Adventure: Running 8/9th gen Coffee Lake CPUs on Z170 motherboard (ASUS Maximus VIII Ranger)

[itsakjt]

No USB BIOS flashback on this board. I understand that was only available on higher end boards than this one.

So about that 9 pin SPI flash header under my BIOS chip: Is there any way to connect the CH341A device to that, with some kind of female molex connector, to flash the BIOS chip?

I had the idea that might be possible from your earlier posts.

The spring clamp gadget that clamps directly onto the BIOS chip looks janky as hell. One of the videos I watched showed the guy breaking off a little resistor beside the BIOS with that clamp, and that was even after he bent the header pins away from the chip to accommodate the clamp. My soldering skills are laughable, so I would much prefer to avoid that road.

Also, I see a bunch of chatter to the effect that all the CH341A programmers are set for 5 volt usage, but BIOS chips like mine demand flashing with 3.3 volts.

There are a bunch of sites showing how to hack the CH341A to yield 3.3 instead of 5 volts.

Do you agree that voltage issue is a significant concern, or did you just roll the dice at 5 volts and hope for the best?

This is how I connected the programmer to my Z270-K.

I used wires like this from an arduino kit. They are not perfect but they worked. Will take a bit of force to install the wires together on the header which will bend some pins slightly as well but you can bend them back. They connect perfectly on the CH341A.

It needs a fixed version of ASProgrammer to flash it properly that I found on Adamant IT channel. I will link it for you. Let me know when you have the programmer.
As for the 3.3V mod, you are right. Initially, I was not aware of it and flashed the Maximus VIII Ranger directly connecting the chip. Same goes for the Z270-K and another Galax board I modded.
So far, I have used the programmer as is (5V on data lines, tried and tested) on numerous BIOS chips and even on the Z270-K directly as you see on the photo and am yet to kill one. LOL.

 

[limited_imagination]

This is how I connected the programmer to my Z270-K.

I used wires like this from an arduino kit. They are not perfect but they worked. Will take a bit of force to install the wires together on the header which will bend some pins slightly as well but you can bend them back. They connect perfectly on the CH341A.

It needs a fixed version of ASProgrammer to flash it properly that I found on Adamant IT channel. I will link it for you. Let me know when you have the programmer.
As for the 3.3V mod, you are right. Initially, I was not aware of it and flashed the Maximus VIII Ranger directly connecting the chip. Same goes for the Z270-K and another Galax board I modded.
So far, I have used the programmer as is (5V on data lines, tried and tested) on numerous BIOS chips and even on the Z270-K directly as you see on the photo and am yet to kill one. LOL.

I have no arduino here, so my wiring will have to be by other arrangements.

I found another site suggesting that the pin pitch on these ASUS headers is 2.0 mm.

That being the case, my plan is to buy this packet of molex connectors, then cut the spring clamp wiring from the CH341A kit, and then plug those wires into one of the 2 mm 2x5 pin female connectors:

10pcs Pitch 2mm 2.0mm 2x5 Pin 10 Pin IDC FC Female Header Socket Cable Connector | eBay

Connector type: IDC. This item is 2mm pitch IDC connector. Pitch: 2mm. Gender: Female. Quantity: 10 Pcs Per Lot.

www.ebay.com

I think that should give me the best contact and continuity for the job.

Is the pin map an identical pattern between the legs of the chip and the pins of the SPI flash header?

What I mean is, looking at my photo, I think pin #1 is the bottom left leg of the chip (with the dot above it), and so is pin #1 on the SPI flash header the bottom left pin of that array of 8 pins (ie. is it immediately right of the hole that's missing a pin in the header)?

And if so, then the rest of the pin numbering would be identical between them, left to right and up and down, correct?

If that's the case and the pattern is identical, then I could just copy the wiring placement from the discarded CH341A spring clamp, and put that wiring pattern into my new 2x5 pin female connector.

[edit - I see your chip and SPI flash header were mounted at 90* to each other, so you couldn't just assume the same orientation and pin map, like I am assuming here on my board, possibly in error.]

If this plan sounds good to you, I will order the gear today.

