TerraRaptor

Another >100% overclock from you, nice chip there:)

@I.nfraR.ed As you were looking bios design guide, here is quite interesting one for amd 762 which also gives some better insight on romsips (or SIP ROM :)). http://www.datasheetcatalog.com/datasheets_pdf/A/M/D/-/AMD-762.shtml Looks that EB/ED thing relates to slew rate and is actually 3-bit entry. So my previous research on replacement values with 0/F is a bit wrong due to varying lenght of some entries.

Strong cpu. What temperature it was?

Well, I have better performance in pifast and wprime with pifast being sensitive to the same things pifast does as well. Please note that bwanasoft uses win7/vista for this super-efficient result - nonsense actually. I beleive that my efficiency in superpi is usually VERY strong and I'm nowhere close. Even with custom romsips I'm behind him. Just for those who are new enough to know why I claim banning all his results (just review the names in that post questioning his results) - I hope that 10 year after, when all the "friends" of bwanasoft are now away from hwbot, action will be taken.

May this result be reviewed by the mods (taking into account the background of bwanasoft)? I can't beat this result with almost the same clocks and much better timings (he has single channel 3-4-4 here and I have dual channel 2-2-2) - the result is obviously bugged/photoshop. Prpbably all the results of bwanasoft should finally be removed from hwbot.

Great score!

Not very likely. Many of bh5 have never been exposed to voltages over 3.2v due to mb limitations with their stock voltage of 3.6v. Another story is that no seller would bother sorting ddr stock to offer bh5 sticks nowadays.

Good score. What temperature coldbug is?

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

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.

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).

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:

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

Can you post a reference low clock result to compare - wprime efficiency was always a challenge for me and I spent lots of time trying to become competitive. I could probably cut another 0.5s here with good run but still thought efficiency is above average.

Старался:)

A16 is reset pin but pcie lanes come from x48 Edit: PCI Express Receive Differential Pair

Oh, it is that simple nowadays when old hardware is dirt cheap and you can have several ram kits, several motherboards etc. Back then, only some people were going mad with extreme hard mods etc and such a handy device was of a great help for people who wanted to keep the warranty etc. Even nowadays I see that vgas are being epowered mostly when they become outdated - and it's understandable. Will one solder ROG Dominus Extreme worth $$$$ paid from own pocket to gain 1% today - don't think so. That's why we have Elite league and Extreme league btw. I'm quite sure some of the bh5 records back then were set with either OCZ or hipro5's booster. Just another example of modern OC hardware - I bought Elmor's EVC2. Can it be replaced with something else? Sure. But it makes things easier as created "by overclocker for overclockers".

Anyone knows what password is used at that server? I need modded K7N2 Delta-L bios ver. 5.8

Nice!

Checked Tzk's ebed bios in dual channel - 275mhz tight is stable in pifast, 277 is not stable in pifast and ends with error, validates @280. Performance sucks. BTW, I don't need that wpcredit tweak with this bios and I feel like there is more scaling with voltage for NB/DRAM (nb is 2.1v and dram is 3.9v for these results). Thanks to Tzk I have now an idea what romsips are and will go deeper now with bios mods. I'm using aggresive cpu interface btw and haven't yet checked optimal.

Probably vid should be set - as I remember some VRMs are set off if Vid is all 0 or 1.

Nope, 2.5-3-3-7 single channel -I was mainly focused on max fsb (as ebed gives very big drop in aida64 numbers - >80ns latency with ebed compared to <65ns with trats-1T). I will try to break max s462 fsb here at hwbot first (although I know that back in the days there were several results with almost 300MHz done). I will put my bh5 under SS this week to check if they are able of 275 tight. I will also test performance/handmade romsips soon to build the best bios for my setup

I have checked EBED bios and it works fine. I was able to pass pifast @275 and it validates 282.5mhz https://hwbot.org/submission/4254667_terraraptor_reference_frequency_a7n8x_e_deluxe_282.46_mhz

Anyone knows if this is going to be fixed? As an option - probably these scripts can run on a different forum (as these were broken after forum migration) and all we need is just use external forum (let it be warp-9, OCN etc) and copy-paste scripts from here?

Great score! For my records - do these CPUs have coldboot/coldbug?