![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
.
Yes, I just want to make sure people understand I'm not bashing on DXR in general, just that we don't have a viable ecosystem at the moment. I agree, at the higher end (2080ti type speeds) it is almost viable, and future games after they figure out how to better work with it, have a great chance of being playable on that hardware. I just have doubts about a 2060 (super) ever getting to the point of being viable, which is what the 5700 competes with. Good on nvidia for trying to push forward with technology and being the first to implement a feature. But just because you are the first to do something, doesn't make it the best implementation. Maybe future games will have better LOD options for raytracing and have better implementations allowing you to see some benefit without to drastic a hit to performance and allow lower level hardware to use it. I hope this is the case, but I'm a skeptical mind
What are you ranting about? Slow down, process, then respond. Ecosystem is getting started, it's not here. I can't go play more than a handful of games with so/so support. The building blocks are in place, and within the next two-three refreshes it will probably be viable to some extent. I don't call a single $1200 card that can kind of run stuff at low resolutions viable/ecosystem. I never said it won't improve nor did I mean to imply that there is nothing at all, just that it's not there yet (but much closer than it has ever been).
No, viable to me is I can buy a mid range card and play a game @ > 40fps (depending on game type, some are ok down to 30, others crap until over 50)... Not a $1200 card to run 60fps @ 1080p with minimum gain in IQ (in most current games, I can't speak to the future, nor can you). Where did your standards go? Up to the stratosphere with this rediculous pricing structure? I am down here on the ground waiting for it to become a reality and it seems we're at/near that tipping point.
Fair enough, but BF4 isn't a great title (from what I've heard, I don't have it just read reviews) to compare as the experience with it on is sometimes ok, sometimes distracting. DXR low in one game makes it look better in some scenes... from the screenshots (again, I don't have, one, just read through a bunch of benchmarks with side/sides) I just don't see that much improvement in general, but I'm not a Hi-Fi type of person (probably because my target is low end, so my bar is so low anything looks good!) I'm not saying you should run out and buy Navi if it doesn't have DXR, nor am I going to run out and buy a 2080ti just because it has it. I was simply explaining my position, not everyone has to share it, but wanted to say that it was from logical reasoning and actual constraints, not just I love a company and everyone else is stoopid. Until it is more useable and more AAA games come out, I put very little thought into it. Once it's more mainstream, I'm sure I'll pick up a midrange card that supports it just to play.
. That was full tracing, not rendering geometry then tracing from those pixels (texels). That doesn't make it better or worse, just different targets (and that was like.. 10+ years ago when programmable GPU's where just coming out and I was getting pre-release hardware from both ATI and Nvidia at the time for doing development). I have a great appreciation for what it's doing, but appreciating something and being able to use it are not equal.Yes, enough people are willing to pay that much, however the mainstream market is still in the low/mid range. What's viable for me isn't necessarily the same for everyone, some can afford more, some can't afford as much. If it's a few % of the overall people that can afford it, and even then a very few % of games, and of those only a few done decently, it's just not there yet for me. If it had come out and improved ALL games currently in existence, awesome feature. When it supports a handful of games with expensive hardware, cool, some can afford it and even fewer utilize it (most use the 2080ti without just for the raw framerate), I don't consider it a viable solution. Again, I'm not trying to knock the technology, because it's pretty awesome to see how close it is to becoming mainstream. Maybe viable isn't the right term, maybe I should be using/saying mainstream instead. I really do hope some awesome games come out that can implement it nicely, but in the mid range I don't think the current gen is going to be usable, which (more than likely) makes AMD's choice to not use it in their mid range a good one. They know they will have to implement it at some point since it will become mainstream at some point, I just hope they don't wait to long, but I would give up DXR for faster performance in my price range easily at this point.
Troll much? No kidding there are performance hits with options, some bigger than others. The question always has and will be is the performance hit worth the IQ. That's honestly up to an individual, some people don't mind slide shows, others are bothered if they can't hold 144fps... With the majority somewhere in the middle (and it depends on the game as well to what's acceptable). The overall point is, I would not waste extra money in the 2060 range of cards for dxr since it would being my fps to low to be useful. To each his own though. We all know AMD is bringing dxr support, but not in the mid-range and I don't blame them one bit. Just as I don't think Nvidia testing waters and getting a head start on development is bad, just that it's premature for most people.
.
I wish Nvidia was more aggressive with G-Sync- they have the superior solution, especially when you get to the edge of performance (which is what Nvidia defines, because well, we're still waiting for 'Big Navi'), and the G-Sync module is very small in cost when you get to monitors that are themselves 'at the edge'.
To be clear: there are still newer, better monitors that are being released with G-Sync first. The feature is far from dead, and likely won't die until there is a consistent performance level from Freesync that can objectively match it. To be further clear: that doesn't exist yet.
...and then lost because they refused to refine their standard.
Which I assume had quite a bit to do with the 'uptake' of Freesync, because building a Freesync monitor that can compete with the performance of a G-Sync monitor ain't cheap
. I'll also unequivocally agree that any VRR is better than no VRR. I'm not going back to no VRR either!
Like I said, I wish they better defined their standard, but I understand their reasoning. They did it b cause it allowed current monitors and chips to simply do a firmware update and have support, versus manufacturing new parts. This meant adoption rates would be much simpler and higher. When competing against someone with more resources, you need to get manufacturers on board. If they went for high standards and proprietary AMD hardware, I believe most would just make gsync monitors. AMD stuck with the public protocol defined by VESA (adaptive sync), rather than creating yet another fragmented segment. I think there needs to either be a better defined standard or a l simple way to define low/mid/high end, so standards categories.I wish Nvidia was more aggressive with G-Sync- they have the superior solution, especially when you get to the edge of performance (which is what Nvidia defines, because well, we're still waiting for 'Big Navi'), and the G-Sync module is very small in cost when you get to monitors that are themselves 'at the edge'.
To be clear: there are still newer, better monitors that are being released with G-Sync first. The feature is far from dead, and likely won't die until there is a consistent performance level from Freesync that can objectively match it. To be further clear: that doesn't exist yet.
...and then lost because they refused to refine their standard.
Which I assume had quite a bit to do with the 'uptake' of Freesync, because building a Freesync monitor that can compete with the performance of a G-Sync monitor ain't cheap
I think this is where most are getting things wrong or maybe not clear. It's not AMDs standard, it's their implementation on the VESA apadtive sync standard first released in HDMI 1.2a. they have since helped better define things in freesync 2 (next version of adaptive sync). Again, AMD didn't make the standard, VESA made it, AMD implemented. They *could* have told monitor manufacturers that they couldn't put the freesync label on a monitor if it didn't meat a specific low end spec, but they wanted adaption (for better or for worse). I think them helping move the industry forward is good and not being locked into a specific brand is good too. I can buy a nice high end freesync monitor and switch between AMD and Nvidia at will. You can't do it the other way around.
Sorry, my memory wasn't quite right on this, you are correct it was DP. Since they use the same signaling it was a no brainier to ext me support. Yes, I admit that AMD did it to get exposure and market. I dont blame them for allowing lower spec as it allows lower cost of entry, I do however wish they had more defined criteria and possibly levels of support like low/mid/high so you could more easy determine what you were getting even if your not a techy person.
Yes sir, sorry about that. They did it in a way that a lot of manufacturers that already supported it on DP could do a firmware update to support. And yes, as I mentioned, they worked with the industry instead of against it.
.