Metro 2033 Exodus and Ray Tracing...

Discussion in 'Video Cards' started by Archaea, Jun 23, 2019.

  1. Factum

    Factum [H]ard|Gawd

    Messages:
    1,718
    Joined:
    Dec 24, 2014
    I guess the OP cannot tell the difference here either:


    OS the question is...is the problem with the eyesight of the OP...or some hardware bug.

    I know to what side I am leaning...
     
    Maddness likes this.
  2. Factum

    Factum [H]ard|Gawd

    Messages:
    1,718
    Joined:
    Dec 24, 2014
    The number or DXR card are not limited to turing...
     
  3. RamboZombie

    RamboZombie [H]Lite

    Messages:
    75
    Joined:
    Jul 11, 2018
    Ok, and your point being what?
     
  4. XoR_

    XoR_ Gawd

    Messages:
    809
    Joined:
    Jan 18, 2016
    Waited for what?

    If you do not have enough money to buy graphics cards and still have have enough for hookers and drugs then perhaps you shouldn't be buying new graphics cards...
     
    Factum likes this.
  5. RamboZombie

    RamboZombie [H]Lite

    Messages:
    75
    Joined:
    Jul 11, 2018
    that is very true, but i happen to have enough for both, but i can still regret a purchase.. that comes with hindsight usually.. and with hindsight being 20/20..

    The only point i was trying to make was: Metro playes the same with and without raytracing. - it does not alter the game in any way.... and well i for one does not think that it's worth it.
     
  6. Factum

    Factum [H]ard|Gawd

    Messages:
    1,718
    Joined:
    Dec 24, 2014
    Trying to find out what some people consider the difference between normal adoptation and what people call a "failure".

    When company A's "failure" sells better than company B's product...are they both a failure?
    Or am I getting to complicated for you know?

    This is funny...you expect light to alter gameplay?

    For sure you are trolling now?!
     
  7. XoR_

    XoR_ Gawd

    Messages:
    809
    Joined:
    Jan 18, 2016
    For all it is and what it does DXR implementation in Metro at least does not reduce performance that much. I would say it is the kind of performance reduction that is pretty acceptable.
    Back in the early days of DX9c games we had bigger performance differences when enabling higher resolution shadows.

    Besides if Metro used different type of global illumination for that last drop of lighting realism that it have I doubt performance would be any better than it is with RTX. VXGI is still pretty expensive. Only advantage to it would be wider support... but at the same time doing such stuff with ray tracing is easier and for now have this advantage that it would then decrease performance presented at Ultra setting in all the benchmarks for that game and with DXR it only affects RTX cards... which are pretty powerful anyway.
     
    IdiotInCharge likes this.
  8. RamboZombie

    RamboZombie [H]Lite

    Messages:
    75
    Joined:
    Jul 11, 2018
    well, i expect the added eye candy to look good, it looked more meeh.. and was imho not worth the added cost for an RTX card - and again no need to reply to this - this forum has an ignore list - welcome to mine..
     
  9. Factum

    Factum [H]ard|Gawd

    Messages:
    1,718
    Joined:
    Dec 24, 2014
    Funny how people complain about performance...in settings you can disable.
    I remember when the "bottleneck" was AF. Going from 4xAF to 8xAF reduced FPS...but it loked better!
    No one whined and called AF a "failure".
    How times have changed
    I remember my first GPU that had the option for AA...but if enabled...the FPS would TANK
    No one raged about how AA "was bad"....we all knew it came at a performance cost...(I guess there were some "I cannot see the value af AA...fps is all I care for"...but they were muppets and got largely ignored).
    Now everyone does AA + 16xAF...and the perfomance "tanking" today is not worth mentioning.

    Now someone mentioned "it did no alter the gameplay"....well DUH!
    Something that would alter gameplay would be GPUGP physics....but oh the whines about the "bad" thing of altering gameplay.
    I suspect it was due to NIVIDA buying AGEIA (PhysX)...and the "brand" came into play.
    Red did not like...green did like....nothing about the tech...all about brand.

    Now we have DXR....a Microsoft expansion of DirectX 12.
    So we should avoid the brand-babble right?

    Oh no....because red is not part of the part...DXR is labed as "bad"...."NIVIDIA crap"...."I cannot tell the difference" crap...again brand....not the technology.
    But it will all change soon (®©), because unlike the "branders" think...DXR is not NVIDA...it's Microsoft (aka it will find it's way into the Xbox-line)...and AMD is planning to join the party:
    Texture based RT: http://www.freepatentsonline.com/20190197761.pdf

    Now I ask the question:
    Will the branders keep having the same stance...or will they sudden have a change of heart?

