too much. so some things have to be connected directly.

Also, i think i'm going to edit my signature because there's too much in there. But right now i don't feel like it.

I think its alright with 2 items, maybe keep it with the original components struck out, because the replacements of those will be less interesting, what is interesting is what you got in the era.

It's still a good enough PC. Good enough for me. If it could run 11 reasonably well or if Win 7 had support,

I think win 10 is the important one, as its the transition from 32 bit to 64 bit, where also you can install 32 bit win 10 which is very different from 64 bit win 10. I think they put extra work in win 10, as its where they get people to transition from 32 bit to 64 bit. some versions of Windows are more outstanding than others, I think XP was particularly good.

That PC never worked too hard but i would leave it on for months at a time. Though, i'm not sure if switching it on/off may be worse?

my guess is that you just need to ongoingly use it. it may not need to be on continually, but eg if you left it off for a few years, it may become problematic.

thus if you use it once a month for a few hours, that may be sufficient to keep it going for a long time.

I dont know how often it has to be used to last the longest.

with inkjets, they soon jam up if not used. Ecotank is much better, where you can leave it unused for a year, and it will still function fine. even HP now has their own version of Ecotank.

I had money at the time so i built a PC that was not totally crap and totally a waste of money. The only thing that was cheaper was the monitor. And it had a dead pixel on day one, and another one two weeks later. In fact it's hard to even find a mention of that Phillips monitor online.

its best to buy new, then action the warranty. if an item is very expensive, its best to have a longer warranty, maybe even pay for an optional extended warranty.

expensive items are best regarded as if they are rented, where the warranty extensions are part of the rental price. you need to evaluate the price as price per year. eg if an item is £600 and has a 12 year warranty, that is £50/year and is "cheaper" than an item which costs £150 with 2 year warranty, as that is £75/year.

the correct MO is to save up for a new one, for when the expensive item eventually becomes unrepairable. this requires some form of budgeting. with the £600 example, you'd need to save up £1 each week towards the new one.

if you buy the right expensive things, its in fact cheaper in the long run than buying cheap items. if you buy things which are too expensive you run into other problems, eg if they last too long, they eventually become obsolete, and are outdone by much cheaper items of the future.

larger firms use this MO, where they have maintenance or service contracts, where the other firm ensures everything works forever.

I had a kitchen appliance, which I got the extended warranty for, renewed each year, eventually one repair, then many months later another, and the warranty price gradually went up by a factor of 3x, then a 3rd repair, and they replaced it brand new for free! where the warranty is now back to the original low amount!

I bought a relatively expensive Epson Ecotank, I think ET-4550, probably 2017, with something like a 5 year warranty, eventually it stopped functioning, probably 2021, and they gave me a refund as a voucher for PCWorld which sells consumer electronics! basically I got it for free for almost 5 years!

the Gigabyte Control Center sounds interesting, beware that gcc also means the GNU C compiler!

going above the 4060 has the risk of needing more cooling which then means noise. with graphics cards, they will make things work with what the main market buys. if you buy things which are too high end, you can end up in a market which is too small where its not worth the companies doing things for, except at a very high price.

in Bristol you see this with cars, that most car park bays are too narrow for larger vehicles such as 4WDs and vans. the bays are designed for the main sizes, where most people go for smaller cars. mine is at the limit of what is alright. now out in the countryside, space is less of a premium, and they have car park bays which are much wider, where you can park a tractor or a truck!

i may as well comment on the PSU. I was sure i was going to buy a SeaSonic. Same again please. Why not, i was satisfied.

brand loyalty in the modern era is questionable. its best to re-assess from scratch.

things which used to have a bad reputation become good, things which used to have a good reputation become bad, sometimes its because a firm gets bought up by another firm with a different attitude. eg Peugeot now seems to be owned by a firm called Stellantis & You, who also own Alfa Romeo, Citroen, Fiat, Jeep, Vauxhall.

instead best to read reviews, do some test drives, etc. dont just buy from the same manufacturer which was good last time.

as regards your signature, is 35below0 the temperature where you live? Fahrenheit or Celsius?