Read more.Quote:
Benchmark result achieved using a heavily overclocked LN2 cooled Galax branded card.
Printable View
Read more.Quote:
Benchmark result achieved using a heavily overclocked LN2 cooled Galax branded card.
Sounds like Futuremark need to work on their optimisations of DXR to me...
Still I guess it doesn't really matter with a benchmark - as long as its consistent it doesn't matter. I'm sure we'll still get some idiots who jump in again try and make out this shows how "bad" the RTX line is etc etc though.
It seems RTX has found itself in the same category as The Big Bang Theory and Nickleback, where it's cool to hate them for no good reason.
I couldn't care less if my card scored 4FPS or 70000FPS in a benchmark, as long as the benchmark is consistent and lets me compare the score against other cards in the future.
Am I right in thinking you've got the 2080? I'm sure I saw you talking about it in the BF V thread.
Yeah you are right - its very much cool to hate on Intel and Nvidia at the moment ;)
I do - following the latest patch it's running at 70-75 FPS, 1440p, DXR on medium, everything else on ultra. Finally at the level where I am happy to leave it on for MP and SP :)
I'm a firm believer that there is no such thing as a bad graphics card, just bad pricing.
After the mining craze the market is rather price sensitive. Perhaps if they had released it next year Nvidia would have had less resistance, but instead they asked customer to hand over the cost of a second hand car for features they can't see in regular use.
Edit: Had a quick look and I see that Overclockers will sell you a Palit 2080ti for "only" £950. I assumed they were still £1500 which is what my wife paid for her second hand MX5 Sport, which is way more fun than any graphics card and all its features are fully supported by all roads and race circuits :D
"2.640 MHz (core clock) and 2.088 MHz (memory clock)"
2.6Mhz impressive... ;)
What it really says to me is that ray tracing is not ready for prime time yet.
Hardware that can really handle it properly is not there. I say year 2020 could be the year for that.
Who cares about the FPS figure in a benchmark? I care about that figure in actual games that is where the card and the new features will be used 98% of the time. At least for me, I run the benchs maybe once or twice a year, if I want to compare the evolution of drivers over time.
I need to set mine into my computer, but for that I need to remove all my loop (it's time already), re-do it and fill it in. Too many social compromises in this dates to do so.
I'm having a comfortable 65-85 FPS count right now in BF V with just 2x980 at 1440p.
I have a query regarding this.. Of the recent 3D Mark releases, which have been possible on a current-gen graphics card at over 60 FPS?
My RTX 2080Ti, with a bit of an overclock, scores 95FPS and 86FPS in the two graphics tests in 3D Mark Time Spy, which is a previous-gen benchmark.
Looking at a GTX 1080Ti, it scores 63FPS and 58FPS respectively, a little higher than the seemingly "terrible" 51FPS in this Port Royal test.
This strikes me as a complete non-story.
This is a single data point which is useless without a wide range of comparisons. That's the whole point of benchmarking - comparison.
You might as well say I have 69. Without knowing what everyone else has, 69 is solely amusing.
As for saying who cares about a benchmark - well that's kind of the scientific method all over. Reproducible testing which can give you valid data with utility.
Sort of. Some benchmarks are gameplay, I like those as it makes GPU companies target optimising for actual games. The problem with synthetic benchmarks is when they make GPU companies spend time optimising for things we don't ever do.
Synthetic benchmarks should be made to report their results in "inches" to make it clear what the nature of the comparison really is.
Benchmarks, schmenchmarks. I play my few games at maxed settings and my graphics card delivers smooth frame rates. That's good enough and all I need to know.
Not if you game every day. Way more fun to own a 2080ti. :) Or two. ;) People buying these, aren't buying 2nd hand cars. My family hasn't bought a 2nd hand car in 30yrs (I had a truck as a kid that was used, my 1st vehicle, last one bought). I bought my last car almost cash (paid off in 3 months). People need to get better jobs, or SAVE longer for better things (thus avoiding interest also). If most people took all interest on debts out of their lives, they'd realize they could probably have had a free house at some point...LOL. We generally put off a purchase for another year etc (who family, many houses) in order to make sure we can pay it off in a month or two if not pure cash. Yeah, we wait longer to get things, but we maximize our income. We were always taught instant gratification (unless you're rich, splurging on top stuff anyway), just buys crap products or lower perf/enjoyment overall :). My dell 24 cost me $640, but I've been enjoying it for over 11yrs. I'm sure all NV's features will be used in the next gen of games say, next xmas after they've had time to support them (with no amd answer, it's the only way forward for devs right now making a new raytracing game etc) :)
Oh yeh, sorry I didn't make it clear I was talking solely about synthetic benchmarks. In-game ones which replicate actual gameplay are extremely useful and give a valuable spot reading that you can interpret on its own.
As you say, synthetic ones reporting in FPS are really just misleading. "Points" are far more descriptive as to the nature of the investigation. Personally, I would have them report in neither "inches" nor "points" and mandate all synthetic benchmarks to report in "nipples/msq".
I know what you mean, Spud. I hate them so much I went and bought an i7 9700K and an Asus RTX 2080 Turbo. Don't know what I was thinking!! : )
The rig is built and next up is OS and software. I'm very excited. Will be interesting to see what I get at standard clocks on this benchmark. Then it'll be tweak time.