notarjy
Also as I expected, Hardware Unboxed did some gaming tests in their Gigabyte Gaming OC review that also posted today. They have a 3080 Ftw3 Ultra on hand so I expect we will see some results from them anyways. I prefer GN's attention to methodology but they are both trustworthy when it comes to benchmarks IMO
 

 
That and Jay's review once it comes out.

In a nutshell :
 
Out of box thermals:
TUF    @37.7DBA   336W   59.8c
FTW3 @43.0DBA   377W   66.0c  (OC Bios)
FTW3 @39.7DBA   377W   70.0c  (Normal Bios)
 
Normalized DBA @40DBA (all fans at same dba):
 
TUF      56.4c
FTW3   64.7c
EAGLE  60.8c
 
Gaming performance:
 
Who cares they are all within 1-3% of each other, to me Thermals and Noise > 1-3% more fps.
 
 
The TUF is also $110 cheaper.  How in the world did Asus achieved water cooled type thermals on air at 56.4c?

macktkau
The obvious point about including performance benchmarks is that you're doing a review of a specific card then you should be doing a review of that specific cards performance because that's an important thing to include in a review of a graphics card. Saying "all these cards at the same" is not the same as providing information about the actual performance. People are watching a video for a review of a card. If you watch that review you learn absolutely nothing about the actual performance of the product is.
 
It just feels lazy, this obsession they have with noise normalised everything makes me wonder if laziness crept into other aspects of the
'review'.

Gaming performance is FPS and smoothness. If that's the same (or within 1 or 2 FPS that is considered margin of error) there is no need to do graphs, it's a waste of time and energy. Especially since in real games, the clocks had to be lowered to avoid crashing, meaning this card is already OC out of box the heck out of it. You won't be able to get better FPS in games. You call him lazy, he's working 100h week in and week out. Not many people are as devoted, and honest as him while being very helpful and have knowledge (tech knowledge or social knowledge, knowing reps, people, etc). 

For this generation of RTX, only place where clocks and power consumption matters is synthetic benchmark, in which things will vary for people as silicon lottery can affect the OC. 
 
Noise normalized is awesome because it shows the coolers effectiveness at some level of noise chosen, you can't cheat out temperatures. This is especially visible when testing AIO Coolers for CPU when I remember Corsair's agressive fans were very very loud and if you reduced to a speed considered adequate for most people (way lower than it was by default) then the cooler was clearly not that good. 

And as long as you're providing charts for both default profile and equalized noise, it's fine, because you can see the difference between the two and choosing which card you want.
 

vegnar
In a nutshell :
 
Out of box thermals:
TUF    @37.7DBA   336W   59.8c
FTW3 @43.0DBA   377W   66.0c  (OC Bios)
FTW3 @39.7DBA   377W   70.0c  (Normal Bios)
 
Normalized DBA @40DBA (all fans at same dba):
 
TUF      56.4c
FTW3   64.7c
EAGLE  60.8c
 
Gaming performance:
 
Who cares they are all within 1-3% of each other, to me Thermals and Noise > 1-3% more fps.
 
 
The TUF is also $110 cheaper.  How in the world did Asus achieved water cooled type thermals on air at 56.4c?

The TUF has the memory and other components cooled separately from the GPU die, with dedicated heatsinks. That helps reducing temperature of all the components and clearly is working very good in this case. 

Lokator
yes, I am the guy slapping a Accelero Xtreme IV with two Noctua NFA12x25 fans on my 2080 Ti XC Black to have a quiet PC, so yes, all I want is a fantastic whisper quiet great GPU.

I'm that same guy, but with a Morpheus II on my 1080Ti.  Right now the TUF, or the STRIX (if you can even get any of them) seems like it might be the best option.  I want as quiet as possible without sacrificing tons of performance.

macktkau

Huntercyril
Gaming performance is FPS and smoothness. If that's the same (or within 1 or 2 FPS that is considered margin of error) there is no need to do graphs, it's a waste of time and energy. Especially since in real games, the clocks had to be lowered to avoid crashing, meaning this card is already OC out of box the heck out of it. You won't be able to get better FPS in games. You call him lazy, he's working 100h week in and week out. Not many people are as devoted, and honest as him while being very helpful and have knowledge (tech knowledge or social knowledge, knowing reps, people, etc). 
 
Noise normalized is awesome because it shows the coolers effectiveness at some level of noise chosen, you can't cheat out temperatures. This is especially visible when testing AIO Coolers for CPU when I remember Corsair's agressive fans were very very loud and if you reduced to a speed considered adequate for most people (way lower than it was by default) then the cooler was clearly not that good. 

