From jda...@ Tue Feb 2 01:06:18 2016 From: jda...@ (John Davis) Date: Tue, 2 Feb 2016 04:06:18 -0500 Subject: [Public WebGL] IE and chrome, webgl2 Message-ID: Any updates? -------------- next part -------------- An HTML attachment was scrubbed... URL: From ash...@ Tue Feb 2 04:41:38 2016 From: ash...@ (Ashley Gullen) Date: Tue, 2 Feb 2016 12:41:38 +0000 Subject: [Public WebGL] IE and chrome, webgl2 In-Reply-To: References: Message-ID: In Chrome Canary you can currently enable a prototype version of WebGL 2 in chrome://flags. For Edge, all we have is this: https://dev.windows.com/en-us/microsoft-edge/platform/status/webgl20 - "Development is likely for a future release". On 2 February 2016 at 09:06, John Davis wrote: > Any updates? > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Wed Feb 3 05:14:07 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Wed, 3 Feb 2016 14:14:07 +0100 Subject: [Public WebGL] Firefox's "Intent to implement and ship: Unprivilaged WEBGL_debug_renderer_info" discussion In-Reply-To: References: Message-ID: Stats for 2015 are complete, Firefox and Safari still do not support WEBGL_debug_renderer_info. -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Thu Feb 4 10:37:41 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Thu, 4 Feb 2016 19:37:41 +0100 Subject: [Public WebGL] WEBGL_debug_renderer survey Message-ID: I would like to collect opinions on WEBGL_debug_renderer_info on this ML and would kindly ask so inclined readers to respond to this thread *once* with answers to the following questions (if possible). Q1: Do you use this extension currently? [yes/no] Q2: Do you plan on using the extension in the future? [yes/no] Q3: Should UAs implement the extension? [yes/no] Q4: Can you briefly describe your use (or planned use)? Q5: If any, Your affiliation? (website, service, company or whatever describes you best) Thank You -------------- next part -------------- An HTML attachment was scrubbed... URL: From sta...@ Thu Feb 4 11:59:43 2016 From: sta...@ (andre v) Date: Thu, 4 Feb 2016 11:59:43 -0800 Subject: [Public WebGL] WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Q1: yes Q2: yes Q3: yes Q4: To see on which card my game has problems on. Q5: Small Indie game developer ( www.istrolid.com ) My biggest issue is still with webGL not working for people. Mostly due to them disabling hardware acceleration some how or having black listed drivers/card. I just don't know. It would be nice to know those 2 things. On Thu, Feb 4, 2016 at 10:37 AM, Florian B?sch wrote: > I would like to collect opinions on WEBGL_debug_renderer_info on this ML > and would kindly ask so inclined readers to respond to this thread *once* > with answers to the following questions (if possible). > > Q1: Do you use this extension currently? [yes/no] > > Q2: Do you plan on using the extension in the future? [yes/no] > > Q3: Should UAs implement the extension? [yes/no] > > Q4: Can you briefly describe your use (or planned use)? > > Q5: If any, Your affiliation? (website, service, company or whatever > describes you best) > > Thank You > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Thu Feb 4 12:59:45 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Thu, 4 Feb 2016 21:59:45 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Tony Parisi: Q1: Do you use this extension currently? NO Q2: Do you plan on using the extension in the future? YES Q3: Should UAs implement the extension? YES Q4: Can you briefly describe your use (or planned use)? 1. Aggregate usage data per graphics driver and cross correlate against performance issues and bugs 2. Collect data for problem troubleshooting for customer support Q5: If any, Your affiliation? (website, service, company or whatever describes you best) VP of Web and Open Technologies, Wevr (http://wevr.com/) -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Thu Feb 4 15:41:28 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Fri, 5 Feb 2016 00:41:28 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Tomas Polach Q1: Do you use this extension currently? [yes/no] yes Q2: Do you plan on using the extension in the future? [yes/no] yes Q3: Should UAs implement the extension? [yes/no] yes Q4: Can you briefly describe your use (or planned use)? We use WEBGL_debug_renderer_info together with UA string to estimate device performance and available VRAM. Based on this information we enable or disable computation expensive features and adapt texture sizes in order to optimize performance and reduce battery consumption. Q5: If any, Your affiliation? (website, service, company or whatever describes you best) Spin-Off ETH (product: online 3d indoor maps) http://archilogic.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From ash...@ Thu Feb 4 16:19:00 2016 From: ash...