Thank you for all your help!

 

[itsakjt]

I have no arduino here, so my wiring will have to be by other arrangements.

I found another site suggesting that the pin pitch on these ASUS headers is 2.0 mm.

That being the case, my plan is to buy this packet of molex connectors, then cut the spring clamp wiring from the CH341A kit, and then plug those wires into one of the 2 mm 2x5 pin female connectors:

10pcs Pitch 2mm 2.0mm 2x5 Pin 10 Pin IDC FC Female Header Socket Cable Connector | eBay

Connector type: IDC. This item is 2mm pitch IDC connector. Pitch: 2mm. Gender: Female. Quantity: 10 Pcs Per Lot.

www.ebay.com

I think that should give me the best contact and continuity for the job.

Is the pin map an identical pattern between the legs of the chip and the pins of the SPI flash header?

What I mean is, looking at my photo, I think pin #1 is the bottom left leg of the chip (with the dot above it), and so is pin #1 on the SPI flash header the bottom left pin of that array of 8 pins (ie. is it immediately right of the hole that's missing a pin in the header)?

And if so, then the rest of the pin numbering would be identical between them, left to right and up and down, correct?

If that's the case and the pattern is identical, then I could just copy the wiring placement from the discarded CH341A spring clamp, and put that wiring pattern into my new 2x5 pin female connector.

[edit - I see your chip and SPI flash header were mounted at 90* to each other, so you couldn't just assume the same orientation and pin map, like I am assuming here on my board, possibly in error.]

If this plan sounds good to you, I will order the gear today.

Thank you for all your help!

Here are the pin config details.
You won't need to connect pin 1, 2 and 9 for the operation. For the rest of the pins, you can find them all on the programmer.
For your board, the connector seems to be upside down so, the top right pin is pin 1 and the single pin on the other end is pin 9. It should go like this when viewed from your previous image.
9 7 5 3 1
. 8 6 4 2
Your idea seems amazing and to be honest, I was looking for something like that back in the day and that way, you can make a proper cable for use in the future as well.

Here is the link to the fixed ASProgrammer

AsProgrammer_V2.1.0.13_XiTongZhiJia.zip

drive.google.com

All the best.

 

[limited_imagination]

Awesome, thanks.

So it looks like what I thought was pin #1, is actually pin #8.

So the SPI header is flipped 180* from the orientation of the chip on my board.

OK, I will order the gear.

Probably it will take it a month to get here from China, but I will for sure follow up and report my progress here at that point.

Thanks again.

 

[limited_imagination]

So the CH341A arrived very quickly, but the 2x5 female connectors are taking their sweet time.

Being impatient by nature, I decided to try using the janky spring clamp cable.

I'm using Asprogrammer 2.1.0.13 on my old Windows 7 laptop, initially with the drivers that auto-installed by Windows, and then later with the drivers that were enclosed with the zip file containing that version of Asprogrammer.

I tried testing both with the motherboard powered up, and also unpowered.

I am pretty sure I got the janky spring clamp into position on the correct pins, but it's hard to be sure.

I have not done the 3.3 volt mod yet, deciding to start off by just seeing whether I could get the device to identify and make a backup from the BIOS chip, before hacking up the CH341A.

The power LED on the CH341A glows a nice bright red when I plug it into the USB port.

And when I click "Detect Chip" in Asprogrammer, I see the "run" LED flash a few times on the CH341A.

But I just get "IC not responding" message in Asprogrammer, every time.

Googling this error message, I see a lot of different suggestions to try different Windows drivers, different programming software, check continuity, cut the cable length to as short as feasible, and one guy said to do the 3.3 volt mod first.

But I'm throwing this out there in case anyone has a brilliant idea that I missed.

If @itsakjt could get it working without the 3.3 volt mod and using Asprogrammer 2.1.0.13 and its drivers, using the pretty long wires shown in his photo, then I strongly suspect that it's a continuity issue with the spring clamp.

I will wait for the 2x5 female molex connectors to arrive, and then try again using the header instead of the clamp, unless someone sees something obvious that I was doing wrong here.

