Ross Scott

Everyone: I recently saw this link about Ubuntu dropping support for 32 bit architecture. Is this going to affect my testing in WINE in the future on Ubuntu Mate? https://www.linuxuprising.com/2019/06/wine-developers-concerned-with-ubuntu.html Veyrdite: The GTA screenshot looks like a classic case of MSAA-only with no alpha texture supersampling. The easiest solution (if possible) is to force full SSAA instead. Although this is assuming the wires are textures and not polygon models. If that's the case, then it's some shader mess intefering with everything.

I'll be having a playsession with fans playing Team Fortress Classic. The instructions are pretty simple. Launch the game in Steam, click "Find Server." Server name: AF Server Password: puppy I plan to primarily be playing "push" maps. The main ones I want to run are dustbowl and warpath, though we could also do avanti and hunted if people are interested. If you've never played TFC before, that's no problem. I recommend playing soldier or heavy weapons guy if you're unfamiliar with the game. The great thing about push-style maps is even if you don't know any of the objectives, if you're just moving in the same direction everyone else is and attacking the enemy, you're still helping the team. The game should be easy to run, but running it in widescreen takes more work. I'm playing TFC over TF2 since I was never crazy about the Pixar-style of TF2, plus I prefer how respawning is instant, so there's no downtime waiting to come back into the game. You'll be able to view the game live tomorrow on June 22nd at 5PM UTC on twitch.tv/rossbroadcast. This is a blog post. To read the original post, please click here »

Freeman's Mind returns, again! I never meant for there to be this long a delay between this episode and the last one. If I had never made the "Games as a service" video, there would have been more out by now, but that felt like a necessary detour for me. There will still be more developments on that later, but I felt like getting the larger issue made more public was a more time-sensitive thing for me. I ran into new errors I've never seen before in Source in this episode; I'm hoping they go away after this one. It's a reminder why I don't plan on doing anything else besides FM2 in the Source engine. This episode I think is the second-longest one also, just by coincidence. There's also something that happens in the episode that wasn't planned at all, but it was a great addition, so I left it in. Ross's Game Dungeon is coming up next, then more Freeman and other videos! This is a blog post. To read the original post, please click here »

Veyrdite: I have an older AMD card, but I initially disqualified AMD from the running since it looks like on Linux, they dropped all support for forcing Antialiasing on the driver level. Downsampling is of course an option, but I'm not sure that applies to older games that may not support modern resolutions (if it does, let me know). Correct me if I'm wrong, but my research before led me to this conclusion: Nvidia drivers = forcing AA only works on some WINE titles, it's hit or miss, have to test it. AMD drivers = forcing AA USED to only work on some WINE title, it was hit or miss, so AMD got rid of the option altogether.

June videochat with fans. Mostly casual questions, got to discuss a few dream game ideas, usual rambling. More Freeman's Mind and Ross's Game Dungeon coming soon, I swear! This is a blog post. To read the original post, please click here »

cgalves: Thanks a bunch for your testing, you may have saved me a bunch of time. I have a few questions, based on your results. •Since AA is a cornerstone of what I'll be testing for, unless the game I'm testing is DX9 or higher AND supports in-game AA options, does it make any sense for me to even use D9VK? Obviously it has its uses, but if it can't force AA, that makes it less relevant for what I'm testing for except for a few games in my testing. •What's the advantage to using Lutris over PlayonLinux for DX5 - DX8 games? Also, many of the games I'm testing are a little less common, I tried typing in a dozen titles of games I was going to test that didn't even have entries in the Lutris database. •Maybe qptain Nemo could assist with this also. Are there any games that give better results with dgvoodoo under DKVK? I think you use it to tell it to render the game under DX11, but whether it works with the specific game is another question. RaTcHeT302: Regarding the fence test, that has only 3 solutions: •Mixed mode MSAA / SSAA (called TSAA by Nvidia, Adaptive AA by AMD, but they may have changed the naming). This uses MSAA for polygon edges and SSAA for alpha textures. Won't clear up all aliasing (like some shader effects), but will clear up the fence (when it works). This technique often doesn't work on most modern games. •SSAA costly in performance, but gives the cleanest picture (when it works) •Downsampling. Similar to SSAA and can be even more costly, but has a much higher compatibility (assuming the game isn't resolution capped) I'll talk a little about this when I make the video.