@ (Ashley Gullen) Date: Fri, 5 Feb 2016 00:19:00 +0000 Subject: [Public WebGL] WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Q1: Yes Q2: Yes Q3: Yes, although possibly disabled in private browsing modes to reduce fingerprinting Q4: Our middleware reveals this data to users for diagnostic purposes. Of particular interest is determining which hardware renderer is in use on dual-GPU laptops. The string returned by WEBGL_debug_renderer serves as the authority on which GPU is in use, verifying that any changed settings have taken effect. Q5: Scirra Ltd: https://www.scirra.com On 4 February 2016 at 18:37, Florian B?sch wrote: > I would like to collect opinions on WEBGL_debug_renderer_info on this ML > and would kindly ask so inclined readers to respond to this thread *once* > with answers to the following questions (if possible). > > Q1: Do you use this extension currently? [yes/no] > > Q2: Do you plan on using the extension in the future? [yes/no] > > Q3: Should UAs implement the extension? [yes/no] > > Q4: Can you briefly describe your use (or planned use)? > > Q5: If any, Your affiliation? (website, service, company or whatever > describes you best) > > Thank You > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Fri Feb 5 02:13:28 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Fri, 5 Feb 2016 11:13:28 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying answer from Kinix (https://www.urbangalaxyonline.com/) We use three.js on our project so the use of actual WebGL commands is > hidden on our case. > However from my experience with desktop OpenGL I believe that this > extension is useful. > So since I cannot answer the questions as I am not aware how (and if) > three.js is making use or would make use of this extension, in general I > would vote for yes, especially of the UA implementation and support. Counting this as: Q1: no Q2: yes Q3: yes Q4: n/a Q5: Kinix (https://www.urbangalaxyonline.com/) -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Fri Feb 5 02:16:23 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Fri, 5 Feb 2016 11:16:23 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying answer from Mike Bond: Q1: No Q2: Hadn't planned on it as we try to run code that works across all devices as equally as possible. Q3: It seems like it might be useful but also a potential privacy concern. As drivers become more mature and WebGL adoption grows, I expect the need to know the specific hardware that you're running on will decrease. So far, I hadn't had a real desire to use this. Most devices that have caused me problems have been blacklisted by Chrome and Firefox anyway. Q4: n/a Q5: Mike Bond (Box Inc. https://www.box.com/ , acquired https://www.verold.com/ last year) -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Fri Feb 5 04:27:25 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Fri, 5 Feb 2016 13:27:25 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Yuri Kovelenov: Q1: yes Q2: yes Q3: yes Q4: Currently we use this extension to disable features for unsupported devices. See https://github.com/TriumphLLC/Blend4Web/blob/master/src/compat.js Q5: Blend4Web -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Fri Feb 5 06:53:25 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Fri, 5 Feb 2016 15:53:25 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from David Evans: Q1: yes Q2: yes Q3: yes Q4: Debugging and allowing developers to communicate better with users of apps, as well as analytics to understand the audience. Q5: PlayCanvas (https://playcanvas.com/) -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Fri Feb 5 06:54:51 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Fri, 5 Feb 2016 15:54:51 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Ben Houston: Q1: We do not use that extension. Q2: I really do not want to vary our WebGL features by GPU model/vendor. I want to vary features based only on WebGL extensions and feature levels. I do not want to have to work around bugs on a vendor and model level. Q3: I do not really need it. If others have strong opinions feel free to add it but I would prefer to avoid its use. Q4: I prefer not to have to use it. I'd rather follow the spec, even if that is hard sometimes. Q5: https://Clara.io and contributor to Three.JS. -------------- next part -------------- An HTML attachment was scrubbed... URL: From eno...@ Fri Feb 5 07:06:23 2016 From: eno...@ (Evan Nowak) Date: Fri, 5 Feb 2016 09:06:23 -0600 Subject: [Public WebGL] WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Q1: yes Q2: yes Q3: yes Q4: We use the data to discover GPU-specific issues. Sometimes this leads to a work-around that fixes the issue for the particular GPU (without regressing functionality for other GPUs). Even if it does not, it is useful to have the information in order to communicate known issues to users. We've occasionally found that issues can be resolved by prompting a user to upgrade their driver on problematic GPUs. It could be useful to have driver version information with the GPU data, as this would help pinpoint where the problem lies in such cases. Q5: Onshape (https://www.onshape.com/) On Thu, Feb 4, 2016 at 12:37 PM, Florian B?sch wrote: > I would like to collect opinions on WEBGL_debug_renderer_info on this ML > and would kindly ask so inclined readers to respond to this thread *once* > with answers to the following questions (if possible). > > Q1: Do you use this extension currently? [yes/no] > > Q2: Do you plan on using the extension in the future? [yes/no] > > Q3: Should UAs implement the extension? [yes/no] > > Q4: Can you briefly describe your use (or planned use)? > > Q5: If any, Your affiliation? (website, service, company or whatever > describes you best) > > Thank You > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tri...