A7N8X-E Deluxe as an alternative for socket 462

[Tzk]

I checked the BPL of the biosfiles you linked above. I read somewhere that there's BPL 3.23 for Nforce2, but all bios files i checked either use 3.02, 3.04, 3.17 or 3.19. I've never seen any BPL >3.19.

Regarding the option rom: My idea was to replace AWDflash.exe with for example memtest86+. On A7N8X you can enter awdflash by pressing ALT+F2 during post. It seems like the awdflash that is integrated in the bios is just the regular exe file which you can run in dos (via bootable floppy). So i tried to find a dos Memtest86+ version or use for example the Memtest86+ DFI ships inside the NF4 Ultra-D bios. However DFI uses two files for this on Ultra-D: one that "prepares" something and memtest itself. So my guess was that this isn't suitable for my NF2 board. I haven't tried any other Memtest file yet as i'm waiting for a USB Nano bios programmer to make testing a lot easier (Link: https://www.ebay.de/itm/NANO-USB-Programmer-for-PC-M-B-BIOS-repairing-with-Economic-shipping/271313593344). Let's see if that little stick is worth the money...

And i found a great source for adding fuctionality to bios files: https://www.lejabeach.com/sisubb/Practical_BIOS_Editing.pdf We'll see if it's possible to add additional options to bios, i guess.

I also discovered this link, but haven't had time to look into the pdf to see if it's of any use to us: https://github.com/pinczakko/BIOS-Disassembly-Ninjutsu-Uncovered

Edited October 12, 2019 by Tzk

[I.nfraR.ed]

That PDF is interesting. I've done most of the things in the first part, but I'm not willing to spend time with the part where he goes into assembly.

I've done reverse-engineering on an android phone kernel to get various calibration data tables of sensors for use in a custom kernel compiled from scratch.

And it is a lot of work. At least you can understand what is going on with the integrated memtest.

PS: My boards are dying one by one. First the AN7 problem with newer CPUs and even if it boots with some CPU I can't save any settings in bios. Perhaps the uGuru chip is bad. Then I had 3 NF7-S stored as working and 2 of them don't react to power button. Changed capacitors and still no sign of life.

I have a willem programmer and have re-programmed some bios chips, but it didn't help. The spare AN7 doesn't do that high FSB like the old one.

Edited October 12, 2019 by I.nfraR.ed

[Tzk]

Are you sure it isn't other hardware killing your boards? Like a faulty psu, gpu or something?

I'm tempted to try the assembly stuff in the pdf, not sure i'm willing to spend enough time on it to get results though. Especially the Cas timing example got me. Imagine you could add drive strength, slew rate and alpha timings to the bios instead of setting them via NF2 Tweaker... I'm aware that this is a fairly advanced task and involves a lot of assembly and reverse engineering. And i'm not a software engineer, so this might be way beyond my abilities.

EDIT:

I investigated Memtest86+ further. The bootable Memtest.bin seems to consist of 3 parts: 1. a sort of bootloader, 2. a setup routine, 3 the program itself. Now, DFI and Biostar split the Memtest.bin inside their bios in 2 parts: the setup routine (memsetup.rom) and the program (memtest.rom) and it looks like the routines are mostly identical to the bootable Memtest.bin.

So the next step is proably to replace awdflash with memsetup.rom, add memtest.rom and see if that works when using ALT+F2. Ironically Awdflash on the latest official bios (1008) for my A7N8X Deluxe v2 seems to be broken... on the penultimate version (official 1007) it works. Still i'll wait till my usb flasher arrives, makes testing a ton easier.

Edited October 12, 2019 by Tzk

[Mr.Scott]

You'll like the Nano programmer. Works great. Well worth the money.

If you manage to find the 3.23 BPL for NF2, please post it or link me to it.

[Tzk]

@Mr.Scott

Someone over at HWLuxx uploaded MSI Beta bios files. It contains NVMM v4.35 instead of NVDAMC v3.19 (what we call "BPL"). No clue why MSI did this or if it's possible to port them over. He just warned that these bios file are unstable and shouldn't be flashed onto a board. But for research purposes it should be fine.

DL Link: http://bierbude.spdns.org:2302/USER UPLOADS/digitalbath/

My guess is that these files are mentionend here: https://forum-en.msi.com/index.php?topic=84715.0

---

A thing made me curious. MSI used NVMM instead of NVDAMC for the latest beta bios... And i wondered why Nvidia should've kept developing their IMC driver/interface (or whatever this is), as NF3 und NF4 for AMD doesn't have a IMC anymore. But there was the NF4 Intel Edition for LGA775. And guess what... Asus P5ND2-SLI bios ships with NVMM v4.90... So i guess MSI took the NVMM from newer boards and tried to backport to NF2. Interesting...

