No, it was a question for Sajin, sorry for any misunderstanding :)

Makalaure
Your power table for the 2 PSU rails are set to 375W each? I'm pretty sure they can only handle ~200W each. I know it will never actually draw 375, but still, why set them above what they can safely draw?

Kinda answered your own question right there.
 

Makalaure
Also, have you noticed what the second table does? I have been unable to find any information on it, other than it controls some "Internal Rail" that no one seems to understand, lol.

What second table are you talking about?
 

In the power tables, the first six refer to:
 
TDP
Internal Rail
PCI-E
PSU Rail #1
PSU Rail #2
Power limit
 
I have not been able to find out what the second one "Internal Rail", or the remaining ones in the table after power limit, for that matter.
 
For reference, if you look at your 1420X bios, you have them set to:
 
450
375
66 (75max)
375
375
350 (425max)
 
I noticed you increased the voltages on all of the lower clock states, was this just to ensure stability when the gpu cycles on and off load?
 
By the way, you can tell me to go away if you like, I won't be offended, I just want to learn as much as possible about this, and I have my own bios about 99% ready (my card has crappy 59% ASIC, but I have managed to keep it stable for a solid 8hours of Valley  [4k/0AA]so far at 1430/7400 @ 1.218V).
 

Makalaure
In the power tables, the first six refer to:
 
TDP
Internal Rail
PCI-E
PSU Rail #1
PSU Rail #2
Power limit
 
I have not been able to find out what the second one "Internal Rail", or the remaining ones in the table after power limit, for that matter.
 
For reference, if you look at your 1420X bios, you have them set to:
 
450
375
66 (75max)
375
375
350 (425max)
 
I noticed you increased the voltages on all of the lower clock states, was this just to ensure stability when the gpu cycles on and off load?
 
By the way, you can tell me to go away if you like, I won't be offended, I just want to learn as much as possible about this, and I have my own bios about 99% ready (my card has crappy 59% ASIC, but I have managed to keep it stable for a solid 8hours of Valley  [4k/0AA]so far at 1430/7400 @ 1.218V).
 

No idea what the internal rail setting is, everyone usually sets it to match the psu rail 1-2 values though. Yes, I increased the lower clock states to ensure stability when the gpu changes power states. Maybe the values after the power limit value have something to do with how the card reacts when in a power saving state.

Ugh, just flashed my latest bios in, and now it's being really weird. GPU-Z reports that it should be at my reported core speeds, but when it is under load it runs at my old default speed (1140) and a slightly higher voltage. This has now stopped making any sense whatsoever and it just seems to do whatever it wants. I have compared mine to yours, and the only key differences are the gpu/vram clock speeds, the voltage limits and a slightly different boost table. It is still stable, and I can use afterburner to still boost to my required core speed (alhtough then GPU-Z reports wildly inaccurate clock speeds. Since I have based my bios off the original hybrid bios, do you know if there is something they have locked in that will prevent the default clock speed from being increased via the bios? That's the only thing I can think of atm.
 
I am tempted to simply take yours and reduce the max voltage to something I feel more comfortable for everyday use.

Makalaure
Since I have based my bios off the original hybrid bios, do you know if there is something they have locked in that will prevent the default clock speed from being increased via the bios?

There isn't anything.

Yeah, I deleted my crappy bios that didn't work and started from scratch off my original one again. I think I had piled edit after edit after edit and probably had one digit off somewhere . Anyway, completely redone, now have everything stable 1405/7400 @ 1.199V with temps in low-mid 40s. No throttling of any kind, have not been able to cause any tdr's so far (which is something considering the latest nvidia drivers are terrible and buggy), although I have still have a load of testing to do. I took a leaf out of your book and basically applied ~+18mV to every clock state to improve stability. Not bad for a 59% asic card. I decided that overvolting wasn't worth it, as I could only max out at 1481 at 1.27V, and that made my temps go up to the high 50s. Even at 1.25V I could only go up to 1450, and temps there were mid-50s. I have to agree with kingpin here, GM200 chips do not like voltage, and 1.2 seems to be the magic number for stability.

also, overvolting the gpu is not the answer, unless your going extreme (LN2 or DICE)
and you can only unlock voltage limits on the 980Ti (not recommended and dont help anyway on air-water) on the Classified or soon to be KPE edition. ( or similar cards from other manufacturers. )

 

So the latest 361 driver seems to be a little more stable than all other drivers atm from the past for the 980 ti. I still get crashes every now and then with the 1420x bios but It is not the bios I assure you guys. It is NVidia Drivers -.- ughgh I just want to be stable at 1420 Mhz some people who have factory oc cards are even crashing with NVidia drivers in the NVidia GeForce forums