    I know here this is going...it's sooooo predictable....but a few branders are setting themselfes up for a major foot-in-mouth-syndrome...which will be hillarious to observe ;)

    TL;DR
    The grapes are sour branders whine bacuse of brands, not technology.
     
    IdiotInCharge likes this.
  10. Archaea

    Archaea [H]ardForum Junkie

    Messages:
    9,179
    Joined:
    Oct 19, 2004
    The primary person whining about brand in this thread is you, Factum.
     
  11. Snowdog

    Snowdog [H]ardForum Junkie

    Messages:
    8,814
    Joined:
    Apr 22, 2006
    So many people gripe about how RTX ruined the cards, made them poor sellers, and it would be so much better without. Well you have that choice. Navi. Let's see how that does in comparison...

    The RTX 2070 has been on the market less than a year and already has a bigger steam share than every single AMD card except the RX 580. That means all the "buy a sale priced Vega claims" went nowhere. Vega with over an extra year on the market is still behind that "overpriced" slow selling RTX 2070, on the market for less than half the time.
     
    IdiotInCharge likes this.
  12. kirbyrj

    kirbyrj [H]ard as it Gets

    Messages:
    24,209
    Joined:
    Feb 1, 2005
    Talk about a straw man. RTX didn't ruin the cards. RTX in and of itself isn't necessarily bad. THE GODDAMN PRICE RUINED THE CARDS. Just so we're clear. People are speculating that RTX could be dropped to LOWER THE PRICE.

    $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
     
    RamboZombie likes this.
  13. Archaea

    Archaea [H]ardForum Junkie

    Messages:
    9,179
    Joined:
    Oct 19, 2004
    In your video Steve says it’s really hit or miss whether anyone can tell RT is on in gameplay in the first 3 minutes, and discloses the in-game benchmarks and cinematic are using RT unrealistically heavy compared to the actual game where in many scenes, RT use in game is very light. Then he proceeds to show both of those two types of extra heavy use cases.

    I’m not very far into the game yet, only about an hour but after watching both your videos, and reading the techup review, (realizing all these videos are probably cherry picking the scenes with the biggest delta) and most of all based on my own comparisons in playing the game myself. Turn it on for a few minutes of gameplay, turn it off for a few minutes of gameplay. Turn it on instantly look around, turn it off look around. I don’t think most people could even accurately claim which is which greater than 50% of the time in Metro 2033 Exodus with randomly selected scenes. I don’t think I could.
    RT is nearly entirely inconsequential in its current implementation in Metro 2033 Exodus—IMO.
     
    Last edited: Jun 28, 2019
    RamboZombie likes this.
  14. XoR_

    XoR_ Gawd

    Messages:
    809
    Joined:
    Jan 18, 2016
    Let me see...
    fjVXR2c.jpg
    My ray-traced crystal ball tells me AMD fanboys will be like RTX ruined RTX cards! and This site received bribe from Evil Corp (Nvidia) and in reality Navi is much faster than these tests show and of course These are early drivers. Performance will improve!!!!!!!!!!!11 and then... go buy RTX 2080 Super :cat:
     
    IdiotInCharge and Factum like this.
  15. XoR_

    XoR_ Gawd

    Messages:
    809
    Joined:
    Jan 18, 2016
    What kind of consequences would satisfy you?
    We can safely assume these cards are nowhere near powerful to do whole scene path tracing in game like this.
    And even if they did such rendering pathway and somehow game did not slow down to single digit fps then people would be like "there is noise, it is better without RTX"...
     
  16. lostin3d

    lostin3d [H]ard|Gawd

    Messages:
    2,000
    Joined:
    Oct 13, 2016
    I have to totally agree with you. The branding of ray tracing has followed suit with what happened with HD, FHD, UHD, and so on in relation to the resolutions that used those labels. With RT, as you said and has been brought up in reviews, it's implementation has been very limited, hybrid, and so far each game has merely taken a slice. Quake II was the first to check all or nearly all the boxes you could do. For Metro the closest anyone will get for real perceptual differences will need to use the Ultra setting as there's quite a bit of things that get turned off on high and even then it's only adding a few more slices of the pie.
     