Ask questions or topics to discuss here for the next videochat with fans on 5:00pm UTC on June 1st at twitch.tv/rossbroadcast. I realize there's been a lot of delays, but more Freeman's Mind and Game Dungeon coming, I promise! This is a blog post. To read the original post, please click here »

Unreal engine games are pretty common. Maybe something like one of the Mass Effect games (or Unreal Tournament 3) would be a good test.

Really just anything that runs on D9VK. Well known games might be best since they're more likely to have been tested. It's not essential that you find one, it would just be nice to have a point of reference to ensure it works. EDIT: If you have a game that has in-game options for AA, but that's disabled, however driver-forced AA is on, that can work too (as long as it's in override mode).

Maybe it's different on Linux, but on Windows, forcing AA has always been a crapshoot on DX9 and above. If you know of any games where it can be forced (in other words, no in-game option to turn on AA, but it does work when forced through drivers), let me know, that could be good to confirm it's working.

Okay, thanks for the advice so far, I'll test it when I get a chance. Question, for D9VK, is the usual method of forcing AA still the same (setting it in the Nvidia control panel + adding the registry setting to the WINE install)?

What I'm looking for is almost a flowchart process. In other words, something like Step 1: Try X software with Y settings. Step 2: If X software is unsuccessful, try Z software with X1 settings, etc. Step 3: If game is DirectX 7, try X2 software with Y2 settings, etc. I'm looking for a "general procedure" to follow that I can apply to any random game of the type mentioned (DX5 - DX9, 3D accelerated, possibly too obscure to have existing documentation). So I'm not expecting one piece of software or one setting to work with everything, but I want to be as prepared as possible with my procedure to get the game working. It's fine if you want to list a different process for a game that already is in an existing database v. one that's not. Now beyond that, there's also game-specific fixes, but that's a deeper level I can document elsewhere that people can help with later if you want (I can email about that). For now, I'm just trying to nail down a general process.

That's alright, you reminded me of another thing I wanted to mention, I'm adding it to the original post.

This is a thread meant for Linux gaming experts who have been helping me via email to help decide on the best configuration for doing old game testing in Linux. I've gotten some conflicting information, so this is an effort to sort it out. Here's a recap of what I'm trying to do: GOALS / SYSTEM INFO: •Test compatibility on a battery of games, all 3D accelerated, from about 1997 - 2010. They'll be DirectX 5 - 9 games with maybe a couple OpenGL titles. A fair portion of these titles will be obscure and unlikely to be in many game databases. None of them will be Steam or GOG copies, only disc originals (some with DRM cracks). •I wish to force mixed mode antialiasing (MSAA+SSAA for alpha textures) or SSAA on as many titles as I can. •I wish to measure the framerate on as many titles as I can. This means disabling vsync for benchmarking, though it would be nice to be able to turn it back on when I'm not also. •The machine I'm using has a FX-8350 CPU and a Geforce 770 GPU. I have a weaker AMD GPU card also I could use, but my initial research showed a wider range of antialiasing compatibility for Nvidia GPUs on Linux. I currently have Ubuntu Mate installed. I was using PlayonLinux and was having disappointing performance on many older titles. I've since been told I should be using Luttris. What I'm looking for is to collectively come up with a COMPLETE guide for everything I should do from a new install to having a solid set up. I was starting to make a checklist when I got all the Lutrris info and discovered WINE was forcing Adaptive VSYNC on me, which sent me back to the drawing board. Here was my checklist so far: Obviously, some of this information is outdated, but I thought it was worth listing here for the parts I did get correct. Anyway, please give me your detailed step-by-step instructions / suggestions for trying to get these games to run. For example, maybe I should try Luttris in DXVK mode FIRST, but then if that doesn't work for a game, THEN try Playonlinux, etc. Be as specific as you need to. Alternately, if you disagree with advice someone else is giving, by all means, feel free to discuss it here. You only need to "dumb down" the final instructions directed towards me, you can be as technical as you like towards each other. Thanks in advance for the help and if you're not particularly knowledgeable about Linux gaming, please refrain from posting in this thread. I'm trying to cut down on the casual chatter here and focus just on the technical aspects of getting things working. WHAT I CURRENTLY NEED HELP WITH •Turning off adaptive Vsync in WINE. It's been suggested to me that I should install libstrangle, which requires me to compile it myself. It's also been suggested I disable Nvidia DRM, though someone else was claiming that would only work for Opteron setups. I have no idea which route I should be pursuing, so I would appreciate some clarification here. •Full installation / setup guide to Luttris for the types of games I'm testing (DirectX 5 - 9, mostly, many of which aren't in a Linux database). Advice on what cases this is likely to be better, worse, etc. •One person suggested dgvoodoo for getting antialiasing working on games it doesn't function on normally. I think initial testing made this look like it was unlikely to help in most scenarios. Any clarification on this + instructions for getting it working if it is in fact, recommended. •For older games that only run at 4x3 resolutions, what is the recommended method to run these in full screen mode on a widescreen monitor with proportional stretching (in other words, the image will be scaled accordingly and not distorted horizontally)? •(low priority) when forcing SSAA, I noticed on at least one title, it led to blurring above and beyond what you normally get from SSAA (all AA has a little bit of blur due to its nature, this was beyond that). In situations this occurs in, are there any "sharpening" graphics options to counteract it? If it helps, the title I remember this happening in was Warhammer: Mark of Chaos.