Maybe I was supposed to select a chip size and page size first in Asprogrammer, but I thought the "Detect Chip" function should have taken care of that by itself.

 

[itsakjt]

So the CH341A arrived very quickly, but the 2x5 female connectors are taking their sweet time.

Being impatient by nature, I decided to try using the janky spring clamp cable.

I'm using Asprogrammer 2.1.0.13 on my old Windows 7 laptop, initially with the drivers that auto-installed by Windows, and then later with the drivers that were enclosed with the zip file containing that version of Asprogrammer.

I tried testing both with the motherboard powered up, and also unpowered.

I am pretty sure I got the janky spring clamp into position on the correct pins, but it's hard to be sure.

I have not done the 3.3 volt mod yet, deciding to start off by just seeing whether I could get the device to identify and make a backup from the BIOS chip, before hacking up the CH341A.

The power LED on the CH341A glows a nice bright red when I plug it into the USB port.

And when I click "Detect Chip" in Asprogrammer, I see the "run" LED flash a few times on the CH341A.

But I just get "IC not responding" message in Asprogrammer, every time.

Googling this error message, I see a lot of different suggestions to try different Windows drivers, different programming software, check continuity, cut the cable length to as short as feasible, and one guy said to do the 3.3 volt mod first.

But I'm throwing this out there in case anyone has a brilliant idea that I missed.

If @itsakjt could get it working without the 3.3 volt mod and using Asprogrammer 2.1.0.13 and its drivers, using the pretty long wires shown in his photo, then I strongly suspect that it's a continuity issue with the spring clamp.

I will wait for the 2x5 female molex connectors to arrive, and then try again using the header instead of the clamp, unless someone sees something obvious that I was doing wrong here.

Maybe I was supposed to select a chip size and page size first in Asprogrammer, but I thought the "Detect Chip" function should have taken care of that by itself.

View attachment 283010

Hi there,

On ASProgrammer, under IC, select the IC make and model. Based on the photo you shared previously, it should be from Macronix or MXIC (you can cross check with the part number).

The flash needs to be carried out when the system is completely shut down and disconnected from the mains. Removing the CMOS battery is also a good idea.
As for the clamp cable, make sure the red wire corresponds to the Vcc pin on the IC and on the programmer, corresponds to the Vcc pin. This will ensure the remaining connections are proper.

There is also one jumper on the CH341A. Can you post a photo of the programmer, one from each side, so that I can compare it with mine?

 

[limited_imagination]

OK, so it sounds like I was not using ASProgrammer correctly. I assumed "Detect Chip" would do the magic and find my IC without me having to manually enter it.

I was lucky enough to get that very clear photo, so I am pretty certain that the part number printed on my chip is as follows:

25Q128A
13EDF
99CHN
63613V

There is no 25Q128A chip that shows up on the IC list in ASProgrammer 2.1.0.13.

And the only manufacturer that has a data sheet with most of that part number is Micron, which has an N25Q128A chip, but no plain 25Q128A.

So I am unsure of the manufacturer and model of this chip to select from the IC list, or whether that list has the equivalent of my chip in it at all.

In Google searches of that part number, I saw this Winraid thread, where there are three people with Prime Z270-A boards trying to do the same thing as me: Cmd8086, kazp3r and ltxLeo. It looks like you participated in that thread quite a bit later on. Unfortunately all three of those people either gave up, or did not report their final success.

[TOOL] Easy automated Mod tool for Coffee Lake bios

Hi! Did you ever get a good BIOS for G20CI to use i7 9900k? I have 9800K

winraid.level1techs.com

kazp3r appears to have read from his BIOS chip, but corrupted it either on reading or writing to it. So I guess I don't want to take a lesson from that episode.

Cmd8086 had a chip with all digits identical to mine until his final "Y" which is a "V" on my chip. Instead of ASProgrammer, cmd8086 used ChProgrammer v.1.35, which had an entry for this chip that he said worked correctly for reads and writes on his 25Q128A chip:

Manufacturer: XMC
Chip name: XM25QH32B
Memory size: 128 MBIT
Size of page: 256
Signature JEDEC: 0x20 0xBA 0x18

That seems odd, and even if correct, that XMC chip does not show up in the IC list of ASProgrammer 2.1.0.13.

