*Official Benchmark Thread* - Post it here or it didn't happen :D

[johnksss]

25 minutes ago, Mr. Fox said:

It would have to be unlocked, at least during the flashing process.

SPD EEPROM unlocked during flashing of the bios? 

[Mr. Fox]

Just now, johnksss said:

SPD EEPROM unlocked during flashing of the bios? 

That's my assumption. I ordered new memory sticks that will be here Wednesday. As soon as I install them I will know for certain that is the case. Given that the issue affects the old (unflashed) BIOS in exactly the same way, I doubt that it is the motherboard firmware. If the new sticks also have no XMP profile, then it will be a mobo RMA. I am hoping it is the memory modules got written with rubbish and not a mobo swap. I think the product name and production date being changed is also an indication of SPD write during the BIOS flash.

 

The release notes saying "fine-tuned memory compatibility" leads me to be suspicious as well. I'm not sure what approach they took to that. It is strange nomenclature, but sometimes Chinese to English translation is not very clean.

[Mr. Fox]

56 minutes ago, Reciever said:

I remember when I was young I used to think that new models should come with better features or in general be overall a better product than the previous year.

 

These days we have measure just how much crap you have to walk through from PR events, Reviews and finally getting the product and figuring out yourself just how much of it is false.

 

I struggle to maintain a positive outlook, and that bothers me.

I struggle with the same thing. I am not a negative person at all. I am always positive and a glass is half full kind of person. However, I'm not going to pretend things are peachy when they suck. It is very difficult to be positive about a lot of things right now. Tech is one of those things. We expect it to get better in time, not worse. In some ways it gets better, but it is like one step forward, two steps backward more often than it should be.

[johnksss]

10 minutes ago, Mr. Fox said:

That's my assumption. I ordered new memory sticks that will be here Wednesday. As soon as I install them I will know for certain that is the case. Given that the issue affects the old (unflashed) BIOS in exactly the same way, I doubt that it is the motherboard firmware. If the new sticks also have no XMP profile, then it will be a mobo RMA. I am hoping it is the memory modules got written with rubbish and not a mobo swap. I think the product name and production date being changed is also an indication of SPD write during the BIOS flash.

 

The release notes saying "fine-tuned memory compatibility" leads me to be suspicious as well. I'm not sure what approach they took to that. It is strange nomenclature, but sometimes Chinese to English translation is not very clean.

Then that sounds like a serious flaw as it should not be wring to SPD locked memory modules unless you unlock them yourself. As they have no way of knowing what memory you have or will install in the board. Ill have to keep that in mind going forward with flashing of this board.

 

I am too suspicious of the "fine-tuned" as well. The board has some pretty good ones already setup with it that do not need flashing....

[jaybee83]

2 hours ago, Papusan said:

I wonder how DDR5 will function in next gen AMD platform. Still a huge disaster for Intel. We all are Guinea pigs for this new tech. Maybe it will be better on older platforms as times goes but I guess they’ll put more (all) work in gen 2 DDR5 platforms (Raptor lake and Intel). Good blessings to all the new coming AM5 platform buyers. But AMD won’t learn anything of the miserable things from Intel. They need to learn it their own hard ways. And it won’t be pretty l. Hail to “new tech”! One more year with sucked and screwed up tech. At least my setup worked well l before I went on summer vacation. Don’t change what’s  works! 

ill be sure to let u know shortly before pulling my hair out @DDR5 on AM5 😅

 

1 hour ago, Mr. Fox said:

That's my assumption. I ordered new memory sticks that will be here Wednesday. As soon as I install them I will know for certain that is the case. Given that the issue affects the old (unflashed) BIOS in exactly the same way, I doubt that it is the motherboard firmware. If the new sticks also have no XMP profile, then it will be a mobo RMA. I am hoping it is the memory modules got written with rubbish and not a mobo swap. I think the product name and production date being changed is also an indication of SPD write during the BIOS flash.

