Use MMtool to insert ffs in Dxe core or Csm dxe in the original bios, then use the old ubu method to avoid having microcodes in the bios
C-Payne described that i told you, severall page ago.
I did as he said
We look forward to writing how to do it, step by step
I did as he said
We look forward to writing how to do it, step by step
Yes, I had written a guide to make a mod bios few months ago,but it was hard to read because the language. I will write an guide in English.
I'm interested to make my own ffs, if i can.
or to make my own efi.
They are variables from one system to another and maybe they do not work that good at all!
or to make my own efi.
They are variables from one system to another and maybe they do not work that good at all!
Thanks to Canonkong I successfully flashed an Asrock x99 Extreme 3.1 mb, now I have an Asus Z10PA-D8 modded bios, but when I use ez flash tool, there is this error "image integrity check failed", I renamed as original bios too, any suggestions are welcomed.
You can use FASM and efi.inc，input the source code and edit in the FASM,and run compiled ，an efi file will build up.
Dufus
Senior member
- Sep 20, 2010
- 675
- 119
- 101
Unfortunately it's in a non-user friendly state and would need some work to make it better, besides I think the point has already been shown and power limits still need to be looked at especially with AVX2, at least that is what I have seen so far with the 2683. I'd really like to test an 18 core at 3.8GHz and also some lower stepping 1 or 0 CPU's but prices are way outside my budget. Perhaps they will drop soon with Thread Ripper being released.
Hey, I have made the V3 mod a long time ago, I have read some pages, but I can't understan correctly, can the new version of mode improove better performance and stability with my 2683v3? For example in AVX (the old verion reduces frequency in AVX in Corona render for example).
Another question is there a new way do do the mod? I have downloaded the CANONKONG bios mod for my Asrock X99M, now I'm confused, do I have to install the v3.efi once more, or I need to instal the BIOS only? And what with the microcodes? Still as the old way?
And once another question. Can somebpdy tell me is this mod possible on the V4 Xeon?? I'm considering what to buy next, a two pieces of 2686v3 xeon and a new MB, or stay with this MB which I have and buy a 2696v4 xeon to mod.
Thanks for the reply
Another question is there a new way do do the mod? I have downloaded the CANONKONG bios mod for my Asrock X99M, now I'm confused, do I have to install the v3.efi once more, or I need to instal the BIOS only? And what with the microcodes? Still as the old way?
And once another question. Can somebpdy tell me is this mod possible on the V4 Xeon?? I'm considering what to buy next, a two pieces of 2686v3 xeon and a new MB, or stay with this MB which I have and buy a 2696v4 xeon to mod.
Thanks for the reply
You don't have to install the v3.efi anymore, you just need to instal the BIOS only. And the microcodes as the old way,most systems can automatic install by the systems itself,you can use tools to update the new one----3A microcode.
I miss a lot....
1. Is there any performance different using ffs v efs?
2. I test payne verssion rly nice.
There are lower voltage. But in AVX I can see throttling to 2,7GHz while v3x2_50_39.efi - I got 3,2GHz-3,3GHz.
1. Is there any performance different using ffs v efs?
2. I test payne verssion rly nice.
There are lower voltage. But in AVX I can see throttling to 2,7GHz while v3x2_50_39.efi - I got 3,2GHz-3,3GHz.
- Update ASUS X99 mainboard wiht FFS MOD BIOS，if lack some of the mainboard models,tell me, I will supplement.
- Thanks to dufus and C_payne share their sourse code,if they don't share the sourse code,we will not have E5V3 tubor boost,thanks again!
- I had made most MOD BIOS for in need.