Worryingly, Petoparac and Lost_N_BIOS at one point suggested that this BIOS chip might require a 1.8 volt adapter for the CH341A to avoid damage, so I guess I am lucky that no bad things happened while I was trying to use "Detect Chip". I don't have a 1.8 v adapter at present.

However, later in the thread, Lost_N_BIOS replied to ltxLeo that if there is a "13" in the second line of the part number, then that is a 3 volt chip (like the part number on my chip), so no 1.8 volt adapter is required.

My second line is 13EDF, so if that comment is true, then I have a 3 volt chip.

At any rate, I have not yet fried my board, and in fact I am typing on that PC right now.

Photos attached, as requested.

 

[itsakjt]

OK, so it sounds like I was not using ASProgrammer correctly. I assumed "Detect Chip" would do the magic and find my IC without me having to manually enter it.

I was lucky enough to get that very clear photo, so I am pretty certain that the part number printed on my chip is as follows:

25Q128A
13EDF
99CHN
63613V

There is no 25Q128A chip that shows up on the IC list in ASProgrammer 2.1.0.13.

And the only manufacturer that has a data sheet with most of that part number is Micron, which has an N25Q128A chip, but no plain 25Q128A.

So I am unsure of the manufacturer and model of this chip to select from the IC list, or whether that list has the equivalent of my chip in it at all.

In Google searches of that part number, I saw this Winraid thread, where there are three people with Prime Z270-A boards trying to do the same thing as me: Cmd8086, kazp3r and ltxLeo. It looks like you participated in that thread quite a bit later on. Unfortunately all three of those people either gave up, or did not report their final success.

[TOOL] Easy automated Mod tool for Coffee Lake bios

Hi! Did you ever get a good BIOS for G20CI to use i7 9900k? I have 9800K

winraid.level1techs.com

kazp3r appears to have read from his BIOS chip, but corrupted it either on reading or writing to it. So I guess I don't want to take a lesson from that episode.

Cmd8086 had a chip with all digits identical to mine until his final "Y" which is a "V" on my chip. Instead of ASProgrammer, cmd8086 used ChProgrammer v.1.35, which had an entry for this chip that he said worked correctly for reads and writes on his 25Q128A chip:

Manufacturer: XMC
Chip name: XM25QH32B
Memory size: 128 MBIT
Size of page: 256
Signature JEDEC: 0x20 0xBA 0x18

That seems odd, and even if correct, that XMC chip does not show up in the IC list of ASProgrammer 2.1.0.13.

Worryingly, Petoparac and Lost_N_BIOS at one point suggested that this BIOS chip might require a 1.8 volt adapter for the CH341A to avoid damage, so I guess I am lucky that no bad things happened while I was trying to use "Detect Chip". I don't have a 1.8 v adapter at present.

However, later in the thread, Lost_N_BIOS replied to ltxLeo that if there is a "13" in the second line of the part number, then that is a 3 volt chip (like the part number on my chip), so no 1.8 volt adapter is required.

My second line is 13EDF, so if that comment is true, then I have a 3 volt chip.

At any rate, I have not yet fried my board, and in fact I am typing on that PC right now.

Photos attached, as requested.

Alright. The chip you have is definitely a 3.3V chip (that's the same with all boards from 100 to 300 series). It should work and remain undamaged after using the CH341A without mod. Can you try the following?
Select Winbond W25Q128JV and try to read the chip? Save the file you get and attach it here so that I can check whether it worked.

 

[limited_imagination]

Thank you. I think I have been successful, but I thought I would post the stages of my effort, to be helpful in case anyone else needs to follow the same thorny track as me.

I tried to select a Winbond W25Q128JV in AsProgrammer as you requested, but there were two sub-types, a -xQ and a -xM.

I tried both of those, and then tried the "Detect Chip" function, but I got a "Search IC" pop-up with a blank list, and an entry in the bottom window, "SPI Flash ID (9F): 000000".

