Modded bios Rampage VI

Quote:

---

Originally Posted by restsugavan

Please kindly sharing experiences friends. We are owner of same CPU i9 7980XE.

After testing on my 2002 modded BIOS with 02000064 microcode with lastest Windows 10 Insider Previews build 18995.1 (lastest kernel).
I think there are some good news for our 7980XE for the new BIOS code. During testing with AIDA64 6.1.5200 Build many times.

1. 7980XE has better bandwidth Read/Write/Copy and lowest latency of DDR4 access since debut.
* * My system got 68.4ns (GSkill 4- channal DDR4 3200 16-18-18-38 2T *64GB) .

2. 7980XE has lower L2 cache latency from 5.5ns (1902 and older bios revision) to 3.1ns (2002).

3. 7980XE also lower L3 cache latency too (but a little bit compare to 1902 and older bios revision ).

However when compare 02000064 microcode to 0200005E or 02000060 microcode , they’re tiny performance gap for 3 revisions.
Personally I love 02000064 microcode with lastest performance optimized for new Windows kernel (1909 and my Insider 20H1 testing) more
than others.

Please waiting new Intel Turbo Boost Max 3.0 new version under testing with W10 1909 and my insider 20H1 a little bit. Final version will
release soon. ***

---

Are u overclocking your system or running it at stock???

Hi restsugavan

Yes, I saw the gain with bios 2002 and Aida64 benchmark

The new Intel Turbo Boost 3.0 is only compatible with the new architecture (CL-X among others) ?

Quote:

---

Originally Posted by tistou77

If you're not interested, do not post
You have trouble understanding when I say that I'm not talking about OC stability but temperatures
The cores have better temperatures with the microcode 49 compared to the microcode since the 50, nothing more to understand
We do not care what you say with the alteration or the stability of the OC, nobody, except you, want to have temperatures at the limit of TJ Max

And one more thing that you post without knowing I think (how to know without testing...!!!!), the microcode updated under Windows does not infuence the temperature and performance (No difference with the 5E I have under Windows)
You mix the microcode "hardware" of the bios, and the microcode "software" of Windows
The 4D that goes back some time and the last in date, the 5E does not change the temperatures either

We will stop there, between your errors on the CPUID of CL-X (new "architecture" of CPU and same CPUID.... :D) and your errors that the last microcodes modifies only the temperarutes in idle (by thinking about it, I do not see how besides, except a magic trick, maybe :D :D)

I'm sure you'll find other topics to tell nonsense ;)

---

I came to this thread initially to post modded bios for the Omega, the board I own, as that is the main topic here. U are the OP and u mod Asus bios to include the latest modules and microcodes using the tool UBU made and updated by SoniX.

I then saw u were speaking about Retpoline in these cpu's and LOL cos everybody knows it doesn't work with skylake and newer cpus, and u even tested the matter to get more LOL.

Then u are used to mod your bios using the latest microcode and your adored 49 and u point out temperatures, saying 49 shows lower temperatures, about 5 degrees. And then we don't understand each other, and I try to explain myself.

If u have a bios with -5 and -5 tjmax, and another with +5 and +5 tjmax, what's the hell of difference??? NONE, since if u don't overclock your system and go stock way, u won't reach dangerous temperatures, as 5 more degrees mean nothing, and if u want to push your cpu manually to get the best of it, u have a 5 degrees margin in your TJMax, then the practical result of having your adored 49 in your system is placebo and more placebo. Don't u really understand it??? Asus and Intel engineers are so so stupid to update and reupdate bios and microcodes so that u get poorer performance??? Guess no... They just do their best to enhance security and don't lose much performance.

Then u say there is a part of hardware and software side in microcodes. The microcodes are software to rule the hardware, and Windows overrides the bios microcode when the bios has an older version, so performance under Windows should be ruled by Windows microcode, which is the same kind of the bios microcode, in fact u can find new microcodes in the mcupdates files for AMD and Intel in the latest Insiders, and these new microcodes are then integrated in the bios's via UBU for example. U can have lower temps (who mean nothing as I said) because of reading sensors, but performance should be ruled by Windows microcode. That's why people use the workaround to force bios microcode over Windows one.

Then u talk about my errors with cpuid. Guy, if I collect every single nonsense u have been saying along your years here, I can make a book. Don´t make me laugh. I just told u the latest ones, such as blaming SoniX for updating the present microcodes in certain bios's, or the retpoline, or the nonsense of temperature=no practical effects, which u can't get into your head!!!

