kevinc313
The FTW3 cards have a high power bios, in fact the 2080 Super FTW3 has higher power limits than the reference pcb EVGA A-chip 2080 Ti 130% bios.
 
The 2080S FTW3 Hybrid for under $800 is a REALLY good deal. Throw an additional fan ($15) on the cooler for push pull.  With proper radiator positioning and a good OC it will be pretty close to a non-A chip 2080 Ti at stock clocks on air (~$1000). 

Can confirm 2080S FTW3 hybrid in push-pull w/ rear placement is running great in my rig with huge OC headroom. I replaced the stock fan. The thing is a beast.

bavor

happydemon
 
Can confirm 2080S FTW3 hybrid in push-pull w/ rear placement is running great in my rig with huge OC headroom. I replaced the stock fan. The thing is a beast.

I upgraded my previous GTX 1080 ti FTW3 Hybrid to push pull fans on the radiator and it made a 3C improvement in maximum temperatures.  My motherboard could run fan speed based on the temperature of the device in PCIEx16 slots.  Its an inexpensive and worthwhile upgrade.

Agreed. It wasn't as inexpensive for me because I went for a Corsair ML120 and a Fractal Design HP, but it was still worth it. I saw a 4C improvement under stress. I was also concerned the radiator was heating up other components of the mobo when just using the stock fan. I'm not starving for power limits being increased because these hybrid models are pretty great as it is.

Sajin

bavor
I was searching about this topic because I had the highest 3dmark firestrike for my system configuration, but someone with a Gigabyte/Aorus card with a higher power limit(300 watts and a higher power limit increase%) recently beat my score.  I'm at the power limit and when Precision X shows that the card hits the pwoer limit, the GPU clocks drop.  That's preventing me form getting a higher firestrike score to beat them.
 
I had GTX 1080 ti FTW3 cards that were replaced udner warranty and EVGA sent me regular reference PCB RTX 2080 Supers as a replacement and not FTW3 models.  So I'm at the limit of what the cards can do due to the power limit.  I guess EVGA will settle for second best reference PCB power limits and let Gigabyte beat them.

Just flash vbios to your card to get a 350w power limit. You can use to perform the flash. Your max power limit will be increased to 351w. 
 
Scared about bricking your card? No problem. You can recover by doing if you brick your card.

Hello!  I've seen you post this suggestion in several threads regarding bios updates to up the power target higher than 112% Re: 2070 Super Hybrid.  
 
To your knowledge, has anyone actually tried the flash?  I've not been able to find a case of this yet on the forums.
 
Thanks!

Sajin
 
I've done it. I've flashed a 2080 ti FTW3 vbios to a 2080 ti xc gaming.

To your knowledge, has anyone flashed the 2070 Super XC Hybrid (08G-P4-3178-KR) with a different bios?

I'd also like to chime in and add to the requests for a 2080 Super XC Ultra bios update!
Do not understand why we're stuck adding 17% to the power :(

Has anyone flashed the ftw3 bios to the XC ultra variant yet?

ac21365To your knowledge, has anyone flashed the 2070 Super XC Hybrid (08G-P4-3178-KR) with a different bios?

 
***Do this at your own peril; I am not responsible for damage this may do to your card***
 
To answer my own question: Yes, I have.
 
I successfully flashed my "2070 Super XC Hybrid 08G-P4-3178-KR" with a "2070 SUPER FTW3 ULTRA 08G-P4-3277-KR" (90.04.76.40.17) bios off of TechPowerUp.
 
A few observations:
- The FTW3 bios' power limit is 260W at +100%, 338W at +130%
- The Super XC (Hybrid) has a total of 4 fewer VRM+chokes than the FTW3
- The stock thermal pads under the Hybrid's VRM+choke heatsink are inadequate
- On the 2070 Super Hybrid -with- the FTW3 bios: +112% power limit allows the card to draw ~300W in furmark @ 1950MHz
- My sample regularly boosts to 2100-2115MHz at 1.08V at lighter loads and ~2085MHz under heavier loads (after tuning in Afterburner)
 
Right now, I'm not inclined to let the board draw more than 300W (250W core).  I am not an electronics engineer, but I feel like the power delivery on the non-FTW3 card is inadequate to *safely provide the full power draw of +130% 338W *with appropriate headroom for overshoot. 
 
I have a thermal probe pressed against the VRM+choke heatisnk that barely got up to 50C @ 300W draw, which prompted me to check the back of the PCB with my IR temp sensor. The back of the PCB was 70C, which spooked me into lowering the power limit for now.  It feels like the VRM's are merrily cooking themselves right along under the stock thermal pads.
 
I did not replace the memory thermal pads nor the heatsink compound, as the core has never been higher than 57C (in part thanks to a pair of Noctua iPPC 2000RPM PWM fans).  Fans are controlled via Corsair Commander and curve is via a thermal probe stuck into the fins at the radiator's intlet.  Pump is powered by a Sata->fan header cable from the PSU.  And of course, I am using two cables from the power supply, each of which have proper gauge to carry the three ground pins of an 8pin connector through only two wires to my PSU, a Seasonic PRIME Ultra Titanium 750W.
 
*edits are marked with an asterisk
 
Thermal Grizzly pads are in, back of card temp tops out at 62C (before heatsink addition), and VRM heatsink warms up rather quickly under load.  1mm TG pads were WAY too thin, I had to double them up and then remove material until everything made sufficient contact.  1.5mm thick TG pads *should* be closer to drop-in than 1mm pads were.  I also added a row of tiny heatsinks to the back of the PCB opposite of each VRM, which helped take the edge off of things.
 
I looked up the specs of the (presumably) 7 VRMs that feed the core; they are rated for 45A max sustained draw.  About the worst case I've seen so far is Furmark: 250W core (300W card) would put 37.5A onto each VRM, assuming all 7 share equal distribution, at 0.950V.  I feel like that leaves enough headroom to run a 300W card limit daily, especially now that the heat is under control.