  17. XoR_

    XoR_ Gawd

    Messages:
    809
    Joined:
    Jan 18, 2016
    1080Ti is on market far longer and compared to Turing cards it was pretty cheap (it is f.cking sad I can say this...) so it is no wonder it sold well.

    How would Turing sales look if they if they did not include RTX capability?
    And how it would look like if Nvidia kept pricing from Pascal generation? They would sell much much better.
    It is safe to say that Nvidia at least sold part of these cards due to ray-tracing and sold hella less cards because of higher pricing.

    With market share they have and performance advantage they can afford pricing cards higher to get less units sold and get more profits per card.
    RTX was an excuse for something they wanted to do all along. And like I said before: far better excuse than "we are poor company that struggle to make ends meet" XD
     
  18. Snowdog

    Snowdog [H]ardForum Junkie

    Messages:
    8,814
    Joined:
    Apr 22, 2006
    And you think the price is too high because of RTX, thus RTX ruined the card, so it isn't a straw man at all.
     
  19. Gideon

    Gideon 2[H]4U

    Messages:
    2,162
    Joined:
    Apr 13, 2006
    Yes and we all know due to the price not a ton of 1080Ti cards were sold, that was always going to be a limited market like most Halo cards are. Yet were far enough into the RTX launch that if they were truly popular and selling well as some like to claim they would easily have far exceeded the numbers of 1080Ti's out there on steam. You can see the largest share is the 1060 due to the price and where most of the market buys at and the 2000 series exceeds that price point at all levels. To me it looks like what they sold to was the market that just has to have the latest thing and most stuck with their older cards.

    If these current prices continue on both sides the market will be damaged as people will loose interest in PC gaming as it will be far more costly then a console. Part of the reason PC gaming has surged again is due to the costs of a gaming rig being quite a bit cheaper then it was at one time. I worry more about that then RT being needed feature for now.
     
  20. MangoSeed

    MangoSeed Limp Gawd

    Messages:
    394
    Joined:
    Oct 15, 2014
    Based on those same steam numbers and time on the market the gtx 1080 (36 months) adoption rate is a whole 13% higher than the gtx 2080 (9 months). And the 2080 is 14% more expensive.

    So accounting for time and price Turing is selling just fine and in line with Pascal. I have no idea how you came to the exact opposite conclusion using the same numbers. But I can guess ;)
     
    IdiotInCharge and Auer like this.
  21. Gideon

    Gideon 2[H]4U

    Messages:
    2,162
    Joined:
    Apr 13, 2006
    One should be showing decline and the other growth and yet thats not happening, amazing how you missed that and I can guess why as well. Also I can walk into any store and the 2000 series is always in stock, Pascal cards were always running out of stock. So yeah it's a painfully obvious difference if you actually look and the fact that Nvidia revenue is way down is also a clue. But dont let facts stand in the way of your feelings that anyone that doesn't like RTX cards must be a AMD diehards even tho I own a 1080 GTX and sold my 290x for it.
     
  22. MangoSeed

    MangoSeed Limp Gawd

    Messages:
    394
    Joined:
    Oct 15, 2014
    So essentially you're ignoring hard numbers and instead ask us to trust in your feelings? None of the statements you're making are supported by facts.

    End of life products are sold out and new products are on the shelves. Based on that you claim the new products aren't selling. Amazing logic.
     
    IdiotInCharge likes this.
  23. purple_monster

    purple_monster Limp Gawd

    Messages:
    386
    Joined:
    Jun 1, 2018
    Hello,

    $1 has been deposited into your account.

    Thank you,
    Nvidia
     
  24. purple_monster

    purple_monster Limp Gawd

    Messages:
    386
    Joined:
    Jun 1, 2018
    oh brother these kind of people...
     
  25. Gideon

    Gideon 2[H]4U

    Messages:
    2,162
    Joined:
    Apr 13, 2006
    upload_2019-6-28_10-30-53.png

    Ok so lets see Jan to May results we should see declines if the 2000 series is selling so well. 1060 increased by 1.43%, 1050Ti increased by .57%, 1050 increased by .03%, 1070 increased by .32%, 1080 increased by .06%, 1080Ti increased by .09% and finally the 1070Ti increased by .11%. So yeah those are some massive declines... Oh wait they all increased in numbers and the inbound Super cards are not going to change that fact either if the rumored prices are true. People simply are not upgrading and the few that did buy those 2000 series many others are buying their used Pascal cards and using them instead if they are still on the 900 series cards or older.
     