 

The release notes saying "fine-tuned memory compatibility" leads me to be suspicious as well. I'm not sure what approach they took to that. It is strange nomenclature, but sometimes Chinese to English translation is not very clean.

fingers crossed papa fox! keep us updated on this. with a bit of luck, the SPD / XMP profiles only got messed up during the bios flash, but wouldnt be touched anymore afterwards.

[Mr. Fox]

58 minutes ago, johnksss said:

Then that sounds like a serious flaw as it should not be wring to SPD locked memory modules unless you unlock them yourself. As they have no way of knowing what memory you have or will install in the board. Ill have to keep that in mind going forward with flashing of this board.

 

I am too suspicious of the "fine-tuned" as well. The board has some pretty good ones already setup with it that do not need flashing....

Just now, jaybee83 said:

ifingers crossed papa fox! keep us updated on this. with a bit of luck, the SPD / XMP profiles only got messed up during the bios flash, but wouldnt be touched anymore afterwards.

There should never be any reason whatsoever for a motherboard BIOS flash to write anything to memory SPD. That is a big no-no. That said, I know one of the "features" of DDR5 is the ability to create your own XMP profiles and save them to the extra empty space. And, you do that within the BIOS. In this situation that part of my BIOS is non-existent now. There is NOTHING on any menus about XMP or custom user XMP profiles. It is totally gone now and AIDA64 doesn't see that on the memory modules either.

 

I wish I knew someone with a DDR5 mobo that could insert these sticks for testing, but I don't know anyone nearby. If it did screw them up, I will keep them and use one of them for BIOS flashing so it doesn't ever happen to the new memory. My Strix Z690 is a DDR4 mobo, so I can't try swapping them. I will have to see if Thaiphoon Burner works with DDR5, but I don't think it does. If it does, then I can fix them.

[Mr. Fox]

26 minutes ago, Mr. Fox said:

I will have to see if Thaiphoon Burner works with DDR5, but I don't think it does. If it does, then I can fix them.

Hmmm. It does work. Look... seems like it wiped out a ton of stuff on the modules. Not just the XMP.

[jaybee83]

3 minutes ago, Mr. Fox said:

Hmmm. It does work. Look... seems like it wiped out a ton of stuff on the modules. Not just the XMP.

"fine-tuned memory compatibility" by equalizing ALL installed modules via total erasure. wth 😵🙄🤦‍♂️

[Mr. Fox]

The EK Quantux Velocity² is nice looking. It is made specifically for LGA-1700. I got this from Titan Rig open box for about $40 below retail. It works better than the OptimusPC Foundation block and has a MUCH higher liquid flow rate. The OptimusPC block is a lot more restrictive. Temps are close to the same, slightly better (3-5°C lower core max in Cinebench).

 

The only downsides are the back of mobo installation and fixed position of inlet and outlet. It fills the empty space between the VRM heat sinks and memory.

 

[Mr. Fox]

On 7/11/2022 at 12:17 PM, Papusan said:

I wonder how DDR5 will function in next gen AMD platform. Still a huge disaster for Intel. We all are Guinea pigs for this new tech. Maybe it will be better on older platforms as times goes but I guess they’ll put more (all) work in gen 2 DDR5 platforms (Raptor lake and Intel). Good blessings to all the new coming AM5 platform buyers. But AMD won’t learn anything of the miserable things from Intel. They need to learn it their own hard ways. And it won’t be pretty l. Hail to “new tech”! One more year with sucked and screwed up tech. At least my setup worked well l before I went on summer vacation. Don’t change what’s  works! 

Hopefully not worse memory latency than Ryzen already has. That will be interesting. Their DDR4 latency is nearly double what it is for the same generation Intel platform. If it gets any worse that will be ridiculously horrible. 

[Mr. Fox]

18 hours ago, Mr. Fox said:

Hmmm. It does work. Look... seems like it wiped out a ton of stuff on the modules. Not just the XMP.

18 hours ago, jaybee83 said:

fingers crossed papa fox! keep us updated on this. with a bit of luck, the SPD / XMP profiles only got messed up during the bios flash, but wouldnt be touched anymore afterwards.

