|
SuperTeeJay posted:Yup, for the UK at least. A supposed pro-tip is to be logged into Paypal and checkout through it rather than struggling to create an Nvidia Shop account when the server is making GBS threads out. Thanks for the tip!
|
# ? Sep 16, 2020 12:53 |
|
|
# ? Jun 6, 2024 08:38 |
|
Why are people talking about store pages going up today? I thought 9am on the 17th was the street date. Stop making me nervous!
|
# ? Sep 16, 2020 13:05 |
|
exquisite tea posted:Why are people talking about store pages going up today? I thought 9am on the 17th was the street date. Stop making me nervous!
|
# ? Sep 16, 2020 13:06 |
|
Incessant Excess posted:Enjoying those blissful final hours of the thread before everyone starts posting shipping updates. oh no
|
# ? Sep 16, 2020 13:07 |
|
exquisite tea posted:Why are people talking about store pages going up today? I thought 9am on the 17th was the street date. Stop making me nervous! because there are like 5 cards for people to buy, good luck.
|
# ? Sep 16, 2020 13:14 |
|
ijyt posted:oh no It's fine all of 3 people are gonna get one threads gonna be flooded with whining instead.
|
# ? Sep 16, 2020 13:15 |
|
pyrotek posted:I'm assuming you mostly use your computer for gaming? The Decima engine (Horizon: Zero Dawn and Death Stranding) uses AVX and it is part of DX12, so I would expect more usage in the future. Prime95 and Cinebench in no way represent a realistic gaming workload. I would recommend turning AVX back to normal, the proper voltage should suffice. It was mostly a stability thing, I know they're unrealistic but wanted to be sure the slight undervolt for temperatures was stable. I'm 5ghz on all cores now and can put AVX back to normal.
|
# ? Sep 16, 2020 13:22 |
|
Carecat posted:Prices are being raised on Overclockers and the FTW was replaced with an XC3, I think the FTW was a listing error. They pulled the same bullshit on the 1080 launch and are notorious for exploiting low stock. OcUK are a trash company imo.
|
# ? Sep 16, 2020 13:46 |
|
There is a lot in common between Fermi, Turing, and Ampere. After all, Fermi made real time ray tracing possible with the GTX 400 series. In addition to real time ray tracing, Fermi provided the NEXT big thing in graphics - hardware tessellation. Left: tessellation OFF Right: tessellation ON How much of your life do you want to live without real time ray tracing and tessellation? Just buy it! See the performance delta between Turing and Ampere: We haven't seen that level of gainz since Fermi arrived on the scene back in 2010:
|
# ? Sep 16, 2020 13:50 |
|
furiously F5-ing Gamers Nexus's page for their review
|
# ? Sep 16, 2020 14:01 |
|
https://www.youtube.com/watch?v=oTeXh9x0sUc
|
# ? Sep 16, 2020 14:02 |
|
https://www.techpowerup.com/review/nvidia-geforce-rtx-3080-founders-edition/
|
# ? Sep 16, 2020 14:02 |
|
pretty muted performance over the 2080 Ti at 1440p (~25% uplift on average)
|
# ? Sep 16, 2020 14:03 |
|
Huzzah! All hail Tech Jesus!
|
# ? Sep 16, 2020 14:03 |
|
drat
|
# ? Sep 16, 2020 14:05 |
|
lol
|
# ? Sep 16, 2020 14:08 |
|
|
# ? Sep 16, 2020 14:10 |
|
I was close to talking myself out of the 3080 but the uplift from a 1080Ti at 1440p varies between 'good' and 'loving massive'.
|
# ? Sep 16, 2020 14:15 |
|
Doesn't hit 4K60 on Control with both RTX and DLSS on
|
# ? Sep 16, 2020 14:15 |
|
ill save most of my laughing for the 3060 Super or whatever reviews, where it trades blows with the 2080ti depending on game or resolution lol however this is still getting a laugh E: those control comparisons are nice though actually, it smokes the 2080ti there even more it seems
|
# ? Sep 16, 2020 14:17 |
|
That's a heckuva incremental increase over last year's $1,700 card for only another $700. Looks like you can lean on it and get a reliable 13 fps increase at 4K nice nice.