EDIT2:

MSI K7N2 Delta2 got NVMM 4.35 in official bios, not our classic NVDAMC ("BPL") v3.19... Question is: can we port NVMM to NVDAMC bios? Might try that when the flasher arrives...

Edited October 12, 2019 by Tzk

[TerraRaptor]

2 hours ago, Tzk said:

why Nvidia should've kept developing their IMC driver/interface

memory controller shouldn't be too different to those in gpu cards - probably the difference is the gpu imc is bandwidth optimized. I can try flashing nvmm 4.35 if you send it to me. 

I'm also close to complete checking romsip registers and will share findings  tomorrow once I compare custom romsip based on those findings  to other popular ones - seems I found something interesting performance-wise

[I.nfraR.ed]

I tried that v4.35 and it works on NF7-S. Can't tell much more right now, needs testing.

[Tzk]

14 hours ago, TerraRaptor said:

I can try flashing nvmm 4.35 if you send it to me.

 Sure! Here's NVMM v4.35 taken from MSI K7N2 Delta2 bios and NVMM v4.90 taken from Asus P5ND0701 (Nforce 4 SLI LGA775). I'd expect that v4.90 won't work on NF2, but i'll probably try it anyways.

nvmm435-cpc-on.bplnvmm490-NF4-cpc-on.bpl 

And here's the other BPLs i collected up to now:

BPL302-cpc-off.bplBPL302-cpc-on.bplBPL304-cpc-off.bplBPL304-cpc-on.bplBPL317-cpc-off.bplBPL317-cpc-on.bplBPL319-cpc-on.bpl

And the Romsips i currently have (just to have everything in a single post):

1008mod3-6-19-1Td26-manta-XT-1TDFI-1-21-1TDFI-12-18-1TRomsip TaiPan [EB][ED]Romsip TaiPan 0.3 (ED]

Edited October 13, 2019 by Tzk

[I.nfraR.ed]

@Tzk Thanks for the collection!

I've tried the 4.90 as well, but it didn't boot, as expected. The 4.35 dropped my max FSB a little. The main difference I noticed are the tighter default timings compared to usual nforce2 bpl. Latencies are good, but nothing extraordinary. And I had problems booting with locked Sempron at 9x multiplier.

This might make it incompatible with certain DDR sticks. Main and subtimings are very close to as tight as you can get with BH-5 on this platform. 

Edited October 13, 2019 by I.nfraR.ed

[Tzk]

You're talking about alphatimings, right? Stock on my Asus is 2-3-5-3-2-2-4 (read from top to bottom in NF2 Tweaker, right column) and i usually try to use something like: 1-3-3-3-2-2-3. What does MSI ship the v4.35 with?

Well, if stability suffers and bandwidth+latencies aren't getting better, we're probably better off with BPL 3.19.

Edited October 13, 2019 by Tzk

[TerraRaptor]

Here is what I checked for romsips with sempron 2200+:

However, trying to combo the above with either 69-4F-22-00-00-EB-FF-FF or 69-41-22-00-00-EF-FF-FF (not sure which one) I run into strange issues - even with stock bios system is not stable anymore and may freeze even in bios setup, spd in DIMM3 gets corrupted on every boot. I hope it may be damaged CPU - haven't checked yet - if "15-18" is a kind of drive strength with FF-FF being an overkill setting then probably changing cpu may resolve it (will check it today).

What surprised me is boost in write/copy with EF - taking into account the theoretical bandwidth of 4000 @250Mhz, those numbers are interesting.

Still, I don't know what has happened to my system and care should be taken to avoid damage to mb/cpu/ram.

 

Edit - should be something described here:

 

Edited October 14, 2019 by TerraRaptor

[Tzk]

So you basically replaced the values per romesip one by one with 0 or F and tested all combinations? Have i got you right on this one? Interesting results nonetheless. The difference for the EB setting (so either E0 or EF) is huge... That's what i more or less observed when testing different romsips.

TicTac posted this regarding the romsips but sadly didn't write down why he does it (besides "multi working again")...