Pushing on, I next tried selecting a Winbond W25Q128BV. This time when I tried "Detect Chip" and the "Search IC" window popped up, there was a chip listed which I selected: "MT25QL128AB (MICRON)". The entry in the bottom window read "SPI Flash ID (9F): 20BA18".

Once I clicked "select" in the "Search IC" popup, the bottom window added the line "<<MT25QL128AB>>".

At that point I figured that I was good to go to read the chip and save it as a .bin file, which I did, and attach it here.

It's not empty, there are a few English words in the hex translation, and it saved as a 16,384 KB file size, which is the same as what I was working with in my Coffeetime 0.99 modded bios file built according to your video instructions.

So while I knew I had good continuity for that janky clip, I considered just going ahead and erasing the chip and writing my Coffeetime modded file to it.

I decided against that so you could review my progress first (thank you for your continued support), and confirm that I have taken a good read of the chip.

The bios chip was not the least bit hot when I removed the spring clip, FWIW.

The board was unpowered during this process, and I did not remove the CMOS battery, deciding that I would only do that when I actually erased and wrote to the chip. My CPU and RAM are still in the slots. I realize this is not best practice, but did it anyway.

The computer booted right up after this experiment, no problem.

Hmm, the bios .bin file was too large to upload, even zipped. I will put it on Google Drive and send you a PM with the link, @itsakjt

I was interested to see what that dumped file looked like if I opened it in Coffeetime, relative to the file that I prepared according to your video.

The one read from the chip does not have any info in the top fields (Vendor, Model, Version, PCH, Date), whereas the one based on your video does have that info.

The one read from the chip has entries for MAC 1 and MAC 2, which are the same number in both fields, and there is no SN or UUID showing.

The one I prepared according to your video has only the MAC 1 number (which the same number as was read from the chip in MAC 1 and MAC 2), but this file has the SN and UUID numbers.

I'm pretty sure the chip was read by the CH341 properly because everything else looks normal in Coffeetime, but those are interesting differences/omissions.

Maybe it was a bad read, or maybe these differences are to be expected?

 

[itsakjt]

Thank you. I think I have been successful, but I thought I would post the stages of my effort, to be helpful in case anyone else needs to follow the same thorny track as me.

I tried to select a Winbond W25Q128JV in AsProgrammer as you requested, but there were two sub-types, a -xQ and a -xM.

I tried both of those, and then tried the "Detect Chip" function, but I got a "Search IC" pop-up with a blank list, and an entry in the bottom window, "SPI Flash ID (9F): 000000".

Pushing on, I next tried selecting a Winbond W25Q128BV. This time when I tried "Detect Chip" and the "Search IC" window popped up, there was a chip listed which I selected: "MT25QL128AB (MICRON)". The entry in the bottom window read "SPI Flash ID (9F): 20BA18".

Once I clicked "select" in the "Search IC" popup, the bottom window added the line "<<MT25QL128AB>>".

At that point I figured that I was good to go to read the chip and save it as a .bin file, which I did, and attach it here.

It's not empty, there are a few English words in the hex translation, and it saved as a 16,384 KB file size, which is the same as what I was working with in my Coffeetime 0.99 modded bios file built according to your video instructions.

So while I knew I had good continuity for that janky clip, I considered just going ahead and erasing the chip and writing my Coffeetime modded file to it.

I decided against that so you could review my progress first (thank you for your continued support), and confirm that I have taken a good read of the chip.

The bios chip was not the least bit hot when I removed the spring clip, FWIW.

The board was unpowered during this process, and I did not remove the CMOS battery, deciding that I would only do that when I actually erased and wrote to the chip. My CPU and RAM are still in the slots. I realize this is not best practice, but did it anyway.

The computer booted right up after this experiment, no problem.

Hmm, the bios .bin file was too large to upload, even zipped. I will put it on Google Drive and send you a PM with the link, @itsakjt

I was interested to see what that dumped file looked like if I opened it in Coffeetime, relative to the file that I prepared according to your video.