@ Fri Feb 5 07:53:12 2016 From: tri...@ (Cedric Pinson) Date: Fri, 05 Feb 2016 16:53:12 +0100 Subject: [Public WebGL] WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Hi, Q1: Do you use this extension currently? yes Q2: Do you plan on using the extension in the future? yes Q3: Should UAs implement the extension? yes Q4: Can you briefly describe your use (or planned use)? When we find bugs in driver/browser we use that to code workaround Q5: If any, Your affiliation? Co-Founder & CTO at sketchfab.com btw thank you very much Florian for your work on webglstats, each time we build feature we use your stats. Cedric ----------------------------------------------------------- You are currently subscribed to public_webgl...@ To unsubscribe, send an email to majordomo...@ with the following command in the body of your email: unsubscribe public_webgl ----------------------------------------------------------- From pya...@ Fri Feb 5 09:09:27 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Fri, 5 Feb 2016 18:09:27 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Luke Groeninger: Q1. No Q2. No Q3. Yes Q4. Off the top of my head, the only thing I would use it for is to detect buggy/broken WebGL implementations so I could disable known broken functionality. Q5. dghost.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From aap...@ Fri Feb 5 14:43:58 2016 From: aap...@ (Austin Appleby) Date: Fri, 5 Feb 2016 14:43:58 -0800 Subject: [Public WebGL] WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: q1: Yes. q2: Yes. q3: Yes. q4: Diagnosing broken-driver-related rendering issues, preemptively moving low-performing clients to fallback renderers. q5: WebGL dev on Google Maps. On Thu, Feb 4, 2016 at 10:37 AM, Florian B?sch wrote: > I would like to collect opinions on WEBGL_debug_renderer_info on this ML > and would kindly ask so inclined readers to respond to this thread *once* > with answers to the following questions (if possible). > > Q1: Do you use this extension currently? [yes/no] > > Q2: Do you plan on using the extension in the future? [yes/no] > > Q3: Should UAs implement the extension? [yes/no] > > Q4: Can you briefly describe your use (or planned use)? > > Q5: If any, Your affiliation? (website, service, company or whatever > describes you best) > > Thank You > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Sat Feb 6 16:10:14 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Sun, 7 Feb 2016 01:10:14 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Xavier: Q1: No, I do not use this extension. My current projects are : - An industrial application which will run on powerfull desktop computers, - An experimental GPU neural network, but it is too early to make it working on every setup, - A 3D viewer for furnitures which leather or fabric like armchairs or couches, but it is clearly written in the specs that it should work with recent enough devices. So I do not need a really accurate detection of the user setup. Q2 : No, I do not plan to use this extension in the future. GPU vendors should provide webgl compliant drivers. With Nvidia or Intel under Linux, I have not any problem. Compatibility problems occurs with mac, mobile, or windows, and they are always caused by a faulty driver implementation. IMHO the webgl developper needs to detect Webgl capabilities (max texture size, max texture units, max framebuffers, etc...) to fit its application with the user setup, and that's all. It the user have outdated drivers, bad graphic hardware, we do not have to detect which is exactly its hardware to make exceptions. And I think this is not a good thing for the user privacy. It we detect the GPU vendor, the browser name and version, the gamepad and audio capabilities, the screen device and resolution, etc... We can have a lot of technical information about the user. It makes easier to track the user. Q3 : No I do not think. And there are so many nice features and extensions which should be great to add to WebGL ! For example GL_ARB_gpu_shader_fp64, it would be great for my neural network :) Q4 : - Q5 : WebGL & Three.JS developper and instructor @ spacegoo.com Teacher in l'ESIPE - University of Paris Est Marne la Vall?e Creator of WebGL Academy : http://webgl.academy Organizer of WebGL Paris : http://webgl.paris -------------- next part -------------- An HTML attachment was scrubbed... URL: From joh...@ Sun Feb 7 07:19:25 2016 From: joh...