Quote

- search for romsip table (address: 0000b300h - 0000b90h )
- u will see there is 6 table here
- change the value with :
6Y AX 3A 02 to ---> 4Y 2X 00 00(multi workin again)
exp:
68 A4 3A 02 to ---> 48 24 00 00(multi workin again)

(Y,X = can be any number..... and it is not y or x on actual bios
so if it is 61 -> 41 if 68 -> 48)

- then save the modded system bios file
- put it back in with Hack version of modbin 6


should we replace the 6x Ax CA 02 to ---> 4x 2x 00 00 too???
yes change that to ...... change it as 4X 2X 00 00
u get me enduracell... nice

http://www.xtremesystems.org/forums/showthread.php?41183-Modding-ecs-nf2-romsip&p=529259&viewfull=1#post529259

I also read about several mod bios with Multi 10 and 10.5 stability issues where these multis had 15 10 for the last two values of the sips. Usually when a modder called these issues fixed, then it was set to for example 16 18 or 15 18. So it seems that the last digit does something to stability. Interesting that you observe no difference with 10 or 1F set.

Regarding your issues: I'd power the system off, swap bios chip, ram, cpu and leave it without battery for a few hours. Sometimes this helps...

Edited October 14, 2019 by Tzk

[I.nfraR.ed]

In my original bioses I've followed that recommendation and changed all these values.

On a side note, the memory timings start from b0/d0/f1 offset 90 (IIRC, at work right now). I've tried to play with some values in the near offsets, without much luck.

Will try your values in the evening, but will change the CPU first, since I'm using one of my best XP-M in the moment.

Tried to find a similar romsip table in VIA KT600 bioses, again without a positive result. Also checked NF3 and NF4 bioses and although there are some tables, it's not the same as in nforce2 bioses.

 

PS: Btw, I think a better option might be a KT880 board, eventually modding it with external PLL, since it has no PCI/AGP lock. The mem controller is better than nF2 and you can also change multiplier in windows, which is useful for super-locked CPUs.

 

Edit:

S2K Bus Disconnect is b0/d0/f0/6F, data set to 1F (or 10). Perhaps we can dig into that direction and the S2K drive strengths are also there and we can control the values runtime. The S2K should be common/similar for all K7 chipsets. But be aware that all documents say it can be dangerous for the CPU and damage it permanently.

I remember using S2KCtl application back in the days, to control the bus disconnect feature. CPUId can also control HALT states and divisors. Have to check if something changes accordingly when you modify the romsip table registers from @TerraRaptor's testing.

 

Edit 2:

ROMSIP from Nforce4 Ultra-D 702-1 bios (BH-5), located in decomp_blk.bin

000061c0: 65 D0 16 2B 00 00 00 00 00 00 00 00 10 4C FF FF    eP.+.........L..
000061d0: 08 00 03 00 00 00 00 0F 1C 70 E0 81 FF FF FF FF    .........p`.....
000061e0: FF FF FF FF FF FF FF FF 00 00 00 00 00 00 00 00    ................
000061f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006200: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006210: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006220: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006230: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006240: 21 41 24 00 00 E4 16 1D 21 41 25 00 00 E4 16 1D    !A$..d..!A%..d..
00006250: 21 41 25 00 00 E4 16 1D 21 41 26 00 00 E4 16 1D    !A%..d..!A&..d..
00006260: 21 41 23 00 00 E4 27 25 21 41 24 00 00 E4 27 25    !A#..d'%!A$..d'%
00006270: 21 41 25 00 00 E4 26 1D 21 41 26 00 00 E4 26 1D    !A%..d&.!A&..d&.
00006280: 21 41 20 00 00 E4 16 1D 21 41 21 00 00 E4 16 1D    !A...d..!A!..d..
00006290: 21 41 21 00 00 E4 16 1D 21 41 22 00 00 E4 16 1D    !A!..d..!A"..d..
000062a0: 21 41 22 00 00 E4 16 1D 21 41 23 00 00 E4 16 1D    !A"..d..!A#..d..
000062b0: 21 41 23 00 00 E4 16 1D 21 41 24 00 00 E4 16 1D    !A#..d..!A$..d..