May videochat with fans. Had some followup commentary on the Games as a service video, and several good questions afterward. More normal videos coming! This is a blog post. To read the original post, please click here »

Since I've made the "Games as a Service" video, several US attorneys have weighed in on the legal portions of my video. There's not a total consensus and in some cases there was minor misunderstanding, but the conclusions all point in the same direction: GAAS is either not fraud or else extremely difficult to prove it's fraud. Furthermore, even if it was established as fraud, it would be on such a minor level under the law, that it may not even carry a penalty. Barring new information, I'm leaning towards declaring the USA a lost cause on this manner and focusing on countries with stronger consumer protection laws. Anyway, here's a list of the legal analyses, and some additional appearances I had in responding. I recommend not watching these unless you're bored or doing something else as most are quite long: "YouTuber Law" video analysis I think this is the best one (also the 2nd shortest). He grasped my arguments well and gave a realistic look at the situation. Leonard French long video analysis A longer look at the laws in the video, I also had some audio appearances in this one where I asked more questions. Leonard French short video analysis A quick look at the laws in the video, he made some conclusions that weren't quite applicable, which prompted the longer analysis Hoeg Law video analysis I thought his legal portion was relatively good, though there was a small misinterpretation on the legal portion and a major misinterpretation on my stance. Hoeg Law audio discussion / debate I appeared with Hoeg Law to go over his rebuttal and debate was was said in the previous video. Discussed the larger issue also and not just the law in this one. Nick Rekieta Law discussion mirror A more casual discussion, he takes a differing view than most other attorneys, but still comes to a similar conclusion, that working within the confines of existing law is unlikely to work in USA. We talk about various other things too. Anyway, sorry to flood the site with all this legal analysis, I swear that's not the long-term direction things are taking, more regular videos coming! ADHD version: Ross was right on some things, wrong on some things, doesn't matter for USA; the situation there is basically hopeless on legal protection against destroying games. This is a blog post. To read the original post, please click here »

What you're saying here flies in the face of what I was told by other developers. Could you clarify? Look at my post at the bottom of page 2, it specifies exactly what I'm talking about. We're not talking about complete code documentation. I think there's not as many barriers as you're making out. Here's the thing: as I understand it, (in the USA anyway) the right to reverse engineer IS legal and protected under federal law as long as it's not using copyrighted material. So say they turn off the DRM + and encryption + release the protocol documentation for "non-commercial and research purposes only, all rights reserved by (name of company)". That language isn't giving explicit permission to reverse engineer, but they don't need to either. It's not fraud under the law though, best I can tell. I agree completely that this shouldn't be legal, but since it is (in the USA), I was looking at it from the perspective of "does this violate any laws?" It looks like no, and while this is still technically a grey area, it's more about subtle definitions rather than seeing any opening to actually challenge this legally. It's less about "finding a reason" so much as finding a precedent. If there's no precedent to protect preservation under legal terms and the existing laws block any interpretation of enforcing laws that way, then literally the only option is to propose new laws, which requires action through representatives in Congress. If that's the case, then it's like in the video: I see that as almost impossible. We can't even get Congress to get lead out of the water in the USA, I sure wouldn't count on this. That said, other countries aren't nearly as hopeless. France especially is showing real promise.