The eclipse p500a from fantex is apremium mid tower that combines airflowperformance and aestheticsa full mesh front and ridiculous amountof fan and radiator configurations withup to seven 120 millimeter fansensure that your components neverthermal throttle the eclipse p500a isavailable in both rgb and non-rgbversions making it a flexible choice foryour next system builder upgradeto see the full list of specs andconfigurations click the sponsored linkin the description belowokay um first things first my eyes arebloodshot my brain is frazzled i've beenbenchmarking for two straight daysover 18 hours of benchmarks because iwant to get these numbers right for youguys and i compared them with variouscardsi'm going to misspeak i'm going to dostupid things that's a normal videoall right 30 80. it's here we unboxed italready we're not going to talk aboutthings i've already talked about in theunboxing in fact i'm going to be likethat youtuber that's like go watch ourprevious video where we already talkedabout theconnector and stuff so you guys cango and watch that video if you want toumright over here don't pop out thereunless i forgetthere we go um we already talked aboutthe cooler the way it potentially comesapart there's no screws on itum the 12 pin plug we talked aboutaftermarket plugs to replace this guy soit's notugly it's all over there i want to getright to the performance videos todaythere's a lot to talk aboutso what testing methodology here and acouple of things about the card as wella lot of people were disappointed inthat last video that we didn't power itup we didn't show you any metrics oranything and that's just unfortunatelythe way the embargoes workedbut let's start with this you know whengenson said and it lights upthe rtx 3080 and it lights up i thinkthere was a collectiveduh they've like all lit not litten upthey've all lit up for like years andyears nowwell my biggest complaint with this guyand i loved this cooler and this isstill one of their favorite coolerdesigns for methis their first double fan cooler isthat when you turn it this way the biggeforce logo lights up greenand i get that that's nvidia's colorgreen that doesn't work with all thebuildsthe titan rtx had a white led and i'llbe damned if this one didn't have awhite led in it too so not only does thegeforce logo on this and the geforce rtxon the side light up whitethe front and the back of the cardinside the heatsink alsowhy do i keep taking this off i want tolet it stand up there and be all ugly sopeople know not to use itanyway the actual bottom side of the carthe x lights up white because it'sextremeand then the top only part of the xlights up because the other side is theback plate for the pcblet's talk about the test we did a quickmethodology run through so you guys knowwhat you're looking atas always please take our benchmarkswith a grain of salt just as you shouldany particular single benchmark listcompare them as many people as you canfindrunning tests formulate your own opinionabout whether or not the card's worth itfor youand do your due diligence don't listento any one reviewer we don't trust invideos number by themselves we don'ttrust amd's number by themselvesdon't trust our numbers by itself get asmuch information as you canwe added more titles we threw out someold ones you guys know i like to use abunch ofmixtures of engines and that gives you areally good indication of what theperformance is likewe use some synthetics we use some livegame plays we use time spysuper position and all that but what wewant to point out is we added a fewtitles in here now that actually use rtxtechnology and those titles specificallybeingmetro exodus we also tested uh shadow ofthe tomb raider with the ray trayshadowsbattlefield v in here which was one ofthe very first rtx in fact i think itwas like the first rtx title that waspublicum and then of course we includedcontrol which is justit's basically a playable benchmark ifyou want to know the truthwith all of the rtx technology availablein there it's got like16 different dip switches basically forturning things up and down but what'simportant to point outis that if it's an rtx title with theexception of control we left them all onif it's an rtx title we tested it bothrtx onrtx off showing you both therasterizationand the taxation of turning on rtx sowith all that out of the wayrun through the benchmarks pause themand thenwell ascend into pcmr because thisgeneration this time around is just theone that just keeps on giving so now that you've seen those benchmarknumbers there's probably two things youdidn't noticeso why don't you go ahead and go backand watch them