000063a0: 65 D0 16 2B 00 00 00 00 00 00 00 00 10 4E FF FF    eP.+.........N..
000063b0: 08 00 03 00 00 00 00 0F 1C 70 E0 81 FF FF FF FF    .........p`.....
000063c0: FF FF FF FF FF FF FF FF 00 00 00 00 00 00 00 00    ................
000063d0: 03 80 80 80 4B 04 00 00 03 80 80 80 4B 04 00 00    ....K.......K...
000063e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
000063f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006400: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006410: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
00006420: 21 41 24 00 00 E4 16 1D 21 41 25 00 00 E4 16 1D    !A$..d..!A%..d..
00006430: 21 41 25 00 00 E4 16 1D 21 41 26 00 00 E4 16 1D    !A%..d..!A&..d..
00006440: 21 41 23 00 00 E4 27 25 21 41 24 00 00 E4 27 25    !A#..d'%!A$..d'%
00006450: 21 41 25 00 00 E4 26 1D 21 41 26 00 00 E4 26 1D    !A%..d&.!A&..d&.
00006460: 21 41 20 00 00 E4 16 1D 21 41 21 00 00 E4 16 1D    !A...d..!A!..d..
00006470: 21 41 21 00 00 E4 16 1D 21 41 22 00 00 E4 16 1D    !A!..d..!A"..d..
00006480: 21 41 22 00 00 E4 16 1D 21 41 23 00 00 E4 16 1D    !A"..d..!A#..d..
00006490: 21 41 23 00 00 E4 16 1D 21 41 24 00 00 E4 16 1D    !A#..d..!A$..d..

Very similar to "CPU Interface OFF" [E4] table from TaiPan 0.3, but the last value is adjusted "higher". There's also a difference in the first part of the table, but don't know if it can be used directly in NF2.

Same "multiplier" table can be found in nForce3-250GB UT bios.

 

Is decomp_blk.rom just the LHA-decompressor? If so, I guess it's no point of interest?

Edited October 14, 2019 by I.nfraR.ed

[Tzk]

6 hours ago, I.nfraR.ed said:

Is decomp_blk.rom just the LHA-decompressor? If so, I guess it's no point of interest?

Afaik yes. However it seems to ship the romsips, sooo.... Good question what it does on top of decompressing the system bios module. Very interesting that NF4 and NF2 share similar romsip tables. 

Edited October 14, 2019 by Tzk

[TerraRaptor]

I have swapped cpu for another sempron 2200+ and the stability problem seems to be gone. Seems EF is not possible with this cpu but works fine with EE and bandwidth is still strong 3950/3980/3750/64.5 - so common romsip tables seem to be good compromise in performance/stability (trats with EB gives 3930/3950/3700/64.7 btw for some reason doesn't clock as good)  - if one needs better performing romsip it's better create one with exact cpu as ultratight romsips won't work with every cpu. Also, I was able to pass 6 loops of 32M @267mhz tight (stopped it manually).

[Tzk]

After trying EBED/ED i also think that Merlin seems to have found near optimal romsips. High bandwidth, low latency and high mhz. There might be better romsips, but i don't think there is much room for improvement left, at least for high fsb. Mantarays XT seems to do a tiny bit better with the downside of limiting the overclock. So i'd consider ED the best option for max. fsb and MantaXT the best option <250Mhz fsb.

The KT880 Infrared suggested above might also be worth a try. But only if there's a way to get AGP/PCI lock. I'm curious how much fsb the Via chipset is able to handle... And i just found this

msi_ms-7047_rev_0a_sch.pdf

https://elektrotanya.com/msi_ms-7047_rev_0a_sch.pdf/download.html

And the same for MS-6570 (K7N2 Delta):

msi_ms-6570_rev_0b_sch.pdf

https://elektrotanya.com/msi_ms-6570_rev_0b_sch.pdf/download.html

Edited October 14, 2019 by Tzk

[I.nfraR.ed]

As far as I can see the max FSB is comparable to nForce2 and that's without the lock. Hooking up external PLL might allow it to go higher.

I've seen reviews that it is a little faster clock for clock and might allow higher stable DDR frequency due to a better mem controller. Also has higher bandwidth.

As for the diagrams, if you're interested you can make the physical FSB-detection switches with cables soldered at the back of the socket (L12-mod). I have them on the old AN7, but it doesn't help when you have a modded bios. Doesn't hurt to have it, though.

I've also made an entire multiplier switch board in an attempt to change the multiplier of a super-locked CPUs, but it didn't work. That's why KT880 would be better.

On Nforce2 you're limited by the FSB and with 9x multiplier (in the case with 2200+ Sempron) you can't go much higher, while on VIA chipset 2600+ is possible. If you find a golden chip you can even go to 3GHz with SS/LN2, I guess. I think it would be rather easy to beat @TerraRaptor's wprime scores even on KT600, but since it is single-channel only I couldn't beat his PI 1M score even at 2.6GHz (came close, but still slower). 

http://fab51.com/cpu/sempron/s10-e.html

Edited October 14, 2019 by I.nfraR.ed

[Tzk]

Then this all depends on the PLL on KT880... MSI seems to use ICS 952930DF on KT880 Delta, but i can't seem to find a datasheet for this chip. Anyways, i'll stick to NF2. Not gonna bench superlocked cpus.