The one read from the chip does not have any info in the top fields (Vendor, Model, Version, PCH, Date), whereas the one based on your video does have that info.

The one read from the chip has entries for MAC 1 and MAC 2, which are the same number in both fields, and there is no SN or UUID showing.

The one I prepared according to your video has only the MAC 1 number (which the same number as was read from the chip in MAC 1 and MAC 2), but this file has the SN and UUID numbers.

I'm pretty sure the chip was read by the CH341 properly because everything else looks normal in Coffeetime, but those are interesting differences/omissions.

Maybe it was a bad read, or maybe these differences are to be expected?

Hi there,

There looks like some corruption is present. I have replied you over DM.
Could you please try dumping the file again, this time with the CMOS battery removed? Also, it is a good idea to remove the battery and wait for about 10 to 15 mins before connecting the programmer.

 

[dumaster]

Congratulations @dumaster.

Hello friend itztrrd, how are you ??? Happy new year and success in 2023 !!!!

I hope all is well with you, thank you again for your tips and patience in guiding everyone who came to your topic like me asking for guidance for modification !!!!

A few days ago my modification completed 1 year, I made the modification and posted the images here and details such as the clock I was using which was 4.7Ghz All Cores, I decided a few weeks later to raise it to 4.9GHz All Cores Sync without push the Vcore and it is running perfectly stable for at least almost 1 year, I could try 5.0 or 5.1 or maybe even 5.2 but I believe the gain would be very low with a substantial increase in Vcore and Temperatures which is not healthy for the cores so I decided to stick with 4.9 All Cores !!!!

I hope my images and details of my modification helped and also motivated others to make this change and enjoy their old 170 boards with new 9000 series processors !!!

Big hugs to all

Changing the subject a bit my friend @itsakjt , I don't know if you've already come across this problem, I have a TL866 recorder and these days I needed to record a 1.8v flash, I know that you need an adapter to reduce the VCC and DATA pins from 3v to 1.8v and the software itself recommends a simple scheme with fews componets for you to build an adapter yourself, follow the photo below !!!!

I got the exact components from my scraps, assembled the adapter and even managed to read the Chip ID, but the read operation failed when comparing the buffer with the chip, without the patience to try to see what could be happening because I had tested all the components I used, the BAT54 diodes ok, the regulator delivering 1.800v, capacitors OK, I decided to buy this adapter that also has an AMS1117 1.8v regulator but for the treatment of clock and data signals it uses a chip caled ALVC164245 that probably does the bidirectional data conversion from 3v to 1.8v in both directions and at the speed that is necessary for recording, follow the adapter images here and PDF of Chip bellow !!!!

https://www.ti.com/lit/ds/symlink/sn74alvc164245.pdf

My question is, is this scheme suggested by the TL866's MiniPro software itself a bad and flawed project that have problems with the almost all 1.8v flahs chips or if is case of the chances of it working with some 1.8v flash chips and not others, or if my diy adapter it could I the had really some problem that I couldn't identify ???

The chip I was recording was a Macronix MX25U12873F and with the new adapter with this ALVC Level-Shifting Transceiver chip the recording was perfect !!!

I would love to know your opinion my friend !!!

Hugs

 

[itsakjt]

Hello friend itztrrd, how are you ??? Happy new year and success in 2023 !!!!

I hope all is well with you, thank you again for your tips and patience in guiding everyone who came to your topic like me asking for guidance for modification !!!!

A few days ago my modification completed 1 year, I made the modification and posted the images here and details such as the clock I was using which was 4.7Ghz All Cores, I decided a few weeks later to raise it to 4.9GHz All Cores Sync without push the Vcore and it is running perfectly stable for at least almost 1 year, I could try 5.0 or 5.1 or maybe even 5.2 but I believe the gain would be very low with a substantial increase in Vcore and Temperatures which is not healthy for the cores so I decided to stick with 4.9 All Cores !!!!

I hope my images and details of my modification helped and also motivated others to make this change and enjoy their old 170 boards with new 9000 series processors !!!