|
# ? Sep 16, 2020 14:17 |
|
Yeah imagine being bitchmade over a $700 card only being 20-30% faster sometimes than the $1,700 card people were buying last month lol
|
# ? Sep 16, 2020 14:20 |
|
https://youtu.be/k7FlXu9dAMU The flood gates have opened
|
# ? Sep 16, 2020 14:23 |
|
i guess the takeaway is that it's a fast card but it won't guarantee you 4K60+ performance w/ RTX and even with DLSS on current games (let alone upcoming games)
|
# ? Sep 16, 2020 14:23 |
|
Interesting that GN is showing a significantly lower performance increase than some of the other reviewers. I wonder if the stupidity of sticking with 3200mhz ram well past its sell date is biting them in the rear end.
|
# ? Sep 16, 2020 14:24 |
|
Looking at my 3200mhz ram...
|
# ? Sep 16, 2020 14:26 |
|
3200mhz ram BAD?
|
# ? Sep 16, 2020 14:28 |
|
there seems to be a lot of variance in benchmarks between the reviewers... some have Control 4K RTX+DLSS >60, others <60 .... it might be highlighting other bottlenecks
|
# ? Sep 16, 2020 14:31 |
|
I'm still on 2666 Mhz anyway those big huge charts from TechPowerUp tell me that even a 2060 would be enough for my 1080p60Hz rear end
|
# ? Sep 16, 2020 14:31 |
|
Rolo posted:3200mhz ram BAD? It's generally fine, but it can certainly be a limiting factor, especially in some of the games that tend to be really heavy on GPUs. A GPU testbench should be doing everything possible to ensure that the GPU is the bottleneck, and it's silly that for a just re-done testbench in late 2020 GN is still using 3200mhz ram, especially on an Intel platform. GN is generally very smart about how they do their testing, but Steve is definitely overly resistant to looking at data from outside sources and then doing testing himself to see if his assumptions are wrong.
|
# ? Sep 16, 2020 14:32 |
|
Also max settings control is something to not really play at for certain settings. Volumetric lighting for example runs way faster at high with no discernible visual difference.
|
# ? Sep 16, 2020 14:32 |
|
Hopeful I can get a 3080 tomorrow to replace my 1080. Seems a little more than double the performance @ 1440p.
|
# ? Sep 16, 2020 14:32 |
|
shrike82 posted:there seems to be a lot of variance in benchmarks between the reviewers... some have Control 4K RTX+DLSS >60, others <60 .... Control RTX is more granular than on/off, are some testing RT Medium (reflections only) and other RT High (reflections + global illumination + shadows + debris)?
|
# ? Sep 16, 2020 14:33 |
|
Twibbit posted:Also max settings control is something to not really play at for certain settings. Volumetric lighting for example runs way faster at high with no discernible visual difference. yeah but people were hoping that the 3080 (and the 3090) would allow them to get 4K60 at max settings for a while doesn't look like it's happening with the 3080 and assuming a 20% delta by the 3090, not for it on upcoming games either
|
# ? Sep 16, 2020 14:34 |
|
Twibbit posted:Also max settings control is something to not really play at for certain settings. Volumetric lighting for example runs way faster at high with no discernible visual difference. Ultra Everything or Trash Bin
|
# ? Sep 16, 2020 14:39 |
|
I imagine ambient temps and HVAC quality accounts for a large chunk of the variation in performance between reviewers, assuming they each got top binned chips. Should really be running reviews with faster ram as mentioned above tho.
|
# ? Sep 16, 2020 14:39 |
|
So we're not getting 3090 reviews until next week, after the 3080 is launched? Ugh, was hoping to see the performance difference to decide if it was worth it to me. There's a lot of games that the 3080 still benchmarks at well under 144fps at 1440p. Helluva upgrade from my 1080ti so maybe I'll try to snag a 3080 tomorrow and not open it until 3090 reviews come out.
|
# ? Sep 16, 2020 14:39 |
|
Rolo posted:3200mhz ram BAD? 3000mhz here. My 1070 is badder than RAM. Looks like the limiting factor for FS2020 is gonna be single core frequency and their crappy non-multi threading. But I can extrapolate that I'll be going from 20-25fps in "high" to a near-locked 60fps@1440p, possibly at ultra settings so I'm happy.
|
# ? Sep 16, 2020 14:40 |
|
Looks like the FE doesn't overclock well at all
|
# ? Sep 16, 2020 14:43 |
|
|
# ? Jun 6, 2024 08:38 |
|
gradenko_2000 posted:I'm still on 2666 Mhz It absolutely is with current games. shrike82 posted:there seems to be a lot of variance in benchmarks between the reviewers... some have Control 4K RTX+DLSS >60, others <60 .... Could also depend on where they are doing the testing. If one place is testing the most demanding spot in the game and the other isn't, that would make a big difference.
|
# ? Sep 16, 2020 14:45 |