I said it was a lazy "review" because it isn't a complete one not that they are specifically lazy people.
 
Open up Far Cry 5 & SOTR and do a 4k ultra settings benchmark run. Run time spy extreme once. Run port royal. Then put the results in the review. It would have taken 30 minutes maximum. You don't even need to add in comparisons, just show what that specific card can do.
 
In Australia noise normalised means nothing to me because in summer if I'm using an intensive game my case fans, aio and GPU fans will be running at 90% because the ambient could be as high as 45c, and the noise of the PC is just more noise on top of a screaming ceiling fan. I have headphones on. My only concerns about noise are idle noise, and max rpm noise.
 
For someone wanting information about how the card performs the "review" is a waste of time.

 
Yes, it doesn't take but a few mins to run in game benchmarks and 3DMark.
 
Even Metro EXODUS has a benchmarking tool.
 
And it's not like you have to be watching it either when they are running.
 
 

notarjy
It does seem like it was rushed, Steve and the crew have definitely been overworking themselves on this release. Unless I missed it they didnt really mention at all the differences in cooler design when covering the temperature results. Sure they covered in the separate disassembly but seems like it would have been worth the 30 seconds to mention. Same can be applied to lots of the stuff omitted. 

 
Yeah they ran tons of performance tests in the live streams.
 
But not a single one in the actual review of the card.
 
How do you review a gaming card without doing any actual gaming tests?

Gamers nexus never runs a benchmark just one time, because that doesn’t even remotely make sense. They always run a specific number of the same benchmark, and then average the results and remove outliers that do not count.

Most reviewers do that.

So, “just run it once” shows that most people don’t understand how much time reviewers spend on each card for reviews.

Also, just a few days ago there was a huge outcry because GN did the tear down on this series FTW3 on the card with 6 SPCaps instead of using the new card, because “that isn’t the most up to date product” yet nothing on the PCB changed except for the two SPCaps being removed and the MLCC array being soldered in place.

Now people are saying that the 6 SPCap card should have been shown in the benchmark.

Yeah, I think the content is exactly what needed to be produced, as I don’t think the exact same benchmark that every single reviewer is producing is necessary. Users have the information they need from other channels, and GN is producing information that other reviewers cannot.

the_Scarlet_one
Gamers nexus never runs a benchmark just one time, because that doesn’t even remotely make sense. They always run a specific number of the same benchmark, and then average the results and remove outliers that do not count.

Most reviewers do that.

So, “just run it once” shows that most people don’t understand how much time reviewers spend on each card for reviews.

Also, just a few days ago there was a huge outcry because GN did the tear down on this series FTW3 on the card with 6 SPCaps instead of using the new card, because “that isn’t the most up to date product” yet nothing on the PCB changed except for the two SPCaps being removed and the MLCC array being soldered in place.

Now people are saying that the 6 SPCap card should have been shown in the benchmark.

Yeah, I think the content is exactly what needed to be produced, as I don’t think the exact same benchmark that every single reviewer is producing is necessary. Users have the information they need from other channels, and GN is producing information that other reviewers cannot.

 
 
Some of us do understand because we used to do it before YT....
 
And the testing took even longer back then because everything was so much SLOWER.
 
Used to take at least a week or more just to test one graphics card.
 
 

macktkau
 
In Australia noise normalised means nothing to me because in summer if I'm using an intensive game my case fans, aio and GPU fans will be running at 90% because the ambient could be as high as 45c, and the noise of the PC is just more noise on top of a screaming ceiling fan. I have headphones on. My only concerns about noise are idle noise, and max rpm noise.
 
 

I just want to say that I'm sorry for your 45c summers.  I'll gladly pass on those.  My condolences.

Bcassiman
I gotta say my card is SINGING. 2100mhz stable in games and up to 2175mhz in 3DMark.

I think you got a unicorn there. Most of the FTW3 Ultra cards I've seen are performing worse than the FE unfortunately 

macktkau

IroesStrongarm

macktkau
 
In Australia noise normalised means nothing to me because in summer if I'm using an intensive game my case fans, aio and GPU fans will be running at 90% because the ambient could be as high as 45c, and the noise of the PC is just more noise on top of a screaming ceiling fan. I have headphones on. My only concerns about noise are idle noise, and max rpm noise.
 
 

I just want to say that I'm sorry for your 45c summers.  I'll gladly pass on those.  My condolences.

Oh man, you have no idea. Last summer there were entire weeks where I couldn't do much gaming because it was so hot, and my system was my 9900k (that I still have) with SLI 980ti's. It was like being in the engine area of a steam train. The temps were brutal.
 