- MOD BIOS TOOLS AND FFS FILES：http://pan.baidu.com/s/1qY0THJQ；
- ASrock X99 mainboard wiht FFS MOD BIOS：http://pan.baidu.com/s/1nvG8Zfr ；
- MSI X99 mainboard wiht FFS MOD BIOS：http://pan.baidu.com/s/1jHHeOvG；
- EVGA X99 mainboard wiht FFS MOD BIOS：http://pan.baidu.com/s/1slJOHDr；
- ASUS X99 mainboard removed Microcode MOD BIOS：http://pan.baidu.com/s/1kVFrs15；
- ASUS X99 mainboard wiht FFS MOD BIOS：http://pan.baidu.com/s/1ge4tWP1；
- GIGABYTE X99 mainboard removed Microcode MOD BIOS：http://pan.baidu.com/s/1hsGJrSS；
- GIGABYTE X99 mainboard wiht FFS MOD BIOS：http://pan.baidu.com/s/1i5OOHbn；
- GIGABYTE NEED EFIFLASH MOD TOOL to flash the MOD BIOS IN DOS：http://pan.baidu.com/s/1dEUMqDN；
Guys, I have ASROCK EP2C612 WS and TWO E5-2683V3 SR1XH and have the following questions:
1. Is this combo known to successfully work to make all cores 3.0ghz?
2. How much hotter does it get and is it feasible to run like this for more than just a very short term?
3. If running dual cpu (14 core x 2), with all cores at 3.0ghz, how would the performance compare to Ryzen 1800x or Threadripper 1950X?
The motherboard and 2 CPUs cost $1240. Threadripper 1950x + ASUS PRIME X399 costs $1345, or 8% more. If I were to resale the former to get the latter, the actual cost would be greater than 8% since I wouldn't likely get what I paid for it but somewhat less. Would it still be worth getting Threadripper 1950x?
1. Is this combo known to successfully work to make all cores 3.0ghz?
2. How much hotter does it get and is it feasible to run like this for more than just a very short term?
3. If running dual cpu (14 core x 2), with all cores at 3.0ghz, how would the performance compare to Ryzen 1800x or Threadripper 1950X?
The motherboard and 2 CPUs cost $1240. Threadripper 1950x + ASUS PRIME X399 costs $1345, or 8% more. If I were to resale the former to get the latter, the actual cost would be greater than 8% since I wouldn't likely get what I paid for it but somewhat less. Would it still be worth getting Threadripper 1950x?
hi, Dufus
If you may share your tools or HOW TO, we will do the test.
There are so many 2696v3 here.
Okay, there's a lot to unpack here. I don't have the specific CPU in question, though I previously had a pair of E5-2666v3 for several months, and currently have a pair of E5-2696v3, using the all-core turbo hack on both.
1) Let others with the exact chip confirm their findings, but my guess is probably not. I'd expect to get 2.8GHz on all cores; beyond 10 cores, the hack seems to have a slightly reduced effect for really high core counts. 12 cores = max - 100, 14 cores = max - 200, and a little less still for 16 and 18. In part, this depends upon the power capabilities of your board, and possibly using a VCC hack in a addition to the turbo ratio hack.
2) Definitely does get hotter, exactly how much depends on a lot of things, however, if you install a decent cooling system, you should be fine. I used my 10 core 2666s for several months, with Hyper212 coolers on them (which are in the segment I'd call "mediocre for serious overclock performance") and would only break 70C with serious non-real-world effort. For my 2696s, I found those didn't quite do it anymore, and now use a pair of Corsair H90 closed loop coolers. I can break into the 70s still with a good load, but that seems fine. Remember that Xeons are, in part, designed for usage in rack-mount systems which have pretty lame 2U height cooling systems in them. Most serious desktop cooling systems are already a step up from that. So, in summary, short answer, treat your cooling seriously, as though you would for any kind of overclock, and you should be fine.