  26. MangoSeed

    MangoSeed Limp Gawd

    Messages:
    394
    Joined:
    Oct 15, 2014
    What's the basis for the assumption that the 2000 series sales would specifically cannibalize Pascal parts? What about Maxwell and Radeon?
     
  27. Gideon

    Gideon 2[H]4U

    Messages:
    2,162
    Joined:
    Apr 13, 2006
    I dont need to post the whole chart to see the same fact, you got eyes and google give them a shot. 580 has increased, R7 shows a tiny decrease, 480 is pretty much flat, the lowly 570 has almost doubled, Maxwell cards show slight declines hell Vega even had a small increase. The $200 range cards are the ones with massive growth, most are holding out for the next gen and the survey supports that notion. As for why, the prices are so high only enthusiasts in the $350+ category have something to possibly upgrade to, the rest are still waiting for something affordable and is actually a upgrade. Hell I usually spend $450 or less and their is nothing for me to upgrade to that will really outperform my 1080. So seeing someone that bought one of these cards for RTX features then feeling underwhelmed by it wont be helping others to decide to spend alot more to upgrade.

    https://store.steampowered.com/hwsurvey/videocard/
     
  28. Araxie

    Araxie [H]ardness Supreme

    Messages:
    6,361
    Joined:
    Feb 11, 2013
    you clearly have no knowledge about statistic and sales.. each new generation of GPU launched doesn't translate directly on a decline on older generation of GPU as those are the points that lot of people wait to buy older generations of high-ends at lower prices, even when pascal launched and GTX 1080 was the fastest high-end GPU ever sold followed by the 1080Ti people kept buying GTX 980Ti as it was (and still is) actually a cheap GPU (heck people are still looking for GTX 980 and 980Ti)... when turing launched people jumped to buy pascal GPUs at a cheaper prices.. is not anything strange to see in any thread when people ask for gpu recomendations to still see gtx 1070 - 1080 as choices the thing is.. RTX are not entirelly needed for 1440P gaming much less for 1080P and those are the performance brackets were most people are, RTX 2080 and 2080Ti are the choice for 4K resolutions and we all know 4K it's still a minority on the gaming market.. that's why it was expected to see still strong sells on pascal post turing launch specially after the mining bubble exploded as the market were flooded of GTX 1060 up to 1080 GPUS.. people are actually still looking to buy GTX 1070 for high refresh rate 1080P or GTX 1080 for 1440P as they are available for cheaper prices.
     
    IdiotInCharge likes this.
  29. AceGoober

    AceGoober Live! Laug[H]! Overclock!

    Messages:
    21,348
    Joined:
    Jun 25, 2003
    For me, taking into account first gen hardware and software implementations, doesn't give me any reason to be excited or wanting to purchase a RTX graphics card. I do care about eye candy but not at a high cost of performance. I'll wait to see where the technology is in a generation or two before deciding whether or not it meets my requirements for gameplay. Until then, I personally have no interest in RTX.
     
  30. Derangel

    Derangel [H]ard as it Gets

    Messages:
    17,526
    Joined:
    Jan 31, 2008
    There has also been a flood of relatively cheap Pascal cards on the market for a while, especially during that time. Whether its a bunch of people offloading cards on the used market or retailers/AIBs trying to clear Pascal stock Nvidia forced them to take on during the mining craze. Its pretty hard to use Steam as an entirely accurate representation of hardware sales in the first place (even if its the "best" option we have as consumers) but the market following the fallout of the mining boom has made it essentially useless for now. If Nvidia discussing Turing in their next quarterly earnings report that should shed some light on how sales are going. Until then, its all pretty baseless speculation.
     
    Gideon likes this.
  31. MangoSeed

    MangoSeed Limp Gawd

    Messages:
    394
    Joined:
    Oct 15, 2014
    Amen.
     
  32. Gideon

    Gideon 2[H]4U

    Messages:
    2,162
    Joined:
    Apr 13, 2006
    Lol you of all people saying that, you defended it as fact with Factum and now it's all garbage and shouldn't be used. Steam survey is nothing more then a very limited look at the market.
     
  33. MangoSeed

    MangoSeed Limp Gawd

    Messages:
    394
    Joined:
    Oct 15, 2014
    Steam numbers are hard facts. I was simply pointing out your interpretation of them is misguided.

    It’s incredible that some people claim that a sampling of hundreds of thousands (millions) of users is a limited look at the market. I assume those people have never taken a statistics class.