I just noticed that besides the data that is either erased or unreadable in the SPD that what remains doesn't even match. Hopefully will have the Vengeance DDR5-6400 modules tomorrow or Thursday. At least I am able to use the computer still. Very weird.

@johnksss

[electrosoft]

37 minutes ago, Mr. Fox said:

I just noticed that besides the data that is either erased or unreadable in the SPD that what remains doesn't even match. Hopefully will have the Vengeance DDR5-6400 modules tomorrow or Thursday. At least I am able to use the computer still. Very weird.

@johnksss

 

Even downgrading the BIOS gives the same results? Luckily ($$) in these situations comparative test hardware usually gives clarity. 🙂

[electrosoft]

KPE 3090ti's are sold out but it was a very limited run so it was to be expected. It would have sold out sooner if queued up members had picked them up.

 

All other EVGA 3090ti models are $1499.99 including the hybrid:

 

https://www.evga.com/products/productlist.aspx?type=0&family=GeForce+30+Series+Family&chipset=RTX+3090+Ti

 

EVGA has pulled all 3090s and 3080s from their combo deal integrator and only 3070ti and below are available.

 

Only three 3090 models are now available to buy for $1299 (FTW3 Black, FTW3 Ultra, XC).  All the others are out of stock including the original KPE 3090.

 

All 3080ti's left are 1099.99. 5 models in stock.

 

All 3080 12gb are 799.99 gobs of those 🙂

 

Clearly they want to push the 3080 12gb as all the 3080 10gb are priced above that.

 

 

 

 

[Mr. Fox]

55 minutes ago, electrosoft said:

 

Even downgrading the BIOS gives the same results? Luckily ($$) in these situations comparative test hardware usually gives clarity. 🙂

Yes. The motherboard has a dual BIOS. Moving the switch to the older BIOS position, the problem is the same. I also tried reflashing the new BIOS and previous BIOS twice on the BIOS switch position I use as the default and nothing changes. I am almost positive the BIOS flash corrupted the memory SPD firmware for that reason. The only other possibility is a hardware failure, and that is unlikely since everything was fine until I flashed the BIOS. I do not think doing that would affect hardware.

[Papusan]

1 hour ago, Mr. Fox said:

Yes. The motherboard has a dual BIOS. Moving the switch to the older BIOS position, the problem is the same. I also tried reflashing the new BIOS and previous BIOS twice on the BIOS switch position I use as the default and nothing changes. I am almost positive the BIOS flash corrupted the memory SPD firmware for that reason. The only other possibility is a hardware failure, and that is unlikely since everything was fine until I flashed the BIOS. I do not think doing that would affect hardware.

Did you flash with a single stick? And tried flash with the second stick?

[Mr. Fox]

46 minutes ago, Papusan said:

Did you flash with a single stick? And tried flash with the second stick?

Flashing was not an issue. There are no errors in flashing and it completes successfully. The problem is that BIOS vA.60 erased data from the memory SPD on both memory modules. The system boots using a DDR5 default memory profile that is part of the BIOS firmware and not using the firmware that used to be on the memory. I do not have any way of flashing a stock SPD or XMP profile onto the DDR5 sticks. I can overclock the memory and set manual timings but the voltage is limited to 1.435V which limits the overclock threshold to 6400 with some errors due to insufficient voltage. The BIOS default is 1.435V max and the memory SPD controls anything higher than that.

[johnksss]

On 7/11/2022 at 2:39 PM, Mr. Fox said:

There should never be any reason whatsoever for a motherboard BIOS flash to write anything to memory SPD. That is a big no-no. That said, I know one of the "features" of DDR5 is the ability to create your own XMP profiles and save them to the extra empty space. And, you do that within the BIOS. In this situation that part of my BIOS is non-existent now. There is NOTHING on any menus about XMP or custom user XMP profiles. It is totally gone now and AIDA64 doesn't see that on the memory modules either.