@ (Johannes Behr) Date: Sun, 7 Feb 2016 16:19:25 +0100 Subject: [Public WebGL] WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: <4C4A90AC-939F-4F8C-BA78-282BAFFC2014@gmail.com> Hi, Q1: Do you use this extension currently? Yes Q2: Do you plan on using the extension in the future? Yes Q3: Should UAs implement the extension? [yes/no] Yes Q4: Can you briefly describe your use (or planned use)? We use the returns of the function as one input for server based client/capacity class matching. This is the uses to control e.g. the memory management setting automatically. Q5: If any, Your affiliation? (website, service, company or whatever describes you best) http://instant3dhub.org, Visualisation as a Service platform to bring large data to any device. Best regards Johannes > On 4 Feb 2016, at 19:37, Florian B?sch wrote: > > I would like to collect opinions on WEBGL_debug_renderer_info on this ML and would kindly ask so inclined readers to respond to this thread once with answers to the following questions (if possible). > > Q1: Do you use this extension currently? [yes/no] > > Q2: Do you plan on using the extension in the future? [yes/no] > > Q3: Should UAs implement the extension? [yes/no] > > Q4: Can you briefly describe your use (or planned use)? > > Q5: If any, Your affiliation? (website, service, company or whatever describes you best) > > Thank You ----------------------------------------------------------- You are currently subscribed to public_webgl...@ To unsubscribe, send an email to majordomo...@ with the following command in the body of your email: unsubscribe public_webgl ----------------------------------------------------------- From nic...@ Sun Feb 7 12:14:47 2016 From: nic...@ (Nicolas Capens) Date: Sun, 7 Feb 2016 15:14:47 -0500 Subject: [Public WebGL] WEBGL_debug_renderer survey Message-ID: Q1: no Q2: yes Q3: yes, but only as a privileged debug feature Q4: If unexpected errors are encountered, I want to subtly make a button appear which when clicked prompts the user to share their WebGL vendor and renderer string. The extension could be amended to suggest or demand the browser to create the prompt automatically when attempting to access the strings (unless disabled or silently allowed in the browser settings). I don't think this extension should be silently enabled by default. As noted in the extension spec, the strings can contain partial personally-identifiable information. Users should explicitly agree to sharing that information for debugging purposes. Not making it generally available also helps avoid vendor-specific content or limitations (best case it's unintentional, worst case it's a net neutrality violation). It's also an improper shift of responsibility if this information were to be used to work around bugs. Browser vendors should implement spec-compliant workarounds and urge driver vendors to fix the underlying bugs. Small and medium-sized application developers can't be expected to (have to) deal with such issues themselves. Large developers of cutting-edge applications might be ok with that, but allowing them to take the matter into their own hands would mask bugs in the browser or driver. Browser and driver vendors would get lazy (or ignorant) and the much more numerous smaller developers would be the victims. Which makes it not in the best interest of the web platform as a whole. Q5: The above is my personal opinion, but I work for Google and I used to work a lot on ANGLE. From: Florian B?sch > Date: Thu, Feb 4, 2016 at 1:37 PM > Subject: [Public WebGL] WEBGL_debug_renderer survey > To: public webgl > > > I would like to collect opinions on WEBGL_debug_renderer_info on this ML > and would kindly ask so inclined readers to respond to this thread *once* > with answers to the following questions (if possible). > > Q1: Do you use this extension currently? [yes/no] > > Q2: Do you plan on using the extension in the future? [yes/no] > > Q3: Should UAs implement the extension? [yes/no] > > Q4: Can you briefly describe your use (or planned use)? > > Q5: If any, Your affiliation? (website, service, company or whatever > describes you best) > > Thank You > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Mon Feb 8 01:49:42 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Mon, 8 Feb 2016 10:49:42 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Chris Deforeit: Q1: Do you use this extension currently? [yes/no] NO Q2: Do you plan on using the extension in the future? [yes/no] NO Q3: Should UAs implement the extension? [yes/no] NO Q4: Can you briefly describe your use (or planned use)? Q5: Your cited affiliation? (website, service, company or whatever describes you best) https://3dthis.com Note: I think device specific infos add useless complexity. A standard should not be modified by devices capabilities. -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Mon Feb 8 08:41:02 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Mon, 8 Feb 2016 17:41:02 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Tarek Sherif: Q1: Do you use this extension currently? [yes/no] No. Q2: Do you plan on using the extension in the future? [yes/no] Yes. (I'll admit, though, that your survey is what got me thinking about using it. Chalk it up to the observer effect.) Q3: Should Browsers implement the extension? [yes/no] Yes. Q4: Can you briefly describe your use (or planned use)? Primarily error reporting. It would be useful to know a user's GPU when troubleshooting reported problems. Many of our users are non-technical, so it would be extremely helpful to have the system able to automatically report that information. Q5: Your affiliation? (website, service, company or whatever describes you best) 3D Software Engineer, BioDigital (https://www.biodigital.com/) -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Mon Feb 8 10:11:52 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Mon, 8 Feb 2016 19:11:52 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Jeff Russell: Q1: Do you use this extension currently? *No.