    The steam survey is the most credible data source we have for PC gamer hardware ownership. It’s certainly far better than wishful thinking.
     
    Factum, Araxie and IdiotInCharge like this.
  34. Gideon

    Gideon 2[H]4U

    Messages:
    2,162
    Joined:
    Apr 13, 2006
    Last reply to you on this as your just all over the place. Steam is like a poll, they ask only a limited amount of users for data and it's been well over a year since I replaced my 290x with a 1080 yet steam still has not surveyed me. It's a pretty small sample size and at best shows trends, not absolute numbers that should be taken as hard facts of market share. Yet you agree with the guy that says it's all baseless speculation but now say it's hard fact, might want to pick one or the other. Net Income for Nvidia is down 31% quarter to quarter while the product margin is way up that alone tells you sale numbers are far lower for the 2000 series. Also operating expenses being way down can show they are manufacturing far less as well. Nvidia's own financials point to my favor.
     
  35. MangoSeed

    MangoSeed Limp Gawd

    Messages:
    394
    Joined:
    Oct 15, 2014
    Thank goodness.
     
  36. Snowdog

    Snowdog [H]ardForum Junkie

    Messages:
    8,814
    Joined:
    Apr 22, 2006
    Steams user base and sample size, is large enough that they don't need to sample the same user again looking for changes. They likely just get a new random sampling of users each time, and get stable results because the sample size, is large enough to get consistent results.

    You are really off a bit on the financials, and that argument really doesn't hold water, plus you can't compare quarters sequentially to each other because GPU sales are something of a seasonal business, so you should compare Year over Year to get a better ideal.

    Which is just as well since revenues were up quarter to quarter, the 31% down you mentioned, was actually the Year over Year, vs Q1 Crypto enhanced numbers from last year.

    NVidia revenues fell mainly from crypto implosion and results returning to reality:

    This supposed big fall in revenue is still the second best Q1 every, only bested by one crypto-enhanced quarter, and margins are down 6.1% Year over year, not up.

    https://www.anandtech.com/show/14360/nvidia-q1-fy-2020-earnings-report-postcrypto-reset

    Basically the narrative of high margins and low sales from RTX cards is NOT supported by the full picture of the financial report.
     
    Araxie, lostin3d and IdiotInCharge like this.
  37. Gideon

    Gideon 2[H]4U

    Messages:
    2,162
    Joined:
    Apr 13, 2006
    https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-first-quarter-fiscal-2020

    Almost 4% gain in margin quarter to quarter is quite a bit, not sure why you want to suddenly include the year ago for this. I did muff the the expenses being down tho, teach me to try and multitask as they are slightly up. As for revenue I went by Net Income not Operating Income since that includes all income. Will just have to disagree on your interpretation as a strong product sells well even in traditionally weak quarters. But I dont want to further derail this thread and your free to disagree with me along with anyone else, but that is the way I see it. Also all the product I see on the shelves is a bit of a indicator as well. As for Steam survey, I find they are as accurate as the polls in the last Presidential race we had here in the USA.
     
  38. IdiotInCharge

    IdiotInCharge [H]ardForum Junkie

    Messages:
    10,347
    Joined:
    Jun 13, 2003
    This is hilarious. Steam actually polls the population, which is in Valve's best interest- polling during the last presidential race failed to poll the population, because it wasn't in the pollster's best interest.
     
  39. Archaea

    Archaea [H]ardForum Junkie

    Messages:
    9,179
    Joined:
    Oct 19, 2004
    I don’t disagree, but I’ve always found it odd. Wouldn’t the liberals have won if they’d (mainstream media, pollsters, talking heads) acted like Trump was definitely in the race or even in the lead and it was imperative the Democrats got out the vote or they’d surely lose? I always wondered how it benefit them to act like Trumps campaign was a non-threatening sideshow and Hillary’s election was a sure thing...
     
  40. IdiotInCharge

    IdiotInCharge [H]ardForum Junkie

    Messages:
    10,347
    Joined:
    Jun 13, 2003
    While the 'math' is the same, political polls are inherently far more political. We'd see more parallels if say the Steam survey (and others) were being influenced by the vendors of said products to some end. It seems none of the major vendors really have a reason to do this, unlike in politics where all coverage increases or decreases public awareness, so bad news is still free coverage. Also, I recommend we drop the political parallel discussion as this isn't the Soapbox ;).