[00:08] He sent a nice letter. [00:08] Going to paste it. [00:08] Hello Jason, [00:08] thanks for your mail. I think it would be possible, but I see some problems: [00:08] 1. Speed: SAE (at least the new version 0.9.0) does require huge processing power and a fast browser. (Chrome, Safari) [00:08] At the moment, mobile- and tablet-users can't run it at full speed and even so some desktop users. [00:08] Not to speak of AGA in real-time, which does require about 2-4x the power of OCS. [00:08] So only a small amount of visitors would be able to play games or show a demo in real-time. [00:08] 2. ROMs: The original kickstart-ROMs are under copyright. That's why I'm using the slow and [00:08] sometimes incompatible kickstart-replacement from AROS on the SAE-site :/ [00:08] I already had contact with M. C. Battilana from "Clonato" (Amiga Forever) about this case some years ago. [00:09] He is not a bad guy. Maybe we can refresh the thematic about roms. [00:09] 3. Games: The situation is very difficult here. Although 20-30 year old, most Amiga-games are still [00:09] under copyright. (by whomsoever ?) e.g. they just remastered "Wings" some months ago or many [00:09] games like "Lemmings" are avail for mobile. 4 Years ago, at the first release of SAE, I had about 20 [00:09] commercial titles in the database. The s-stom was huge :) [00:09] So titles you may provide could be demos, intros and freeware-games. [00:09] The integration of SAE itself should be easy, just a few lines of code. [00:09] I could do that or help you doing it with example-code. [00:09] My main-machine died. I'm currently on Win without IRC-client. [00:09] Will see what I can do. [00:15] -- [00:15] By the way, all the more-than-one item emulators are snapped in place. [00:37] Also, this is interesting. [00:37] https://archive.org/details/Kaverns_of_Kfest_1.1 [00:37] Implies something is wrong with the .dsk image. [00:38] Solved [00:38] Ignore [04:11] ----- [04:11] The speed of SAE is actually rather nice and the compilation abilities as time goes on will increase notably, so I'm fine with the fact that it will run slower for older browsing machines - this is all cutting edge and cutting edge efforts always take a little rough calls in the beginning. Having used your SAE implementation on a variety of places and platforms over the past month convinced me it was rea [04:11] dy (with caveats) for prime time. [04:11] Thanks, but I think you are talking about the old version 0.8.3. Have you tried the new version 0.9.0, which is live since about two weeks? [04:11] It's almost a rewrite of 0.8.3 and does require more computing-power. But if the speed of the new version 0.9.0 is okay for you, then it's also okay for me :) [04:11] So, we've been working with the SAE emulator as it is on your github, and it's a strangely designed thing, where we've had to begin to spend time really trying to understand all the information it needs to be fed. You wouldn't have a list of this information, would you? We could then feed it what it needs, or hard-set it for now. That's why I was looking for you to hop in. [04:11] Yep, it's still beta. That's why config.js is such a mess. You could take a look at the function getDatabaseConfig() in index.js as a starting-point. [04:11] Problem is, that some games or demos does require special attributes like blitter-immediate, original floppy-speed or manual disk-change. (const DBF_* in index.js) [04:11] So, just passing the ADFs will not always work, it depends on the title. [04:11] The principle is even simpler than under 0.8.3: [04:11] - Call the new method cfg.setModel() [04:11] - Fill rom-/floppy-data. [04:11] - Fill additional-settings like the output-div element or the assignment of the joystick buttons. [04:11] - Call sae.start(), sae.stop(), etc [04:11] I'm just setting some things up here, so if you want, I can do some example-code tomorrow night. [04:12] (He rewrote it from 0.8.3 to 0.9.0 two weeks ago) [05:01] so if it's not fast there may not be much reason to prefer it over mame [05:05] interesting [05:05] yeah how much slower are we talking here? the old version seemed to run reasonably well at least for the one or two things we tried [05:17] I found iot runs the same [05:18] He's being humble. It's fine. [05:23] hmm I guess mame is pretty slow http://fos.textfiles.com/dfjustin/amiga/a500p/ [05:25] Can you throw a demo in it, like state of the art? [05:27] http://spaceballs.untergrund.net/files/spb-sota.dms [05:28] you think I don't have it? :) [05:30] hmm guru meditation [05:31] arbee won't be happy about that [05:36] put a different one in there [05:45] Well, I think it rude to go "add this" and then not cpoint to where to get it. [05:47] That's quite slow. [05:51] He says he'll whip