But then u come here with replies from Intel and Microsoft contingent workers to make me laugh and laugh and laugh, as is the case of Retpoline about which u even brough a link in which they say it is not compatible. Everybody makes errors, but u saying others they make errors is the big LOL!!!

And I don't recall a 50 microcode version, which, by the way, are in hex...

And one final thing, I have tested lots of mobos and processors, and found differences in temps with the same processor and different Asus boards. For example, the Asus RVE and RV10 had a 7-10 degrees difference which was not explainable!!! So the cpu temperatures readings is a matter who can be used to write a book!!!

Ok, if having 5C (or +) more in load do not bother you, why use a watercooling ? A Direct Die ? Etc... ?!
Because on one side you want to gain "in temperature" but on the other side you let a microcode increase the temperatures
A little contradictory :D

I do not necessarily speak for you ;)

Me and that's only me, I use a microcode that has a TJMax planned initially for the first generation of Skylake-X (7xxx)
The microcode with the TJmax 110C being released at the same time as the 2nd generation of Skylake-X (9xxx) I believe and maybe there is no difference with these 9xxx
the 7980XE and the 9980XE, for example don't have the same TJunction and probably a difference "initially" at the TJmax

And I was posting bios with microcode 49 because some people asked, otherwise I would not post it

Quote:

---

Originally Posted by tistou77

Ok, if having 5C (or +) more in load do not bother you, why use a watercooling ? A Direct Die ? Etc... ?!
Because on one side you want to gain "in temperature" but on the other side you let a microcode increase the temperatures
A little contradictory :D

I do not necessarily speak for you ;)

Me and that's only me, I use a microcode that has a TJMax planned initially for the first generation of Skylake-X (7xxx)
The microcode with the TJmax 110C being released at the same time as the 2nd generation of Skylake-X (9xxx) I believe and maybe there is no difference with these 9xxx
the 7980XE and the 9980XE, for example don't have the same TJunction and probably a difference "initially" at the TJmax

And I was posting bios with microcode 49 because some people asked, otherwise I would not post it

---

Guy, please read my posts, I guess u are not reading/understanding. I said I tried every single microcode with my cpu to its limits, and I didn't find 49 allowed higher overclocks. If your theory is true, as I said, u can get higher OC's (or make them stable) with 49, but that is not the case.

I have never tested and will not test an OC higher than the one I have (4.6ghz) ;)
the only thing I see is that my CPU heats less with microcode 49 (with the same OC) and therefore the TJMax (105°) originally planned for 7980XE (which is normal by the way)

Quote:

---

Originally Posted by tistou77

I have never tested and will not test an OC higher than the one I have (4.6ghz) ;)
the only thing I see is that my CPU heats less with microcode 49 (with the same OC) and therefore the TJMax (105°) originally planned for 7980XE (which is normal by the way)

---

All the mainboard manufacturers have changed TJMax to 110. Gigabyte did it from the very first moment...

Nowadays Intel cpu's won't get burned for high temperatures. Voltage and amperes sustained in time are way more dangerous!!!

@vmanuelgm, On my testing both F5 ASUS load optimized default and OC to 4.5GHz the result showing those improvement together.

@tistou77 , Hi master tistou77 nice to see you hear again. Intel Turbo Boost MAX 3.0 October 2019 update will backward compattible with our SKUs as well master. The only different between Cascadelake X and our Skylake X SKUs was Intel Performance Maximizer Tool October 2019 update.

@restsugavan, nice to see you too and yes, always there, as long as I can help with my modest contribution :D

Ok for TBM, thanks for the information
I would surely test the 10980XE, especially if it is soldered :D

There are some good new to our X299 platform master tistou77 . :)

Today Intel had launch both XEON Cascade Lake W and X together. Both X and W Cascade Lake coming with better optimized for all X299 motherboards.
The XEON W Cascade Lake LGA 2066 support Optane DC persistence memory upto 3TB per system. (Upto 6 x 512GB Optane DC DIMM and 2 x DDR4 Module) for 1 socket SKUs. ( It will support upto 6TB for 2 Socket system.)
Attachment 82390
Attachment 82389

For Cascade Lake X first debut the same feature may temporaly not availible yet. However in longterm before EOL Q2 2020 Intel may grant support Optane DIMMs as well.
Attachment 82391