I'm not reading that wall of text.
 
Vulcan and Kev are seemingly concerned about 2 things a 2 things only:
1) Power drawn in watts vs a 2080ti
2) The price
 
Well, I've already complained about the price in the price reveal thread and I'm not nearly concerned about power draw the way you are. I dont care that it needs more power IM FOCUSED ON ABSOLUTE PERFORMANCE ONLY
So stay on topic
 
I posted that port royal score and you guys got genuinely upset. Since your so upset about his shunt mod,Lets use another score then by a standard 3090 on air at stock settings:
 
port royal: 13,613.
 
User posts over at Reset Era and has provided plenty of proof to the legitimacy of all his numbers
He also managed a small overclock of +75 as well as + 200 memory and posted some things regarding his OC.
 
now compare this to supped up 2080ti's via redit:
the best scores under water from users with high overclocks are at 9100 with majority of users scoring in the 8000 range
 
https://www.reddit.com/r/nvidia/comments/adx6ew/port_royal_benchmark_6600k_2080ti/
 
The 3090 is scoring more than 50% higher compared to the 2080ti. This is a huge performance jump and was not expected upgrading from 2xxx. ppl thought that they would see 30% at the top end.
3090 represents a large performance jump. 
 
Interestingly, this user has a gaming x trio with 370W Default Power Limit with a Maximum Power Limit of 380W. This is rather disappointing and represents the lower end of the spectrum 
 
The more impressive cards from evga will preform much better

clcorbin
I suspect if OCing is going poorly right now it is more due to overall process quality (they ARE refining the process as more lots are made and more data is created showing where they have holes) than due to all the good bins going to 3090s.  And it WILL get better as production is refined and the masks are eventually respun.

I feel like the TU102 picked up a solid 5% performance in the first 6 months of production.

Intoxicus

ehabash1

kevinc313
 
HE PUT A WATER BLOCK ON IT TO GET THAT SCORE.  AND A SHUNT MOD.  NOT STOCK AT ALL.
 
Barely beat a FTW3 3080 on a stock bios on LN2.

your confirmation bias is showing Hardcore.
It literally beat a professional overclocker who was running tests for over 5 hours with a HIGHER power limit bios over 800 watts and LN2 pushed to its near limit temperature threshold and got so cold they were worried about motherboard cracking.
 
And "barley beat? your talking about beating a world record by over 300 points thats a huge margin
 
This dude made a ghetto ass waterblock that is not officially supported and his vrms are getting too hot on the backside. Hes reporting temps over 60 core as well and that might as well be air temps
 
his power usage wasnt even close to 550 watts. The board he used it literally a base model with poor power stages simply not designed to handle it; and thermally wasn't allowed to even get close to the threshold 

Don't feed into Vulcan and Kevin's biased mental gymnastics. No fact will sway them and they've clearly decided what to believe regardless of facts

This is a personal attack. 
 
Vulcan and I are both more technically competent regarding this, that is why you cannot understand what we are talking about.  It has nothing to do with mental gymnastics.