I now have a much cooler 3080, undervolted it with an aggressive fan curve. It hasn't broken 65c since. I also swapped from a 1050w gold PSU to a 850w Titanium that runs cooler too.

 
I understand, used to live in Texas, USA so yeah I know what hot is.
 
I feel for you really. 
 
And I have been in the Desert and more than one with 54C temps.

z1nonly

jankerson

z1nonly

macktkau
For someone wanting information about how the card performs the "review" is a waste of time.

They said it performs similar to all the other 3080's with the exception of the parameters that they went into detail on.

 
They can SAY anything.....
 
They can tell you the sky is red with purple Pooka dots too....

-Same can be said for a benchmark graph they put up on a youtube video.

 
Yes I know.
 
But at least with graphs we have something to compare to, to see if they are real.
 
Never trust anything blindly. 
 
And even with the so called review I am not personally seeing the same temps he said in the video.
 
Even with the case closed up, not on an open air test bench like he is using.
 
So I have nothing to compare to...

vegnar

jankerson

z1nonly

jankerson

z1nonly

macktkau
For someone wanting information about how the card performs the "review" is a waste of time.

They said it performs similar to all the other 3080's with the exception of the parameters that they went into detail on.

 
They can SAY anything.....
 
They can tell you the sky is red with purple Pooka dots too....

-Same can be said for a benchmark graph they put up on a youtube video.

 
Yes I know.
 
But at least with graphs we have something to compare to, to see if they are real.
 
Never trust anything blindly. 

Steve has already said several times that they are no longer showing gaming performance as they all perform within 1-3 % of each other.

 
Well, show it then so we can all see it.
 
Not saying they have to run a ton of games at every resolution and OC the cards etc.
 
It's lazy not to include any gaming tests for a gaming card in a review of the said card.
 
 

vegnar
In a nutshell :
 
Out of box thermals:
TUF    @37.7DBA   336W   59.8c
FTW3 @43.0DBA   377W   66.0c  (OC Bios)
FTW3 @39.7DBA   377W   70.0c  (Normal Bios)
 
Normalized DBA @40DBA (all fans at same dba):
 
TUF      56.4c
FTW3   64.7c
EAGLE  60.8c
 
Gaming performance:
 
Who cares they are all within 1-3% of each other, to me Thermals and Noise > 1-3% more fps.
 
 
The TUF is also $110 cheaper.  How in the world did Asus achieved water cooled type thermals on air at 56.4c?

My last, and current build, had a great deal of Asus components, mainly because I wanted all of these to Sync with Asus ARUA lighting (the monitor does not, but I already had 34" ultrawide Gysnc, not giving it up anytime soon).  Monitor, motherboard, Strix Flare Keyboard, Mouse, and a Strix 2080TI 011G (their top card in the 2080 ti series).  Everything was great, until it wasn't.  Six months in and half the RBG lighting on the keyboard was way dimmer, with some switches totally burnt out.  I immediately started an RMA, can't find a phone number (easily) on Asus's site so I did the online RMA request.....six times.....no response.  Then a few months later my z390 Maxiumus Hero Wifi started having issues, I did not even bother trying to contact them, because I bought the 4 year extended warranty from Amazon which I used, they didn't even try to repair or replace it and just refunded me my $350 so I bought an MSI board this time around.  
 
I have owned MANY Asus products over the years, and NEVER had any issues, this is the first time I ever even tried to use their warranty, and in a nutshell....their support SUCKS, and I am far from the only one with that opinion.  
 
My GPU is great, rock solid and overclocks well, but, if you buy any Asus GPU and you get unlucky good luck getting Asus to resolve the issue for you.  I'm not totally bashing Asus (really), but I have not read or heard about anyone complaining about EVGA as far as standing behind their products, just something to think about.

parasitegod
If anyone doesn't like GN content, then go watch someone else. They have already said it several times, that it's pointless to show the same graphs (with few percentage of deviation here and there) in every video. It's a sensible thing to do. "Oh look, this game has 2fps less with this card. Looks like I won't be buying this garbage" - said no one ever.

I am a big fan of GN, for many reasons. When it comes to card testing I enjoy their usual attention to detail in methodology and how they really try to explain all the caveats of controlled tests and how to read charts properly. As scientist myself, this all speaks to me greatly. That does not mean they dont put out reviews occasionally that I feel gloss over questions that are worth asking. If this card has such different cooling and frequency results, why not run some tests to see if they really do produce the same gaming results. That would have been my approach in this specific situation, even after determining that in majority of situations all aib parter cards will perform about the same. In the end I will continue to watch GN because I still have loads of respect for the way they do their work.