* Q2: Do you plan on using the extension in the future? *"No", in that I have no direct plans, but "probably" in that future development might need it.* Q3: Should Browsers implement the extension? *Yes.* Q4: Can you briefly describe your use (or planned use)? *Might be useful at times to predict performance or turn off certain features to dodge driver bugs.* *To be honest though, things like video memory size might be even more useful to know, or even* *better yet, upper ceilings for memory in general on mobile platforms (this has been a big headache for us).* *I'm well aware these sorts of requests go against the grain of web dev thinking today though, particularly* *as it relates to security and user identification.* Q5: Your affiliation? (website, service, company or whatever describes you best) *I'm a founding member of Marmoset - http://www.marmoset.co .* -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Mon Feb 8 11:06:48 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Mon, 8 Feb 2016 20:06:48 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from David Catuhe: Q1: Yes Q2: Already using Q3: Yes Q4: I?m using it to detect the GFX driver for debugging purpose. Also really useful for IE/Edge to detect when the engine switch to software rendering Q5: www.babylonjs.com. I?m working for Microsoft as principal Program Manager -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Tue Feb 9 04:42:21 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Tue, 9 Feb 2016 13:42:21 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Felix Palmer: Q1: Do you use this extension currently? [yes/no] > NO > > Q2: Do you plan on using the extension in the future? [yes/no] > YES > > Q3: Should UAs implement the extension? [yes/no] > YES > > Q4: Can you briefly describe your use (or planned use)? > Debugging issues specific to certain hardware locally as well as sending debug info up to vendors to fix potential issues on their end > > Q5: Your cited affiliation? (website, service, company or whatever > describes you best) > Procedural Ltd (http://www.pheelicks.com/) -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Tue Feb 9 09:39:49 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Tue, 9 Feb 2016 18:39:49 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from James Austin: 1. Yes we do 2. Yes but not critical 3. Yes 4. Currently using to detect certain combinations of browser/webgl renderer to implement workarounds for bugs in those implementations. Ideally we wouldn't need this if all renderers etc were fully standardised and worked. It can be useful mainly for identification of client setup to reproduce bugs for debugging purposes. 5. Turbulenz (https://ga.me/) -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Wed Feb 10 02:51:36 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Wed, 10 Feb 2016 11:51:36 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Patrick Cozzi: Q1: Do you use this extension currently? Not in Cesium, but we report it in webglreport.com. Q2: Do you plan on using the extension in the future? No current plans. Q3: Should Browsers implement the extension? Yes, if they don't see it as a privacy issue. Q4: Can you briefly describe your use (or planned use)? Workaround bugs and maybe use fast paths if we really really had to. I could also see it being useful to get stats about users. Q5: Your affiliation? cesiumjs.org -------------- next part -------------- An HTML attachment was scrubbed... URL: From juj...@ Wed Feb 10 07:02:53 2016 From: juj...@ (juj j) Date: Wed, 10 Feb 2016 17:02:53 +0200 Subject: [Public WebGL] How to dive into http://webglstats.com WebGL adoption rate? Message-ID: Hi all, the "Visitors who have WebGL" rating at http://webglstats.com shows that 83.1% of tracked users have "webgl" and 9.5% have "experimental-webgl", with a total of 92.6% adoption. Is there a way to cross reference these numbers against OS and browser and their versions? I am trying to answer the following questions: - which browsers were the experimental-webgl people on? - why don't the remaining 7.4% of people have webgl? What were the browser version and OS version distribution on these unsupported browsers? How much of the unsupportion was due to user remaining on old browser version and not having updated? Which browsers were these? - how much of unsupportion is due to driver having been blacklisted (due to a bug), and how much of it was due to user running hardware that just does not have the support? Does there exist statistics to make conclusions about the above? Thanks, Jukka -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Wed Feb 10 07:42:14 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Wed, 10 Feb 2016 16:42:14 +0100 Subject: [Public WebGL] WEBGL_debug_renderer_info survey results Message-ID: I have created a summary of the survey (and overview of the