again and then maybewatch them one more timeand again if you just really like the . i don't know there's a patternand that pattern really stuck out to meafter about 10 hours of testingand that's the fact that the 2080 ti at1080pwas roughly the same fps as the 3080 at1440.let that sink in this is like a wholeresolution tier higherat the same fps on average than the20dti it didn't always hold up sometimesit was lessbut it was close enough to where youwouldn't notice itbut speaking of fps none of these titlesthat featured dlsshad dlss turned on native hardwareretracing retracement ray tracing anyoneright now that's on an rtx cardor any card that features oh i guess rtxcard is what you need for rate racingand dlss butif you've seen how good dlss 2.0 is andthe amount of people are like dls 2.0was amazing we criticized dlss when itfirst came out for me it was a smudgymess i didn't like itand i just threw it aside and didn'tbother with it againmy two 2080 ti's at home don't have itturned on i just was like i don't likeit doesn't look good to mebut phil was like no bro you need tolook at dlss 2.0 and then he showed mecontrol which is one of the titles thatwe have in here with dlss 2.0 on and itwas absolutely insane how good it isspeaking of insane if i look at controlhereat 4k with the 3080 fehe was getting 35 fps that's nativeray tracing no dlss once we turn dlss2.0 onwhich is 2.99 that's the latest versionover 70 fps double the frame ratein control and i went yeah but is it allsmudgy yeahget off my lawn like i was i was abitter old yeah i probably lookedand i was just like wowit looks so good he was getting 100 fpswith the 3080in 1080p ray tracing now 1080pum you can't really run dlss for 1080pso that's one of the reasons why wedidn't turnon dlss on these titles because ifyou're running 1080pit doesn't make any sense the horsepoweris too high you actually have aperformance hitif you turned on dlss with a 1080p panelandyou know card running 1080p but 1440p68 fps hardware he was getting over 120fps with dlss 2.0 on with ray tracingin control all the functions turned onbruh 2080 ti and 2080when it was launched in gamescom in 2018was a very controversial launch a lot ofpeople were very mad because we didn'tsee the typical 25 to20 to 30 more like 20 to 22 and a halfpercentuptick in performance or upliftingperformance we're used to seeinggeneration on generationbecause this was the first card of a newera of graphics cards if you will thisis 2 000all over again with the geforce 256rasterization that's the way we're doinggames now and thenyou could i love that this is like thefifth time i pointed this outall those articles from like a non-techpc mag and all of thempc gaming they're all still publiconline you can go and read the reviewsof what people said about the way theold cards performedthe things everyone's like oh no wedon't care about rtxjust leave it the way it was is the samething they said 20 years agofast forward and it's the way we dogames now but what happened with the 20series is like anything else any newemerging tech which now has rt cores andtensor cores and a whole newmanufacturing process that's differentthan the way they were doing it it'sactually got worse yields in terms ofwaferwaste and such it was expensiveso it not only did it not conform to thetypical 700-ish dollarprice of a 1080 ti it was 500 more thanthat at 1200so it gave you likethat what 40 percent more cost for like20 more performance that just was likethat doesn't make any senseso people lost their minds rightfully soit just seemed like a move in the wrongdirectionin terms of pricing the common thing youalso heard people say was just wait forthe second gen rtx cardsbehold your savior the 30 80.this card is 400 less than 2080 ti andit is up to 35to forty percent faster in some of ourtests i say some because nono one test is indicative of how theyall all engines react differentlydifferent architecturesit's got more rt cores more tensor coresfaster memoryless of it 10 gigs versus 11. butit's insane how they're able to get thismuch extra performance and i foundmyself going holy crap 30more at 400 less my brain can't handlethisand then i dropped the real bombshellwhen i told phil i'm doing this wrongwell you guys tell me that all the timebut i'm doing this wrong because i'mcomparing it to a 2080 tithe story here is the way it compares toa 2080that was a 800 card when it launchedthat's an 800 card nowthis is up to 90 fasterat 0 more moneywe have not seen that level ofgenerational uptickfrom one generation to the next fromnvidiaever that's the story here there was achart going around we showed it in ourprevious video