The 4k difference does appear to be close to 40% at 4k with their tests. (Maybe more recently)
40% would have been a good generational improvement...at the same price point.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
AMD Navi 23 ‘NVIDIA Killer’ GPU Rumored to Support Hardware Ray Tracing, Coming Next Year
- Thread starter Gregster
- Start date
- Status
The 4k difference does appear to be close to 40% at 4k with their tests. (Maybe more recently)
40% would have been a good generational improvement...at the same price point.
Soldato
- Joined
- 28 May 2007
- Posts
- 10,202
I agree, at the same price it would have been decent with what we are used to these days.
You want to measure how much faster something is you need to calculate how much it would take to get there, not what the percentage ''''in points''''' difference. In this case that would be 28%, its not that, if you start with the slower card and calculate how much performance it needs to gain to match the faster card, so 72 + 28% = 92. That's how you would check your equation and 92 is wrong, it should be 100.
As @TheRealDeal demonstrated you divide the faster card by the slower to get the difference, which is 39%.... round it up its 40%. to check it, 72 + 39% = 100.08.
You want to measure how much faster something is you need to calculate how much it would take to get there, not what the percentage ''''in points''''' difference. In this case that would be 28%, its not that, if you start with the slower card and calculate how much performance it needs to gain to match the faster card, so 72 + 28% = 92. That's how you would check your equation and 92 is wrong, it should be 100.
As @TheRealDeal demonstrated you divide the faster card by the slower to get the difference, which is 39%.... round it up its 40%. to check it, 72 + 39% = 100.08.
I see that, yes. The number was around 30% at launch. (I posted a link that showed ~30% @4k last page) I had 30% as the running number, calculated the two percentages incorrectly, and didn't give it a second thought when the result was close to the 30% I have seen before.
My mistake.
Yes definitely something going on. Notice a nervous scratch of the head immediately after the question is asked
They know something we don't. I think I watched that when it came out and didn't read much into it but with the console information it now looks a lot more interesting.
Associate
- Joined
- 22 Jun 2018
- Posts
- 1,701
- Location
- Doon the watah ... Scotland
First time I watched it I noticed it and always thought "I want to be part of that discussion after the video ! " It's stuck with me since.
Also AMD in the server side of things are pushing an infinity fabric idea which I think we will see spread, where devices can independently talk to each other at very high speed and scale well.
Also AMD in the server side of things are pushing an infinity fabric idea which I think we will see spread, where devices can independently talk to each other at very high speed and scale well.
Don't look at 1080P performance, with a card as fast as the 2080TI it tends to be CPU bottlenecked at 1080P, look at 4K results.
1080P: 1080TI vs 2080TI 81% - 100% (+23.5%)
4K: 1080TI vs 2080TI 72% - 100% (+39%)
Soldato
- Joined
- 8 Jun 2018
- Posts
- 2,827
If up to 128 CU variants are true only for CDNA. I wonder what the potential performance would be for gaming 
.
It's been rumored a while back they have cards with CU counts higher then 80.
. It's been rumored a while back they have cards with CU counts higher then 80.
If up to 128 CU variants are true only for CDNA. I wonder what the potential performance would be for gaming
It's been rumored a while back they have cards with CU counts higher then 80.
128 CU variants?
And watch them flip flop and toss out every possibility close to launch.
15% faster at a reasonable price (650-700 max) would be a good area for amd to hit.
Associate
- Joined
- 31 Aug 2017
- Posts
- 2,209
Cant see it being the biggest one at 15% - however it maybe one of the lower tiers.
We just dont know right now, but i suspect the new top nvidia card to be 40 to 50% faster than the 2080ti and the top AMD card to be similar.
Who will be fastest, no sodding idea.
We just dont know right now, but i suspect the new top nvidia card to be 40 to 50% faster than the 2080ti and the top AMD card to be similar.
Who will be fastest, no sodding idea.
Source please.
What will the prices be, no sodding idea.
Latex glove territory most likely, hopefully not though.
Yeah where did this come from?
Associate
- Joined
- 24 Jun 2016
- Posts
- 852
- Location
- Hartlepool
LolJensen's derriere
Big Navi being 15% faster than a 2080 Ti is pathetic if true. 40 RDNA 1 CUs is only 30% ish slower than a 2080 Ti, double the number of RDNA 2 CUs is barely 50% faster?
I know AMD's GPUs often underperform, but I sincerely doubt this is true.
I know AMD's GPUs often underperform, but I sincerely doubt this is true.
And watch them flip flop and toss out every possibility close to launch.
15% faster at a reasonable price (650-700 max) would be a good area for amd to hit.
Agreed. There is no such thing as a bad GPU (ignoring AMD's driver problems), just a bad price.
- Status