Big hugs to all

Changing the subject a bit my friend @itsakjt , I don't know if you've already come across this problem, I have a TL866 recorder and these days I needed to record a 1.8v flash, I know that you need an adapter to reduce the VCC and DATA pins from 3v to 1.8v and the software itself recommends a simple scheme with fews componets for you to build an adapter yourself, follow the photo below !!!!

View attachment 285899

I got the exact components from my scraps, assembled the adapter and even managed to read the Chip ID, but the read operation failed when comparing the buffer with the chip, without the patience to try to see what could be happening because I had tested all the components I used, the BAT54 diodes ok, the regulator delivering 1.800v, capacitors OK, I decided to buy this adapter that also has an AMS1117 1.8v regulator but for the treatment of clock and data signals it uses a chip caled ALVC164245 that probably does the bidirectional data conversion from 3v to 1.8v in both directions and at the speed that is necessary for recording, follow the adapter images here and PDF of Chip bellow !!!!

https://www.ti.com/lit/ds/symlink/sn74alvc164245.pdf

View attachment 285901View attachment 285902

My question is, is this scheme suggested by the TL866's MiniPro software itself a bad and flawed project that have problems with the almost all 1.8v flahs chips or if is case of the chances of it working with some 1.8v flash chips and not others, or if my diy adapter it could I the had really some problem that I couldn't identify ???

The chip I was recording was a Macronix MX25U12873F and with the new adapter with this ALVC Level-Shifting Transceiver chip the recording was perfect !!!

I would love to know your opinion my friend !!!

Hugs

Hello @dumaster - Long time and wish you a happy new year 2023.
Glad that your system is running rock stable.
As for your question, I don't have much experience with 1.8V chips. The software can very well be erratic.
One thing you can check is that when you attempt to read the chip, is the voltage dropping below 1.8V?
The reason I ask this is because I have seen the CH341A to behave this way especially when connected to some motherboards directly and for one such motherboard, I had to supply 3.3V externally to aid the CH341A since the voltage was dropping to 2V while it was attempting to read causing corruption of data.

 

[thegooddoctor]

Asus Strix E Z270 Gaming motherboard successfully modded with Xeon E-2286M processor. I couldn't reliably write to the onboad bioS chip 25Q128A; had a W25Q128.V BIOS chip shipped with the processor. Swapped out the BIOS chip (hot air station) to the W25Q128.V and flashed it with the CH341 programmer (modded the bios by following the video

). My first modded BIOS didn't work (latest BIOS) so I went with an earlier BIOS with ultimate success. Used Winx64 and Debian to perform the flashes/verification.
Great reading this thread ... I hope my information can help others!

 

[aQi]

Asus Strix E Z270 Gaming motherboard successfully modded with Xeon E-2286M processor. I couldn't reliably write to the onboad bioS chip 25Q128A; had a W25Q128.V BIOS chip shipped with the processor. Swapped out the BIOS chip (hot air station) to the W25Q128.V and flashed it with the CH341 programmer (modded the bios by following the video

). My first modded BIOS didn't work (latest BIOS) so I went with an earlier BIOS with ultimate success. Used Winx64 and Debian to perform the flashes/verification.
Great reading this thread ... I hope my information can help others!

Thats brilliant as I read that z170/z270 had issues when trying to run xeon processors (6th/7th gen)

What are the performance uplifts you are getting ?

Edit: So the processor W-2286M which is a mobile unit and is coffeelake indeed. Care to elaborate how did you manage to get it ? Hard guess is chinese website ?

 

[thegooddoctor]

CPU Mark score 16246
Which places this setup just below a i9-9900

Intel Xeon E-2286M vs i9-9900 [cpubenchmark.net] by PassMark Software

Words of advice: you can only overclock the ram slightly (I read somewhere that 2900MHz is about max but that can vary less system to system); I opted for 2400MHz.
Had to disable the auto CPU overclock as well; haven't tried OC'ing the CPU.
Had issues with the iGPU until I updated to Intel's most recent drivers.