on reddit where somebodyactually did like the performance upliftof one generation to the next and it'slike and thenliterally ampere is just like off thetop of the chart well it didn't hit thetop of the chart but it showedjust how absolutely ridiculous that onegenerationjump was typically they'll go back youknow a couple generations because theythey believe most people will update onthat like two year cycle or that twogeneration cycle excuse meso it's up to four to five years toninety percent more performance at zeropercentextra cost price for price so there'smultiple ways you can look at this youcan be like oh okay you knowtop tier versus top tier well this isn'ttop tier 30 90s top tierthat one won't have quite the uhcinderella story i guess becausethat's seven hundred dollars more thanthis one and you're not going to get 700more performance which is you knownearly double the cost that's just theyou want the best it has an exponentialcost associated with it and that's finei will be running 3090 in my system i'mfairly certain i'm going to buy the cardbecause i don't believe we're beingseated any 3090 fe cardsand i really really like the way the3090 fe looks and i want to build asystem around iti'm going to be a customer just like youbuying itso there's some other things to talkabout here though the pattern that isaidearlier is that 1440p fpson the 3080 was like roughly the same as1080pperformance on the 2080ti it was weirdwhich really made me go wait a minutethat's with an importantan important asterisk on all thesechartsis that the 2080ti saw a massiveimprovement when we did our benchmarkingand i wentwhat is going on here so i've got mybenchmarkspreadsheets here and i've got tabs forall the games we do and i've got everycard we've ever tested formulations andstuff on cost of performance and upliftdrivers that we've used and all thatsort of stuffbecause one of the things we look for islike did we see any reduction inperformance with a certain driver on acertain card does that hold likeplanned obsolescence plannedobsolescence nvidia slows down yourcards to try and make you buy new onesthat'sthey've never done that linus has talkedabout it i've talked about it gamer'snext talked about itwe've all proven if anything we see aperformance increaseover time in the generations and thematurity of the driversbut this one i couldn't explain becausei was seeing big improvements like 20fps improvements in like 1080p on the2080 tiso i spent a while trying to figure thatone out and then it dawned on me whenthe 2080ti came out and we benchmarked it it wason my 8th gen8700k system and the benchmarksfor this launch were re-performedbecausewhen it's that old we always re-performthe benchmark benchmarks because like isaid things changewe saw a major uplift by going to the10900kwhich tells us that 1080p the 2080ti ifyou're on an 8700k or something aroundthat area or olderyou could potentially be bottleneckingyourself pretty strongly in thatlike i said we saw up to 20 fpsimprovement on some titles which isinsaneso that major gap between these two iswith that uplift accounted forotherwise you can't have a 10 900k atlaunch and theni could have been like here were thenumbers that launched versus numbers atlaunch but that wouldn't be indicativeof real-life performance becausethe performance available today withthat card and the cpus and if you wouldupgrade it would have been the sameimprovementso it actually brought the two cardscloser together but it's still anamazing gap thereso then i went okay if you have 2080 tiand you're selling it to get a 3080 andyou're on like a 8 700 kare you gonna further bottleneckyourself phil and i talked about thisand i wasn't sure whether or not ishould recommend that you go out and buya new cpu and motherboarddepending on how how old your system isi i might recommend that if you're onlike a 47 70k or 4790k like a devil'scanyon or something yesit's time to upgrade because putting a3080 on thatthere there's that's even unless you'rerunning 4keven at 1440p you're going to seebottlenecking it's insane how fast thiscard is but what we felt wasprobably the more cost effective andbeneficial upgrade is actually to yourpanelyes 1080p is still the dominantresolution on the steam umstatistics you know system statswhatever whatever they call itso unless you're an esports gamer thatneeds like the 360 fps like the newpanels and stuff from nvidia thatfeature reflex technologydoing things like starcraft league oflegends umcs go valorent rocket leagueoverwatch bottom line is unless you'rean esports gamer that needs that ultrafast fpsyou'd be running a fast enough cpuanyway so probably more