 

I wish I knew someone with a DDR5 mobo that could insert these sticks for testing, but I don't know anyone nearby. If it did screw them up, I will keep them and use one of them for BIOS flashing so it doesn't ever happen to the new memory. My Strix Z690 is a DDR4 mobo, so I can't try swapping them. I will have to see if Thaiphoon Burner works with DDR5, but I don't think it does. If it does, then I can fix them.

Usually memory exchanges are far easier than anything else. Send them back and they send you a new pair.

 

 

[johnksss]

5 hours ago, Mr. Fox said:

I just noticed that besides the data that is either erased or unreadable in the SPD that what remains doesn't even match. Hopefully will have the Vengeance DDR5-6400 modules tomorrow or Thursday. At least I am able to use the computer still. Very weird.

@johnksss

That does not work as I just downloaded the free version and tried to read my memory and I get the exact same as you do....

[johnksss]

1 hour ago, Mr. Fox said:

Flashing was not an issue. There are no errors in flashing and it completes successfully. The problem is that BIOS vA.60 erased data from the memory SPD on both memory modules. The system boots using a DDR5 default memory profile that is part of the BIOS firmware and not using the firmware that used to be on the memory. I do not have any way of flashing a stock SPD or XMP profile onto the DDR5 sticks. I can overclock the memory and set manual timings but the voltage is limited to 1.435V which limits the overclock threshold to 6400 with some errors due to insufficient voltage. The BIOS default is 1.435V max and the memory SPD controls anything higher than that.

And they locked the bios memory voltage to 1.435V? Seriously?

[Mr. Fox]

26 minutes ago, johnksss said:

Usually memory exchanges are far easier than anything else. Send them back and they send you a new pair.

 

 

I may do that. I will have to reinstall the original heat sinks and RGB diffusers and hope they don't have a problem with that (or hope that they not notice). Then I can sell the replacements (or hold onto them as backup parts) unless the Vengeance 6400 kit I ordered is inferior. Luumi seems to be a fan of them. I had actually ordered this kit once before and returned it unopened for a refund because I changed my mind about wasting $350+ for a potentially very modest (at best) performance improvement. Ditching the RGB crap will be a special bonus feature, LOL. I deliberated  selected Vengeance versus Dominator because I do not want the RGB. (Price was about the same.)

25 minutes ago, johnksss said:

That does not work as I just downloaded the free version and tried to read my memory and I get the exact same as you do....

OK, that is good to know. Thank you for checking that. The erased XMP profiles, modified product name on one module, production date and other modified data that shows in AIDA64 and CPU-Z are definitely an issue and not the same on both memory sticks. That is what I originally discovered before trying to look at the SPD using Thaiphoon Burner, so there still seems to be an issue with memory firmware modification. Of course, this is subject to verification when I receive the replacement memory tomorrow. I am crossing my fingers and hoping it is not something that the flash caused to go haywire with the mobo.

18 minutes ago, johnksss said:

And they locked the bios memory voltage to 1.435V? Seriously?

Yes, unfortunately. If it were not for that problem the memory would still function the same as it did before other than the information and labeling data being messed up and XMP profiles missing. But, the 1.435V limit is a huge impediment. I was running 6800 CL30 with 1.475V stable as a daily driver memory overclock and the 1.435V limit is capping me out at 6400 CL30. If I try to set anything manually higher than 1.435V is resets to Auto.

The discovery of the issue was immediate after flashing. On the first boot after flashing A.60 I was greeted with an error message that said "memory has been changed" and I do not see that message after flashing. When I entered the BIOS after seeing that message to apply my custom settings is when I discovered the missing XMP profiles and 1.435V cap applied. When you look at the SPD in the BIOS, it is also different there for each memory module and the XMP info does not show in the SPD viewing it in the BIOS. In fact, you cannot find anything anywhere in the BIOS where XMP is mentioned.

[johnksss]

35 minutes ago, Mr. Fox said:

I may do that. I will have to reinstall the original heat sinks and RGB diffusers and hope they don't have a problem with that (or hope that they not notice). Then I can sell the replacements (or hold onto them as backup parts) unless the Vengeance 6400 kit I ordered is inferior. Luumi seems to be a fan of them. I had actually ordered this kit once before and returned it unopened for a refund because I changed my mind about wasting $350+ for a potentially very modest (at best) performance improvement. Ditching the RGB crap will be a special bonus feature, LOL. I deliberated  selected Vengeance versus Dominator because I do not want the RGB. (Price was about the same.)

