Great news- LostNBios' C621E bios mod has run great on several sets of dual xeon processors including non-ES as well.
If anyone has 50652h CPUID samples, none of the ASUS BIOSes for C621E will run it, though some run 50654h
The two machines I run right now on SAGE boards are

- dual QLQ9 '6149' or '6143' 2x 16 core/32 thread, 2.8GHz base, 3.3 GHz no OC. With Lost's modded BIOS, the ASUS board runs them and also
overclocks to level 1 plus other overclock features. Performance test9 CPU score (dual) is ~34.900. Geekbench 4 CPU score Win10 29,000, Linux Debian 65,000

- dual QLKN '8175' or '8168' 2x 24core/48thread. Performance test 9 CPU score (dual) is ~38,700. Geekbench 4 CPU score Win10 29,000, Linux Debian 100,000
Cinebench R20 15876, Cinebench R15 6882

Neither of above systems will post with ASUS C621E BIOS, only the modded 5503 from Lost.
Lost's mod of ASUS 5503 flashed and worked with the ES CPUs in place.
Flashed right away, then I tried other ASUS BIOSes which failed again, then reflashed his modded 5503 BIOS, worked immediately again.
All features work, and benchmarks within 2% of production performance. THANKS!!!

[Will post separately about a new problem with Supermicro dual boards not running samples anymore.... ]

@maxdata, Where can I find the bios image?
I guess there have been updates since this thread was started; where do I find the most recent file?

I'm a bit confused by what you said about 50652h cpus not working with any bios.. Do you mean that they won't work with the modded bios either?

My 50652h samples (QLQ9, QKLN) would not run on any old or new BIOS downloadable from ASUS:
https://www.asus.com/Commercial-Servers-.../HelpDesk_BIOS/

Lost's mod of ASUS C621E 5503 (link provided at post #241 in this topic) flashes and works immediately. This is the only modded BIOS that has worked on ASUS SAGE with my 50652h samples. For me it flashed with no problem over any of the ASUS C621E BIOS versions that I tried up through 6003

Lost has a link to his BIOS mod at post #241 of this topic (woops, I should have noted that in my first post above - sorry!)

Hi everyone. i have a xeon 6136 engineering sample which works fine under windows. but when i use it in linux, i get a hardlock seemingly at random. dmesg shows this error: [ 0.101234] [Firmware Bug]: TSC_DEADLINE disabled due to Errata; please update microcode to version: 0xffffffff (or later). has anyone else come across this issue?

a quick edit; is there a reason that we are using bios 5503? did something change in later versions or is there just a lack of demand
-------------------------------

sorry @Lost_N_BIOS for the ping but i see the thread has been quiet for some time now so not sure if you get updates for it

@Gurney_ - Microcode version 0xffffffff is not a thing, so that may be generic or not valid error message (ie latest microcode may not fix issue anyway). What is your CPU's CPUID?
Last time I updated the mod BIOS I used the latest BIOS and latest microcodes too, so unless there is new BIOS and microcodes out now, then the last one I posted would have all the latest
Yes, I see now there is several BIOS updates since then, I guess no one asked for a new mod BIOS yet is all. Do you have flash programmer, in case newer BIOS versions do not work for this mod? If not, does this board have USB Flashback, I forgot?

Zitat von Lost_N_BIOS im Beitrag #352

---

@Gurney_ - Microcode version 0xffffffff is not a thing, so that may be generic or not valid error message (ie latest microcode may not fix issue anyway). What is your CPU's CPUID?
Last time I updated the mod BIOS I used the latest BIOS and latest microcodes too, so unless there is new BIOS and microcodes out now, then the last one I posted would have all the latest
Yes, I see now there is several BIOS updates since then, I guess no one asked for a new mod BIOS yet is all. Do you have flash programmer, in case newer BIOS versions do not work for this mod? If not, does this board have USB Flashback, I forgot?

---

@Gurney_ @levelud @maxdata @paranoid_android @mitdal369 @avd @Onyxx1606 @myio @Branen @malucul @Kaboooom2000uk @dahuoxia @TechStream @XeonScalable
@Scotchmist @sochotek @nhan72nn @gloobox @trespasser2 @adhss_phil @SH_LEE @coldwove @lnwlf2k @rulik006 @Funeral_YAR @Irgen @samthekiller @dgjoma @Airo @Babar

6102 Update - This is stock BIOS (So DO NOT FPT), BIOS is flashback ready, w/ ES ucodes added back in and all updated to latest as of 3/28/2020


http://s000.tinyupload.com/index.php?file_id=35068615341062024964
http://www.filedropper.com/asus-ws-c621e...-flashbackready
https://www.sendspace.com/file/l3cnzk >>> On pause, please do not use unless you plan to join in the troubleshooting testing until we sort it out >> * Edit - Working 6201 (yes 6201) version posted at Post #405 for anyone reading later

* Edit - Sorry to everyone I tagged! Please wait, I got anxious to share with you since I didn't realize until now there was updated BIOS.
This one linked above is not good, please wait, I will try to sort out with someone in PM. Anyone that wants to help with that, if you have programmer for recovery let me know (mainly so we can test older ME FW if necessary, not so much about recovery)
It may be simple issue I noticed with file, and have sent levelud a test BIOS in PM, if it's that and we resolve quickly, I will put new links above shortly

Hey Lost thanks for your work, I habe tried it for now and I stuck at Q-Code 79 CSM Init.
I have a pair of QL1K's running.

Thanks for report! I don't know reason for that, I do not have this boards It would possibly be old ME FW is needed instead of new, or old version of SECCore in 5503 that @avd mentioned initially
Do you have flash programmer? If yes, we can test old ME to see if this fixes it, if not then we have to wait on someone else in the above list to test and have programmer in hand so they can test a few other BIOS I make.
Sorry for this, I assumed there would be no issue, probably they tried again to block somehow.

Very nice to see you're working on the latest bios updates. I just got my board 3 days ago and a pair of ES platinum chips. Flashed your original bios mod and it worked right away. Very happy with the work you are doing!

@Babar - Thank you, and good to hear the OG BIOS is working for you! Hopefully we can get the latest one sorted out here shortly too
I may have to pick up one of these boards just so I can keep up on this for you guys, since there is so many users that rely on this motherboard!
Anybody know of one for sale with dead/bricked BIOS

* Edit - Wow, I don't even see WS W621E Sage on amazon, but I found WS C621E Sage and it's $600 new / $400 used
* Edit - Ohh, I see that (WS C621E) is the exact correct model - $600+ new on ebay, no used And damn, even not working with trashed CPU sockets it sells for up to $300 on ebay completed sales
With prices like that for the board, and then add CPU's on top of that, by the time prices are low enough that I could grab something like this to better help you guys it will be obsolete and you'll all be using something else instead

It sure seems like a pretty awsome motherboard, though I have barely started to test stuff, still waiting for more hardware. Excited to see if bifurcation works like it should when I get that far.

Haha, yeah, its NOT a cheap board. But my hope is that it will replace 3-4 gamers and a couple servers all in one.
And the price is of course why we need to make ES chips work on it, or it would be 10x as expensive :)

sorry @Lost_N_BIOS, sorry i forgot i posted here. would the CPUID be different for a ES cpu compared to retail? wikichip says my 6136 is 0x50654 but obviously is non ES. i've got the board packed up so if i need to boot it back up let me know