something up for us tomorrow, we'll see how it works in the framework [05:55] nice [06:16] https://archive.org/details/Kansasfest2016JasonScottTheEmularityDeepDive [06:25] is that a wolf? [06:25] a whale? [06:26] oh good, they killed it [07:20] *** Coderjoe has quit IRC (Read error: Operation timed out) [07:27] *** Coderjoe has joined #jsmess [09:04] SketchCow: good presentation [09:39] *** Coderjoe has quit IRC (Read error: Operation timed out) [10:04] *** Coderjoe has joined #jsmess [11:40] *** Coderjoe has quit IRC (Read error: Operation timed out) [11:44] *** Coderjoe has joined #jsmess [11:58] *** joepie91_ is now known as joepie91 [12:36] *** Coderjoe has quit IRC (Ping timeout: 260 seconds) [12:52] *** Coderjoe has joined #jsmess [13:30] Everything is a book! [14:25] yes, that's a good way to look at the site [15:07] *** GLaDOS has joined #jsmess [15:34] *** Coderjoe has quit IRC (Ping timeout: 260 seconds) [16:02] *** Coderjoe has joined #jsmess [16:42] *** pfallenop has joined #jsmess [17:35] He's gotten a client going and is joining us. [17:37] *** naTmeg has joined #jsmess [17:54] Everyone welcome naTmeg! [17:55] naTmeg: Do you know much about how our system works? (I don't expect you to, just wanted to know.) [17:58] Hi there. No, i now nothing about it. [17:58] That's perfectly fine. [17:58] So, for the purposes of the Archive, at least currently, it's likely we would use sae as an "engine" that has a set of pre-formed machines on it. [17:59] So we'd make a couple kinds of Amigas, represented by pre-formed settings, and run against that. [18:00] yes. you could do that via the config-object [18:00] Let me ping bai and db48x who do a lot of the loader work. [18:01] maybe i should write a quick introduction to the API and how to config. [18:04] Absolutely. We can start with a basic machine and then go from there. [18:05] i'm normaly not into web-stuff. is fact, sae is my first js/html5-project, so i'm not realy helpfull in terms of integration, cms and that stuff.. [18:05] The previous version, we combined your javascript into one js.gz item for loading. I think that worked well. (Our system needs one single js blob) [18:05] We've done a lot of work with it. Your item would live inside a canvas, accessing a filesystem we'd provide the access to. [18:05] BrowserFS basically. [18:06] But once it works, it's trivial for us to slam it against thousands of amiga floppies, screenshot the results, and then tune individual settings. [18:07] ok ic. i make an simple example now. [18:10] Great. [18:10] Is the javascript one .js file or is it multiple? [18:11] currenty there are multiple files. i plan to make one big file some day. [18:21] We were able to make it one file for now. [18:21] The system will allow us to snap in improvements, so I'm not worried. [18:21] Do you know of closure, by the way? (I assume you do) [18:23] Anyway, make an example for us, I'm sure we can work from that. [18:33] regarding closures. i tried to cache as much as possible. reloading does a reset to the defaults. [18:52] Well, I mean Google Closure, a set of tools for optimizing Javascript. [18:53] Anyway, let us focus on one thing at a time, so having an example and working from that would be good. [19:09] oh ic, i never used any 3rd-party script or tools yet. well except GA. we can discuss that later. [19:09] Exactly. [19:10] Your emulator is actually rather fast. I'm sure these tools will clean things up, or become a part of the toolchain of "OK, now bake a special version for websites" [19:11] ok we should give them a try then [19:47] I'm always working on things, but don't hesitate to have any questions or requests - getting the amiga emulator working with Internet Archive is very important to me. [19:47] And I'd love to bring thousands in to play old stuff on it. [19:47] (And obviously, work with Mike B. about getting a special license for the original BIOS/ROM) [20:00] sounds great. that was exactly my intention, make amiga emulation available for websites. i really appreciate the news about the roms. [20:01] *** Coderjoe has quit IRC (Read error: Operation timed out) [20:02] *** SketchCow changes topic to: http://github.com/jsmess/jsmess | http://www.emularity.com | Look out Amiga [20:18] *** Coderjoe has joined #jsmess [20:33] i have a first alpha-example running, need some more work... [22:05] Go for it [22:05] excellent [23:01] ok, i'm done with a basic and minimal description of the new API. http://scriptedamigaemulator.net/example.htm [23:03] i'll do a detailed explanation of the config-object in html later. meanwhile, if something is unclear, just ask. [23:51] updated, please reload [23:58] *** Coderjoe_ has joined #jsmess [23:58] *** Coderjoe has quit IRC (Read error: Operation timed out)