• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

EVGA GeForce RTX 3090 Ti FTW3 Ultra

A real reviewer would of downclocked the 3090ti to the same speed as the 3090 which would of given people a better idea if the ti is Significantly faster than OC models of a base card......
This kind of downclocking hasn't worked for years, because of NVIDIA's Boost algorithms and different power limit settings. The only way to do this properly is to get the original BIOS and run it on a card that's compatible with this BIOS. Of course I've tried my industry contacts to find a 3090 Ti FE BIOS, but no luck. Might be buying a 3090 Ti FE to answer that question, but got other more important reviews to work on right now.
 
Hi,
Surprised the three options aren't
Normal
OC
Mining
:laugh:
 
thank you for the review

But i think like others said it would have been better to include 3090 custom cards at 450-480w (123% power limit increase), you can't compare 3090 Fe vs 3090ti custom it doesnt really make sense in my opinion and review is then biased. Comparing a 3090 reference card at 350w vs a 3090ti at 480w/500+w is a bit strange.
If you check others reviews, most use 3090 custom cards vs 3090ti custom cards, and so it show only 5-7% difference which is then not significant at all
 
Last edited:
thank you for the review

But i think like others said it would have been better to include 3090 custom cards at 450-480w (123% power limit increase), you can't compare 3090 Fe vs 3090ti custom it doesnt really make sense in my opinion and review is then biased. Comparing a 3090 reference card at 350w vs a 3090ti at 480w/500+w is a bit strange.
If you check others reviews, most use 3090 custom cards vs 3090ti custom cards, and so it show only 5-7% difference which is then not significant at all
You can read TPU's review of the ASUS 3090 STRIX OC. The relative performance at 4K is below:

1648659066477.png


Compare it to this card:

1648659126817.png


The difference seems to be around 7%.
 
I am bit of confused by some results of AMD cards. For Example in CP 2077 benchmark at 1440/Ultra I get 92.57fps avg and 59.75fps min. With Aorus RX 6800 XT with stock clocks, fast timing on vram and +15% power limit and 22.3.1. Drivers. In Horizon Zero Dawn at 1440p/Ultimate benchmark it averages at 162 fps.
 
when evga found out one out of three vram phases needs not to be cooled by contacting the mosfet with thermal pad on 3090.
they went all out on 3090ti with three out of three vram phases uncovered. and they also found out four is unnecessary.
last but not least, kemet smd caps are also no needed. taiwanese apaq is enough.

for the warm win 3!
 
Last edited:
I am bit of confused by some results of AMD cards. For Example in CP 2077 benchmark at 1440/Ultra I get 92.57fps avg and 59.75fps min. With Aorus RX 6800 XT with stock clocks, fast timing on vram and +15% power limit and 22.3.1. Drivers. In Horizon Zero Dawn at 1440p/Ultimate benchmark it averages at 162 fps.
I don't use integrated benchmarks, these are usually completely wrong and don't represent actual gameplay

How's your FPS running around in the city?
 
I don't use integrated benchmarks, these are usually completely wrong and don't represent actual gameplay

How's your FPS running around in the city?
Fps in game coresponding with what I get in build in benchmarks. Maybe SAM wasnt enabled or old drivers were used?
I don't use integrated benchmarks, these are usually completely wrong and don't represent actual gameplay

How's your FPS running around in the city?
Even those test does not represent actual gameplay, they use exact the same conditions even You run them milion times. They are 100% acurate and does not suffer from human error. If You test in game, few looks on ground or sky, few more blasts here and there and Your fps is all over the place. 5 runs each with slightly different result.
My ingame fps corespond to what I have seen during gameplay. Averaging around 85-95fps just runing/driving around with some dips to 70-75fps during intense combat.
 
Get me the numbers for 10 runs please
Challenge accepted When I get home from work and if kids gove me at least 30 minutes to do it.
 
Hello team,

What about the Eth hash rate mining average?

Ty
I can't say for sure on what the hash rate is, but I've read that the 3090Ti is not LHR. I thought the 3090 was around 120....I'd venture to guess the 3090Ti should be around 130ish.
 
I can't say for sure on what the hash rate is, but I've read that the 3090Ti is not LHR. I thought the 3090 was around 120....I'd venture to guess the 3090Ti should be around 130ish.
Well, the thing is if you have 130MHs but a high power consumption, miners will run away from this card I suppose. Not sure eh
 
Well, the thing is if you have 130MHs but a high power consumption, miners will run away from this card I suppose. Not sure eh

I haven't looked into the power draw folks are dinking around with trying to get on it, but I saw a few posts making mention of around 135 hashrate on the 3090Ti. So I'm not sure if the small gains over a 3090 would be worth it if the power draw is a lot higher.
 
I haven't looked into the power draw folks are dinking around with trying to get on it, but I saw a few posts making mention of around 135 hashrate on the 3090Ti. So I'm not sure if the small gains over a 3090 would be worth it if the power draw is a lot higher.
Exactly, so probably the stock will be better than before with old models. Thank you for your info.
 
I am very thankful to Microcenter for being able to purchase two 3090 FTW3 for just $1799 ($1955 with tax) during the release of Cyberpunk. I sold my 2080Ti - and I could clearly see the 3090's benefit over that 2080Ti immediately in not only Cyberpunk but Counterstrike and DCS.

I will definitely skip over the 3090Ti. My case currently isn't big enough, my CPU is aging and I can wait till the 4090 before upgrading my entire system. I need a 13th gen intel 13900K and DDR5 RAM.
 
Back
Top