I really like how much the processor downclocks (seen low as 800MHz); of course it will seamlessly upclock with load.
Bought on An onLIne chinese website; the one I bought had an IHS with liquid metal under the IHS; also came with a BIOS chip preprogrammed for a particular board (don't get just the file of the bios shipped; get the chip - it helped a lot as per my previous post).

Asus Strix Z270E Gaming motherboard (for best compatibility use Gigabyte - there are some CPU listings that compare different motherboard vendors and use of these chips)

All in all, an 8C16T upgrade; plan to use this as a proxmox server.

Here is CPU-Z
The board was using an i7-7700 (non-K) previously; this is an excellent upgrade!

 

[itsakjt]

CPU Mark score 16246
Which places this setup just below a i9-9900

Intel Xeon E-2286M vs i9-9900 [cpubenchmark.net] by PassMark Software

Words of advice: you can only overclock the ram slightly (I read somewhere that 2900MHz is about max but that can vary less system to system); I opted for 2400MHz.
Had to disable the auto CPU overclock as well; haven't tried OC'ing the CPU.
Had issues with the iGPU until I updated to Intel's most recent drivers.

I really like how much the processor downclocks (seen low as 800MHz); of course it will seamlessly upclock with load.
Bought on An onLIne chinese website; the one I bought had an IHS with liquid metal under the IHS; also came with a BIOS chip preprogrammed for a particular board (don't get just the file of the bios shipped; get the chip - it helped a lot as per my previous post).

Asus Strix Z270E Gaming motherboard (for best compatibility use Gigabyte - there are some CPU listings that compare different motherboard vendors and use of these chips)

All in all, an 8C16T upgrade; plan to use this as a proxmox server.

Here is CPU-Z
The board was using an i7-7700 (non-K) previously; this is an excellent upgrade!

Excellent.
This is basically a 9900, R0 revision.
CPU ID is 906ED.
This will work with the BIOS modding guide here.
My i7 9700K at 5.2 GHz scores 619 on CPU Z single thread and 4800+ on multi thread. The RAMs are clocked at 3600 MHz on an ASUS Maximus X Formula (Z370)

The i7 9700KF scores 603 and 4700+ in the same benchmarks on the Maximus VIII Ranger (Z170) equipped with RAMs at 3200 MHz.

You can try OCing the RAM frequency. If you use it for gaming, it will give you a nice performance boost.
Great work and kudos for the share.

 

[ArisAris]

Have a look my friend

 

[itsakjt]

Have a look my friend

Hello there,

Looks like this programmer software is not the fixed version.
Please download the one I linked. That contain some fixes and is the only one I found to properly work with the CH341A.

 

[ArisAris]

but when try to erace... Both leds are light. no flashing on programmer

I use the one you mentioned on youtube, but let me check again...

 

[itsakjt]

but when try to erace... Both leds are light. no flashing on programmer

Yes I am aware. Please try with the exact file I linked you via the Google Drive link.

 

[ArisAris]

Yes from XiTongZhi. Thats the one I use...

 

[itsakjt]

Yes from XiTongZhi

Correct. It has some fixes. I even tried with the latest version and it doesn't work. Only that one works and perfectly as well.

 

[Solaris17]

fwiw, I wrote a guide in the GPU section to show people how to do this.

Guide: Flashing with a hardware programmer

Lets get physical So; you tried flashing your GPU and you didn't backup. Now you have some dude on the forum linking you a BIOS but you have no way to apply it because you don't have a spare GPU either. Well no worries friend. In this guide we are going to cover the simple procedures of...

www.techpowerup.com

but looking at the screen shot, it looks like the chip is in write protect mode. So make sure you are following its advice and hitting the unlock button.

Though generally speaking this can also crop up as a issue running the software as non-admin on some machines that have strict UAC, additionally and in most cases, improperly connecting the chip. There are x24 and x25 positions on the ch341 make sure you have it in the right location.

 

[ArisAris]

can read the chip, I can do nothing else.... I will check your guide my friend and try again.....

 

[Solaris17]

can read the chip, I can do nothing else.... I will check your guide my friend and try again.....

doesnt look like its reading the chip, its all FF unless you know you have a blank chip thats a bad read.