OK, that is good to know. Thank you for checking that. The erased XMP profiles, modified product name on one module, production date and other modified data that shows in AIDA64 and CPU-Z are definitely an issue and not the same on both memory sticks. That is what I originally discovered before trying to look at the SPD using Thaiphoon Burner, so there still seems to be an issue with memory firmware modification. Of course, this is subject to verification when I receive the replacement memory tomorrow. I am crossing my fingers and hoping it is not something that the flash caused to go haywire with the mobo.

Yes, unfortunately. If it were not for that problem the memory would still function the same as it did before other than the information and labeling data being messed up and XMP profiles missing. But, the 1.435V limit is a huge impediment. I was running 6800 CL30 with 1.475V stable as a daily driver memory overclock and the 1.435V limit is capping me out at 6400 CL30. If I try to set anything manually higher than 1.435V is resets to Auto.

Okay, so I went ahead and flashed and all went well. XMP is still there. And voltage is not locked....

 

So Something definitely went wrong with your flash...

[Mr. Fox]

2 minutes ago, johnksss said:

Okay, so I went ahead and flashed and all went well. XMP is still there. And voltage is not locked....

 

So Something definitely went wrong with your flash...

Well, this is not the first time we saw that with the Unify-X. Remember my first one when the first mobo was brand new? I guess flashing MSI firmware is an unsually risky matter that should be avoided when possible. There was no error of any kind and the flashing process appeared to occur normally. I don't normally even consider firmware updates, but with DDR5 being new and evolving technology, the temptation exists where it normally would not.

[johnksss]

7 minutes ago, Mr. Fox said:

Well, this is not the first time we saw that with the Unify-X. Remember my first one when the first mobo was brand new? I guess flashing MSI firmware is an unsually risky matter that should be avoided when possible. There was no error of any kind and the flashing process appeared to occur normally. I don't normally even consider firmware updates, but with DDR5 being new and evolving technology, the temptation exists where it normally would not.

This is true...

 

When I flashed i made sure to have memory back to full stock first though. And I never used that user XMP profile thing either, so that may or may not have played a role in things not going according to plan....Speculations of course.

[Mr. Fox]

1 hour ago, johnksss said:

This is true...

 

When I flashed i made sure to have memory back to full stock first though. And I never used that user XMP profile thing either, so that may or may not have played a role in things not going according to plan....Speculations of course.

Yes, I always go into the BIOS and reset firmware defaults for everything before flashing after making sure I saved my old profile to USB in case I want to go back to the old firmware. I do not use that user-XMP thing either. I looked into it to see what is involved and it did not make any sense to me. With the ability to save complete BIOS profiles that cover everything, including things totally unrelated to overclocking, it seems more like a silly gimmick than a useful feature to me.

 

I do set the memory XMP profile first and tune from that point, at least until I have it dialed in. Then I write down or screenshot everything and start over without using the memory XMP profile. I also start with the voltage higher than I expect and gradually go down until I get errors in testing or no boot, then go back up a bit. Not remarkably different the CPU or GPU overclocking other than all of the timings that take time to get dialed in.

 

I got an update from UPS saying the Vengeance kit should be here for sure tomorrow, so we shall see if  installing that solves the dilemma. These are what I ordered. I wanted to get them on Amazon, but delivery would have not happened until Sunday from Amazon for some reason, and the price was identical. 

CORSAIR Vengeance 32GB (2 x 16GB) 288-Pin PC RAM DDR5 6400 (PC5 51200) Intel XMP 3.0 Desktop Memory Model CMK32GX5M2X6400C38

 

[Talon]

@Mr. Fox

 

Do you not have the "Special PMIC Unlock" option in the BIOS to bypass the 1.435v limit on locked PMICs?  Sorry to hear about your bad luck with the memory.