• 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

Status
Not open for further replies.
No, because you don't know if the benchmarks showing performance parity are not the outliers lol

Well, you bring up a good point but it doesn't negate mine. It's STILL good science to remove outliers - but you have a point that you have to be sure what the outliers actually are. I don't know what the sample size is that Nabla is using, but if I assume there's 10 games he's included and 2 he's removed - that's fairly reasonable to assume they're the outliers. If it was like 5 games total and he removed 2, then it wouldn't be fair to remove them.
 
That's not how stats are summarized.. we can't suggest population level parameters with such a small sample, and that too not well behaved

Well, many gamers replay games such as the NFS series that are dozen on their own, F1 starting from version 2010 onwards, the Colin McRae Rally series, StarCraft, Diablo, Warcraft, Fortnite, CS all of them, Crysis 1, 2, 3, Far Cry 1, 2, 3, 4, 5, Quake - all of them, etc, all the smaller and less popular games - there are literally hundreds of games and yes, the sample size is very small with only 2020 or 2019 games.
 
No, because you don't know if the benchmarks showing performance parity are not the outliers lol

Fine, I will provide another perspective..This time with no outlier adjustment..Would consider medians rather than averages. Medians are generally considered robust to outliers (but not at such low sample size of 13)
  • There's 50% chance that Rx takes more than 1ms time than Rtx to render a frame
  • Rtx median frame rendering time is 19ms
Calculate your percentages however you want to.

Edit: Avg frame rendering times can be found in previous comment
 
Last edited:
Well, many gamers replay games such as the NFS series that are dozen on their own, F1 starting from version 2010 onwards, the Colin McRae Rally series, StarCraft, Diablo, Warcraft, Fortnite, CS all of them, Crysis 1, 2, 3, Far Cry 1, 2, 3, 4, 5, Quake - all of them, etc, all the smaller and less popular games - there are literally hundreds of games and yes, the sample size is very small with only 2020 or 2019 games.

There were just 13 games released on pc during 2019-20?
 
@4K8KW10 I can cherry pic a bunch of games and make the 5700XT look at least as good as an RTX 2080, the 5700XT seems to do much better than Turing with newer titles so its pretty easy.

You're working really hard picking your way through the internet to find things to use to chip off every percent you can get and then argue about those few percent as if its somehow a huge issue for AMD's competitiveness, it isn't, it doesn't make the blindest bit of difference, its just deranged.
 
@4K8KW10 I can cherry pic a bunch of games and make the 5700XT look at least as good as an RTX 2080, the 5700XT seems to do much better than Turing with newer titles so its pretty easy.

You're working really hard picking your way through the internet to find things to use to chip off every percent you can get and then argue about those few percent as if its somehow a huge issue for AMD's competitiveness, it isn't, it doesn't make the blindest bit of difference, its just deranged.

I agree, the 5700/XT was disruptive if it had come in at the price it dropped down to it would have seemed a better deal hopefully now as there is no Gcn hareware in second wave of Navi the drivers will be better from the start.

Amd still need to make money as they got to try and complete with Intel and Nvidia.
 
And that makes artificially limiting stock to drive up prices and therefore actively lying about MSRP somehow OK? Turing gave us more than Pascal, Pascal gave us more than Maxwell. Are Turing, Pascal and Maxwell prices acceptable then? You clearly don't get what's going on, or you don't care, and that's fine. That also means theres zero reason to continue this particular subject of conversation.

The 2080Ti was the worst price performance...ever maybe?
performance-per-dollar_3840-2160.png


Turing was unacceptable.
 
Hardware unboxed is like jayz2cents. Both fanboys of different brands. They take every chance to **** on Nvidia while praising AMD. All their benchmarks / videos since turing release have been nothing but biased garbage, using games where AMD fares better and ignoring the rest, especially in their ‘over time’ videos.

So if you bash jayz but praise HUB you’re a hypocrite. Both are at the same level. Just that one is more upfront about it ( jayz being a ma$$ive nv fanboy ), while HUB trying to seem more professional and ‘not biased’.

Funny how HUB had no video pointing out the massive downside of the rx5700xt which is no raytracing. Both consoles + rdna2 will have it while perf in consoles at least being lower than turing. So the excuse that ‘but it aint worth it yet! Perf hit too high’ doesn’t fly anymore. Next 7 years or so, maybe 3 if we get a mid-gen refresh will be pioneered by RT in console games mostly.

That being said, its funny how people choose to bash consoles in general, while people on PC who pretend to enjoy pristine IQ above consoles with ultra settings... will be using rx5700xts while having their games look worse than nextgen games cause no RT. Meanwhile, turing users, yes even those crappy rtx2060s will have RT. Sure they might have to drop the resolution a bit, or maybe use DLSS but they’ll have the option unlike the others. So the ‘fine wine’ is more like ‘sour milk’ in this case.

Bring on the ‘pff dont care bout RT’ excuses while pretending to care about ‘pc masturr raec’ ultra details better than console lulz.
 
Last edited:
That’s the pretending not to be biased. An occasional slap on the wrist while full on beating the other doesn’t count in my book sorry.

either you’re fully impartial and **** on turing’s price, initial DLSS perf etcetc but do point out the advantages, while also bash AMD’s lack of raytracing, ML etc or you don’t do either. They always make excuses for AMD : ‘RT isnt worth it yet, DLSS was bad.. but its ok now but only used sparingly, so again no problem if AMD doesnt have it etcetc’

can’t wait for the inevitable ‘RDNA2 Hub’ reviews in which suddenly raytracing is worth it even though perf will be same or lower than turing/ampere.
 
That’s the pretending not to be biased. An occasional slap on the wrist while full on beating the other doesn’t count in my book sorry.

either you’re fully impartial and **** on turing’s price, initial DLSS perf etcetc but do point out the advantages, while also bash AMD’s lack of raytracing, ML etc or you don’t do either. They always make excuses for AMD : ‘RT isnt worth it yet, DLSS was bad.. but its ok now but only used sparingly, so again no problem if AMD doesnt have it etcetc’

can’t wait for the inevitable ‘RDNA2 Hub’ reviews in which suddenly raytracing is worth it even though perf will be same or lower than turing/ampere.

Wow, there's utterly no bias in you is there? :D
 
That’s the pretending not to be biased. An occasional slap on the wrist while full on beating the other doesn’t count in my book sorry.

either you’re fully impartial and **** on turing’s price, initial DLSS perf etcetc but do point out the advantages, while also bash AMD’s lack of raytracing, ML etc or you don’t do either. They always make excuses for AMD : ‘RT isnt worth it yet, DLSS was bad.. but its ok now but only used sparingly, so again no problem if AMD doesnt have it etcetc’

can’t wait for the inevitable ‘RDNA2 Hub’ reviews in which suddenly raytracing is worth it even though perf will be same or lower than turing/ampere.
Not that I've watched every one of their videos but I find it hard to believe they've never once mentioned the lack of RT on AMD hardware.
 
Status
Not open for further replies.
Back
Top Bottom