3) Obviously this depends a bit on if you're talking single core or multi-core performance. For single core performance, you're going to be a bit behind AMD offerings, even for things that use a small handful of cores (e.g. gaming). In a lot of real-world situations, whether or not that makes a difference you care about is up to you. Obviously Intel vs AMD comparison can vary based upon the specific workload, but most of what I've seen suggests that IPC of Ryzen is somewhere in the Haswell-broadwell range on average, or in other words, clock speeds are [close to] comparable. For an 1800x you'd be looking at 4.1Ghz (max) x 8 cores, or 1950x is 3.9Ghz (at best) x 16 cores. There is some performance loss in a dual socket system as compared against a single package solution, but even still, 2.8Ghz x 28 cores is likely more than enough to overcome that in heavily threaded workloads.
Would you please make another BIOS with FFS MOD for ASUS X99-E WS/USB 3.1 with vcore offset -60 vcach -20. with -70 i have blue screen. I tried to make one by myself but after flashing the BIOS my pc does not post.
Last edited:
Can you help me with a bios ffs for a single CPU, Asrock X99 OCFormula that is like this: system agent offset voltage, vcore additional offset & cpu cahe voltage offset all to -70?
Or teach me how to create an ffs!
Last edited:
please add Asus dual mainboard series, Z10PA-D8
https://www.asus.com/us/support/Download/5/35/0/1/XmK8IaNbIeBunwQA/10/
As I know, not modded 2683v3 in dual CPU can make nearly 3200pts in Cinebench, OCed Threadripper 1950x can do it with 3450pts. Single 2686v3 can make 1750pts without mod, and 2100pts with mod - what is 20% more. So if You mod a dual 2683v3 You can get ~3800pts in cinebench. You have 10% more than OCed Threadripper. Only one cone, is that dual CPU consumes more power. But with dual CPU, You can also buy a 2696v3 and there will be much more advantage comparing to Threadripper. EPYC is the only better way for me
CANONKONG where I can get newer microcodes? I only have the old ones. And the second question, will it work in nearly future with V4 Xeons? I'm very interested in it.
Please make a mod bios with ffs mod for ASUS X99-E WS/USB 3.1 with offset -50 .
Last edited:
As I know, not modded 2683v3 in dual CPU can make nearly 3200pts in Cinebench, OCed Threadripper 1950x can do it with 3450pts. Single 2686v3 can make 1750pts without mod, and 2100pts with mod - what is 20% more. So if You mod a dual 2683v3 You can get ~3800pts in cinebench. You have 10% more than OCed Threadripper. Only one cone, is that dual CPU consumes more power. But with dual CPU, You can also buy a 2696v3 and there will be much more advantage comparing to Threadripper. EPYC is the only better way for me
CANONKONG where I can get newer microcodes? I only have the old ones. And the second question, will it work in nearly future with V4 Xeons? I'm very interested in it.
Thanks! Makes sense then that 2 x 2683v3 with mod would therefore surpass OC'd Threadripper 1950x in that benchmark. But is the mod safe to use 24/7/365?
As for another con, I guess 64 PCIe lanes of TR 1950x vs 40 PCIe lanes on c612 2683v3. I guess those extra 24 lanes might help those who want to do quad SLI or something (although from what I understand it's not supported in gfx driver, at least on Nvidia side, anymore). Either could adequately support dual GPU + at least 2 m.2 drives.
Last edited:
Dual CPU boards usually route PCIe lanes out from each cpu, so depending on the number of Wellsburg chips, could be 40 (CPU Integrated IO) + 8 (Wellsburg PCH) per CPU. Typically you only see one Wellsburg [edit: per board, not per CPU] though - multi-PCH setups are very unusual even in the server space. A typical dual CPU board would have 40+40+8 lanes, although the PCH lanes are usually used for low bandwidth devices like IPMI/video/gigabit, and only the CPU lanes are routed to PCIe slots.
The CPU's PCIe/DMI port 0 can be run as a PCIe x4 port, but I've never seen a board that routes them that way. When you have so many PCIe lanes, routing a rather bandwidth limited x4 port out isn't worth it.
Edit: correct errors.
Last edited:
kjboughton
Senior member
- Dec 19, 2007
- 330
- 118
- 116
TRENDING THREADS
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 9K
-
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 7K
-
Facebook
Twitter