extension support and use). http://codeflow.org/entries/2016/feb/10/webgl_debug_renderer_info-extension-survey-results/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Wed Feb 10 08:03:43 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Wed, 10 Feb 2016 17:03:43 +0100 Subject: [Public WebGL] How to dive into http://webglstats.com WebGL adoption rate? In-Reply-To: References: Message-ID: I'm the maintainer of webglstats. On Wed, Feb 10, 2016 at 4:02 PM, juj j wrote: > Is there a way to cross reference these numbers against OS and browser and > their versions? I am trying to answer the following questions: > The drilldown bar on the left lets you select a subset and drill down to platform, operating system and browser. Not the version unfortunately. > - which browsers were the experimental-webgl people on? > See the sidebar. > - why don't the remaining 7.4% of people have webgl? > Blacklists mostly, and some old browsers. > What were the browser version > Can't be queried currently. > and OS version distribution on these unsupported browsers? > Can't be queried currently. > How much of the unsupportion was due to user remaining on old browser > version and not having updated? Which browsers were these? > Can't be queried currently. > - how much of unsupportion is due to driver having been blacklisted > (due to a bug), and how much of it was due to user running hardware that > just does not have the support? > That's an answer I cannot provide as the reason for no support (beyond old browser) wouldn't show up in the data. > Does there exist statistics to make conclusions about the above? > Not yet, but I'm working on an update to the site and data handling that would make that kind of question (insofar as there is data about it) possible. -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Wed Feb 10 15:19:46 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Thu, 11 Feb 2016 00:19:46 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Ben Adams: Q1: Do you use this extension currently? no Q2: Do you plan on using the extension in the future? yes Q3: Should UAs implement the extension? yes Q4: Can you briefly describe your use (or planned use)? - User issue diagnostics; details for vendor bug reports; workarounds - Device performance detection (recommended/default effect level choices) - Issues with Higher DPI vs GPU power; resolution has increased but installed GPUs power hasn't always matched. - Feature detection doesn't always equate to a good user experience; just because a device can do things, switching everything on many result in a poor experience; interfere with networking etc. - Association settings suggestion; rather than lengthy setup benchmarking (e.g. 60% of people with this GPU use these settings) Q5: If any, Your affiliation? (website, service, company or whatever describes you best) Co-Founder & CTO, Illyriad Games, www.ageofascent.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From kbr...@ Wed Feb 10 17:39:27 2016 From: kbr...@ (Kenneth Russell) Date: Wed, 10 Feb 2016 17:39:27 -0800 Subject: [Public WebGL] WEBGL_debug_renderer_info survey results In-Reply-To: References: Message-ID: Thanks for putting together this information Florian. On Wed, Feb 10, 2016 at 7:42 AM, Florian B?sch wrote: > I have created a summary of the survey (and overview of the extension > support and use). > http://codeflow.org/entries/2016/feb/10/webgl_debug_renderer_info-extension-survey-results/ > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Thu Feb 11 02:30:32 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Thu, 11 Feb 2016 11:30:32 +0100 Subject: [Public WebGL] Re: WEBGL_debug_renderer survey In-Reply-To: References: Message-ID: Relaying response from Marcin Ignac: Q1: Do you use this extension currently? [yes/no] NO Q2: Do you plan on using the extension in the future? [yes/no] YES Q3: Should UAs implement the extension? [yes/no] YES Q4: Can you briefly describe your use (or planned use)? I think it's critical for bug reporting. It would be amazing to just get this information when something is not working. Especially that I mostly work with http://plask.org, and test in Safari. Chrome and FF only for big client projects. Q5: Your cited affiliation? (website, service, company or whatever describes you best) http://variable.io - data driven art generative design studio -------------- next part -------------- An HTML attachment was scrubbed... URL: From pya...@ Thu Feb 11 07:09:39 2016 From: pya...@ (=?UTF-8?Q?Florian_B=C3=B6sch?=) Date: Thu, 11 Feb 2016 16:09:39 +0100 Subject: [Public WebGL] WEBGL_debug_renderer_info survey results In-Reply-To: References: Message-ID: Thanks. as a reminder, the survey isn't closed, I'll still update the blog post as new responses come in. On Thu, Feb 11, 2016 at 2:39 AM, Kenneth Russell wrote: > Thanks for putting together this information Florian. > > > On Wed, Feb 10, 2016 at 7:42 AM, Florian B?sch wrote: > >> I have created a summary of the survey (and overview of the extension >> support and use). >> http://codeflow.org/entries/2016/feb/10/webgl_debug_renderer_info-extension-survey-results/ >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: