Nvidia's 3000 Series RTX GPU [3090s with different memory capacity]

4K ULtra results.
Starting at the high-level overview where we average performance across all nine games in our test suite, this is a good view of what's to come. The RTX 3080 Founders Edition blew away the competition: It's not even close. It ended up 27% faster than the Titan RTX, 32% faster than the RTX 2080 Ti, 57% faster than the RTX 2080 Super it replaces at the $700 price point, and 69% faster than the RTX 2080 FE. If you're hanging on to a 10-series Nvidia card, the 3080 also beat the 1080 Ti by 87% and ended up nearly 150% faster than the GTX 1080 FE.

AMD's best showing at 4K ultra is still the Radeon VII, which the 3080 beats by 79% — though, of course, we're more interested in seeing how Big Navi compares. The RTX 3080 also beat the RX 5700 XT by 95% overall. AMD has talked about 50% better performance-per-watt for Navi 2x, which might be enough to catch the RTX 3080 (depending on how AMD gets that 50% boost — a 300W card that's 50% more efficient should do it).
...
Stepping down to 4K medium settings, the RTX 3080 maintains its pole position, and the margin of victory was basically unchanged. It's still 26% faster than the Titan RTX (at less than one third the price) and 69% faster than the RTX 2080 FE. That's the great thing about 4K: Even at medium quality, it's still typically GPU bottlenecked (unless you're using a very slow CPU).
https://www.tomshardware.com/news/nvidia-geforce-rtx-3080-review
 
Roughly 30% faster as a 2080Ti while sucking 25% more power @4K.
https://www.techspot.com/review/2099-geforce-rtx-3080/

Yep, I little bit more efficient than the 2080TI, but only slightly.

upload_2020-9-16_15-30-17.png
*stolen from eurogamer

Wow the performance/watt different is quite big between the 2080 and the 2080TI in some games, I'd have expected them to be mostly equal.

edit: yeah, I think I'll be waiting for the 3070

edit2: :oops:

upload_2020-9-16_15-40-32.png
 
Last edited by a moderator:
Currently available Ampere RTX Reviews. Check Videocardz link for updates.

Media Type NVIDIA GeForce RX 3080 Model
ComputerBase
NVIDIA GeForce RTX 3080 Founders Edition


Thanks to Videocardz!
https://videocardz.com/newz/nvidia-geforce-rtx-3080-founders-edition-review-roundup

According to ComputerBase's review, 3080 is 25% faster than 2080Ti at same power consumption (270W), but gains only +4% going from 270W to 320W(+19% power), 4K resolution. Seems like pushed tad higher than efficiency curve of the process.
pp3080pyjg8.png

No negligible difference between PCIe 3.0 vs 4.0 overall, except 2 titles 5% in CoD:MW, 6% in Control.
pcie43080y4ksu.png

Lower DRAM and CPU temperatures compared to Asus Strix 2080Ti@320W.
dram3080ngjdt.png
 
Roughly 30% faster as a 2080Ti while sucking 25% more power @4K.
https://www.techspot.com/review/2099-geforce-rtx-3080/
This made me thinking, what would happen if 3080 is on the same node as 2080Ti
It seems that Nvidia pushed their power usage to create a big enough gap to the previous gen.
If I'm purchasing a GPU right now, my target would be 2060 super or 2070 because I believe you can have both cards with only a single 8pin power.
Sadly, even 3070 will need to have more than what a single 8pin can provide. Hopefully we can have 3060 with only a single 8pin power (or RDNA2 equivalent).
 
Its high performing card, thats for sure, but comparison with 2080 is completely misplaced. This thing should be compared to Titan RTX OC, because thats how much it pulls from the wall.
 
Comparisons are made at the same price point, not same power consumption.
That is a good point, although 980 > 1080 > 2080 followed "expected" path, then came Turing which hiked the prices so high that we needed Ampere to be released at "normal" price to show how overpriced this thing was.
 
The GeForce RTX 3080 provides a surprise in terms of efficiency. Yes, it works more efficiently than the GeForce RTX 2080 Super and the GeForce RTX 2080 Ti, measured in FPS per watt, but with a 10 percent advantage over the Ti, the progress is surprisingly small. However, the comparison of the GeForce RTX 3080 with the GeForce RTX 2080 Ti at 270 watts shows that with a little less clock, a much better result would have been possible. Nvidia has leveraged an astonishingly large portion of manufacturing and architecture efficiency improvements to gain more performance. The fact that the GPU on the graphics card clocks close to the maximum also indicates this.

Why Nvidia sacrificed another 50 watts of TDP for the four percent performance gain determined in the test course, if the jump in performance would have been remarkable even without, remains open for the time being.
https://www.computerbase.de/2020-09/geforce-rtx-3080-test/7/#abschnitt_preisperformanceverhaeltnis
 
1080, 2080 and 3080 launched at the same price.
Looking at the wiki, nope, you're wrong.
2080 was priced like 1080Ti with 2080Ti $300 more expensive vs 2080. 1080Ti is only $100 more expensive vs 1080.
980 price was $50 cheaper than 1080.
So basically 1080Ti = 2080 = 3080
*I'm comparing launch prices.

The problem with 2080 is that it isn't always a win against 1080Ti. So during the 2080 launch, basically those that have 1080Ti, their upgrade path would be paying $300 more for the 2080Ti version which is ridiculous.
 
Looking at the wiki, nope, you're wrong.
Yeah, I'm not.

2080 was priced like 1080Ti
Because 1080Ti launched a year later than 1080.

with 2080Ti $300 more expensive vs 2080
So? A top end card can have whatever price they feel like slapping onto it.

980 price was $50 cheaper than 1080.
It was $150 cheaper considering that 1080 was available as FE only at first. Pascal was the actual point where NV has increased the prices, not Turing.

So basically 1080Ti = 2080 = 3080
So basically this is right only at the moment of 1080Ti launch which wasn't the moment of 1080 launch.
 
Seems to me that this gpu really shows that flops isn't everything. I've no proof of course but I believe that the more you need fp32, the more this gpu will shine. In others situations, it will be bottleneck elsewhere (bandwitdh ? Rops? Even drivers ? A mix of everything I guess). In a weird way it reminds me some old ati gpu :D
 
Yeah announcing 30 Tflops of shader power made everyone go crazy. But of course with apparently double the FP32 performance of 2080ti, it manages around 30-40% improvement in actual games. What a surprise?
 
Looks like the 3080 is running up against VRAM and/or bandwidth limitation with RD2 at 4K, when compared to 2080 TI (3fps difference).
kHmPSvX.jpg
 
Back
Top