It does depend on the scene. Hardware unboxed found about a 10% increase over the 2080ti at 1440p (high quality), and about 20% over dannyboy75's 1080ti.
https://youtu.be/csSmiaR3RVE?t=296
It does depend on the scene. Hardware unboxed found about a 10% increase over the 2080ti at 1440p (high quality), and about 20% over dannyboy75's 1080ti.
https://youtu.be/csSmiaR3RVE?t=296
Seems like Hexus will have the Asus TUF and MSI Gaming X reviews up soon enough, looking at Scan's listing - the links are already there
You are reading that correctly! Or you could scale back to 2080Ti performance and get improved power efficiency - but not both.
I *knew* Nvidia were pulling a bait-and-switch. The "exclusive preview" was pushing specific games at the highest resolutions with particular settings and filtering, and their perf/watt graph was an obvious exercise in misdirection. They've improved in perf/dollar, but only because Turing was such a rip-off; they're basically back to where they were with Pascal, which is to say worse than Maxwell but better than a kick in the teeth.
The biggest revelation for me was how much it performs like a doubled-up RX 5700. If AMD pull off their claimed 50% perf/watt increase and Big Navi has 80 CUs, then they'll be able to provide greater-than 3080 performance at around 330W or 3080-equivalent performance at 280W (likely both, given the usual XT / Pro split on chip bins). I'm now excited for AMD's response!
LTT reckons that there are a couple of issues here. They used NV's own power monitoring system and found that the 3080 easily pulled 350w despite being rated at 320 watts and that rumours are the 3090 will easily need 400w. For a die shrink that show they have pushed it hard to get a performance increase at the expense of power draw.
1080p gaming is old hat - not saying that you shouldn't game at 1080p but the 30xx cards aren't aimed at that at all same as the new consoles are aimed at 4k. Again a few sites/reviews saying 1080p gaming a 30xx isn't worth buying. The same sites are also saying that you might as well buy a complete new system to get the best out of a 30xx card because anything older than a year struggles and changing the card out makes little sense unless you upgrade the display/system
Heck NV is pushing DLSS/Raytracing etc. so hard and none of them really make a difference down at 1080p to be brutally honest. This generation is pushing 4k hard with the pc side and consoles
Old puter - still good enuff till I save some pennies!
Pleiades (17-09-2020)
I've been saying since the 8nm news leaked that we're about to get another Fermi: a gargantuan die overclocked to its limits and drawing obscene amounts of power to eke out a lead against a smaller AMD competitor. The big difference is this one isn't actually late, hasn't been hacked to bits to scrape out a tolerable yield, and Nvidia have clearly put more effort into planning for the colossal TDP.
All AMD have to do is execute on their claims for RDNA 2. Fingers crossed they do, because we'll finally get competition at everything but the ultra-high-end - and personally I couldn't care less about £1400 GPUs.
For some titles, yes the 3080 and 3090 (presumably), will make 4K gaming a possibilty - where 60 fps is acceptable. Anything FPS based though I really cant imagine playing below 100fps these days. It's going to take the next gen and a reduction in price for 4K / 144Hz panels before 4K gaming really hits the sweet spot.
Until then I'll keep at 1440p and enjoy 120-144 fps.
Not true in the slightest.
If you are gaming at 1440p, or even more, 1440p ultrawide, then the 3080 makes sense and is a huge upgrade over the old 2080...35-40% or more, with a bigger jump if you are on an ultrawide and play at higher FOVs.
The numbers look pretty spot on to me compared to the hype and unveiling - pretty close to what they implied with their % gains and I don't really see where the bait and switch that other posts mention has happened.
It's still worth waiting to see AMD's retort before buying anything really, but the numbers on the 3080 look really good so far. Great upgrade from the original 2080FE.
Last edited by Spud1; 17-09-2020 at 01:43 PM.
dannyboy75 (17-09-2020)
Thanks, interesting link. Unless I’m reading it incorrectly though, the 3080 averaged 48fps in their test @1440p, versus 37 for my 1080Ti. So the 3080 is giving roughly a 30% boost? I know FS is more cpu-limited than most modern games, but that still seems a decent increase to me. Especially after watching the slideshow of New York last night!
RIP Scan's website.
Kalniel: "Nice review Tarinder - would it be possible to get a picture of the case when the components are installed (with the side off obviously)?"
CAT-THE-FIFTH: "The Antec 300 is a case which has an understated and clean appearance which many people like. Not everyone is into e-peen looking computers which look like a cross between the imagination of a hyperactive 10 year old and a Frog."
TKPeters: "Off to AVForum better Deal - £20+Vat for Free Shipping @ Scan"
for all intents it seems to be the same card minus some gays name on it and a shielded cover ? with OEM added to it - GoNz0.
OCUK is struggling too!
You're comparing new-for-new, though - that comment specifically said second-hand. If the 2080Ti drops to anything below £600 second-hand (big if, to be sure) then it would provide similar or better performance per dollar to the 3080 at a lower price point while consuming less power. That's assuming anyone can get a 3080 for £649 - I'm seeing prices up to £900 on some models, at which point you could spend £750 on a 2080Ti and still get a "good deal".
That depends if you consider 20-25% similar or not, I don't really. But yes, I am comparing new for new.
Scans website seemd dead, OCUK probably is but they are doing their usual price scalping trick by the looks of things. Nvidia still have not put the FEs up for order from what i can see.
Currys have them available, website working:
https://www.currys.co.uk/gbuk/rtx-30...-criteria.html
Half my reply got cut off...
Bait and switch was on:
Efficiency - perf/watt has barely moved at the card's actual performance levels; 3080 owners won't see the quoted "1.9x" gain.
Peak performance - they quoted max FP32 performance knowing full well it's unattainable in games and is deeply misleading when compared to Turing.
In-game performance - they got DF to launder their claims using specific titles, settings and resolutions, knowing that's the absolute best-case scenario. If you don't play at 4K with DLSS, you'll never see that "up to 2x performance" over a 2080.
There are currently 1 users browsing this thread. (0 members and 1 guests)