of a moot pointbut it'd be more beneficial honestly ifyou're running that old 1080p from 2010to go you know what go buy yourself a300 1440p high refresh rate panel withthe money that you seewith the money that you save yeahabsolutelywith the money that you save for the3080 versus a 2080 tithat 400 is enough to buy a high-end cpuwe feel like the beneficial upgradewould actually be to your panel you'llgeta much nicer looking image you'll getmore modern techyou'll get the full horsepower whichmeans you can keep your older cpubecause you're offloading more of thatload to the gpuspeaking of load let's talk temperaturesbecause that's the other thingthe real story is how does that get thecooler on this2080 ti has a back plate has heat padsit touches things andholy crap does it get hot soimmediately people's concerns because ofthe heat of this guy is what's thisgoing to do for your cpu cooler ifyou've gotthis fan right here pulling air throughwell that way and blowing it up into thecpu cooler areawhat's that going to do for yourtemperatures surely that's going to heatup your cputhe air that comes through here iswarmish yeah i said it warmishit's not hot enough to heat up your cpuperiod now i know other youtubers areprobably already talking about that intheir review today we're gonna do aseparate video on that because theproblem is our test bed is actually awater cooled systemso we don't have a cpu tower sucking airthrough itbut i can tell you the air that's comingthrough there is nowarmer than the air you would feelcoming through aioand if that were the case no one wouldever mount a front-mounted aiobecause it would just heat up yoursystem and make it far too hot in fact ihave a video about that evenmore than one of those videos showing afront mounted aioon your system does not hurt your systemwhatsoeverfight me bro it doesn't hurt your systemneither will this because even at thefactory fan curve which runsdead silence the amount of air movingthrough it is enough to cool the gpubut it's not enough to cause a massiveheatbubble above it it just dissipates intothe airand you have to like get your hand realclose to it to start to feel any of thatwarmthand once i sped up the fan curve andused a user-defined fan curve where itwas sitting more or less like 52instead of 40 under load you can feelthe air coming through and it's just alittle bit warmer than ambientnot enough to hurt anything in yoursystem in facti would much rather have slightly warmair blowing on my ramthan no air at all if you're running aneio if anything i feel like this couldhelp your systemso we're going to do full testing onthat but as for the corewith the stock fan curve on our open airtest bench mind you people always liketo fight us on that we don't want to putit in a caseand have the case determine what ourtemperatures are because there's no caseout there that's going to matcheveryone's use case the use case casewhich means that no matter what theresults will not be indicative of yourexperienceso we open air test bench so the onlything limiting the cores temperature isthe cooleritself don't make me have to keeprepeating it it doesn't take rocketsurgery to understand thisokay if you're not understanding whati'm saying then maybe we just can't befriendsif the cooler is the only thingaffecting the temperatures that tells ushow good the cooler doesif i took the same card and threw it inthat case that steve did a review ofwith like the 15 fans that blow intonothingof course the cooler is going to looklike it sucks so we don't want that toaffect itwith that rant out of the way with thestock fan curvewe never saw it get any higher than 72celsiusunder hours and hoursof testing today that car and wellyesterday it saw a solid six hours ofrun timebefore we switched other cards and thentoday it saw another solid two and ahalf hoursof run time without a break because wewant to get the card nice and heatsoaked we don't want to have a cold cardstart a benchmark and then look realgood at the beginning and then drop weheat the card upthen we run our test6 72c and when i put the fan the fancurve to the factoryor the user defined fan curve you knowyou go into after burner you click fanthere's a button you click it brings outa drop down you can set the fan curve ijust click default user defined whichhas a nice ramp to it61c the aibs have their work cut out forthembecause the saving grace for them onthis card was that it looked prettybut the flow was terrible in my opinionthis looks greatand it has the flow to go with it sowith that out of the wayi just have one question how the helldid they do ityou


Post a Comment

Previous Post Next Post