Edited October 14, 2019 by Tzk

[TerraRaptor]

26 minutes ago, I.nfraR.ed said:

KT880 would be better.

Quite a while ago I had A7V880 and clock to clock it was slower than A7n8X. I remember I threw it away as it has stability issues over 250mhz but now I understand it was romsip issue. So Kt880 may be a solution but these are very rare boards.

[TerraRaptor]

1 hour ago, Tzk said:

AGP/PCI

Is it really essential? On s370, I had overclocks of 2x for AGP/PCI/IDE and it was okay with proper hardware (if we stick to pure 2D). Not sure if V-Link is the weak point though

[I.nfraR.ed]

Shouldn't be a problem with strong hardware. It's not a problem for the videocard. If the integrated controllers don't drop at high FSB then external PLL might not be needed.

My KT600 stops at ~240MHz with mods, but haven't messed with bios yet.

[I.nfraR.ed]

I have bought the frankenstein board Asrock K7Upgrade-880. The bios is not very tweakable, with most voltages missing (or having something like +10%, normal, high).

Will need voltmods for everything.

The good things:

- Unlocked Sempron 2200+ boots with its native multiplier x9 (and not the highest detected), which will help reaching high frequencies, that are not possible on nForce2.

- I can control the multiplier within Windows, which was the original goal

- Board works at 250MHz FSB no mods, could go to 260, but it crashed. Memory divider set to 166 right now.

- There's S2K drive strength in bios

- There are FID jumpers, which might be useful in some cases

Will have to check the bios if I can mod it somehow, unlock hidden options, etc.

This was the only KT880 board I could find. Came with a Sempron 2600+ and a cooler.

Edited October 16, 2019 by I.nfraR.ed

[Tzk]

I'm still waiting for my NanoFlasher and i need a new PSU for further bios modding/testing... Had to pull a psu from a working machine up to now

250FSB with 6:5 Divider means you're running 208Mhz on the ram? Also 82Mhz AGP and 41Mhz PCI and no issues with onboard components?! Usually i ran into issues like OS corruption, random freezes and HDDs refusing to work above ~37Mhz PCI. I also noticed that Asrock used AMI Bios on their NF2 boards, so you might have to build another toolchain to mod their bioses.

Depending on the bios layout you might also need a romsip mod, if Via does it like Nvidia.

Edited October 17, 2019 by Tzk

[I.nfraR.ed]

Yes, that's correct. AGP frequency is not a problem, the card can work@100+ MHz.

Don't know how much more the PCI could handle, but I guess it might not be a problem.

The board is a PITA to mod. It is using separate controller for the CPU, can mod it easily (bios limited to +10% which results in 1.775V for the test Sempron 2200+). I can go up to 2V via software in windows though. And there's a FB pin, so this should be an easy mod.

The rest (Vagp, Vnb, Vdimm and Vsb) are controlled by a single LM324. Vagp and Vnb are tied together, so incresing Vagp will increase chipset voltage as well. Vsb is 2.7V.

Vdimm is problematic. I guess something else is fed with Vdimm and when the DRAM votlage is increased to 3+ the board gets unstable and hardly even POSTs when 2 DIMMs installed.

If it boots, then it freezes randomly. The only solution I see is to use separate power for DRAM, e.g. lift the mosfet leg and get power from a DC-DC buck converter, but I will have to order some. "High" setting for Vdimm from bios translates to around 2.64V, but when 2 modules installed the voltage drops to 2.44V. I have tried direct 3.3V rail mod and it behaves the same. Also tried OCZ Booster (which is the same, but source is 5V rail) and still freezing, although I can't really use it for benching on this board, because the slots are very close to each other and it interferes with one of the slots used for dual-channel configuration.

On top of that, the mod area is very crammed with all the slots and elements around it. The LM324 chip is between first AGP slot and the expansion card slot.

As you might guess 2.44V is not really enough for anything with BH-5 and tight timings. Running different divider than 1:1 means performance loss. So..waiting for the buck converter.

Can't figure bios modding out. It's AMI, but all amibcp tools I tried say "bios checksum incorrect" and refuse to load. There's only one good bios for this board (Performance bios M1.00), which has the timings unlocked. It's a "beta" from Asrock.

Edited October 18, 2019 by I.nfraR.ed

[Tzk]

Have you tracked VTT? It's usually supposed to be half of Vdimm. I'd guess that raising Vdimm without VTT will get you some instablilities. With 2.44V VDimm you could use Infineon BE-5 or CE-5 to test high clocks. IIRC Infineon worked best at extremely low volts or a bit higher volts (2.4V or 2.9V).